.^; 
 
 COLUMBIA UNIVERSITY 
 DEPARTMENT OF PHYSIOLOGY 
 THE JOHN G. CURTIS LIBRARY 
 
Digitized by tine Internet Archive 
 
 in 2010 with funding from 
 
 Open Knowledge Commons (for the Medical Heritage Library project) 
 
 http://www.archive.org/details/translationsoffoOOburd 
 
TRANSLATIONS 
 
 OF 
 
 FOREIGN BIOLOGICAL MEMOIRS 
 
 BURDON-SA.NDERSON 
 
ITonbon 
 
 HENRY FROWDE 
 
 Oxford University Press Warehouse 
 Amen Corner, E,C. 
 
» \ U 1 
 
 TRANSLATIONS 
 
 OF 
 
 FOEEIGN BIOLOGICAL MEMOIRS 
 
 I. 
 MEMOIRS 
 
 ON THE 
 
 PHYSIOLOGY OP NERVE, OF MUSCLE 
 
 AND OF THE 
 
 ELECTRICAL ORGAN 
 
 EDITED BY 
 
 J. BURDON-SANDEESON, M.D., RR. SS. L. & E. 
 
 WAYNFIiETE PEOFESSOE OF PHYSIOLOGY 
 IN THE TJNIVEESITY OF OXFOED 
 
 AT THE CLARENDON PRESS 
 1887' 
 
 [_ All rights reserved ] 
 
V(o, 
 
 T68 
 
PREFACE. 
 
 The following pages are intended to give the reader such 
 preliminary hints as to the subjects of the Memoirs which are 
 included in the present series of translations, as may aid him in 
 using the volume either for the purpose of extending his know- 
 ledge of the branch of physiology to which they relate, or of 
 obtaining information on special questions. It will be seen by 
 glancing at the Table of Contents, that the Memoirs refer to 
 three principal subjects, and that with one exception, that of 
 Prof, du Bois-Reymond's research on Malapterurus, all have 
 appeared within the last five years. 
 
 The purpose of the investigation recorded in the first Paper 
 is to obtain a new proof of the well-known law established by 
 Pfluger in his classical research on Electrotonus, published in 1859, 
 relating to the changes of excitability of a nerve produced by 
 the flow along it of a voltaic current. In 188 1 the Author invented 
 and described a new instrument, by which it is possible to subject 
 a nerve to instantaneous mechanical excitation, of which the in- 
 tensity can be measured and varied at will. When this is accom- 
 plished by the falling on the nerve of a weighted lever, it is found 
 that the nerve is so little damaged thereby, that the excitation 
 may be repeated as often as necessary without impairment of the 
 effect, and that its strength may be graduated by the observer 
 with an exactitude scarcely inferior to that which is attainable 
 in the use of induction currents. As a means of measuring the 
 excitability of a nerve, mechanical excitation has this great 
 advantage as compared with excitation by induction currents, 
 that it is free from those ' after-eflects ' of which the nature is so 
 fully studied in the series of researches contained in Part II 
 
IV PREFACE. 
 
 of this volume; and secondly, that by the mechanical method 
 it is possible to explore the condition of the nerve in the 
 ' intra-polar ' region, i. e. in that part of its course which lies 
 between the inJflow and outflow of the current. It may be stated 
 generally that the observations of Tigerstedt made by so different 
 a method, while they extend those of Pfliiger, afford additional 
 evidence of their accuracy, the only difference between them 
 being that the state of diminished excitability at the cathode 
 after opening of the current is found not to be so marked when 
 tested mechanically as when tested electrically. 
 
 The second paper contains the results of a new investigation of 
 the ' Law of Contraction,' with special reference to the conditions 
 which determine the state of excitation that occurs at the anode 
 at the moment that a current led throug-h a motor nerve is 
 broken. The author shows that these conditions correspond 
 with those which produce ordinary negative polarisation in the 
 nerve, that is, a state of the nerve in which, if the seat of outflow 
 of the current be investigated immediately after the circuit is 
 broken, it is found to be negative to all other parts ; and he 
 thinks that this correspondence is so close, that the two concomi- 
 tant phenomena, viz. the contraction of the muscle and the polari- 
 sation of its nerve, must have the same cause. He is thus led to 
 regard the make and break contractions as phenomena which are 
 alike produced by the closing of a current, but which differ from 
 each other as regards the way in which this current itself origin- 
 ates. This research should be read along with that of Griitzner 
 (No. 4) on the same subject, who by different experimental 
 methods has arrived at the same explanation of the anodic con- 
 traction. He finds that by far the greater number of the known 
 instances in which the phenomenon can be experimentally in- 
 vestigated admit of more complete and satisfactory explanation 
 as resulting from negative polarisation than on Pfliiger's theory. 
 The special cases which he has investigated are — (i) that in 
 which the anodic excitation produced by breaking a current 
 which is led through a nerve in a direction opposed to the nerve- 
 current, is found to be dependent on certain complicated but 
 
PREFA.CE, V 
 
 well-ascertained relations between the strength of the former 
 and that of the latter (see Introduction to Griitzner's paper); 
 and (2) the numerous cases in which excitation at the anode 
 results, not from break of the current, but from sudden diminu- 
 tion of its strength. In these the varieties of effect which are 
 observed are, in Griitzner's opinion, such as point to polarisation, 
 not subsidence of anelectrotonus. How far this way of looking 
 at the subject is justified, the reader will be able to judge after 
 studying the criticisms of Prof. Hermann and Prof, du Bois- 
 Reymond. 
 
 In No. 5, Prof. Hering has investigated the excitation of motor 
 nerves by voltaic currents from an entirely different point of 
 view. More than thirty years ago du Bois-Reymond showed by 
 a remarkable experiment (see p. 128), in which he used no 
 apparatus excepting 'physiological clay,' that is, clay kneaded with 
 ' normal ' salt solution instead of water, that if the cut and natural 
 surfaces of a nerve-muscle preparation are bridged, the muscle 
 contracts when the bridge is completed, and again when it is 
 broken. Starting from this observation, the author proceeds to 
 illustrate, by a variety of new modes of experiment, the exciting 
 influence exercised on nerve by its own current. He shows, for 
 example, that the mere falling of the end of a nerve on a moist 
 conductor (p. 130) produces contraction when the contact is of 
 such a nature as to bridge the two surfaces, that by this means, 
 if the contact is repeatedly made and broken, a tetanus ' without 
 metals ' may be produced, and that a real ' secondary twitch from 
 nerve ' is obtainable when one of two nerves is excited, of which 
 the two ends are applied to each other with their cut surfaces in 
 the same plane and looking in the same direction. When this is 
 the case, the muscle to which the unexcited nerve belongs re- 
 sponds to excitations of a distant part of the other by currents of 
 minimal intensity. In this respect the phenomenon may be con- 
 trasted with du Bois' ' paradoxical twitch,' which, as is well 
 known, only appears when the exciting currents are strong, and 
 led in in pretty close proximity to the contact between the 
 two nerves. He finally shows how the effects of opening and 
 
VI PREFACE. 
 
 closing of external currents are modified by the excitatory- 
 influence of the nerve-current, and how impossible it is to judge 
 of or interpret these effects correctly without taking this 'in- 
 terference ' into account. It is to be noted that Prof. Hering is 
 led by his observations to a conclusion as to the nature of the 
 opening contraction, which, although by no means identical with 
 that arrived at by Tigerstedt and Griitzner, is in intelligible 
 relation with it. 
 
 The second Part begins with three independent investigations 
 of the same subject by three of the most distinguished living 
 physiologists — the first by Prof, du Bois-Reymond, of Berlin; 
 the second by Prof Hering, of Prague ; the other by Prof. Her- 
 mann, of Zurich, now of Konigsberg — all of which were published 
 within a few months of each other. The purpose of du Bois' 
 paper is to give a more full account, based on new experimental 
 investigations, of the phenomena which the author discovered 
 many years before and designated by the term 'positive polarisa- 
 tion.' In muscle and nerve there are, according to him, two 
 kinds of polarisation, of which one is identical with physical 
 polarisation of porous bodies steeped with electrolytes, the other 
 is purely physiological, that is, peculiar to living excitable 
 structures. As it manifests itself in muscle and nerve, positive 
 polarisation consists in this, that when either of these structures 
 is traversed axially by a voltaic current of very short duration, 
 the galvanoscopical after-effect observed on opening, instead of 
 being, as in ordinary polarisation, opposed in direction to the 
 led-in current, is in the same dii'ection with it, varying according 
 to the density of the external current and other well-defined 
 conditions which need not here be referred to, but exhibiting 
 throughout the most intimate association with the functional 
 activity of the organ. 
 
 A few months later in the year in which du Bois presented 
 his Memoir to the Berlin Academy, a communication on the 
 same subject was submitted to the Academy of Vienna by 
 Hering, in which he described the same phenomena under 
 another designation, the main fact being that when a current 
 
PREFACE. VU 
 
 of considerable intensity and short duration is led through a 
 muscle, and the condition of the muscle is immediately after 
 investigated galvanoscopically, it is found that a sudden change 
 as regards its electrical condition takes place, of such a kind 
 that the current led off from the organ, at first feebly 'negative,' 
 that is, in the opposite direction to that of the external current, 
 becomes strongly ' positive,' that is, in the same direction. 
 
 Similarly, Prof. Hermann, in a Paper published in Pfluger's 
 Ai'chiv in January 1883, gave an account of the same pheno- 
 mena, which, as regards muscle, closely accorded with that of 
 Hering, adding new proof of the polar localisation of the electro- 
 motive changes observed ; and he further extended his researches 
 from muscle to nerve so as to include all the subjects investi- 
 gated by du Bois, with the exception of that of the electrical 
 organs of fish. Prof. Hering's paper ('Ueber Veranderungen 
 des electromotorischen Verhalten der Muskeln in Folge electri- 
 scher Reizung ') we have not thought it necessary to translate, 
 preferring to present to the reader the more advanced view 
 of the subject contained in his second paper, which appeared at 
 the same time with it, under the title, ' Ueber du Bois-Reymond's 
 Untersuchungen der secundarelectromotorischen Erscheinungen 
 am Muskel.' The reader will find in the three papers satis- 
 factory evidence that all these distinguished observers have 
 had before them the same facts, though from difference of method, 
 or from other causes, they may have been presented to them 
 under different aspects. But in all that belongs to the doctrine 
 of ' secondary electromotive phenomena ' it will be equally clear 
 to him, that while Hering and Hermann tend in the same direc- 
 tion, their views are utterly at variance with those of the great 
 founder of Electro-physiology. 
 
 Du Bois holds that just as in the ordinary polarisation of 
 porous bodies soaked with electrolytic liquids, the change affects 
 all parts through which branches of current pass, so in polarisa- 
 tion of nerve and muscle, whether negative or positive, the 
 change affects all the parts traversed in a degree which depends 
 upon the current density, and that as regards positive polarisa- 
 
VIU PKEFACE. 
 
 tion, the action of the current is directly exerted on certain 
 hypothetical carriers (Trager) of electromotive forces which are 
 distributed throughout the polarisable structure, and which the 
 current is capable of ' directing ' (see concluding paragraph of 
 No. 6). 
 
 Hering rejects this theory on the ground that the changes of 
 which ' positive polarisation ' is the expression, have their seat 
 exclusively at the anode, and explains the experimental facts 
 used by du Bois in proof of ' internal polarisation ' (that is of 
 a polarisation affecting all traversed structures, and consisting in 
 the coming into existence of electromotive forces all along the 
 line of flow, and not merely at anodes or cathodes), on the sup- 
 position that he employs structures in which, in consequence of 
 their heterogeneity, such sudden transitions from less to greater 
 density occur in the course of the current-paths, that the effects 
 of anodes are produced. On the assumption that positive polari- 
 sation has its seat at the anode exclusively, he has no difficulty 
 in referring what is observed to the well-recognised facts — 
 (i) that the effect of opening a current led through a muscle 
 or nerve is to produce excitation at the anode ; and (2) that the 
 characteristic expression of the excitatory state is negativity ; 
 and accordingly, while recognising the essential relation of the 
 change in question to the functional activity of the organ in 
 which it manifests itself, he prefers to speak of it as ' anodic after- 
 effect' rather than as ' positive polarisation.' 
 
 Hermann objects to the term 'positive polarisation' even more 
 decidedly than Hering, on the ground that the phenomenon 
 is not analogous to phj^sical polarisation,' He, like Hering, 
 believes it to be excitatory, and holds that it is due to the 
 setting up of a persistent change of this nature in the neigh- 
 bourhood of the spot at which the current has entered. Among 
 the various experiments which lead to this conclusion, those are 
 of most importance which show ( i ) that in muscle it is confined 
 to the anodic region ; (2) that it is annulled whenever the 
 current is led into the muscle or nerve through a dead part ; 
 and (3) that the anode becomes relatively negative not only 
 
PREFACE. IX 
 
 to the ' intra-polar ' region, but also to extra-polar parts in its 
 neighbourhood. Hermann is aocordingly led to regard all 
 secondary electromotive phenomena as resulting from the inter- 
 ference of the excitatory change which occurs at the anode in 
 consequence of the opening of the current, with ordinary galvanic 
 polarisation, it being understood that the physical change always 
 anticipates the vital one as regards the order of its occurrence ; 
 so that whenever the conditions (strong current and short 
 closure) are such as to enable these two antagonistic influences 
 to manifest themselves together, the process consists of a more 
 transitory physical effect, which is followed by a more lasting 
 physiological one. 
 
 If the views of Hermann and Hering are accepted, the reader 
 will see that the form given by Tigerstedt and Griitzner to their 
 theory as to the essential nature of the opening contraction must 
 be modified. Tigerstedt assumes that at the break of a current, 
 the anode becomes, in consequence of negative polarisation, 
 virtually a cathode, an assumption which takes no account of 
 the facts of positive polarisation. On this point the reader will 
 compare with advantage du Bois' criticism of Tigerstedt in his 
 sixteenth paragraph, with that of Hermann in the paragraph 
 which precedes his conclusion. 
 
 If it is true that ' positive polarisation ' of a nerve or of a 
 muscle is in reality a purely physiological, or, in other words, an 
 excitatory effect of the leading through it of a voltaic current, it 
 becomes of interest to enquire what is the relation of the elec- 
 trical after-efiects so fully studied by du Bois-Reymond, Hering, 
 and Hermann, to the more mechanical effects which form the 
 subject of the first part of this series. In general it is, as every 
 physiologist knows, extremely difficult to demonstrate in detail 
 the correlation which undoubtedly exists between the electro- 
 tonus of Pflliger, and the galvanoscopic phenomena which were 
 originally investigated by du Bois and treated of by him under 
 the same title. The chief reason of the difficulty lies in the 
 circumstance that until recently it has been possible to investi- 
 gate the former only in muscle, the latter only in nerve. But 
 
X PREFACE. 
 
 from the moment that it was shown by Hermann and Hering's 
 researches, that the well-known electromotive after-effects of the 
 current in nerve (du Bois' electrotonus) have their counterparts 
 in muscle, that is, that the electrical state of a muscle which 
 is traversed axially by a current is subject to polar variations 
 which correspond to an- and cathelectrotonus, it at once became 
 of interest to enquire how these variations are linked to con- 
 comitant changes in what must always be regarded as the 
 essential function of muscle — that of contracting in response 
 to stimulation. 
 
 In 1 88 1 it was discovered by Engelmann that the anodic and 
 cathodic localisation of the closing and opening excitatory effects 
 could be demonstrated in the heart of the frog, and Hering 
 showed how, with the aid of the double myograph, as much 
 could be done with a striated muscle. Electrotonus of muscle 
 was thus in a certain sense demonstrated, but to complete 
 the proof it was necessary to show that as cathelectrotonus 
 is represented by contraction, so anelectrotonus is represented 
 by relaxation. This was accomplished by Prof. Biedermann, 
 Hering's coadjutor at Prague, in 1883^ It was necessary 
 for the purpose to employ as subject of experiment a muscular 
 structure capable of persistent tonus, for it was only in such 
 a structure that relaxation could be observed. The required 
 condition was found to be fulfilled by the heart of the snail 
 (Helix pomatia), which, as Biedermann discovered, can generally 
 be thrown into spasm of sufficient duration for the purpose, 
 by the simple expedient of allowing the ventricle to contract 
 against unwonted resistance. By the observation that the 
 closure of a battery-current through the heart leads to relaxa- 
 tion which takes its start from the anode, just as in the relaxed 
 heart, the closing contraction takes its start from the cathode, 
 Biedermann was able to prove that in that structure, relaxation 
 of muscle already contracted is as characteristic of anelectro- 
 tonus, as contraction of relaxed muscle is of cathelectrotonus. 
 
 * W. Biedermann, Ueber das Herz von Helix pomatia. (Ein Beitrag zur ver- 
 gleichenden Physiologie der Muskeln.) Wiener Sitzungsberichte, vol. Ixxxix. ig-fiS- 
 
PREFACE. XI 
 
 In the paper of which a translation is introduced into our 
 series of Memoirs, Biedermann extended the same mode of 
 experiment to a striated muscle of the frog. In order to 
 bring about a similar condition of permanent spasm to that 
 which, in the heart of Helix, is produced so easily by mere 
 distension, he availed himself of the action of veratrine, and 
 found that in this, as in the other case, the cathodic contraction 
 is replaced by anodic relaxation, and the anodic contraction 
 by cathodic relaxation. Here, then, an opportunity seemed to 
 be afforded for the first time, of determining experimentally 
 whether electrical and functional electrotonus occur under 
 the same conditions — whether, as the opening contraction 
 corresponds with the anodic after-effect (of which, according 
 to Hering, positive polarisation is but the expression), the 
 cathodic break-relaxation is marked by a cathodic after-effect, 
 which, although of opposite sign, i.e. positive, as regards the 
 condition of the cathode itself, is of the same sign galvano- 
 scopically. To ascertain whether it was so or not, Biedermann 
 prepared the sartorius by subjecting one end of it to the action 
 of veratrine, leaving the other intact, and then induced cathe- 
 lectrotonus in the poisoned, that is, tonic end, leading a battery- 
 current through the whole riband for a couple of seconds. The 
 poisoned end was at once thrown into veratrine spasm and the 
 cathode became negative ; but when (after compensating) the 
 current was again closed, its after-effect was in the opposite 
 direction, that is, in the direction of the exciting current. In 
 other words, relaxation of the veratrine spasm was, as was 
 expected, accompanied by diminution of the negativity which 
 is its galvanoscopic expression. When the same observation 
 was made with the modifying current in the opposite direction, 
 that is, with the anode at the veratrine end, it was similarly 
 found that a current led through during spasm was reversed, 
 that is, that the anode became positive. 
 
 The general conclusion derived by Biedermann from these 
 experiments is that the state of muscle varies, not merely 
 between the state of excitation and the normal, or unexcited 
 
Xll PBEFACE. 
 
 state (the former being indicated galvanoscopically by the pre- 
 sence in the tissue of a current from the excited to the unexcited 
 part, and mechanicall}^ by the presence of more or less spasm, 
 the latter by the opposite effects), but that muscle varies in 
 the direction of greater or in that of less fitness for func- 
 tional activity. By employing such terms as 'up' or .'down' 
 to denote these two tendencies, it is possible to express his 
 conception thus : — A voltaic current produces at the anode 
 when it is closed an ' up ' state, of which the galvanoscopic 
 sign would be relatively positive, and the mechanical sign 
 axial extension but aptness to contract ; at the cathode a 
 ' down ' state, associated conversely with ' negativity,' and axial 
 shortening. The assumption of a state intermediate between 
 these two is suggested by the indications of what, for want 
 of a better name, is called by Biedermann ' reaction,' that is, 
 of a tendency (of which the behaviour of all excitable tissues 
 suggests the existence) to assume a condition opposite to that in 
 which it actually is, by virtue of which the ' down ' state is 
 produced at the anode at make, the ' up ' state by the cathode 
 at break. In Biedermann's experiments it was strikingly seen 
 that the same external cause produces contraction and negativity 
 when acting on muscle in the ' up ' state, relaxation and posi- 
 tivity when acting on the same muscle in the ' down ' state. 
 This being so, there may be between contraction and complete 
 relaxation an intermediate state in which the two opposite 
 tendencies neutralise one another. If such a state exists, it does 
 not follow that it is the normal state of the tissue, for by the 
 word normal the physiologist understands simply the state 
 which best fits the structure for its function, and cannot be 
 guided in his use of it by the correlation of its elementary 
 properties. 
 
 Part III contains Prof, du Bois-Reymond's investigations in a 
 field of research in which up to the present moment he has been 
 almost the only worker — that of the electrical phenomena of fish 
 which are possessed of special electrical apparatus. The Memoir 
 on Malapterurus, which appeared a few years ago in the Author's 
 
PREFACE. xm 
 
 Collected Papers, has been translated at his suggestion. It con- 
 tains a full description of the methods by which the difficulties 
 of the investigation were in the first instance overcome, and 
 the reader will find it to be an indispensable introduction to the 
 other two Memoirs on Torpedo, which contain his more recent 
 observations. As each of the three memoirs is complete in itself, 
 and is preceded by a summary of the subjects discussed in each 
 paragraph, no introduction is needed. It may not, however, be 
 out of place to draw attention to the very remarkable discoveries 
 recorded in the concluding paper relating to the fact of ' irre- 
 ciprocal conduction,' which the author has proved to exist in a 
 remarkable degree in the columns of the electrical organs of 
 Torpedo. 
 
 The Editor regrets that the publication of the volume has 
 been unduly delayed, in consequence of the unexpected difii- 
 culties experienced by the translators in their work. The 
 subject is so new to English scientific literature, that much 
 time is unavoidably lost in deciding between the two alterna- 
 tives, of using Germanised phraseology and of coining English 
 equivalents for the forms of expression used by German writers. 
 In general, the latter alternative has been preferred, but not 
 without a heavy sense of responsibility. Whether successfully 
 or not, great pains have been taken to select such equivalents 
 as appear most accurately to express the meaning of the original, 
 and the translators have, as a rule, consented to employ the same 
 equivalents for the same words. 
 
CONTENTS, 
 
 PART I. Researches relating to the Law of Contraction. 
 
 Author. Title. Translator. Page. 
 
 1. TiGEKSTEDT . Changes of excitability produced 
 
 in nerves by a constant cur- 
 rent Dr. James Niven . . 5 
 
 2. „ Contribution to the theory of 
 
 the Break Contraction ... „ „ 
 
 3. „ On the Internal Polarisation of 
 
 Nerves ,, ^^ 
 
 4. Grdtzner . . The Nature of the Electrical 
 
 Break-contraction .... Mr. J. S. Haldane 
 
 5. Hering . . On Nerve-excitation by the 
 
 Nerve-current Dr. Aug. Waller 
 
 49 
 
 77 
 
 93 
 
 127 
 
 PART II. Researches relating to Secondary Electromotive Phenomena. 
 
 6. Du Bois-Rey- On Secondary Electromotive 
 
 MOND. Phenomena in Muscles, 
 
 Nerves, and Electrical Organs. Mrs. Lauder Brunton . 163 
 
 7. Hering . . On Du Bois-Reymond's Re- 
 
 searches on Secondary Electro- 
 motive Phenomena of Muscle. Dr. Aug. Waller . 
 
 8. „ On the Positive After- variation 
 
 of the Nerve-cjurrent conse- 
 quent on Electrical Stimula- 
 tion . • 
 
 '.). Hermann. . On so-called Secondary Electro- 
 motive Phenomena in Muscle 
 and Nerve . Mr. Francis Gotch . 
 
 229 
 
 25.5 
 
XVI CONTENTS. 
 
 Author. Title. Translator. Page. 
 
 10. BlEDEEMANN , On the Phenomena of Inhibition 
 
 produced by the Electrical 
 Excitation of Striated Mus- 
 cles, and on Positive Cathodic 
 Polarisation Miss Edith Prance . 331 
 
 PART III. Eesearcbes on the Electrical Phenomena of Malaptervirus 
 
 and Torpedo. 
 
 11. Du Bois-Rey- Observations and Experiments 
 
 MONO. on Malapterurus Miss Edith Prance . 369 
 
 12. „ Living Torpedos at Berlin. First 
 
 Communication Editor 417 
 
 13. „ Living Toi-pedos at Berlin. Se- 
 
 cond Communication . . . Miss Edith Prance . 479 
 
CORRIGENDA. 
 
 p. 6, note I, for fortschritte read Fortschritte. 
 
 P. 7, note 4, for nouvelles read noxivelle. 
 
 P. 23, § 2, for irritability read excitability. 
 
 P. 56, line 4 from top, /or Marcy read Marey. 
 
 P. 56, Exp. II, heading, ybr .... read the break-contraction. 
 
 P. 57, Exp III, heading, /or .... read the break-contraction. 
 
 P. 57, Exp. IV, heading, /or .... read the break-contraction. 
 
 P. 94, line 13 from bottom, /or break read make. 
 
 P. 123, line j8,for Tliermopyle read Thermopile. 
 
 P. 166, line 15, /"or through read in. 
 
 P. 168, line 1 9, /or polarisation-current read eflFects. 
 
 P. 187, line 13 from bottom, /or would read should. 
 
 P. 203, line 19 from bottom, /or Sect. 2 read p. 163. 
 
 P. 204, line 13 from bottom, /or that read of. 
 
 P. 204, line II from bottom, /or Sect. 9 read Sect. 10. 
 
 P. 249, line 4, for Sect. 9 read Sect. 8. 
 
 P. 277, note I, for p. 404 read p. 343. 
 
 P. 280, line 2 from bottom. /or Sect, i read Sect. 21. 
 
 P. 297, line 11, for or read in the. 
 
 P. 343, note, for No. V read No. VI. 
 
 P. 471, line i,for positive read negative. 
 
 P. 490, line 6 from bottom, /"or No. V read No. VI. 
 
 P. 504, note, for p. 224 read p. 457. 
 
Part I. 
 
 RESEARCHES 
 
 RELATING TO THE 
 
 "LAW OF CONTRACTION." 
 
I. 
 
 CHANGES OF EXCITABILITY 
 
 PKODUCED IN 
 
 NERVES BY A CONSTANT CURRENT. 
 
 By Dk. EOBEET TIGERSTEDT. 
 
 Communicated to the Swedish Eoyal Academy of Sciences, April 12, 1882. 
 
 B 2 
 
I. 
 
 In his famous investig-ations into the physiology of electrotonus, 
 Pfliiger, in the year 1859, submitted the changes of excitability in 
 nerves elicited by a constant current of electricity to an experi- 
 mental examination which will stand to all time as a model for 
 similar investigations, as regards both the number and variety of 
 the methods of research employed and the completeness of the 
 experimental criticism. As test-stimulus he employed partly the 
 constant current, partly the induction-shock, and in part also he 
 used chemical stimuli. The most important actual relations 
 established by this investigation are, very briefly, as follows. In 
 the region of the positive pole of the current the excitability of the 
 nerve is diminished, as well in the extrapolar as in the intrapolar 
 direction ; and the diminution attains its definite amount gradually, 
 being not so great for some moments after the current is closed 
 as it afterwards becomes. When the current is opened there 
 occurs immediately an increased excitability, which may last for 
 a longer or shorter period. In the region of the negative pole 
 of the current the excitability of the nerve is increased as well 
 in the extrapolar as in the intrapolar direction. This increase 
 occurs directly after contact is made. When contact is broken, 
 diminished excitability is exhibited if the current is weak or 
 moderate, which, however, after an exceedingly brief interval 
 gives place to a condition of increased excitability lasting a 
 longer or shorter time. 
 
 It is at the poles that the change in excitability is greatest, 
 diminishing as we leave these both in the extrapolar and intra- 
 
6 CHANGES OF EXCITABILITY IN NERVES 
 
 polar directions. The greater the strength of the polarising 
 current, the greater is the change of excitability and its extrapolar 
 extension. Between the poles the diminution of excitability- 
 extends as the strength of the polarising current increases ; or, 
 to put it otherwise, with increasing strength of current, the in- 
 difference-point moves gradually from the positive to the negative 
 pole. 
 
 The greater the separation of the electrodes the greater is the 
 tract of nerve over which the extrapolar change of excitability 
 extends. The behaviour of the intrapolar tract is absolutely the same 
 whatever the distance between the electrodes. 
 
 To test the local changes of excitability in the intrapolar tract 
 Pfliiger used chemical stimuli ; the induction-coil he employed only 
 to investigate the total excitability. He abstained from examining 
 the changes of excitability with sharply localised electrical stimuli 
 fearing an escape of electricity between the constant and test 
 currents. 
 
 An important, though perhaps not unexpected, addition was 
 made to these results two years after their publication by Obernier 
 in a research carried on under Pfliiger's guidance. Obernier 
 proved by experiments that, with the application of strong polarising 
 currents (6-8 Groves), there occurs at the negative pole in the first 
 moments after contact is broken a very considerable diminution of 
 excitability ^. 
 
 Before this, however, attacks had been made on the correctness of 
 Pfliiger's results — attacks which have since been from time to time 
 renewed by various enquirers. The most notable fact, however, is, 
 that all those who have thus assailed Pfliiger, with few exceptions, 
 have conducted their counter-researches with nothing like the same 
 care as Pfliiger employed. What Meissner says of one of the first 
 of them may be applied with justice to them all. ' Inasmuch as 
 Pfliiger has grounded all his enunciations on investigations made 
 with every kind of efficient stimulus, one has a right to demand 
 from a criticism which is directed to prove these enunciations in 
 part untenable, that it be pursued on a like basis and "^dth a similar 
 use of every species of stimulus^. 
 
 To this category of works we must refer those of Schiff^, of 
 
 * Obernier, Archiv fiir Anatomic und Physiologie, 1861, pp. 268-278. 
 ^ Meissner, Bericht iiber die fortschritte der Anatomic und Physiologie, 1S63. 
 Leipzig, 1864, p. 359. 
 
 3 Schiff, Lehrbuch der Nerven-physiologie, Jahr 1858-59, p. 94. 
 
PRODUCED BY A CONSTANT CUREENT. 7 
 
 Budge ^, of Valentin 2, of SehifF and Herzen^, and of Lautenbach*. 
 Thus we have the employment of metallic instead of unpolarisable 
 electrodes ; no measures of precaution are mentioned as being- used 
 to prevent escape of current from one circuit into the other ; ex- 
 periments are carried out on one and the same nerve in such a way 
 that the direction, as well of the polarising as of the test current, 
 is many times reversed ; and so on. Under such circumstances it 
 was impossible to arrive at clear and unambiguous results. Schiff 
 avows too ' that to bring together all tolerably similar results of ex- 
 periment toonld seem to joroduce a perfect chaos'^.' 
 
 Small as was the impression which the works I have mentioned 
 were able to make on Pfliiger's results, these were equally far from 
 being upset by the adverse conclusions obtained from experiments 
 on living human beings. Fick^ found no change elicited in the 
 excitability of nerves by the constant current. Eulenburg'^ con- 
 firmed Pfliiger's laws. Erb ^ obtained at first quite opposite results, 
 but having had his attention directed by Helmholtz to the likelihood 
 of this having been caused by a peculiarity in distribution of the 
 current in the uninjured human body, he repeated his experiments^, 
 and now attained results completely agreeing with those of Pfliiger 
 and Eulenburg. Runge ^^ on the other hand arrived at quite opposite 
 conclusions. It would seem pretty clear, however, that the com- 
 plicated, not to say undeterminable, conduction-relations which exist 
 in the living human body must cause a considerable difference 
 between these results and those which rest on experiments made 
 with isolated preparations of nerves, and that results of this kind 
 cannot be adduced against Pfliiger as proofs — so long at least as 
 no question is taken into consideration excepting that of the way in 
 which the normal excitability of a nerve is changed by a constant 
 current of electricity. 
 
 * Budge, Arcliiv fiir patholog. Anatomie, xxviii, 1863, pp. 282-301. 
 
 * Valentin, Die Zuckungsgesetze des lebenden Nerven und Muskels, Leipzig, 
 1863, pp. 14, G5-86; Physiologische Pathologie der Nerven, i, Leipzig, 1864, 
 pp. 134, 135 ; Zeitsclirift fiir Biologic, viii, 1S72, pp. 210-238; Moleschott's Unter- 
 suchungen, xi, 1873, pp. 169-181 ; Zeitschrift fiir Biologie, x, 1874, pp. 153-176. 
 
 ^ SehifF nnd Herzen, Moleschott's Untersuchungen, x, 1867, pp. 430-446. 
 ^ Lautenbach (under SchifTs direction), Archives des sciences physiques et 
 naturelles, nouvelles periode, Ivii, 1877, pp. 88-99. 
 •' SehifF, Moleschott's Untersuchungen, x, p. 436. 
 
 * Fick, Medicinische Physik, Zweite Auflage, 1866, p. 377. 
 
 ' Eulenburg, Deutsches Archiv fiir Klinische Medicin, iii, 1877, pp. 1 17-142. 
 " Erb, ib. iii, 1867, pp. 271-273, and 513-524. 
 ' Erb, ib. iii, pp. 525-528. 
 " Runge, ib. vii, T870, pp. 356-384. 
 
8 CHANGES OF EXCITABILITY IN NERVES 
 
 Of more weig-ht than these investigations are those of Bilharz 
 and O. Nasse, of ]\Iunk, of Griinhagen, of Wundt, and of Hermann 
 and Bernstein, which have been conducted with all the care that 
 researches of this nature demand. They also confirm in all 
 essentials Pfliig-er's results, but serve at the same time partly to 
 complete and partly to amend them, or to g-ive a different theoretic 
 significance to the facts discovered by him. A short glance at 
 these vnW indicate their importance. 
 
 Under the direction of du Bois-Reymond, Bilharz and O. Nasse 
 tested Pfliiger's laws with mechanical stimuli ^ For this purpose 
 they availed themselves of Heidenhain's tetanomotor. The in- 
 crease or diminution of the tetanus produced by it, when a 
 constant cuiTcnt was passed in a given direction, served as a test 
 of the change of excitability which then occurred. 
 
 Of course the test-stimulus injured the nerves to a very con- 
 siderable extent. The nerve was hammered for six seconds 
 continuously, and after each excitation got one to two minutes 
 to recruit ; yet in most cases it did not bear more than 6-8 
 successive stimulation experiments. So far as one can gather 
 from the brief descriptions of the authors, they passed the polar- 
 ising current in one direction only throughout each series of 
 experiments. 
 
 With fresh nerves they established by this procedure the complete 
 correctness of Pfiiiger's laws. As soon as the polarising current 
 was closed a diminution took place at the positive pole in the 
 tetanus elicited by the tetanomotor, at the negative pole an increase. 
 When the experiment was continued the reverse relation set in ; 
 first there was no change when the current was closed, and after 
 further application of the mechanical stimulus closing the current 
 yielded precisely the opposite result, namely at the positive pole a 
 strengthening, at the negative a weakening of the tetanus. The 
 authors obtained just the same result when they produced local injury 
 of the nerves with chemical substances. Afterwards Bilharz found ^ 
 that other modes of injuring a nerve, for example the transmission 
 of a very strong constant current through it, or the bringing of a 
 red hot wire near it, &e. gave occasion to the same deviation from 
 Pfiiiger's results. 
 
 From the brief account which the authors have published of 
 individual experiments it is impossible to gather on what these 
 
 * Bilharz nnd 0. Nasse, Archiv fiir Anatomic und Physiologic, 1862, pp. 66-83. 
 ^ Bilharz, Archiv fiir Anatomie und Physiologic, 1862, pp. 84-89. 
 
PRODUCED BY A CONSTANT CUliliENT. 9 
 
 deviations depend. Any how they cannot tell against Pfliiger's laws, 
 inasmuch as experiments on fresh nerves exhibited a complete 
 accordance with them, as the authors expressly state ; and assuredly 
 no sig-nificance can be attached to the adverse results given by 
 nerves damaged in every imaginable way. 
 
 Moreover, a nerve is so delicate an organ that it is only with the 
 greatest caution that we can draw conclusions from the phenomena 
 which it presents after serious injury ; to which maybe added that in 
 investigations on the changes produced in nerves by a constant 
 current a multitude of sources of error may easily disturb even the 
 best arranged and most thoroughly thought out experiments. 
 
 Some years after, H. Munk came forward with considerations 
 adverse to Pfliiger's results, and on this ground, that the latter had 
 in his investigations not taken into consideration the internal 
 secondary resistance in nerves. Of the great work in which Munk 
 was to have presented his experiments and his new doctrine ^ there 
 has aj)peared only the first part, which deals with investigations on 
 the secondary resistance in nerves. The question of changes of 
 excitability evoked by the constant current Munk accordingly treats 
 of only in a provisional communication ^. 
 
 Introducing as he did in the path of the current so great a resist- 
 ance that the changes in resistance occurring in the course of 
 experiment became relatively insignificant, he came to quite 
 different conclusions from Pfliiger. These and the theoretical 
 view at which he has arrived he formulates in these words. ' The 
 excitation of the nerve in consequence of the electrical current 
 is due, firstly, to the direct displacement of the nerve-fluid in the 
 direction of the current, and, secondly, to the return of the fluid, 
 on interruption of the current, to those parts of the nerve from 
 which it has been displaced.' . . . . ' The muscular contraction due 
 to a given displacement of nerve-fluid is increased by a previously 
 existing displacement of the same, when both displacements are in 
 the same direction, but diminished when they are in opposite 
 directions.' 
 
 Thus, according to Munk, the extrapolar excitability during 
 closure of a descending current, is increased at the negative pole 
 when the test-current is descending, diminished when it is ascend- 
 ing ; and so on. 
 
 ^ H. Munk, Untersuchungen iiber das Wesen der Nervenerregung, i, Leipzig, i8<58. 
 ^ H. Munk, Archiv fiir Anatomie und Physiologic, 1866, pp. 369-390. 
 
10 CHANGES OF EXCITABILITY IN NERVES 
 
 To estimate the value of these conclusions is clearly impossible as 
 Munk has adduced no proofs to substantiate them. In his criticism 
 of Piliiger he has moreover entirely overlooked the fact that the 
 latter had also obtained by means of chemical stimuli the same 
 results as he had got by testing" with an electrical stimulus ; and 
 just as little does he take into account that Bilharz and O. Nasse had 
 also found these confirmed by mechanical stimulation applied to fresh 
 nerves. Under these circumstances we cannot regard his results as 
 conclusive. 
 
 With all possible care and observance of minutiae Hermann^ 
 examined the changes of excitability, chiefly in the intrapolar region, 
 with the object of substantiating the doctrine based on his investi- 
 gation of the action-current in polarised nerves. ' Nerve excitability,' 
 he says, ' increases in amount as we proceed to spots stronger in 
 positive, or weaker in negative polarisation ; in the reverse case it 
 diminishes.' 
 
 In the introduction to his paper he directs attention to a cir- 
 cumstance, in itself clear enough, but which has been strangely 
 overlooked by enquirers in this subject, namely, that, as the strength 
 of the test-stimulus is increased, a point is reached where the 
 influence of the constant current on the excitability of the nerve 
 suddenly vanishes ; and this point corresponds to such a strength of 
 stimulus as evokes a maximal muscular contraction. With such a 
 strength the muscular contraction can no longer be increased by 
 katelectrotonus, for the very reason that it is at a maximum ; its 
 diminution through anelectrotonus also is trifling, and vanishes 
 with such a strength of the stimulus as exceeds by only an in- 
 significant amount that at which the maximal contraction occurs. 
 
 Further on in his paper Hermann gives an account of the in- 
 vestigations by which he supports his own doctrine and opposes 
 Pfliiger, — that is, not Pfliiger's experiments, but his theoretical 
 explanation of them. I pass over here one of Hermann's experi- 
 ments, as to which he afterwards himself discovered that the 
 result might be referred to a defect of arrangement 2, and shall 
 only mention the others. These aim at proving : 
 
 I. That if the polarising current along with the seats of stimu- 
 lation is brought nearer the muscle, while nothing else is altered, 
 a diminution ensues in the influence of the polarising current for 
 every stimulus applied on the near side of the indifference-point, 
 
 * Hermann, Archiv fiir die ges. Physiologic, vii, 1S73, pp. 323-364. 
 '■' Hermann, ib. vii, 1873, pp. 497-498. 
 
PRODUCED BY A CONSTANT CURRENT. 11 
 
 while its influence increases for every stimulus applied on the 
 far side. 
 
 2. That at seats of stimulation lying* far up, an approximation of 
 the polarising" current to the muscle is accompanied, first by a 
 diminution, then by an increase of influence on the excitability. 
 
 3. That with strong polarising- currents inhibition takes place 
 at the neg-ative pole ; and if the excitatory efiect has to pass this 
 pole in the course of its propagation, muscular contraction must fail, 
 and this as well with an ascending current and extrapolar stimula- 
 tion as with a descending- current and intrapolar stimulation. The 
 last case should furnish a real experimentum crucis against Pfliiger's 
 view. With the first two it is not so, as other circumstances 
 mig-ht influence the result, although such phenomena cannot be 
 foreseen on Pfliiger's theory. 
 
 The experiments instituted in proof of (i) were completely 
 successful. When the fixed system of electrodes was broug-ht 
 nearer to the muscle, it appeared that with an ascending' current 
 the katelectrotonic increase of contraction became greater and its 
 anelectrotonic diminution less, as also that with a descending- 
 current the relation was reversed. 
 
 With intrapolar stimulation Hermann found occasionally at the 
 upper seat of stimulation a diminution, and at the lower an increase 
 of muscular contraction. This, which would certainly form a 
 decisive proof against Pfliig-er's theory, may however be explained 
 on the ground that we have to deal with the same sources of error 
 as in the experiments previously referred to, namely with the 
 presence of nerve branches to the thigh, whereby, in the more 
 distant position of the system of electrodes, the real indifierence- 
 point came closer to the proximal electrode than it did when the 
 system of electrodes was brought nearer to the muscle. 
 
 In proof of (2) Hermann did not arrive at definite results. On 
 the other hand, in the experiments which he instituted to obtain 
 proofs of (3) he found, quite correctly, that with strong currents 
 the muscular contractions fell off", and were entirely annulled when 
 the excitation had to pass the negative pole. With an ascending 
 current this could be explained on Pfliiger's theory by inhibition 
 at the positive pole, but this theory could not possibly explain 
 why the influence of a stimulus applied in the neighbourhood of 
 the negative pole should be diminished or annulled by a de- 
 scending current. 
 
 Besides this it was only with currents of moderate strength that 
 
12 CHANGES OF EXCITABILITY IN NERVES 
 
 Hermann found a diminution in the influence of an intrapolar 
 stimulus at the positive pole of an ascending current ; with weak 
 currents, on the contrary, he obtained a reinforcement. 
 
 Particulars as to the streng'th of the polarising current, as well 
 as with regard to the distance between its electrodes and those 
 of the test-current, are, it is true, wanting in Hermann's paper; 
 but the facts disclosed by him do nevertheless tell against Pfliiger's 
 theory, and cannot be explained in accordance with it, unless one 
 assumes that the intrapolar test-electrodes were not near enough 
 to the respective poles to lie in the anelectrotonic tract when the 
 currents were weak ; and, similarly, that with strong currents the 
 anelectrotonus had extended to their position, although they lay 
 tolerably near the negative pole. Ob^'iously this cannot be decided 
 without new experiments, at least so long as we have no detailed 
 particulars on which to found conclusions. 
 
 Shortly after Hermann, Bernstein^ took up this question. He 
 employed maximal induction-shocks, and by their means investi- 
 gated the change of excitability in the tract lying between the 
 polarising current and the muscle. He moved the test-electrodes 
 as close as possible to the nearest electrode of the polarising 
 current ; the distance between the electrodes of the latter being, 
 on the other hand, considerable, namely 25 mm. The strength of 
 the current was great, ranging from 4 Daniells with a derivation 
 circuit of 10.000 rheochord units, up to 6 Groves without deri- 
 vation. He sums up the result of his experiments thus — ' When 
 a constant current is passing along a nerve, the liberation of 
 excitation is impeded at the positive pole, so that weak stimuli 
 exercise less influence than under normal circumstances ; but the 
 maximum of the excitation elicited by powerful stimuli is greater. 
 At the negative pole, on the other hand, the liberation of excitation 
 is facilitated, so that weak stimuli have greater efiect ; but the 
 maximum of excitation which can be Liberated by strong stimuli 
 is less.' 
 
 By a closer sifting of his results Bernstein endeavoured to show 
 their accordance with Pfliiger's theory, and expressly declared that 
 they were no way in antagonism with its chief points, but rather 
 served to confirm it. 
 
 Hermann 2 soon came forward in opposition to Bernstein. He 
 had repeated Bernstein's experiments without succeeding in cou- 
 
 * Bernstein, Archiv fiir die gesammte Physiologic, viii, 1874, pp. 40-60, 
 
 * HermanB, Archiv fiir die ges. Physiol ogie, viii, 1874, pp. 258-275. 
 
PRODUCED BY A CONSTANT CURRENT. 13 
 
 firming his results, and he conjectured therefore that they were 
 due to defective insulation, inasmuch as strong currents might easily 
 enough have in part found their way to the nearest electrode 
 of the test-current by conduction in the moist chamber, in which 
 case its direction would of necessity be reversed. For further 
 information as to the controversy the reader is referred to the 
 originals^. 
 
 All the investigations which I have so far referred to were in 
 the main conducted by the same experimental methods. About 
 the same time Wundt and Griinhagen sought to throw light on 
 the question from a new point of view. 
 
 These enquirers had independently proposed to themselves to 
 test the excitability of the nerve during the first moments after 
 opening and closing the polarising current. With this object 
 Wundt ^ availed himself of the pendulum myograph, and applied 
 the test-stimulus at a short and accurately measured time after 
 opening or closing of the polarising current. The essential part 
 of his results is as follows — ' Directly after the closing of the 
 polarising current the excitability begins to rise through the whole 
 length of the nerve. This increase of excitability continues, in the 
 neighbourhood of the kathode, into the persistent katelectrotonic 
 increase. In the neighbourhood of the anode the excitability 
 mounts to a maximum and then again sinks gradually to make way 
 for the anelectrotonic diminution. During a certain period, there- 
 fore, after closure of the constant current excitability is increased 
 along the whole course of the nerve. This stage of increased 
 excitability in both phases of electrotonus exceeds notably in dura- 
 tion the course of a muscular contraction with its associated stage 
 of latent stimulation. At the moment the circuit is broken, with 
 strengths of current which are not such as to introduce diflTerences 
 of conduction due to inhibitory processes, there occurs an excitation 
 of which the intensity is the same whether on the kathodic or 
 anodic side, and whether inside or outside of the poles. Along with 
 this excitation, however, an inhibition sets in, which in part is 
 occasioned by the remanent inhibition at the anode at closure of 
 the current, and in part comes into existence at opening, in eon- 
 sequence of the occurrence of changes at the kathode which are 
 
 ' Bernstein, Archiv fiir die ges. Physiologie, viii, pp. 498-505. Hermann, ib. ix, 
 1874, pp. 28-34. 
 
 ^ Wundt, Archiv fiir die ges. Physiologie, iii, 1870, pp. 437-440 ; — Untersuehungen 
 zur Mechanik der Nerven und Nervencentren, i. Erlangen, 1871. 
 
14 CHANGES OF EXCITABILITY IN NERVES 
 
 of the same nature as those which occurred at the anode at 
 closure.' 
 
 In a provisional communication on his investigation, which, so 
 far as I know, was not j)ublished, Griinhagen ^ very briefly made 
 known the results he had obtained relative to the duration of 
 the chang-es in excitability produced by a constant current. In 
 this research he employed tetanus and its changes on closing and 
 opening the current. He applied the test-stimulus between the 
 polarising current and the muscle, and found that the tetanus 
 in anelectrotonus fell off and in katelectrotonus increased, in such 
 a way that in the first case the changes of excitability were longer 
 in fully pronouncing themselves than in the latter. On the other 
 hand, he could make out no change of excitability at the negative 
 pole consequent on opening. An increase of excitability at the 
 anode during the first moments after opening of the current, 
 such as Wundt had observed, Griinhagen could not establish. 
 Finally, he found an immeasurably short duration of increase in 
 excitability at the kathode after interruption of the current. On 
 these disagreements with Wundt' s result we can lay no stress 
 to the disadvantage of the latter, since the author has neither 
 here nor since given a full statement of his experiments. The 
 chief fact is that Griinhagen has confirmed Pfliiger's laws in the 
 most essential points. 
 
 The researches then which have been instituted since the ap- 
 pearance of Pfliiger's work, to determine the changes of excitability 
 in a fresh nerve elicited by a constant current, have, in essentials, 
 confirmed Pfliiger's results, and have, further, supplemented them 
 with new investigations on the changes of excitability occurring 
 directly after closure and opening of the current (Wundt, Griin- 
 hagen). The theoretical difference between Pfliiger and Hermann 
 is however still undecided ; and the experiments instituted by the 
 latter on the inhibitory influence of the negative pole with 
 powerful currents constitute for the present very weighty ob- 
 jections to Pfliiger's theoretical view. 
 
 Under such circumstances, and as hitherto no investigation of 
 the changes of excitability has been instituted with purely me- 
 chanical stimuli, I considered that there were sufficient reasons 
 for taking up the enquiry once more from this point of departure, 
 the more so that the mechanical method of stimulation, with 
 
 ^ Griinhagen, Arcliiv fiir die gesammte Physiologie, iv, 1871, pp. 547-55°* 
 
PRODUCED BY A CONSTANT CURRENT. 15 
 
 the perfection to which it has recently attained, seemed to offer 
 many and weighty advantag-es for experiments of this kind. In 
 conducting these one has not, as with electrical stimuli, to fear 
 an escape of electricity from the polarising to the test-current, or in 
 the opposite direction. One can repeat the experiments in close suc- 
 cession without the nerve being thereby rendered useless, at the 
 same time taking as measure of change of excitability individual 
 muscular contractions, and not, as with chemical stimuli, exclu- 
 sively tetanus ; and, what is perhaps most important, one can 
 get as near as one likes to the poles of the polarising current with 
 mechanical stimuli and thus investigate it more satisfactorily than 
 was ever possible with electrical or chemical stimuli. To this may 
 be added that the mechanical stimulus is so completely different in 
 its nature from the stimulus — the constant current — whose opera- 
 tion on the nerves it is the purpose of these investigations to 
 determine. 
 
 Inasmuch as the changes of excitability outside the poles of 
 the polarising current had been already so many times investi- 
 gated by different enquirers, and as the results obtained — in so 
 far as the methods used were free from error — had invariably 
 confirmed those of Pfliiger, I determined to apply myself specially 
 to the chang-es in excitability of the intrapolar tract, and there 
 again particularly to direct my attention to the pole nearest the 
 muscle, that is, with an ascending current to the positive, and 
 with a descending current to the negative pole, because in this 
 way results could be attained in the simplest form and with 
 least disturbance from other circumstances. At the same time 
 I did not neglect to examine also the changes of excitability at 
 other parts of the nerve, that there too I might test the correctness 
 of Pfliiger's laws by this method. My investigation separates 
 itself accordingly into the following ten divisions. 
 
 I. The changes of excitability at the negative pole of an ascend- 
 ing current — extrapolar. 
 
 3. The changes of excitability at the j^ositive jaole of an ascending 
 current — extrajwlar. 
 
 3. The changes of excitability at the negative jjole of an ascend- 
 ing cur r en t — intrapolar. 
 
 4. The changes of excitability at the positive pole of an ascending 
 curren t — intrapolar. 
 
16 
 
 CHANGES OF EXCITABILITY IN NERVES 
 
 5. The change in site of the indifference-point with increasing 
 strength of an ascending current. 
 
 6. The changes of excitdbility at the negative pole of a descending 
 current — extrapolar. 
 
 7. The changes of excitability at the positive pole of a descending 
 current — extrapolar. 
 
 8. The changes of excitability at the negative pole of a descend- 
 ing current — intrapolar. 
 
 9. The changes of excitability at the positive pole of a descending 
 current — intrapolar. 
 
 10. The change in site of the indifference -point with increasing 
 strength of a descendhig current. 
 
 The experimental arrangements were as follows (see Fig. i). 
 
 ^ig. I. 
 
 From the battery {B.^ the current passes to the electro-magnet 
 {E)^ which elevates the lever of my new mechanical stimulation- 
 apparatus, which I employed in these experiments^ (see Fig. 2). 
 This current then passes to a Pohl's reverser without the cross wires. 
 If the arms of the reverser dip into the mercury cups i, the current 
 passes directly to the electro-magnet, and the lever is held sus- 
 
 * Of. Tigerstedt, 'En ny metod for mekanisk retning af nerver' Nordiskt, Medi- 
 cinskt Arkiv, xiii, i88i, Nr. 12. A description of this method will shortly appear in 
 the Zeitschrift fiir Instrumentenkunde. 
 
PEODUCED BY A CONSTANT CURRENT. 17 
 
 pended so that it does not rest on the nerve. If, however, the 
 arms of the reverser are placed in the cups of mercury 2, the current 
 is compelled, before reaching the electro-magnet, to pass through 
 the metronome (M), which by its oscillations opens and closes 
 the circuit to the electro-magnet. Each time the circuit is broken 
 the lever falls on the nerve and stimulates it. When the circuit 
 is closed by the swing of the metronome in the opposite direction, 
 the lever rises again from the nerve. The direction of the current 
 is indicated in the figure by the arrows. 
 
 The polarising current I obtained from a battery of Meidinger's 
 Elements (i?i). First it was carried to a du Bois-Reymond's 
 rheochord (E/i), where it divided, then to a Pohl's reverser, from there 
 to a du Bois-Reymond's key (K), and then to the nerve. The current 
 was closed and broken by means of the key, its strength being 
 graduated in the usual way by means of the rheochord. 
 
 The current was, of course, led through the nerve in the ordinary 
 way by unj^olarisable electrodes. To make them properly suit my 
 apparatus, I gave them a different form from what they usually 
 have ; and I take occasion here to describe them, as they seem to 
 me more practical and handier for stimulation experiments than 
 most of the others in use. 
 
 Each consists simply of a box of ebonite 2^ centims high, the 
 base of which is i centim square. Underneath it is provided with 
 two steel prongs screwed in, by means of which it can be fixed on 
 the cork plate of my lever-apparatus, outside the bed for the nerve. 
 The inside of the trough is filled with paraffin to prevent any 
 contact between the clay and the steel prongs. The clay, having 
 been previously soaked with a 0-6 °/o solution of common salt, 
 is shaped with the hands into long rolls, which are then 
 placed in the trough, with their ends bent forwards. These 
 ends can afterwards have such a form given them as may best 
 suit each particular experiment. In the clay is stuck a glass 
 tube, which is filled with a concentrated solution of sulphate of 
 zinc ; and into this is introduced a well-amalgamated slip of zinc, 
 which is connected by means of a conducting wire with the wires 
 of the battery (Fig. i -Sj). 
 
 In certain experiments, where it was of primary consequence 
 to bring the mechanical exciter quite close to the electrodes, I 
 carried the current to the nerve through a strip of filter-paper 
 soaked in a o-6 ^ solution of common salt, which was attached 
 to the unpolarisable electrodes just described. 
 
18 
 
 CHANGES OF EXCITABILITY IN NERVES 
 
 To keep it moist, the nerve was placed with the tract to be 
 subjected to the mechanical excitation lying" on a strip of filter- 
 paper soaked in a o-6 % solution of common salt. When the 
 intrapolar tract was being excited this strip included nearly the 
 whole distance between the electrodes. By this arrangement, 
 it is true, the branch of the current which passed through the 
 nerve was somewhat w^eakened, but this could be made up for by 
 means of a stronger current. Besides, in this wav not onlv did 
 
 Fig. 2. 
 
 tigerstedt's instrument for mechanical excitation of nerves. 
 [The femur of the ' nerve muscle preparation ' is held by the clamp CI. The nerve 
 lies on the anvil a. Whenever the current through the electro-magnet E is broken, 
 the edge of the hammer L falls on the nerve. The degree of excitation thus produced is 
 regulated by the position given to the sliding weight W. The distance of the seat 
 of excitation from the muscle can be varied by means of the spiral screw S. The whole 
 instrument is enclosed in a glass case so that the preparation is in moist air.] 
 
 I succeed in keeping the nerve moist, but I obtained also a certain 
 measure of security against the sources of error which, in con- 
 sequence of the varying thickness of the nerve, attended Hermann's 
 experiments on the changes of excitability produced by a constant 
 current ^. In some experiments specially indicated, where it was 
 important to have the polarising current as strong as possible, 
 I put no paper under the nerves. 
 
 In those experiments in which paper was used the nerve was 
 never stimulated outside the sheet on which it lay. False results, 
 produced by the different strengths of the current in the rest of 
 
 ' Hermann, Archiv fiir die ges. Physiologic, vii,- 1873, pp. 497-498. 
 
PRODUCED BY A CONSTANT CURRENT. 19 
 
 the nerve, and also in the extremely short piece of nerve lying 
 between the paper and the electrodes, are therefore precluded. 
 
 The stimulation experiments were arranged in the following 
 order. It was ascertained what strength of mechanical stimulus 
 would produce a contraction of a suitable magnitude, that is, a 
 moderately strong contraction with stimulation in the neigh- 
 bourhood of the positive pole, a moderate or weaker one with 
 stimulation near the negative pole. This having been found, the 
 metronome was set in motion, and the muscle had now to trace 
 out about half a dozen contractions on the cylinder of the 
 myograph. Thereupon, the metronome continuing to open and 
 close the current to the electromagnet, the polarising current was 
 at once closed, and the muscle had again to trace out five or six 
 contractions. After this the polarising current was opened, and 
 half a dozen muscular contractions recorded uninfluenced by the 
 polarising current. Following the same order, I continued the 
 experiment for some time, the number of repetitions depending, 
 partly on the endurance of the nerve, partly on the purpose for 
 which the experiment was undertaken. At the same time the 
 strength of the polarising current was sometimes varied. Every 
 experiment relating to change of excitability under the in- 
 fluence of the current, was preceded and followed by an experiment 
 in which the mechanical excitation acted alone. 
 
 Throughout each experiment the mechanical excitation continued 
 to act in the order given to it. By this arrangement I could 
 study the course of the changes of excitability more closely 
 than in any other way at my disposal, as well while the current 
 was closed as when it was open. The metronome interrupted the 
 current to the electromagnet three times in two seconds on the 
 average. I was unable to determine more minutely the relations of 
 excitability during the first moments after the opening and closing 
 of the current. This, however, lay outside my purpose, which 
 was to investigate the persistent changes of excitability during 
 closure of the current and after its discontinuance. 
 
 To show that the nerve completely tolerates mechanical ex- 
 citation when thus conducted, I made a number of experiments 
 with mechanical stimulation alone. These proved that the nerve 
 possesses a much greater toleration of mechanical stimulation than 
 has hitherto been supposed, and greater than my own earlier ex- 
 periments had shown ^. Besides, they showed that the contractions 
 
 * Tigerstedt, Studieniiber mechanische Nervenreizung, i, Helsingfors, 1880, p. 34. 
 
 C 2 
 
20 CHANGES OF KXOITABILTTY IN NERVES 
 
 decline regularly and slowly when mechanical excitation is con- 
 tinued. This proves that mechanical stimulation is in reality quite 
 specially adapted for the study of chang*es of excitability in nerves. 
 In consideration of the importance of this matter, I shall here cite 
 one or two of these experiments. 
 
 Experiment A. Distance from the muscle of the point of stimula- 
 tiou (plexus) 39 mm. ; position of the sliding weight on the arm of the 
 lever 22 ; 3 stimulations in 2 seconds; number of stimulations 420. 
 
 Experiment B. 1. Distance from the muscle of the spot stimulated 
 (plexus) 42 mm. ; position of the sliding weight on the lever arm 25 ; 
 no stimulations in one minute ; number of stimulations 330. 
 
 2. The same nerve. Distance of the spot stimulated from the 
 muscle 26 mm. ; rhythm and strength of the stimuli as before ; 
 number of stimulations 139. 
 
 The muscular contractions were in some experiments recorded on 
 a registering apparatus of Marey's, provided with Foucault's regu- 
 lator ; in others on a recording cylinder constructed by Loven, which 
 last was worked by a water-motor with the greatest regularity. 
 
 I used an elastic recording lever constructed by Lov^n, which 
 magnified the muscular contractions ^'^ times. 
 
 The experiments were continued from October 1881 up to 
 January of this year, and were made on specimens of E-ana 
 esculenta which had been sent in autumn from Berlin. Through- 
 out each experiment the polarising current was constantly in one 
 direction. In these investigations it was important to enquire 
 into the changes of excitability of the nerve evoked by a current 
 with a certain fixed direction. In order that the result should 
 come out as clearly as possible, it was imperative that the direction 
 of the current should remain unaltered for each experiment. 
 Considering how delicate and easily destroyed the nerve substance 
 is, one must use the greatest circumspection to avoid having 
 it more destroyed or injured than is indispensably necessary for 
 the purposes of the experiment. If, however, the direction of a 
 constant current is frequently changed, it may easily happen that 
 the results are disturbed. I am convinced for my part that the 
 numerous irregularities which many enquirers have found, depend 
 to a considerable extent on their having too frequently changed 
 the direction of the current. In every experiment the conditions 
 must be made as simple as possible, and it is better to use several 
 nerves than seek to get everything from one nerve. Such ex- 
 periments may often give no results. 
 
PRODUCED BY A CONSTANT CURKENT. 21 
 
 In the description of the experiments these abbreviations are 
 used. 
 
 + P = the distance of the positive pole from the muscle. 
 — P = the distance of the negative pole from the muscle. 
 E = the distance of the spot stimulated from the muscle. 
 Hh = the number of rlieochord units in the derivation circuit *. 
 7 = the position of the sliding-weight on the levei-. 
 
 Partly to represent the results more clearly, partly to give 
 sufficient proof of the superiority of the mechanical method of 
 stimulation for investigations relating to the general physiology 
 of nerves, I have appended to the description of the experiments 
 reproductions of my curves in fairly large number^. 
 
 1. The extrapolar changes of excitability at the negative 
 pole of an ascending current. 
 
 With weak and moderate currents the excitability is considerably 
 increased. This increase attains its full strength directly after 
 the current is closed. An intensity of the stimulus which shortly 
 before called forth quite feeble muscular contractions or none at 
 all, now evokes very strong ones. The increased excitability 
 is observed during the whole time the current is closed. Never- 
 theless the contractions diminish in course of time, though to a 
 very slight extent, sometimes indeed scarcely appreciably, and in 
 many cases not at all. If the current is opened the contractions 
 resume their previous magnitude. No inhibition showed itself with 
 the conditions under which my experiments were conducted ; on 
 the contrary a slight augmentation of excitability set in after the 
 current was opened. This increase, however, as already pointed 
 out, is trifling and is not always distinctly apparent. It lasts a 
 longer or shorter time and gradually disappears. 
 
 During the closure of the current the increase is greatest at 
 the negative pole, and from thence stretches upwards. The greater 
 the strength of the polarising current, the greater also is the 
 change of excitability, and the more distinctly do its attendant 
 phenomena appear. At the same time it extends over larger 
 tracts of the nerve. 
 
 If the experiment is repeated in such a way that the polarising 
 current is opened and closed in regular succession, while its 
 
 ^ Cf. Hermann, Aichiv fiir die ges. Physiologic, vii, 1873, p. 333. 
 ^ These have been omitted, as it appeared that the experimental results could 
 be understood without them. 
 
22 CHANGES OF EXCITABILITY IN NERVES 
 
 strength with that of the stimulus remains all the time unaltered, 
 the muscular contractions evoked under the influence of the polar- 
 ising* current remain of uniform amount through the whole of 
 the long series. The effect due to exhaustion of the nerve in 
 consequence of stimuli following each other in close succession 
 is by no means so pronounced when the current remains closed 
 as when the nerve is exposed to mechanical stimulation only. 
 
 The increase of excitability occurs also with perfect distinctness 
 even when the current produces make- and break-contractions, 
 I have not made experiments with stronger currents, as no ob- 
 server has questioned that with stimulation in the presence of 
 a powerful ascending current the muscular contractions fall off 
 and disappear. It is only the theoretical explanation of the 
 phenomenon about which there is disagreement. 
 
 Examples. 
 Experiment 1. 
 
 + P = 24; — P = 40; ^ = 44; 3 Meidinger, Uh. = 45, 
 
 The mechanical stimulus by itself generated no contractions. The 
 constant current produced a strong make- but no break-contraction. 
 Under its influence the mechanical stimulus called forth strong 
 contractions of uniform height, which disappeared after the current 
 was opened, but returned in even greater strength when it was again 
 closed. 
 
 Experiment 2. 
 
 -\-I' — 14 ; —P = 30 ; E = 2>S'-> I Meidinger, Eh. = 160 ; y = 30. 
 
 a. The mechanical stimulus evoked only small contractions. The 
 constant current produced a make- but no break-contraction. Under 
 its influence the muscular contractions elicited by the mechanical 
 stimulus at once gained double strength, and during the whole time 
 the current was passing kept at nearly the same height. Wlien the 
 current was opened, the contractions were very slightly stronger at 
 first, but soon declined to their former strength. 
 
 b. If the nerve was now stimulated somewhat nearer to the pole 
 (^=34), other conditions being the same, the mechanical stimulus 
 evoked no muscular contractions, but, under the influence of the current, 
 these showed themselves much stronger than in a, though not of the 
 same height all the time. When the current was interrupted they 
 at once disappeared. 
 
PRODUCED BY A CONSTANT CURRENT. 23 
 
 Experiment 3. 
 
 + P = lo ; —P = 19 ; E —2^; 1 Meidinger, Eh. z=. 450 ; y = 20. 
 
 a. The mechanical stimulus produced no muscular contractions ; 
 the constant current produced a make- but no break-contraction. 
 Under the influence of the current there occurred remarkably large 
 contractions of uniform strength, which instantaneously disappeared 
 when the current was opened, to return when it was again closed. 
 
 h. When the nerve was subjected to the same intensity of me- 
 chanical stimulation somewhat higher up {E = 26), the contractions 
 that took place during the passage of the current were not nearly 
 so strong as in a ; and on the other hand they underwent a slow 
 decrease during each period that the current was closed. 
 
 c. When the polarising current was strengthened (2 Meidinger, 
 Rh. = 1,450), so that contraction took place when it was opened 
 the contractions were notably stronger on stimulation of the spots 
 above mentioned than they previously were, or even than in a. The 
 slow decline of the contractions with continuance of the current can 
 here be clearly recognised. In each experiment of these series, («, 6, c,) 
 however, the contractions elicited during the action of the constant 
 current are always of nearly the same height. Thus in c one can 
 make out no diminution of the conti'actions at the end as compared 
 with those at the beginning, but on the contrary the latter are a little 
 less than the former ; and yet the nerve had been subjected during 
 nearly 160 seconds to at least 240 stimulations. 
 
 2. The extrapolar changes of irritability at the positive pole 
 of an ascending current. 
 
 The excitability of the nerve is considerably diminished. An 
 intensity of the stimulus, which a short time previously elicited very 
 larg-e contractions, now causes quite small ones or none. This 
 diminished excitability lasts all the time the current is closed. In 
 the circumstances under which my experiments were conducted the 
 diminution set in directly the current was closed. At the same time 
 the first contraction was sometimes greater than those following it. 
 Occasionally, on the other hand, while the current continued to flow 
 through the nerve, the muscular contractions increased from their 
 first minimum, only however to a trifling extent and often scarcely 
 perceptibly, in some cases not at all. When the current is opened 
 the muscular contractions return to their original strength, without 
 at the same time any recognisable inhibition exhibiting itself in 
 the first moments. On the contrary, there appears in the nerve a 
 
24 CHANGES OF EXCITABILITY IN NERVES 
 
 slight increase of excitability, which, however, is not very great and 
 in many cases is quite absent. It persists for a longer or shorter 
 time and then gradually disappears. 
 
 During the passage of the current the diminution of excitability 
 is greatest at the positive pole, and falls off as we move from this 
 point lower down. The stronger the polarising current, the greater 
 also is the diminution of excitability, and the greater the nerve- 
 tract over which it extends. 
 
 If the experiment is repeated in such a manner that the polaris- 
 ing current is closed and opened in regular succession, while the 
 strength of the current and that of the stimulus remain all the 
 time constant, the contractions evoked under the influence of the 
 polarising current in most cases get weaker and weaker, while those 
 produced by the mechanical stimulus alone keep throughout at 
 their original height, and do not decline in consequence of ex- 
 haustion of the nerve or muscle nearly so quickly as under ordinary 
 circumstances. 
 
 Examples. 
 Experiment 4. 
 + P = 26; — P = 37; E=()\ 3 Meidinger, ^/i. = 4500 ; 7 = 30. 
 
 a. The mechanical stimulus produces very strong muscular con- 
 tractions : the polarising current both make- and break-contractions. 
 Directly after the current is closed the contractions fall to less than 
 half their previous magnitude, and rise somewhat with the con- 
 tinuance of the current, but to a veiy trifling extent and quite slowly. 
 When the current is opened they return to their former strength, 
 which they retain until it is again closed, when once more a 
 diminution occurs. In the course of the experiment the contractions 
 produced under the influence of the polarising current become feebler 
 and feebler gradually, though not quite regularly, while those evoked 
 by the mechanical stimulus alone retain their original magnitude, 
 and this although the number of stimulations in this experiment 
 amounts to 120. 
 
 h. Under conditions otherwise quite similar, but with a weaker 
 strength of the polarising current (3 Mcidinger, Rib. = 1500), which 
 however still elicits make- and break-contractions, the nerve is 
 stimulated nearer to the positive pole {E = 15). The contractions 
 produced by the mechanical stimulus, which are of the same strength 
 as in a, disappear completely under the influence of the current, and 
 again reach their original height as soon as the current is opened. 
 
PRODUCED BY A CONSTANT CURRENT. 25 
 
 Experiment 5. 
 
 4-P=32-5; — P = 36; E=ig; 3 Meidinger, ^/^. = 500 ; y = 20. 
 
 a. The mechanical stimulus generates moderately strong con- 
 tractions ; the polarising current both make- and break-contractions. 
 Under its influence the muscular contractions of the first two series 
 vanish instantaneously. Later on they no longer vanish when the 
 current is closed, but they become at once quite small and remain so 
 during the whole time the current is passing ; when the current is 
 opened they again reach their former height, but there is no further 
 rise. Even in the last series of all the contractions retain close upon 
 their original magnitude, although the number of stimulations 
 amounts to about 140. 
 
 h. The strength of the polarising current is considerably increased 
 (3 Meidinger, Bh. = 10,000), and the nerve is stimulated somewhat 
 nearer to the muscle {U = 16). Under the influence of the current 
 the contractions completely vanish. 
 
 c. The stimulus is brought still nearer to the muscle (^ := 14), and 
 the strength of the polarising current modified till it coincides with 
 that in experiment a (3 Meidinger, Rh. = 500). No change occurs in 
 the strength of the contractions either when the current is closed or 
 when it is opened. 
 
 3. The intrapolar changes of excitability at the negative pole 
 of an ascending current. 
 
 With weak and moderate currents the excitability of the nerve is 
 considerably heightened. This increase appears in its full strength 
 directly after the current is closed. An intensity of stimulation, which 
 shortly before called forth only slight contractions or none, now pro- 
 duces strong ones. This increased excitability exhibits itself the whole 
 time the current is passing ; it is however noticeable, especially in the 
 case of nerves which have undergone repeated experiments accord- 
 ing to this arrangement, that with the current closed the contrac- 
 tions very slowly fall off; the decrease is, however, unimportant, and 
 does not always occur. As soon as the current is opened the 
 contractions return to their original height. In the circumstances 
 under which I have conducted my experiments, I have not been able 
 to make out any inhibition in the period immediately following. 
 On the contraiy, there is often exhibited a heightening of the 
 excitability, which, however, is never particularly great ; this too can 
 often not be clearly proved. 
 
 During closure it is at the negative pole that the increase of 
 
26 CHANGES OF EXCITABILITY IN NERVES 
 
 excitability is greatest. It increases with the strength of the 
 current till the latter has reached a certain grade. 
 
 If the experiment is pretty frequently repeated the enhancement 
 of excitability remains much the same during each closure of the 
 current, even when the contractions elicited by the mechanical 
 stimulus alone fall off. 
 
 Examples. 
 Experiment 6. 
 
 + P = 20 ; —P = 40 ; E = 37-5 ; 3 Meidinger, Rh. = 92 ; y=iT. 
 
 a. The mechanical stimulus produces weak muscular contractions ; 
 the polarising current only a make-contraction. Under its influence 
 the contractions grow to more than double their size, and during the 
 passage of the current show an exceedingly slight falling off. As soon 
 as the current is opened the contractions fall off, but are still some- 
 what greater than they originally were. The same phenomenon is 
 repeated if the experiment is gone through several times in suc- 
 cession. 
 
 h. The nerve is stimulated somewhat farther up with the same 
 strength of mechanical stimulus {E =■ 38). The polarising current is 
 now stronger (3 Meidinger, Rh. = 150). With this arrangement the 
 contractions rise from a very low point to au equal height with 
 those in a. On opening the current they disappear completely, to 
 reach the same strength as before when the current is closed. In 
 other respects the phenomena of a are here repeated. 
 
 Experiment 7. 
 
 -f P = 10 ; — P = 30 ; E = 2g; I Meidinger, H/i. = 1950 ; y — S. 
 
 The mechanical stimulus produces exceedingly slight contractions ; 
 the polarising current only make-contractions. Under its influence 
 there occur strong, almost maximal, contractions, which keep through- 
 out at the same height, and instantaneously cease when the current is 
 opened. When the current is closed the strong contractions set in 
 afresh. After the nerve has rested for a minute the experiment is 
 repeated from the beginning in the same way. There now appear, 
 after the current has been twice closed and opened, small muscular 
 contractions, showing that the excitability of the nerve has increased. 
 
 Under the influence of the polarising current the same strong 
 contractions as before take place ; they are now however not of so 
 uniform a height as at the beginning of the experiment, but display 
 slight irregularities. 
 
PRODUCED BY A CONSTANT CURRENT. 27 
 
 4. The intrapolar changes of excitability at the positive pole 
 of an ascending current. 
 
 The excitability of the nerve is considerably diminished with all 
 strengths of the current, even with the weakest. Frequently this 
 diminution does not show itself on the instant, but the muscular 
 contractions gradually fall off under the influence of the current, 
 so that the first are somewhat greater than those which follow. 
 This diminished excitability is exhibited all the time the current 
 is closed. It pretty often happens that meanwhile the muscular 
 contractions begin gradually to rise, but this rise is always 
 inconsiderable, and in certain cases is imperceptible. With 
 the current open the contractions instantaneously increase. At 
 the same time no inhibition can be demonstrated. On the contrary 
 there occurs a trifling increase of excitability, which is however 
 sometimes quite unnoticeable. It lasts a longer or shorter time, 
 and then gradually disappears. 
 
 The diminution of excitability when the current is closed is 
 greatest at the pole itself. It increases with the strength of the 
 polarising current. 
 
 If, under the same external circumstances, repeated experiments 
 are made, the diminution of excitability generally increases, though 
 very inconsiderably. Sometimes, too, it may be observed that 
 the diminution falls ofi", but at the same time the contractions 
 evoked by the mechanical stimulus alone have become stronger. 
 In general these maintain their original height or undergo a very 
 slight increase ; now and then it may happen that they fall off", 
 but this is very unusual. 
 
 Examples. 
 Experiment 8. 
 
 + P = 8 ; —P = 34 ; JE = g ; 2 Meidinger, Bk = 50 ; y = 20. 
 
 a. The mechanical stimulua produces weak contractions. The 
 polarising current evokes neither make- nor break-contractions. 
 Under the influence of the current the contractions instantly vanish 
 and reappear only when the current is opened. They are then a 
 trifle weaker. 
 
 b. The same nerve is stimulated somewhat higher up, ^ = 10, with 
 the same strength of mechanical stimulus, and the polarising current 
 is made still weaker (2 Meidinger, Eh. = 42). Even then the 
 contractions cease under its influence and resume their previous 
 strength only when the current is again opened. 
 
28 CHANGES OF EXCITABILITY IN NERVES 
 
 c. The current is still weaker (2 Meidinger, Bh. = 30). The same 
 spots are stimulated as in b, and under the same circumstances. 
 Now the contractions no longer cease, but they fall off perceptibly in 
 strength. On opening the current they regain their former intensity. 
 If it is still further weakened, the current produces no change in the 
 contractions. 
 
 Experiment 9. 
 
 + P = 15 ; -P = 34', U= ly; 2 Meidinger, R/i. = 44 ; y = 30. 
 
 a. The mechanical stimulus excites weak contractions. The 
 polarising current produces neither make- nor break-contractions. 
 Yet under its influence the muscular contractions instantaneously 
 vanish, and reappear in their former magnitude directly after the 
 next break of the current. 
 
 b. The nerve is excited with the same strength of mechanical 
 stimulus, but somewhat further from the pole {E ■=■ 1 9). "With the 
 same intensity of the polarising current the muscular contractions 
 now do not disappear, but they fall off considerably in height. 
 When the current is broken they again reach their former 
 magnitude. 
 
 Experiment 10. 
 
 -|-P = i7; —P^Q^S; E = 20; 3 Meidinger, Bh^ = 5000 ; 7 = 30. 
 
 The mechanical stimulus excites tolerably strong contractions ; 
 the polarising current produces both make- and break-contractions. 
 Under its influence the contractions disappear. Slight traces of them 
 may, however, be remarked during the flrst moments after inter- 
 ruption of the current. As soon as the current is opened the con- 
 tractions reappear and they are then somewhat stronger than they 
 were before closure, but sink by degrees to their original strength. 
 
 Experiment 11. 
 
 -1-P = 15 ; — P = 34 ; ^= 22-5 ; 3 Meidinger, Bh. = 4530 ; y = 30. 
 
 a. The mechanical stimulus excites moderate contractions ; the 
 polarising current make- and break-contractions. Under its influence 
 the contractions decline almost to complete evanescence. The decrease 
 is not so great directly after the current is passed as it is later on. 
 On opening the current the contractions reappear in their original 
 intensity. The current is again closed, and now the muscular 
 contractions comiiletely disappear ; but in the further course of the 
 experiment, the current continuing to pass, minute contractions 
 
PRODUCED BY A CONSTANT CURRENT. 29 
 
 again occur. When the current is opened, these for some time 
 grow more minute. Finally, they again mount so that they overstep 
 their original height not a little. When the polarising current 
 is closed, they do not altogether disappear, but merely decline to a 
 minimum. 
 
 h. The nerve is excited with the same strength of stimulus, but 
 nearer to the positive pole (^ = 21). The muscular contractions are 
 now fairly strong. Under the influence of the polarising current 
 they wholly disappear, but show themselves as large as before so 
 soon as it is again opened. If the experiment is repeated in this 
 way several successive times, the muscular contractions evoked by 
 the mechanical stimulus alone become a little greater than they 
 originally were. 
 
 5. The change of position of the indifference-point with 
 increasing strength of an ascending current. 
 
 In order to investig-ate how the excitability at diiferent points 
 of the intrapolar tract alters with varying strengths of the polaris- 
 ing current, I proceeded in the following way, I first investigated 
 how a very weak current (1) changed the excitability of the nerve 
 at a spot [A) in the neighbourhood of the positive pole. If it 
 appeared that at this spot the excitability was heightened, I raised 
 the strength of the polarising current until the excitability was 
 here visibly lowered by the current (2). Thereupon I examined 
 how the excitability was changed by the current (2) at another 
 spot (-5) higher up. If the excitability there showed itself un- 
 changed or heightened, the strength of the current was again 
 raised till a diminution of the excitability set in. After that I 
 ascertained what relation the excitability at a spot (C) lying still 
 higher up bore to the strength of the current (3) last mentioned. 
 
 If Pfliiger's view that with an increasing strength of current the 
 indifference-point shifts from the positive towards the negative 
 pole represented the real state of the case, the excitability at a 
 point not far fropi the positive pole would be raised by very 
 weak currents ; w^hile, if the strength of the current increased, 
 a diminution of excitability would display itself at this spot, at 
 the same time that another spot lying higher up would still be in 
 the stage of increased excitability ; and so on. 
 
 If therefore, as I gradually raised the strength of the current 
 the seat of stimulation gradually shifted from the positive to 
 the negative pole, I was necessarily in a position to prove that 
 
30 CHANGES OF EXCITABILITY IN NERVES 
 
 the condition of diminished excitability extended more and more 
 over the intrapolar tract of the nerve. Whether the decrease 
 of the contractions which took place with a given strength of 
 current on stimulation of a given spot (A) depended on a hindrance 
 occurring at the positive pole to the conveyance of excitation to 
 the muscle, or on an actual decrease of excitability at this spot, 
 I thought mig'ht be decided by stimulating a spot [JB) higher up, 
 which, supposing the last hypothesis confirmed, would necessarily 
 manifest a heightened or unaltered excitability. The justness of 
 this assumption was established in the most convincing manner. 
 
 With regard then to the changes of excitabilit}'^ in the intrapolar 
 tract, in so far as these depend on the strength of the polarising 
 current, Pfliiger's law holds absolutely. The more the strength 
 of the polarising current is raised, the greater is the portion of 
 the intrapolar tract over which the condition of diminished ex- 
 citability extends, and the shorter is that portion of the tract which 
 exhibits an increase of excitability ; or, in other words, with in- 
 creasing strength of the polarising cm'rent the indifference-point 
 moves from the positive toward the negative pole. With quite 
 weak currents the condition of diminished excitability shows itself 
 only in the immediate neighbourhood of the positive pole, while 
 far the larger portion of the intrapolar tract manifests an increased 
 excitability. With stronger currents, on the other hand, the ex- 
 citability is lowered over the greater part of the intrapolar tract, 
 and only the smaller part exhibits enhanced excitability. 
 
 In consequence of the inhibition which sets in at the positive pole 
 with strong polarising currents, the investigation into the increase 
 of excitability at the negative pole can, of course, not be pursued 
 beyond certain strengths of the polarising current. 
 
 Examples. 
 Experiment 12. +P=ii; —P = ^2; 2 Meidinger ; y = 10. 
 
 a. Eh. ^ 220 ; U = i^. The mechanical stimulus produces weak 
 currents. The constant current excites a make- but no break-con- 
 traction. Under ita influence the magnitude of the contractions 
 elicited by the mechanical stimulus is unaltered. 
 
 b. Rh. =: 220 ; ^ =: 18. Here the mechanical stimulus excites only 
 small contractions, but these increased considerably under the influ- 
 ence of a current of equal strength with that in a. 
 
PRODUCED BY A CONSTANT CURRENT. 31 
 
 c. Rh. = 300. With increased strength of the current, on the 
 other hand, no change is produced in the magnitude of the con- 
 tractions. 
 
 d. Rh. = 400. When the current is still further increased it 
 quite annuls the muscular contractions. Here be it observed that 
 in the experiments h, c, d the same part of the nerve was used 
 (^ == 24). 
 
 e. The nerve is now stimulated somewhat higher up (E = 24). 
 Under the influence of the constant curx-ent, which is now of the 
 same strength as in d, the muscular contractions are, it is true, 
 diminished, but this diminution does not by a long way reach the 
 same amount as in d. 
 
 f. Stimulation is now applied nearer to the negative pole {E = 25). 
 The mechanical stimulus now produces only slight contractions, 
 which under the influence of the same current as in d and e increase 
 by a very little. Throughout the experiment the strength of the 
 mechanical stimulus has been unaltered. 
 
 Experiment 13. -|-P=ii;— P=:33;2 Meidinger. 
 
 a. E r= i^\ Rh. ■= 90. The mechanical stimulus causes only small 
 contractions. The constant current produces a make- but no break- 
 contraction. Under its influence the contractions elicited by the 
 mechanical stimulus vanish forthwith. 
 
 h. ^ = 18 ; Rh. = 90. The mechanical stimulus causes only feeble 
 contractions. These are not modified by a constant current of the 
 same strength as in a. 
 
 c. E = iS ; Rh. = 200. Stimulation is continued at the same spot, 
 but the strength of the polarising current is increased. On closure, 
 the muscular contractions at once vanish, to reappear when the 
 current is opened. 
 
 d. E =: 2g ; Rh. = 200. The mechanical stimulus here excites fairly 
 powerful contractions, which are not in the least altered by the action 
 of a current having the same strength as in c. 
 
 e. E = 16; Rh. := 60. Finally the nerve is stimulated close to the 
 positive pole. The strength of the polarising current is quite in- 
 considerable. When this is closed no change occurs in the strength 
 of the contractions. Throughout the experiment the mechanical 
 stimulus remained of the same strength. 
 
 6. The extrapolar changes of excitability at the negative 
 pole of a descending current. 
 
 The excitability of the nerve is considerably increased. This 
 
32 CHANGES OF EXCITABILITY IX NERVES 
 
 increase is shown in its full amount directly after the current 
 is closed. An intensity of stimulus which previously elicited 
 very slig-ht contractions or none, now causes very strong- ones. 
 This increased excitability lasts the whole time the cu.rrent is 
 passing. At the same time the muscular contractions g-et smaller, 
 although only to a slight extent, and often there is no diminution. 
 On the cessation of the current the contractions decline to their 
 previous size. 
 
 In the circumstances under which my experiments were per- 
 formed no diminution of excitability set in after the current was 
 opened. A slight increase sometimes occurs, but in general the 
 contractions return to their previous height after the current is 
 opened. As long as the current is closed the rise in excitability 
 is greatest at the negative pole, and from there it falls off towards 
 the muscle. 
 
 The stronger the polarising current the greater also is the 
 increase of excitability, and the greater the tract of nerve over 
 which it extends. 
 
 If the experiment is repeated in such a way that closure and 
 opening succeed one another regularly, while the strength of the 
 stimulus and of the current remain constant throughout, the 
 contractions exhibited under the influence of the polarising 
 current remain steadily uniform through the whole series, even 
 though the contractions evoked by the mechanical stimulus alone 
 might gradually fall off through exhaustion of the nerve. 
 
 Examples. 
 Experiment 14. 
 
 -f P = 30 ; —P = 17 ; E = 16; 3 Meidinger, E/i. = 85 ; y = 32. 
 
 The mechanical stimulus occasions moderate contractions : the 
 constant current make- but not break-contractions. Under its 
 influence the contractions evoked by the mechanical stimulus rise 
 to about the maximum. At the same time the first contraction 
 produced after passage of the current is a little stronger than its 
 successors, which undergo an exceedingly slow decline. After the 
 current is opened the contractions return directly to about the same 
 height as they previously had. 
 
 Experiment 15. 
 
 -I-P = 36; — P =24; E = 22; 3 Meidinger, Rh. = 100 ; y = 15. 
 The mechanical stimulus produces very strong muscular con- 
 
PRODUCED BY A CONSTANT CURRENT. 33 
 
 tractions, as also does the constant current, but only when it is 
 closed, and not when it is opened. Under the influence of the constant 
 current the contractions produced by the mechanical stimulus assume 
 their maximum directly, and keep at this strength as long as the 
 current is closed. When it is opened they diminish till they have 
 reached their former grade. Under the influence of the polarising 
 current the contractions remain constantly alike on repetition of the 
 experiment, although the mechanical stimulus by itself elicits even 
 weaker contractions, which gradually fall off, quite according to the 
 law established by experiments on nerve-exliaustion. 
 
 Experiment 16. +P = 40 ; — P = 28 ; 3 Meidinger ; y = 22. 
 
 a. U = 2^; Kh. = 25. The mechanical stimulus occasions no con- 
 tractions ; the polarising current none either. Under its influence, 
 however, weak contractions are caused by the mechanical stimulus, 
 which throughout the closure of the current keep at the same height, 
 and when the current is opened straightway vanish. 
 
 b. ^ = 23 ; Rh. =: 119. The mechanical stimulus brings about no 
 contractions ; the constant current a make- but not a break-contraction. 
 Under its influence the mechanical stimulus causes weak contrac- 
 tions, about the same as in a ; these have a uniform height while 
 the current is closed, and vanish instantaneously when it is opened. 
 
 c. E ■=■ 21. The mechanical stimulus by itself causes no con- 
 tractions. Under the influence of a current of the same strength 
 as in h contractions occur which are somewhat smaller than the 
 former. 
 
 7. The extrapolar changes of excitability at the positive pole 
 of a descending current. 
 
 The excitability of the nerve is considerably lowered. Such an 
 intensity of stimulus as but a short time previous evoked quite 
 strong- muscular contractions, now produces only insignificant con- 
 tractions or none at all. In the circumstances under which I under- 
 took my experiments this diminution showed itself directly after 
 closure of the current ; at the same time the first contraction was 
 sometimes a little greater than its successors. If the current is led 
 through the nerve for some length of time the muscular contractions 
 gradually increase, but this takes place very slowly and some- 
 times scarcely appreciably, in many cases not at all. When the 
 current is opened the contractions sink to their previous height. At 
 the same time no inhibition can be demonstrated. On the contrary 
 a slight increase of excitability takes place, which is yet never of 
 consequence and is often altogether absent. 
 
34 CHANGES OP EXCITABILITY IN NERVES 
 
 While the current is closed, the chang-e of excitability is greatest 
 at the positive pole, and falls off in the direction of the spinal 
 cord. The stronger the polarising- current the greater is the 
 diminution of excitability, and the greater also is the tract of 
 nerve over which it extends. If the experiment is repeated in 
 such a way that the polarising current is closed and opened in 
 regular succession while the strengths of the stimulus and of the 
 current remain the whole time unmodified, the muscular con- 
 tractions in each group become smaller and smaller, while the 
 contractions elicited by the mechanical stimulus alone retain all 
 the time their original height, and do not decline nearly so 
 quickly as they usually do in consequence of nerve exhaustion. 
 
 Examples. 
 Experiment 17. 
 -\-P = ^^l —P = 7 ; E = 41 ; 3 Meidinger, Pk. = 160 ; y = 42. 
 
 The mechanical stimulus occasions strong contractions ; the constant 
 current make- but no break-contractions. Under its influence the 
 contractions drop to half their initial amount. The experiment is 
 repeated time after time, and now the contractions produced under 
 the influence of the constant current grow always slighter, while 
 those occasioned by the mechanical stimulus alone retain the whole 
 time near upon their original height. Finally, they cease completely 
 when the current is passed. In several series it is observed that 
 the first stimulation after closure of the current produces a greater 
 effect than any of its successors, but that, on the other hand, later on, 
 with the continued flow of the current along the nerve, the contrac- 
 tions increase quite slowly. 
 
 Experiment 18. +P = 28 ; — P = 14 ; 3 Meidinger ; y = 20. 
 
 a. E = 2B '1 ^^^- = 4°- The mechanical stimulus produces tolerably 
 strong contractions ; the polarising current make- but no break- 
 contractions. Under its influence the contractions elicited by the 
 mechanical stimulus diminish to half their original strength. As 
 the experiment proceeds the contractions elicited under the influence 
 of the constant current fall off more and more, and at last disappear, 
 while those occasioned by the mechanical stimulus alone retain all 
 the time their initial magnitude. 
 
 b. ^ = 38 ; Eh. = 100. Under the influence of a stronger current 
 than that in a, which yet produces only make-contractions, the 
 contractions caused by the mechanical stimulus fall to half their 
 
PRODUCED BY A CONSTANT CURRENT. 35 
 
 previous size. When the experiment is repeated they finally dis- 
 appear. At the same time the contractions are not instantaneously 
 annulled, but during the first moments after closure of the current 
 small contractions still show themselves, disappearing with continu- 
 ance of the current. 
 
 c. E = ^i ; Eh. = 100. With the same strength of polarising 
 current as in h the diminution of the contractions is far less, the 
 distance of the spot stimulated from the pole being greater. 
 
 d. U = 40; Kh. = 270. An increased strength of current, on the 
 other hand, makes the contractions at once disappear at a spot lying 
 only I mm. nearer the pole. 
 
 e. ^ = 42 ; Rh. = 270. This is also the case 2 mm. higher up. 
 After a little small contractions show themselves for a time, the 
 current remaining closed. 
 
 8. The intrapolar changes of excitability at the negative pole 
 of a descending current. 
 
 The excitability was raised with every strength of current tried, 
 i. e. up to 10 Meiding-er's elements inclusive. This rise instantan- 
 eously assumes its full amount when the current is closed. An 
 intensity of the test-stimulus, which shortly before evoked quite 
 small contractions or none, now produces quite strong ones. This 
 increased excitability continues the whole time the current is 
 closed. Yet it is to be observed that • the contractions gradually 
 decline during closure of the current, especially with nerves on 
 w^hich repeated experiments have been performed with reference to 
 this point. This decrease, however, is in most cases exceedingly 
 slight, and often does not appear at all. When the current is opened 
 the contractions fall to their original height ; when weak currents 
 were opened no diminution of excitability presented itself in the 
 circumstances under which I carried on my experiments. On 
 the other hand, in the first moments after a stronger current is 
 opened, an evident diminution of excitability is exhibited. To this 
 in many cases succeeds a rise of excitability, which is yet never of 
 consequence and often scarcely recognisable. The increase of ex- 
 citability during closure of the current is greatest at the negative 
 pole. It rises along with the strength of the current up to a 
 certain limit. 
 
 If the experiment is frequently repeated, the increase of the 
 contractions remains about the same during each closure of the 
 
 D 3 
 
36 CHANGES OF EXCITABILITY IX NERVES 
 
 current, even when the contractions called forth by the mechanical 
 stimulus alone gradually fell off. 
 
 Examples. 
 Experiment 19. 
 
 + P = 39; — P = 24-5; Ez=2'j; 3 Meidinger, i?/^. = 760 ; 7 = 27. 
 
 The mechanical stimulus by itself occasions no contractions ; the 
 polarising current produces both make- and break-contractions. 
 Under its influence very strong contractions take place, which 
 remain at the same height all the time the current is closed, and 
 disappear when it is opened. 
 
 Experiment 20. 
 
 -f P = 44 ; —P =24; E = 26', 3 Meidinger, E/i. = 132 ; y = 30. 
 
 The mechanical stimulus causes small muscular contractions ; the 
 polarising current make- but not break-contractions. Directly after 
 the current is closed the contractions produced by the mechanical 
 stimulus become very powerful. They keejD at nearly the same height 
 all the time the current is closed ; yet they do gradually fall, though 
 the decrease is very trifling. "When the current is opened they 
 return to their initial magnitude ; sometimes they rise beyond tliis. 
 The diminution caused in the contractions is pretty much the same 
 after repeated experiments, although the contractions produced by 
 the mechanical stimulus alone gradually get weaker. 
 
 Experiment 21. 
 
 -f P = 44 ; -P = 26 ; ^ = 31 ; y = 34 ; 3 Meidinger. 
 
 The mechanical stimulus calls forth strong contractions. The strength 
 of the polarising current is raised after each experiment. With Bh. = 
 90, the reinforcement of the contractions is still quite inconsiderable, 
 though fully apparent. The stronger the current applied, the greater 
 increase is there in the contractions as compared with those elicited 
 by the mechanical stimulus alone. This increase of excitability is 
 exhibited all the time the current is closed ; the contractions occurring 
 under its influence, however, slowly decline from the first one after 
 closure, which is the strongest. After interruption of a current of low 
 intensity they resume their fonner height, but with a certain strength 
 of current {Eh, = 450) and upwards a lowering of excitability sets 
 in directly after the current is opened, at first scarce recognisable, 
 but always more clearly marked as the strength of the current in- 
 creases. At the end of this diminution a moderate increase of 
 excitability takes place. 
 
PRODUCED BY A CONSTANT CURRENT. 33^ 
 
 Experiment 22. +P = 30 ; — P = 16; E = 16; y = S', 
 
 10 Meidinger, JRA. = 20,000 ; no paper under the nerve. 
 
 The mechanical stimulus produces excessively slight contractions ; 
 the constant current only make-contractions. Under its influence the 
 contractions excited by the mechanical stimulus suddenly become of 
 maximal intensity, one might even say tetanic ; after interruption of 
 the current they instantly vanish. In order to apply the mechanical 
 stimulus quite close to the negative pole in this experiment as well 
 as in the two following, I led the current to the nerve through a 
 strip of filter-paper moistened with a o-6 0/0 solution of common salt. 
 
 Experiment 23. +P = 30 ; —P =13-5; ^ = 13 ; y = 12 ; 
 10 Meidinger, Pk. = 20,000 ; no paper under the nerve. 
 
 The mechanical stimulus produces quite small contractions, the 
 constant current make- and break-contractions. Under its influence 
 the contractions evoked by the mechanical stimulus are very strong, 
 almost tetanic. With interruption of the current they again decline, 
 but are still a little greater than they previously were. The same 
 action occurs if the current is once more closed. 
 
 Experiment 24. +P = ^^; — P=i4; ^=14; to Meiding-er, 
 P/l. = 20,000 ', y = ^. 
 
 The mechanical stimulus elicits weak contractions ; the constant 
 current only a make-contraction. Under the influence of the constant 
 current the contractions produced by the mechanical stimulus reach 
 maximal height, and retain this all the time the current is closed. 
 After it is opened there appears first a diminution of the excitability, 
 after which the contractions resume their original height. 
 
 9. The intrapolar changes of excitability at the positive pole 
 of an ascending current. 
 
 The excitability' of the nerve is lowered with all strengths of 
 the current, even with the weakest. This diminution does not 
 assume its full amount all at once, but the contractions fall off 
 gradually under the influence of the current, so that the first con- 
 tractions after contact is made, though considerably smaller than 
 those preceding them, are somewhat greater than their successors. 
 This lowered excitability lasts as long as the current is in action ; 
 yet it happens that the contractions meanwhile gradually increase, 
 though the increase is only trifling and often does not occur at 
 all. If the current is opened the contractions recur forthwith to 
 
38 CHANGES OF EXCITABILITY IN NERVES 
 
 their previous height. No accompanying diminution of the excit- 
 abiKty can be made out. On the contrary, in many cases an in- 
 crease of the contractions sets in, which is yet never great, and 
 occasionally is not at all noticeable. The greatest diminution 
 during closure of the current takes place at the pole itself. It 
 increases with the strength of the polarising current. If the 
 experiment is repeated time after time, other conditions being the 
 same, the diminution of excitability generally increases each time, 
 though only to a very inconsiderable extent. The contractions 
 elicited by the mechanical stimulus alone in most cases remain 
 uniform throughout the experiment. 
 
 Examples. 
 
 Experiment 25. +P = 39 ; — P = 34-5 ; E = ^6 ; 3 Meidinger, 
 B/i. = 1000 ; y = 27. 
 
 The mechanical stimulus causes strong contractions, the polarising 
 current make- and break-contractions. Immediately after contact 
 is made the contractions sink to half their former height and remain 
 thus as long as the nerve is traversed by the current. When the 
 current is opened they regain their previous height. With repeated 
 making and breaking of contact, the contractions produced under 
 the influence of the polarising current decline gradually, whereas 
 the mechanical stimulus by itself evokes contractions of tolerably 
 uniform strength during the whole period. 
 
 Experiment 26. -f P = 44 ; — P = 24 ; ^ = 41 ; 3 Meidinger, 
 R/i. = 132 ; y = 33. 
 
 a. The mechanical stimulus occasions moderately strong contrac- 
 tions ; the polarising current make- but not break-contractions. 
 Under its influence the contractions elicited by the mechanical 
 stimulus disaj)pear ; the moment after closure, however, there is a 
 trace of contraction. As soon as the current is opened the conti'ac- 
 tions exhibit their initial strength, with perhaps a slight increase. 
 
 h. E =■ 39*5. Other conditions being the same, the nerve is 
 stimulated somewhat farther from the pole. If a polarising current 
 is applied of the same strength as in a, the muscular contractions fall 
 off to a considerable extent, but not so much as in a. The first con- 
 traction after the current is thrown in is a trifle gi^eater than its 
 successors. Directly after the current ceases the contractions regain 
 their original strength. 
 
PRODUCED BY A CONSTANT CURRENT. 39 
 
 Experiment 27. 
 
 + P = 34 ; —P = i6 ; 3 Meiding-er, Rh. = 20 ; y = 27. 
 
 a. jE^ = 33. The mechanical stimulus causes moderately strong 
 contractions ; the constant current neither make- nor break-con- 
 tractions. Under its influence, however, the contractions become 
 notably diminished, tliis decrease being less directly after the current 
 is closed than afterwards. After interruption of the current the 
 contractions return to their former strength arid even go a little 
 beyond. 
 
 b. U = 30. Under the influence of the same polarising current 
 the diminution of excitability takes almost the same course as in a ; 
 the increase of diminution attending the continuance of the current is 
 here quite clearly exhibited. 
 
 c. E =■ 28*5. Here the diminution of excitability elicited by the 
 same polarising current is considerably less than in a and h ; all the 
 circumstances attending it, however, make their appearance here just 
 as much as in the foregoing series. 
 
 d. E = 25*5. Here the polarising current causes a scarcely notice- 
 able diminution of the contractions. 
 
 10. The change of position of the indiflference-point with 
 increasing strength of a descending current. 
 
 When I had convinced myself by means of the foregoing' in- 
 vestigations that, up to the strongest polarising currents which 
 were at my disposal, there was in the intrapolar tract a decrease 
 of excitability at the positive, and an increase at the negative pole, 
 it became important to examine in what way different strengths 
 of current aflfected the relation between the length of the intrapolar 
 portion of nerve exhibiting increased excitability, and that of the 
 similar portion exhibiting decreased excitability ; or, in other 
 words, according to what law the indifference-point changed its 
 position with varying strengths of the polarising current. 
 
 For this purpose I adopted the following procedure. I stimu- 
 lated a given spot lying at a certain distance from the negative 
 pole, and examined how the excitability there varied, when, begin- 
 ning with quite weak currents, I progressively increased their 
 strength. When I selected a spot not far from the negative pole, 
 these contractions exhibited in the first place an increase, which 
 became greater and greater the stronger the current used. When 
 however its strength was still further raised, there showed itself 
 instead a considerable diminution of the contractions. That this 
 
40 CHANGES OP EXCITABILITY IN NERVES 
 
 did not depend on inhibition of excitation setting* in at the negative 
 pole itself was proved by the experiments given under 9, in which 
 the nerve was stimulated directly at the negative pole, and where 
 stronger currents were applied than in the case above mentioned. 
 
 With increasing strengths of the polarising current, therefore, 
 the condition of diminished excitability extends over a correspond- 
 ingly increasing" extent of the intrapolar tract, and in the same 
 proportion the part showing increased excitability decreases. In 
 other words, as the strength of the polarising" current increases, the 
 indifference-point moves from the positive toward the negative 
 pole. 
 
 With quite weak currents the condition of diminished excitability 
 is exhibited only in the neighbourhood of the positive pole, while 
 the greater part of the intrapolar tract shows increased excitability. 
 With strong" currents it is only in the immediate neighbourhood 
 of the negative pole that an increased excitabiKty can be demon- 
 strated ; a lowered excitability presents itself over all the rest of 
 the intrapolar tract. 
 
 Examples. 
 
 Experiment 28. +P = 44 ; — P = 33 ; ^ = 27. 
 
 a. 2 Meidinger, i27i. := 19,360. The mechanical stimulus produces 
 weak contractions ; the constant current make- and break-contrac- 
 tions. Under the influence of the current the contractions evoked by 
 the mechanical stimulus increase considerably, and when it is opened 
 they return to their original size. 
 
 6. 4 Meidinger, Rh. •=■ 19,360. The polarising current occasions 
 make- and break-contractions ; under its influence the contractions 
 evoked by the mechanical stimulus show a j)ersistent increase. When 
 the current is interrupted a slight increase of the contractions sets in. 
 
 c. 6 Meidinger, Rh. = 19,360. The polarising current causes make- 
 and break-contractions. Through its action the muscular contrac- 
 tions are completely arrested. Traces of them are visible only for 
 a short time after the current is passed. On opening the current, 
 however, the contractions are for a time greater than before. 
 
 d. 2 Meidinger, Rli. = 360. The polarising current j)roduces make- 
 and break-contractions. The mechanical stimulus, which was of the 
 same strength all through the experiment, now excites no contrac- 
 tions, in consequence of exhaustion of the nerve. Under the influence 
 of the polai'ising current, however, contractions still show themselves, 
 though only slight ones. 
 
PRODUCED BY A CONSTANT CURRENT. 41 
 
 Experiment 29. +P = 30 ; — P = 13 ; U = 16; y — 25. 
 
 a. 3 Meidinger, RTi. = 360. The mechanical stimulus occasions 
 weak muscular contractions ; the constant current only make-con- 
 tractions. Under its influence the contractions notably increase, and 
 sink to their former size when the current is opened. Presently these 
 quite disappear, but the contractions elicited under the influence of 
 the current retain their original strength nearly unaltered. 
 
 h. E = 155 ; 3 Meidinger, Rh.=- 2,360. The nerve is stimulated 
 at a spot only -5 mm. distant from the preceding. The mechanical 
 stimulus here causes rather weak contractions, the polarising current 
 make- and break-contractions. Under its influence there follows a 
 distinct increase of the contractions. After the current is shut off 
 the contractions fall to their original amount. 
 
 c. E = 16-5 ; 3 Meidinger, Rh. = 19,360. If the nerve is stimu- 
 lated I mm. higher up than in h, and therefore -5 mm. higher than in a, 
 while the strength of the current is considerably raised, the latter now 
 exerts an insignificant and indeed hardly perceptible influence in 
 strengthening the contractions. 
 
 d. E = 15-7 ; 6 Meidinger, Rh. — 19,360. The nerve is acted on 
 by the stimulus at a spot in close proximity to that previously 
 excited, and the strength of the current is so far raised as now to 
 cause a make- contraction and only a small break-contraction. In this 
 case the muscular contractions instantly disappear when the current 
 is closed. • 
 
 Experiment 30. -f P = 34 ; — P = 16 ; y = 13. 
 
 a. E = 22; 3 Meidinger, Rh. = 340. The mechanical stimulus 
 produces rather weak muscular contractions, the constant current 
 make- but no break-contractions. Under its influence the con- 
 tractions elicited by the mechanical stimulus increase very consider- 
 ably. When the current is opened they decline to their original 
 height. 
 
 h. E=22; 3 Meidinger, Rh. = 2,000. The polarising current 
 brings about make- and break-contractions. Under its influence a 
 very considerable increase occurs in the contractions produced by the 
 mechanical stimulus, though by itself this calls forth only slight ones 
 in consequence of exhaustion of the nerve. 
 
 c. E = 22; 3 Meidinger, Rh. = 20,000. With this greater strength 
 of the polarising current the contractions increase still further than 
 in a and b. When the current is opened there is a transitory rise in 
 the height of the contractions as compared with that which they 
 attained before contact was made. 
 
42 CHANGES OF EXCITABILITY IN NERVES 
 
 d. E ■=■ 21-8; 4 Meidinger, Rh. = 20,000. The nerve is stimulated 
 • 2 mm. nearer the negative pole. The strength of the polarising 
 current is considerably raised, but still produces both make- and break- 
 contractions, though the latter are feebler than in h and c. Under 
 the influence of the current the muscular contractions vanish 
 instantaneously. With interruption of the current they recur, and 
 then they exhibit a distinct but gradually declining rise as compared 
 with their original height. 
 
 e. ^ = 21-8; 4 Meidinger, Rh. = 1,880. With diminished strength 
 of current, which, however, still produces both make- and break-con- 
 tractions, there again occurs under the influence of the current a rise 
 in the height of the muscular contractions. 
 
 These investigations have therefore sufficiently confirmed the 
 results at which Pfliig-er arrived by means of electrical and chemical 
 stimulation relative to the chang-es generated in nerve by a constant 
 current. If a constant current is passing through an uninjured 
 nerve the following laws hold good for all the modes of excitation 
 at present available. 
 
 The excitability of the nerve is augmented at the negative pole, as 
 well outside as inside, in whatever direction the current ^x^sses along 
 the nerve. 
 
 The excitaViUty of the nerve is diminished at the positive pole, as 
 well outside as inside the pole, in whatever direction the current passes 
 along the nerve. 
 
 At both poles the change of excitability, whether inside or outside, 
 increases with the strength of the polarising current until, hi con- 
 sequence of that change, the muscular contraction has either attained 
 its w.aximum or is completely annulled. Further changes in the ex- 
 citability of the nerve can, of course, not be proved by an alteration 
 in the magnitude of the muscular contractions. 
 
 It is at the two poles that the change of excitability is greatest, and, 
 from each of these it diminishes in both directions, i.e. both outwards 
 and inwards. 
 
 At each pole the change of excitability extends outside the pole over 
 tracts which are greater the stronger the polarising current. 
 
 In the intrapolar tract the indifference-point shifts from the neigh- 
 bourhood of the positive pole toioards the negative pole with increasing 
 
PRODUCED BY A CONSTANT CURRENT. 43 
 
 strength of the polaiising current and that qiiite indej^endently of the 
 direction in which the current flows along the nerve. 
 
 N.B. An exception to the above occurs in the case of the diminution 
 and final cessation of the muscular contractions when with an ascend- 
 ing current the nerve is stimulated in the region of the negative pole. 
 For the explanation of this phenomenon see further on. 
 
 How the change of excitability is developed during" the first 
 moments after contact is made, and in what manner it declines after 
 interruption of the current, is more difficult to demonstrate than 
 are the persistent changes of excitability of the nerve. The former 
 changes are of a more fugitive nature than the latter, and for 
 their demonstration special experimental arrangements are requisite. 
 The proper aim of my investigation was not to enquire into this, 
 but its chief object was to examine the persistent changes of 
 excitability which establish themselves while the current is closed. 
 
 My investigations have, however, yielded results relating to the 
 duration of the change which confirm those previously arrived at on 
 this subject. Thus in general the following statements hold good 
 as well for mechanical as for other kinds of stimulation processes, 
 when the nerve is traversed by a constant current. 
 
 At the negative pole the change of excitability on closure appears 
 immediately in its full strength. 
 
 At the positive pole the change of excitahility on closure attains 
 its maximum only hy degrees. 
 
 At the negative pole, after interruption of the current, supposing the 
 current not tootoeaJc, there is exhibited at first a diminution of excit- 
 ability, tohich is afterwards followed by an increase. 
 
 At the positive pole, on opening, a rise of excitability occurs 
 directly. 
 
 Here it should be observed that both the increase of excitability 
 occurring at the negative pole after the current is opened, and the 
 diminution of excitability perceptible there at the very first, are 
 not nearly so sharply marJced with mechanical stimulation as with 
 electrical, and often cannot be made out at all. 
 
 The following statements hold good for the changes of excitability 
 when the current is in action for several seconds. 
 
 At the negative pole in many cases the muscular contractions decline 
 after having first reached a maximimi. This decrease is for the most 
 part exceedingly slight, and proceeds very slowly. 
 
44 CHANGES OP EXCITABILITY IN NERVES 
 
 At the positive pole the muscular contractions in many cases rise 
 from the minimum to which they had at first sunk, Itut this rise is 
 exceedingly trifling and very slowly takes place. 
 
 These phenomena admit of but one explanation, namely, that the 
 strength of the polarising current, when it has traversed the nerve 
 for a certain time, is lowered by internal polarisation of the nerve ^. 
 When the change of excitability produced by the polarising current 
 is so great that the excitation of the nerve becomes in conse- 
 quence stronger than is required to evoke maximal contraction, or 
 weaker than the minimal strength required to produce a response, 
 the diminution in strength of the polarising current produced by 
 polarisation is of course unobservable. Accordingly the phenomena 
 in question are not always exhibited, only appearing w^hen the 
 excitation of the nerve resulting from the change of excitability has 
 either not risen above the strength required to generate maximal 
 contractions, or not fallen below the strength necessary to call 
 forth minimal contraction. 
 
 These phenomena do not present themselves with very minute 
 strengths of the polarising current, presumably because in that case 
 the internal polarisation cannot rise to an observable degree. 
 
 Finally, the following statements hold good with respect to the 
 repeated action of a polarising current passing constantly in the 
 same direction along the nerve. 
 
 The muscular contractions originating at the negative pole remain for 
 a long time at nearly the same height, even when, for other reasons, the 
 contractions elicited by the test-stimulus would have fallen off. 
 
 The muscular contractions originating at the positive p^ole become 
 gradually smaller, even when those produced by the test-stimulus alone 
 remain throughout at nearly the same height. 
 
 In the foregoing exposition I have spoken of the increased and 
 diminished excitability of the nerve in accordance with the faulty 
 mode of expression in general use, instead of keeping closely to 
 the actual results, and speaking of the occurrence under particular 
 circumstances of an increase or decrease in the size of the muscular 
 contractions. 
 
 If the view propounded by Hermann is right, the diminution 
 or increase of the muscular contractions does not depend on a 
 
 ' See du Bois-Reymond, Gesammelte Abhandlungen, II. p. 171. 
 
PRODUCED BY A CONSTANT CURRENT. 45 
 
 local change of excitability in the nerve, but on the decrease or 
 increase in strength of the excitation during its propagation along 
 the nerve to the muscle. 
 
 It still remains then to examine whose theory, Pfliiger's or 
 Hermann's, best agrees with the facts, so far as they are at 
 present known to us, relating to the changes in extent of the 
 muscular contractions while a constant current is passing along 
 the nerve. 
 
 In this connection special weight must be assigned to the above 
 recorded experiments on the character of the muscular contractions 
 when with an ascending current the nerve was stimulated close to the 
 positive pole and inside it, and to those where with a descending 
 current it was stimulated quite close to the negative pole. 
 
 These experiments have shown that, if the stimulus is made to 
 operate near enough to the pole, the muscular contractions decrease 
 in the former case under the influence of the polarising current, 
 however weak it may be, and increase in the latter case however 
 strong the polarising current. 
 
 According to Hermann's theory, the negative pole must be sup- 
 posed to exert an inhibitory action on the propagation of the excita- 
 tion, otherwise it could not explain the reason of the diminution and 
 cessation of contraction when with a strong ascending current the 
 nerve is stimulated above the negative pole. Here an experimen- 
 tum crucis seems possible. If this phenomenon depends on inhibi- 
 tion, not at the positive, but at the negative pole, then plainly the 
 effect of such inhibition should be to weaken and finally to annul the 
 muscular contractions when the nerve is stimulated inside the pole 
 with the current descending, provided that the intensity of the 
 current is sufficiently great. This, however, is not the case. If 
 with a descending current the nerve is stimulated sufficiently near 
 the negative pole, the muscular contractions do not disa2:»pear, even 
 when a polarising current of lo Meidinger's elements is applied. 
 They increase, on the contrary, and this to such a degree, that a 
 strength of stimulus which before closure evoked np contractions 
 at all, is now, under the influence of the current, able to evoke 
 maximal contractions. 
 
 Not only therefore does the negative pole exert no inhibi- 
 tory action on the propagation of a stimulus to the muscle, but, 
 quite the reverse, the excitability of the nerve is very considerably 
 increased. 
 
 On the other hand, a much smaller strength of current is sufficient 
 
46 CHANGES OF EXCITABILITY IN NERVES. 
 
 completely to arrest the action of the stimulus when exercised in 
 the region of the positive pole. 
 
 Hermann's theory, therefore, cannot explain all the phenomena 
 which attend the changes produced by a constant current in the 
 magnitude of the muscular contractions. They may, however, be 
 explained by Pfliiger's theoretical \iew in accordance with which 
 the cause of the changes in size of the muscular contractions is 
 to be sought i/i the local changes of exc'dahility produced hy the 
 current hi the nerve. Until new facts are discovered which neces- 
 sitate a modification of this \-iew, it may be accepted. 
 
 No such fact has presented itself in the foregoing investiga- 
 tion. 
 
II. 
 
 CONTRIBUTION TO THE THEORY OF 
 THE BEEAE-CONTEACTION. 
 
 By Dk, ROBERT TIGEESTEDT. 
 
 Communicated to the Swedish Royal Academy of Sciences, September 13, 1882. 
 
II. 
 
 Through the indefatigable labours of du Bois-Reymond, of 
 Pfliiger, and of several other observers^ abundant material has been 
 accumulated, consisting of carefully-ascertained facts relative to 
 the exciting action of the electric current on nerves. These facts 
 may be summed up very briefly in the following propositions, 
 (i) It is not to the absolute value of the current density at any 
 moment that the motor nerve responds by contraction of its muscle, 
 but to the change of this value from one moment to the next, in 
 such sense that the excitation to movement following these changes 
 is the more considerable, the more rapidly they have proceeded 
 when of equal amount, or the greater their amount in a unit of 
 time (Law of du Bois-Reymond^). (2) When the current is closed, 
 it excites at its negative pole ; when it is opened, the nerve is 
 stimulated at the positive pole ^. (3) While the current is closed, 
 the excitability of the nerve is weakened in the neighbourhood of 
 the positive pole, strengthened in the neighbourhood of the nega- 
 tive pole 2. (4) In the first moments after opening of the current, 
 the excitability of the nerve is strengthened at the positive pole, 
 and weakened at the negative pole^. The greater part of these 
 facts Pfliiger summed up in his well-known principle, 'A given tract 
 of nerve is stimnlated hy the appearance of katelectrotonus and the 
 disapjoearance of anelectroto7ins, not however hy the disapjiearance of 
 katelectrotonus nor by the ap2)earance of anelectrotonus ^.' He further 
 
 * Du Bois-Eeymond, Untersuchungen liber thierische Elektricitat, Berlin, 1848, 
 pp. 258-259. (The original paper appeared in 1845.) 
 
 ^ Pfliiger, Untersuchungen tiber die Physiologic des Electrotonus, Berlin, 1859; 
 V. Bezold, Untersuchungen iiber die electrische Erregung der Nerven und Muskeln, 
 Leipzig, 1 86 1. 
 
 ^ Pfliiger, \oc. cit. p. 456. 
 
 E 
 
50 CONTRIBUTION TO THE 
 
 showed that from this principle the law of contraction as he had 
 found it could be deduced with the greatest ease. 
 
 The fact, that when the current is closed, excitation occurs at 
 the negative pole, connects itself with a number of purely physical 
 phenomena. Thus, for example, in the induction spark the electric 
 light shows itself only at the negative pole. In vacuo, the 
 negative pole is entirely surrounded with a beautiful violet light, 
 while the positive pole emits only a luminous crown of purple- 
 coloured rays. The temperature of the violet is higher than that 
 of the purple light. In the voltaic arc the light constantly appears 
 first at the negative pole, while the positive is yet dark ; presently 
 this too gets warmed up, small particles begin to pass over, and so 
 on (Neef, Ruhmkorff, Despretz, Moigno ^). 
 
 We have therefore analogies derived from other sources, which 
 teach that, at least in some cases, the electric current acts most 
 powerfully at the negative pole. There are also facts which might 
 serve as analogies for the changes of excitability occurring at both 
 poles during the closure of a current. Thus Matteucci ^, and after 
 him Fleischl ^, found that, when nerves are treated with weak 
 alkaline solutions (-08 o/^), their excitability rises, and that, on treat- 
 ment with weak acids (-075 ojo), it falls. Now we know that the 
 current gives off acids at its positive pole and alkalies at its nega- 
 tive. We have here, therefore, a complete agreement between the 
 changes of excitability elicited by the constant current and those 
 generated by the corresponding electrolytes. To these facts we may 
 add an old observation of Gustav Crusell's, who found that the or- 
 ganic tissues are coagulated at the positive pole, probably through the 
 formation of acids, and that the coagulum formed is again liquefied 
 by the negative pole, probably through the alkalies liberated there, 
 or in other words, the positive pole makes the structure of organic 
 tissues firmer, while the negative pole exerts the contrary action"*. 
 
 * Daguin, Traite ^l^mentaire de Physique, iii, Paris and Toulouse, 1879, PP- 
 
 472-473- 
 
 ^ Matteucci, Comptes Eendus de TAcad^mie des Sciences, torn. 65, p. 155, 1S67. 
 
 * Fleischl, Archiv f. Anatomic und Physiologic, 1882, p. 17. 
 
 * G. Crusell, Uber den Galvanismus als chemisches Heilmittel gegen ortliche 
 Krankheiten, St. Petersburg, 1841, p. 14 et seq. 
 
 Among Crusell's experiments the following may be given here : ' If a galvanic 
 current from a small compound battery is led into the eye of a living animal by means 
 of a needle introduced through the cornea, after the current has been in operation 
 half a minute, a whole minute, or longer, according to the strength of the battery, 
 there occurs, often at once, at latest after some days, a clouding of the pupil, some- 
 times associated with a partial coagulation of the aqueous humour. If the galvanic 
 
THEORY OF THE BREAK-CONTRA.CTION. 51 
 
 Analogy, however, does not necessarily imply identity. Though 
 we may point to phenomena in other provinces similar to those 
 with which we have become acquainted in studying the action 
 of the current on nerves, we cannot regard the two as equivalent ; 
 nor form by their aid any sufficient and complete theory of the 
 phenomena which occur at the moment that the current is closed 
 and during the period of its flow along a nerve. Yet these analogies 
 and others of a like kind are perhaps not without value, since at 
 any rate they suggest a possibility, however slight, of getting 
 beyond the bare facts of experiment. 
 
 It is quite otherwise with the break-excitation and with the 
 other phenomena attending opening of the current. We can, so 
 far as I know^ point to no similar facts in other provinces of know- 
 ledge, and Pfiiiger^s famous formula, that the disappearance of 
 anelectrotonus stimulates, is merely a very precise and exact word- 
 ing of the facts. It states, however, nothing more, and advances 
 us not one step towards understanding the essence of the pheno- 
 menon. Even at the earliest period in the history of electro- 
 physiology, the break-contraction had arrested the attention of 
 enquirers; and even Volta attempted to find for it a theoretical 
 explanation. Neither he nor his followers, however, succeeded in 
 presenting a theory which satisfied the requirements of science; 
 and his views, as well as those of Pfaff, Lehot, Erman, Marianini, 
 and the earlier ones of Matteucci, can now claim merely an historical 
 interest \ 
 
 Peltier, whose views appear to be more important than these, 
 explains the break-contraction in direct contraction as due to 
 the polarisation of the muscle, expressing himself thus : ' On doit a 
 M. Ritter' — he says — 'puis a M. De La Rive la connaissance de ce 
 fait ; qu'un arc metallique, formant un circuit hydro^lectrique par 
 
 current is led from the eye in the same way, there occurs at once a copious disengage- 
 ment of gaseous bubbles, and, if the eye has been previously exposed to the action of 
 the in-going current, there occurs usually a manifest diminution of the clouding and 
 of the coagulum, both of which nearly disappear in a few weeks, especially if the 
 experiment is accompanied with only a slight amount of inflammation ' (pp. 11-19). 
 
 I repeated the experiment in the following way. I used a battery of 5 or 6 
 Meidinger's elements. When I applied the positive pole to the cornea of a frog 
 which had just been killed, and the negative to any part of the trunk, after a few 
 moments there appeared a distinct clouding of the cornea. When I changed the 
 direction of the current, so that the negative pole rested on the cornea, there was at 
 once a copious disengagement of minute bubbles of gas ; these being removed, the 
 cornea appeared almost as clear and transparent as at first. 
 
 ^ See the complete account by du Bois-Reymond, Untersuchungen uber thierische 
 Elektricitat, Berlin, 1848, pp. 307-402. 
 
 E 2 
 
52 CONTRIBUTION TO THE 
 
 son immersion dans deux liquides separes, devient un couple vol- 
 taique et produit un courant en sens inverse d'autant plus energique 
 que le metal est plus inalterable. On avait pense d'abord qu'une 
 polarisation moleculaire de tout Tare etait la cause de ce contre- 
 courant ; mais il a ete reconnu depuis qu'il n'y avait que les bouts 
 immerges qui jouissaient de cette faculte, et qu'elle etait due a une 
 couche d'oxigene au pole positif, et une d''hydrogene au pole nega- 
 tif. La meme cause produit les memes efFets sur la grenouille ; la 
 patte positive se charge d'oxigene, et la negative d'hydrogene. Le 
 contre-courant se demontre en plongeant les pattes dans deux tasses 
 ou aboutit le fil d'un galvanometre tres-sensible. . . . Plus la pile 
 ^sera forte, plus le temps du courant sera long, plus les pattes seront 
 chargees. . . . Ce qui est vrai pour une grenouille entiere Test 
 encore pour un muscle, pour une portion de muscle,' &c. ^ 
 
 Du Bois-Reymond, who established the correctness of Peltier's 
 facts b}' investigations of his own 2, urged against his explanation of 
 the break-contraction that it was not easy to see how it could be 
 thus explained, since these electrical charges of Peltier would seem 
 to require a complete circuit to produce a current, a condition lost in 
 the very act of interruption ^. It may however be said, in answer 
 to this objection of du Bois-Eeymond, that the equalisation of the 
 differences of tension which exist in the muscle at the moment 
 when the external circuit is broken gives rise to internal currents. 
 
 All this holds only for the break-contraction elicited by direct 
 excitation of the muscle, but may however be applied to the break- 
 contraction resulting from stimulation of the nerve, as to all appear- 
 ance it is identical with the other. 
 
 Peltier's view was taken up by Matteucci, but carried no further. 
 He contented himself with pointing out that the break-contraction 
 
 * 'We owe to M. Hitter,' he says, ' and after him to M. De La Rive, the knowledge 
 of this fact, that a metallic arc, forming a hydroelectric circuit by its immersion in 
 two separate liquids, becomes a voltaic couple and produces a current in the reverse 
 direction, which is the more energetic the more unalterable the metal. It was at 
 first thought that a molecular polarisation of the whole arc was the cause of this 
 counter-current ; but it has since been recognised that it is only the immersed ends 
 which possess this property, which is due to a layer of oxygen at the positive pole and 
 to one of hydrogen at the negative pole. The same cause produces the same effect in 
 the frog ; the positive foot gets charged with oxygen, the negative with hydrogen. 
 The reverse current shows itself when the feet are plunged into two cups, in which 
 
 the wires of a very sensitive galvanometer terminate The stronger the batter}' 
 
 and the longer the duration of the current, the more highly charged the feet become. 
 .... What is true of a whole frog is true of a muscle, of part of a muscle,' &c. 
 
 ^ Du Bols-Reymond, loc. cit. i. pp. 377-382. 
 
 * Ibid. loc. cit. p. 381. 
 
THEORY OF THE BREAK-CONTRACTION. 53 
 
 could be explained by the susceptibility of the nerve to internal 
 polarisation, without making- the slightest attempt to show how all 
 the phenomena characterising" and attending the break-contraction 
 could be deduced from it^. 
 
 Moreover, the internal polarisability of nerves had not been studied 
 sufficiently to admit of a true parallel being- drawn between it and 
 the break-contraction. It is easy to explain, therefore, why this 
 statement has received no further attention from physiolog-ists. 
 
 Then came Pfliiger's investigations, which have to so great an ex- 
 tent impressed the nerve physiology of to-day. Pfliiger's conception 
 of the break-contraction has since then been accepted almost without 
 objection, although, as already explained, it supplies no explanation 
 of the phenomena, but only an elegant statement of their laws, 
 
 A remarkable theory was put forward about the same time by 
 Chauveau. According to him the physiological action of electricity 
 is 'the result of a mechanical disturbance communicated to the 
 molecules placed in the path of the current 2.' 
 
 The constant current in general excites only by means of ' an 
 accessory instantaneous current of high tension, with which it 
 commences ; ' and similarly by means of ' an accessory instantaneous 
 current of high tension, with which it terminates^.' Of these 
 currents the first flows in the same direction as the stimulation- 
 current, and the second in the opposite direction. The action of 
 the last is assisted by the polarisation-current^. 
 
 In later times, so far as I know, only H. Munk has attempted to 
 advance a real theory of the break-contraction. His theoretical 
 view starts from the phenomenon of Porret, and is thus presented 
 by the author himself : ' The excitation of the nerve is due in the 
 first place to the direct displacement of the nerve fluid in the 
 direction of the current, and similarly, by the return of the fluid, 
 when the current is opened, to those parts of the nerve whence it 
 had come. Inasmuch as these movements of the nerve-fluid extend 
 over the extrapolar tract to the muscle, the muscular contraction is 
 produced by the motion of the nerve-fluid — independently of its 
 direction — provided that the motion is propagated to the muscle 
 with sufficient intensity and velocity^.' 
 
 Munk therefore would regard both the make- and the break- 
 contraction as caused by a mechanical excitation. It is, however, 
 
 ' Matteucci, Comptes Renclus, torn. 65, 1867. 
 
 ^ Chauveau, Journal de la Physiologie, iii. No. ix. p. 52, January, i860. 
 ^ Ibid. loc. cit. pp. 61-66. * Ibid. loc. cit. p. 70. 
 
 ' H. Munk, Archiv fiir Anatomic, Physiologie, &c., 1866, p. 383. 
 
54 CONTRIBUTION TO TEE 
 
 impossible to conceive that the displacing power of a very weak 
 current, such as is just able to produce a muscular contraction, 
 would be sufficiently great and could occur with sufficient sudden- 
 ness to exert such an action. Moreover, the facts made out by 
 Munk in this connection relative to the changes of excitability pro- 
 duced by a constant current, which were to have supported his 
 conception, do not agree with the results of foregoing and succeeding 
 investigations. Besides, he has never published the complete account 
 of his theory. We need therefore give it no further consideration. 
 
 II. 
 
 Certain phenomena, which I observed in the course of my inves- 
 tigations upon the changes of excitability produced in nerves 
 by the constant current, led me to enquire whether the internal 
 polarisation of nerves might not be the true cause of the break- 
 contraction. Accordingly, I devoted myself in the first place 
 to a thorough investigation of the internal polarisation of nerves, 
 and soon found that it affi)rded in a remarkable degree the con- 
 ditions required for the production of a break-contraction. 
 
 The chief laws of the internal polarisation of nerves are these : — 
 T. With a current whose strength is not greater than that pro- 
 duced by three Meidinger's elements polarisation is. as near as may 
 be, directly proportional to the strength of the current. 
 
 2. If the polarising current acts on the nerve for unequal and 
 increasing lengths of time, other conditions being the same, polari- 
 sation increases ; it rises more rapidly at first, and afterwards more 
 slowly, finally approaching its maximum with extreme slowness. 
 
 3. When the polarising current is opened, polarisation instantly 
 attains its highest value, and after this declines continuously. This 
 decline proceeds at first very rapidly, afterwards however more 
 and more slowly, so that polarisation is still present for a long time 
 after the opening of the polarising current, and approaches the zero 
 point only asymptotically. 
 
 The two chief conditions on which the strength of polarisa- 
 tion depends are identical with those which determine the break- 
 contraction, as to which it has long been known that it is in 
 the main a function of the strength and duration of the po- 
 larising current. But the third law of polarisation also tells in 
 the occurrence of the break-contraction in a very remarkable way. 
 Since polarisation lasts for an appreciable time after interruption of 
 the polarising current, the break- contraction, if it has once occurred 
 
THEORY OF THE BREAK-CONTRACTION. 55 
 
 on application of a strong current A, must also occur with a weaker 
 durrent JB, if onl}^ this current is closed soon enoug-h after the 
 stronger current is opened; For, to the polarisation produced by 
 the current £, which of itself is not sufficient to cause a break- 
 contraction, there is added thd polarisation remaining over from the 
 current A, and the sum may now be strong enough to liberate a 
 contraction. Further, under such circumstances, the break-con- 
 traction must the more certainly appear, and must be the greater 
 in extent the sooner the current B is closed after the opening 
 of^. 
 
 These phenomena are very easily observed in excitation experi- 
 ments. The description of them by Biedermann is in accordance with 
 the preceding considerations. He found in fact that ' directly after 
 a break-contraction liberated by a strong current, iveak currents, pre- 
 viously only effective at make, exercise an excitatory ^action at break 
 which is almost as intense as that of the strong current. This excita- 
 tion-effect diminishes, hozvever, in a short time, and disappears completely 
 after a feiv minutes of rest if the nerve is sufficiently vigorous. 3f ore- 
 over, under conditions othenoise the same, this piecuUar effect is the more 
 persistent the greater the period of time during which the stronger current 
 has operated, and the less active the nerve is ^.' 
 
 The full parallelism between the break-contraction and polarisa- 
 tion may be very elegantly demonstrated with one and the 
 same strength of current on an ordinary nerve-muscle prepara- 
 tion. I performed such experiments in this way. I passed a 
 current of a given strength through the nerve for periods of 
 various duration. If the period was very short, only a make-contrac- 
 tion occurred ; if it was longer, there was also a break-contraction. 
 If, directly after the break-contraction appeared, I again closed 
 the current and opened it again very soon, repeating the experiment 
 in this way for several successive times, I obtained a whole series of 
 break-contractions, which by-and-bye fell off in size, and at last 
 quite ceased, to reappear as soon as I allowed the current to act on 
 the nerve for a longer time. In these experiments therefore there 
 appeared in the break-contractions all the phenomena which I had 
 previously become acquainted with by means of the galvanometer. 
 In order to throw more light on the question I shall here com- 
 municate some examples of experiments. 
 
 In the following records, T^ = the duration of the closures ; 
 
 1 Biedermann, Wiener Sitzungsberichte Math. Part iii. vol. 83, p. 328, 1881. 
 
56 
 
 CONTRIBUTION TO THE 
 
 T- = the intervals during* which the current remained open ; C = 
 the size of the break-contraction. 
 
 The muscle traced its contractions by means of a spring lever on 
 the cylinder of Marcy's reg^istering" apparatus, which moved with 
 the g-reatest regularity. They were magnified about ^'^ times. The 
 electrodes used were, of course, unpolarisable. 
 
 Experiment I. — 2 Meidingers, 500 rheochord-units in the deri- 
 vation circuit ; current ascending. 
 
 No. 
 
 T\ 
 
 J12 
 II 
 
 II 
 
 No. 
 
 II 
 
 II 
 
 C. 
 II 
 
 No. 
 
 TK 
 
 rp2 
 II 
 
 C. 
 
 1 
 
 0-5 
 
 0.4 
 
 
 
 9 
 
 6.4 
 
 o.S 
 
 3-5 
 
 17 
 
 0.4 
 
 0-5 
 
 
 
 2 
 
 0.4 
 
 0-5 
 
 
 
 10 
 
 I-O 
 
 0-7 
 
 2.1 
 
 18 
 
 1.9 
 
 0.7 
 
 
 
 3 
 
 0-4 
 
 0.4 
 
 
 
 11 
 
 0-6 
 
 0-6 
 
 2-2 
 
 19 
 
 5-0 
 
 0.7 
 
 2.9 
 
 4 
 
 0.4 
 
 0.4 
 
 
 
 12 
 
 0-5 
 
 0-6 
 
 2-0 
 
 20 
 
 0-6 
 
 0.6 
 
 2.9 
 
 5 
 
 1.8 
 
 0.7 
 
 
 
 13 
 
 0-5 
 
 0.6 
 
 
 
 21 
 
 0-5 
 
 0-5 
 
 
 
 6 
 
 3-5 
 
 0.7 
 
 
 
 14 
 
 0-5 
 
 0-6 
 
 
 
 22 
 
 0-5 
 
 05 
 
 
 
 7 
 
 3 
 
 0.6 
 
 
 
 15 
 
 0-5 
 
 0-5 
 
 
 
 23 
 
 0-5 
 
 0-5 
 
 
 
 8 
 
 5-4 
 
 
 
 
 16 
 
 0-4 
 
 
 
 
 24 
 
 7.2 
 
 
 5-5 
 
 0.7 
 
 0-5 
 
 Experiment II. — 2 Meidingers, rheochord 800 ; ascending 
 current. 
 
 
 The \ under column C denotes that . 
 
 . . . is 
 
 just noticeable 
 
 
 
 No. 
 
 T\ 
 
 T\ 
 
 C. 
 
 No. 
 
 Ti. 
 
 T-. 
 
 C. 
 
 No. 
 
 yi. 
 
 T\ 
 
 C. 
 
 1 
 
 0-3 
 
 " 
 
 
 
 14 
 
 0-3 
 
 " 
 
 II 
 1.2 
 
 27 
 
 II 
 
 0-5 
 
 " 
 
 2-(> 
 
 2 
 
 0-2 
 
 0-3 
 
 
 
 15 
 
 0-3 
 
 0.4 
 
 2.2 
 
 28 
 
 0.4 
 
 0-5 
 
 1-9 
 
 3 
 
 0-2 
 
 0-3 
 
 
 
 16 
 
 0.2 
 
 0.4 
 
 i-i 
 
 29 
 
 0-3 
 
 0-5 
 
 1.9 
 
 4 
 
 0-2 
 
 0-3 
 
 
 
 17 
 
 0.2 
 
 3-0 
 
 
 
 30 
 
 0-3 
 
 0-5 
 
 2.2 
 
 5 
 
 0-2 
 
 0-3 
 
 
 
 18 
 
 0-2 
 
 0.6 
 
 
 
 31 
 
 0-2 
 
 0-5 
 
 1.8 
 
 6 
 
 0-2 
 
 0-3 
 
 
 
 19 
 
 0-3 
 
 0.6 
 
 
 
 32 
 
 0-3 
 
 0-5 
 
 0-5 
 
 7 
 8 
 
 1-6 
 
 5-1 
 
 0-3 
 06 
 
 t 
 
 2-8 
 
 20 
 21 
 
 0.2 
 0-2 
 
 0.6 
 0-6 
 
 
 
 
 
 33 
 34 
 
 0.2 
 0.2 
 
 0-4 
 0-5 
 
 0-5 
 + 
 
 9 
 
 0-5 
 
 0.7 
 
 2 
 
 22 
 
 0.? 
 
 0.6 
 
 
 
 35 
 
 0.2 
 
 0-5 
 
 t 
 
 10 
 
 0.4 
 
 0-5 
 
 2.6 
 
 23 
 
 1-7 
 
 0.6 
 
 
 
 36 
 
 0.2 
 
 0-5 
 
 09 
 
 11 
 12 
 
 0-3 
 0-3 
 
 0-5 
 0.4 
 
 2-3 
 
 1.8 
 
 24 
 25 
 
 3-9 
 0.4 
 
 0.7 
 0.6 
 
 2-3 
 
 1.8 
 
 37 
 38 
 
 CI 
 
 0-3 
 
 2-5 
 0-9 
 
 
 0.9 
 
 13 
 
 0-3 
 
 0-4 
 0.4 
 
 1-5 
 
 26 
 
 0.4 
 
 0-5 
 0-3 
 
 15 
 
 39 
 
 0-2 
 
 4-4 
 0.8 
 
 0.9 
 
THEORY OF THE BREAK-CONTKACTION. 
 
 57 
 
 No. 
 
 TK 
 
 40 
 
 0-3 
 
 41 
 
 0.4 
 
 42 
 
 0-3 
 
 mi 
 
 0-4 
 
 C. 
 o 
 
 o 
 
 t 
 
 iV^o. T'. 
 
 43 o.'s 
 
 44 0-3 
 
 45 0-3 
 
 0.4 
 0-4 
 0-4 
 
 C. 
 It 
 
 0.7 
 
 t 
 o 
 
 46 I '6 
 
 47 0-9 
 
 rp2 
 0.7 
 
 c. 
 
 i-o 
 
 Experiment III. — 2 Meidingers, rheochord 500 ; ascending current. 
 
 The t under column C denotes that .... is just noticeable. 
 
 No. 
 
 II 
 
 IJ}-2, 
 
 II 
 
 C. 
 II 
 
 No. 
 
 2". 
 II 
 
 Ji2 
 
 c. 
 
 No. 
 
 TK 
 
 /p2 
 
 C. 
 
 1 
 
 0-3 
 
 0-5 
 
 
 
 12 
 
 0-2 
 
 0-5 
 
 t 
 
 23 
 
 0-2 
 
 0-4 
 
 0-5 
 
 2 
 
 0'3 
 
 o-S 
 
 
 
 13 
 
 0-2 
 
 0.4 
 
 t 
 
 24 
 
 0-2 
 
 0-4 
 
 0.4 
 
 3 
 
 0-2 
 
 0-5 
 
 
 
 14 
 
 0-2 
 
 0-5 
 
 
 
 25 
 
 0-2 
 
 0-4 
 
 0-5 
 
 4 
 
 0-2 
 
 0-5 
 
 
 
 15 
 
 0-2 
 
 0-5 
 
 t 
 
 26 
 
 0-2 
 
 0-4 
 
 t 
 
 5 
 
 1.9 
 
 
 t 
 
 16 
 
 0-2 
 
 
 
 
 27 
 
 0-2 
 
 
 
 
 6 
 
 6.8 
 
 0-8 
 0.6 
 
 1.8 
 
 17 
 
 0-2 
 
 0.4 
 
 0-5 
 
 
 
 28 
 
 0-2 
 
 0.4 
 0-4 
 
 
 
 7 
 
 0-3 
 
 0-5 
 
 
 
 18 
 
 0-2 
 
 0-5 
 
 
 
 29 
 
 0-2 
 
 0-4 
 
 t 
 
 8 
 
 0-2 
 
 0-5 
 
 
 
 19 
 
 0-2 
 
 0-5 
 
 
 
 30 
 
 0.2 
 
 0-4 
 
 
 
 9 
 
 0-2 
 
 
 
 
 20 
 
 II.6 
 
 
 7-1 
 
 31 
 
 0-2 
 
 
 
 
 10 
 
 14.9 
 
 0-6 
 0-8 
 
 5-0 
 
 21 
 
 0-3 
 
 0.6 
 0-4 
 
 2.7 
 
 32 
 
 0-2 
 
 0.4 
 0-4 
 
 
 
 11 
 
 0-3 
 
 0.6 
 
 1-5 
 
 22 
 
 0.2 
 
 0-4. 
 
 0.9 
 
 33 
 
 0.2 
 
 • ^ 
 
 
 
 Experiment IV.— 2 Meidingers, rheochord 1000; ascending current. 
 
 
 l 
 
 he t un 
 
 der CO 
 
 umn C 
 
 denote 
 
 3 that . 
 
 ... is just 
 
 noticeable. 
 
 
 No. 
 
 TK 
 
 ip2 
 
 c. 
 
 II 
 
 No. 
 
 II 
 
 y2_ 
 
 C. 
 
 No. 
 
 T'. 
 
 m2 
 II 
 
 C. 
 
 1 
 
 0.4 
 
 0.4 
 
 
 
 12 
 
 0-2 
 
 0.4 
 
 I-O 
 
 23 
 
 0-3 
 
 0-5 
 
 t 
 
 2 
 
 0-4 
 
 0-4 
 
 
 
 13 
 
 0-2 
 
 0.4 
 
 t 
 
 24 
 
 0-2 
 
 0-5 
 
 t 
 
 3 
 
 0-3 
 
 0-4 
 
 
 
 14 
 
 02 
 
 0.4 
 
 t 
 
 25 
 
 0-2 
 
 0-5 
 
 t 
 
 4 
 
 0-3 
 
 0-4 
 
 
 
 15 
 
 0-3 
 
 0-4 
 
 t 
 
 26 
 
 I-I 
 
 08 
 
 t 
 
 5 
 
 0-3 
 
 0.4 
 
 
 
 16 
 
 0-2 
 
 0-4 
 
 + 
 
 27 
 
 0.7 
 
 0.8 
 
 t 
 
 6 
 
 0-3 
 
 0.4 
 
 
 
 17 
 
 2.7 
 
 0.7 
 
 2-2 
 
 28 
 
 I-O 
 
 0-8 
 
 i-i 
 
 7 
 
 0-3 
 
 0.4 
 
 
 
 18 
 
 0.6 
 
 06 
 
 1.4 
 
 29 
 
 13.2 
 
 0.7 
 
 5-4 
 
 8 
 
 0-3 
 
 0-4 
 
 
 
 19 
 
 0-5 
 
 0.4 
 
 1-3 
 
 30 
 
 0-2 
 
 0-4 
 
 1-6 
 
 9 
 
 3-4 
 
 0.8 
 
 1-0 
 
 20 
 
 0-5 
 
 0.6 
 
 0.6 
 
 31 
 
 0-2 
 
 0-5 
 
 II 
 
 10 
 
 10.9 
 
 0.6 
 
 2.6 
 
 21 
 
 0.6 
 
 0-5 
 
 1. 1 
 
 32 
 
 0-2 
 
 ^ 
 
 0-^ 
 
 t 
 
 11 
 
 0-3 
 
 0.4 
 
 I-O 
 
 22 
 
 0-5 
 
 0-5 
 
 0.8 
 
 33 
 
 O-I 
 
 u 
 
 t 
 
58 
 
 CONTRIBUTION TO THE 
 
 Experiment V. — 2 Meidingers, rheochord 1970 ; ascending 
 current. 
 
 
 The t under col 
 
 amn C denotes that . 
 
 . . . ii 
 
 > just noticeable 
 
 
 
 No. 
 
 T\ 
 
 T'. 
 
 C. 
 // 
 
 No. 
 
 TK 
 
 
 C. 
 
 No. 
 
 II 
 
 7-2 
 
 C. 
 
 1 
 
 0.4 
 
 0-5 
 
 
 
 12 
 
 0-3 
 
 0.4 
 
 
 
 23 
 
 0-3 
 
 0-4 
 
 t 
 
 2 
 
 0-3 
 
 0-5 
 
 
 
 13 
 
 0'3 
 
 0.4 
 
 
 
 24 
 
 0-3 
 
 0-4 
 
 I-O 
 
 3 
 
 0-3 
 
 0-5 
 
 
 
 14 
 
 0-3 
 
 0-4 
 
 
 
 25 
 
 0-3 
 
 0-5 
 
 I-O 
 
 4 
 
 1-4 
 
 0.8 
 
 
 
 15 
 
 8-3 
 
 0.9 
 
 3-4 
 
 26 
 
 0-3 
 
 0-4 
 
 T.4 
 
 5 
 
 3-2 
 
 I-O 
 
 i.S 
 
 16 
 
 0-5 
 
 0-5 
 
 51 
 
 27 
 
 0-3 
 
 0-4 
 
 
 
 6 
 
 I-O 
 
 0.7 
 
 2-2 
 
 17 
 
 0-4 
 
 0-5 
 
 3-2 
 
 28 
 
 0-3 
 
 0-5 
 
 t 
 
 7 
 
 0-3 
 
 0.7 
 
 2-2 
 
 18 
 
 0.4 
 
 0-5 
 
 2.8 
 
 29 
 
 0.2 
 
 0-5 
 
 t 
 
 8 
 
 0.4 
 
 4-7 
 
 1-3 
 
 19 
 
 0.4 
 
 0-5 
 
 1.6 
 
 30 
 
 0-2 
 
 0-5 
 
 t 
 
 9 
 
 0-3 
 
 0-5 
 
 
 
 20 
 
 0-3 
 
 0-5 
 
 1-4 
 
 31 
 
 0-2 
 
 0-5 
 
 t 
 
 10 
 
 0-3 
 
 0-5 
 
 
 
 21 
 
 0-3 
 
 0.4 
 
 1-4 
 
 32 
 
 0-2 
 
 0-5 
 
 1-7 
 
 11 
 
 0-3 
 
 0.4 
 
 
 
 22 
 
 0-3 
 
 0.4 
 
 1-3 
 
 33 
 34 
 
 0-2 
 0-2 
 
 0-5 
 
 t 
 t 
 
 We have thus seen that, as regards the fundamental phenomena, 
 there is a complete parallel hetween polarisation and the break- 
 contraction. But characteristics of secondary importance also show 
 remarkable agreement. 
 
 Biedermann^ found that when nerves are treated with an alcoholic 
 solution of common salt, not only was the break-contraction stronger, 
 but also that it made its appearance with a strength of current 
 which before was quite inadequate to produce it. I also have found 
 that, when the nerve is treated with an alcoholic solution of common 
 salt, its susceptibility to polarisation is increased by half, as the 
 following experiments show. They were conducted after pre- 
 cisely the same method as my earlier experiments on internal polari- 
 sation. I used a o"6 "/, solution of common salt mixed with 10 °/o of 
 alcohol, and took the greatest care that the contact should be the 
 same both before and after the treatment with alcohol. 
 
 * Biedermann, loc. cit. p. 321. The retardation of the break-contraction discovered 
 by Biedermann at the same time does not for the present concern us. 
 
THEORY OF THE BREAK-CONTRACTION. 
 
 59 
 
 Experiment VI. — 3 Meidingers, rheochord 1000 ; duration of 
 closure 5" ; distance between the electrodes S^^. 
 
 I. Before application of the alcohol: — 
 
 No. 
 1 
 2 
 3 
 4 
 5 
 
 Mean 
 
 Deflection. 
 
 2-8 
 
 2-95 
 3 
 
 2-95 
 3 
 
 Mean variation 0-064. 
 2. After application of the alcohol for i minute 
 
 2-94 
 
 No. 
 
 6 
 
 7 
 
 8 
 
 9 
 10 
 
 Mean variation o-ig2. 
 
 Mean 
 
 Deflection. 
 
 3-8 
 4.4 
 
 4-35 
 4-45 
 4.2 
 
 4.24 
 
 Experiment VII. — 3 Meidingers, rheochord 1000 ; duration of 
 closure 5"; distance between the electrodes 20™™. 
 
 Before application of the alcohol : — 
 
 No. 
 1 
 2 
 3 
 4 
 5 
 
 Mean 
 
 Deflection. 
 3-r 
 3-2 
 3-1 
 3-2 
 3-4 
 
 3-2 
 
 Mean variation 0-08. 
 2. After application of the alcohol for i minute :- 
 
 No. 
 
 
 Deflection 
 
 6 
 
 
 ... 4-8 
 
 7 
 
 
 4-9 
 
 8 
 
 
 ^•o 
 
 9 
 
 
 4-7 
 
 10 
 
 
 5.0 
 
 
 Mean 
 
 4-88 
 
 Mean variation 0'094. 
 
60 
 
 CONTRIBUTION TO THE 
 
 Experiment VIII. — 3 Meiding-ers, rheochord 1000 ; duration of 
 closure 5"; distance between the electrodes 8°^". 
 
 I. Before application of the alcohol : — 
 
 No. 
 1 
 
 2 
 3 
 4 
 5 
 
 Mean variation 0-144. 
 
 Mean 
 
 Deflection. 
 
 5-2 
 5-2 
 5-2 
 
 4.9 
 
 47 
 
 2. After application of the alcohol 
 No. 
 
 7 
 
 8 
 
 9 
 
 10 
 
 Mean 
 
 Deflection. 
 7.2 
 6.7 
 6-8 
 6.8 
 6-6 
 
 6.82 
 
 Mean variation 0'i53. 
 
 Here, again, the changes in polarisability are parallel with the 
 occurrence of break - contractions. The same phenomena that 
 Biedermann observed in the muscular contractions after treatment 
 of the nerve with alcohol present themselves when we examine 
 with the galvanometer the polarisation of the nerve. 
 
 In a paper which well deserves attention Griitzner has recently 
 considered, amongst other questions, the occurrence of the break- 
 contraction when different parts of a nerve are stimulated ^. "VVe 
 shall, for the present, pass over Griitzner's conclusions relative to 
 the significance of the ' nerve current-' in the production of the 
 break-contraction, and only note that in his experiments, when 
 the plexus was stimulated, the break-contraction occurred as well 
 with an ascending as with a descending current, if the slider of his 
 rheochord (he employed a rheochord with one wire, of the same 
 construction as that used for compensation experiments) stood at 
 from 80 to 140™°^ from the zero point. If, however, the nerve was 
 stimulated in the middle, below where the branches to the thigh 
 are given off, above its separation into the two end divisions, and 
 in the same way if it was stimulated near the knee, in most cases 
 
 ^ Griitzner, Arcliiv fiir die gesammte Physiologie, vol. 28, pp. 130-178, 18S2. 
 
THEOKY OF THE BEEAK-CONTRACTION. 
 
 61 
 
 both make- and break-contractions failed to appear even when the 
 slider stood at looo™"^ from the zero point ^. 
 
 If the break-contraction is determined by a polarisation of the 
 nerve, then, ceteris paribus, the susceptibility of the plexus to 
 polarisation must be greater than that of the lower tracts of the 
 nerve. 
 
 And this is actually the case. In experiments on this question 
 the two sciatics of one and the same frog* were fastened together, 
 and different tracts of the nerve submitted to the polarising current. 
 As the same electrodes and the same parts of them were used in all 
 the experiments, and as at the same time the distance between 
 them was constant, it is clear that the differences in the extent 
 of the deflections, which were produced wheu different parts of the 
 nerve were experimented on, are entirely dependent on local con- 
 ditions. In other words, the polarisability of a nerve differs at 
 different points of its course. What this depends on does not affect 
 the question before us, which is whether any parallelism exists 
 between polarisability and Griitzner's results. I proceed to give 
 some examples of experiments. I examined the j)olarisation of 
 each nerve preparation at three different places, namely, (i) at the 
 plexus, (2) where the branches to the thigh were given off, (3) in 
 the lowest part of the nerve. 
 
 Experiment IX. — 3 Meidingers, rheochord 1000 ; duration of 
 closure 5''; distance between the electrodes 8™™. 
 
 I. 
 
 No. 
 1 
 2 
 3 
 4 
 5 
 
 Mean variation 0-056. 
 
 Mean 
 
 II. 
 
 No. 
 6 
 
 7 
 
 10 
 
 Mean 
 
 Deflection. 
 
 2-6 
 
 2-95 
 3-0 
 2-95 
 3-0 
 
 2.94 
 
 Deflection. 
 
 !•! 
 
 JC-25 
 1.25 
 I-20 
 
 Mean variation o-o^ 
 
 ^ Griitzner, loc. cit. pp. 163-165. 
 
 1. 16 
 
62 
 
 CONTRIBUTION TO THE 
 
 III. 
 
 No. 
 
 11 
 
 12 
 
 13 
 
 14 
 
 15 
 
 Mean variation 0*05: 
 
 Mean 
 
 Deflection. 
 
 1-05 
 
 I-I5 
 
 I-O 
 
 I-O 
 
 i-i 
 
 i-o6 
 
 Experiment X. — 3 Meiding-ers, rheoeliord 1000; duration of 
 closure 5"; distance between the electrodes lo""™. 
 
 No. 
 1 
 2 
 3 
 4 
 5 
 
 Mean variation O'l 36. 
 
 No. 
 
 6 
 
 7 
 
 8 
 
 9 
 10 
 
 Mean variation o-o8. 
 
 No. 
 
 11 
 
 12 
 
 13 
 
 14 
 
 15 
 
 Mean variation 0-073. 
 
 Mean 
 II. 
 
 Mean 
 III. 
 
 Mean 
 
 Defection. 
 4.8 
 
 5-25 
 
 5-1 
 
 5-15 
 
 5.12 
 
 Deflection. 
 3-1 
 3-3 
 3-25 
 3-2 
 3-4 
 
 3-25 
 
 Deflection. 
 
 I-I5 
 1-40 
 
 1-35 
 1-40 
 
 1-35 
 1-33 
 
 Experiment XI. — 3 Meidingers, rheochord 1000; duration of 
 closure 5"; distance between the electrodes 10"^™. 
 
 I. 
 
 Deflection. 
 
 No. 
 1 
 
 4-7 
 4.9 
 
 4.9 
 
 4.9 
 
 5-0 
 
 Mean variation 0-072. 
 
 Mean 
 
THEORY OF THE BREAK-CONTRACTION. 
 
 63 
 
 II. 
 
 No. 
 
 6 
 
 7 
 
 8 
 
 9 
 10 
 
 Mean variation 0-144. 
 
 Mean 
 
 III. 
 
 No. 
 11 
 12 
 13 
 14 
 15 
 
 Mean variation 0-024. 
 
 Mean 
 
 Deflection. 
 3-45 
 3-7 
 4.0 
 
 3-65 
 3-8 
 
 3-72 
 
 Deflection. 
 
 2-1 
 
 2.15 
 2.1 
 
 2-1 
 2-15 
 
 2-12 
 
 Experiment XII. — 3 Meidingers, rheochord looo; duration of 
 closure 5''. 
 
 I. 
 
 No. 
 
 
 Deflection. 
 
 1 
 
 
 4-35 
 
 2 
 
 
 4-35 
 
 3 
 
 4 
 
 5 
 
 
 4-35 
 
 4-2 
 4-2 
 
 
 Mean 
 
 4-29 
 
 ation 0-073. 
 
 II. 
 
 
 No. 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 
 Deflection. 
 2.4 
 2.25 
 
 2-5 
 
 2.5 
 2.4 
 
 Mean variation 0-073. 
 
 Mean 
 
 III. 
 
 No. 
 11 
 12 
 13 
 14 
 15 
 
 2.41 
 
 Deflection. 
 
 I-3 
 
 1-45 
 
 1.50 
 
 1-75 
 1-75 
 
 Mean variation o-i6o. 
 
 Mean 
 
 1-55 
 
64 CONTRIBUTION TO THE 
 
 These experiments may suffice to show the agreement between 
 Griitzner's results and the polarisability of different tracts of the 
 nerve. On what this difference of polarisability depends does not, 
 as I have already pointed out, concern the question with which we 
 are now occupied. It is probably due, mainly at least, to the 
 greater conductivity of the thicker parts of the nerve. I have not 
 attempted to determine whether other circumstances also exert an 
 important influence. 
 
 We have found, therefore, a most remarkable parallelism between 
 the break-contraction and polarisation, not merely in the funda- 
 mental phenomena, but also in those of secondary importance. I 
 may therefore venture to regard polarisation as the cause of the 
 break-contraction, which, accordingly ^ is neither less nor more than 
 a viake-contraction dejwndent on the existence of a polarisation current. 
 
 It is self-evident that the absolute magnitude of the polarisation- 
 current cannot be regarded as an objection to this theory. The 
 exciting action of a current is conditional, not on its strength alone, 
 but also on the abruptness of its commencement. It is for this 
 reason, for example, that the effect of an induction-shock on the 
 nerve is so much greater than that of the constant current. The 
 current of polarisation, already in existence during the passage of 
 the polarising current, must come into action when the latter is 
 opened, with an abruptness almost equal to that of an induction- 
 shock, and so produce an excitation in spite of the smallness of its 
 absolute strength. From this quarter, therefore, so far as I can 
 see, no objection can be made to my theory. When the circuit of 
 the polarising current is broken, that of the polarisation current is 
 closed through the physiologically inactive nerve-sheaths, indepen- 
 dently of any external derivation. In order to test still further 
 this theoretical view, as well as its working capabilities, I shall 
 now enquire whether we can obtain a simple and unstrained ex- 
 planation of all the well-established phenomena presenting themselves 
 on interruption of the polarising current. 
 
 III. 
 
 The fundamental conception in my theory of the break-contraction 
 is the reference of all the phenomena appearing on interruption of 
 a current to those which are exhibited on its closure. This once 
 made good, it then only remains to explain scientifically the make- 
 contraction and its attendant phenomena, t© enable us to construct 
 a satisfactory theory of the action of the electric current on nerve. 
 
THEORY OF THE BREAK-CONTRACTION. 
 
 65 
 
 In short,, we have now to examine whether^ in reality, all the facts 
 we find when the current is broken are really in accordance with 
 those which present themselves when it is closed. 
 
 Just as the polarising- current produces its exciting effect on 
 closure at the negative pole, so the seat of stimulation at break is 
 at the positive pole, which now becomes the negative pole of the 
 polarisation-current. 
 
 And, just as the positive pole of the polarising current exerts a 
 blocking action, so at break there is a tendency to blocking at the 
 negative pole, for this is now the positive pole of the polarisation- 
 current. Starting from this, we can with the greatest ease establish 
 theoretically the law of contraction. 
 
 Most physiologists of the present day are agreed in expressing 
 the law of contraction in its most general form after Pfliiger's well- 
 known scheme. 
 
 LAW OF CONTRACTION 1. 
 
 Strength of current. 
 
 
 Ascending 
 current. 
 
 Descending current. 
 
 Weak current | 
 
 Make : — 
 Break : — 
 
 Contraction 
 Rest 
 
 Contraction 
 Rest 
 
 Moderate current j 
 
 Make : — 
 Break : — 
 
 Contraction 
 Contraction 
 
 Contraction 
 Contraction 
 
 Strong current •< 
 
 Make : — 
 Break : — 
 
 Rest 
 Contraction 
 
 Contraction 
 Weak contraction 
 or rest 
 
 If my interpretation of the make-contraction is correct, the same 
 phenomena must occur when a descending current is broken as 
 when an ascending- current is closed, and, conversely, we shall 
 encounter the same phenomena when an ascending current is 
 broken as when a descending current is closed. The first stage 
 of Pfliiger's law will be the only exception, as the current has 
 not yet attained the intensity required to produce polarisation 
 of sufficient strength. The following scheme offers a survey of 
 these relations, the first stage being omitted for reasons already 
 given. 
 
 ' Pfliiger, Untersuchungen iiber die Physiologic des Electrotonus, Berlin, 1859, 
 P- 454- 
 
 F 
 
66 
 
 CONTRIBUTION TO THE 
 I. 
 
 Strength of current. 
 
 Break of a 
 descending current. 
 
 Make of an 
 ascending current. 
 
 Moderate current 
 
 Contraction 
 
 Contraction 
 
 Oi . Rest or weak 
 btron^ current , , ■ 
 ° contraction 
 
 Rest 
 
 II. 
 
 Strength of current. 
 
 Break of an 
 ascending current. 
 
 Make of a 
 descending current. 
 
 Moderate current 
 
 Contraction 
 
 Contraction 
 
 Strong current 
 
 Contraction 
 
 Contraction 
 
 We see, therefore, that the agreement is as complete as the 
 theory requires. 
 
 In order now to deduce the law of contraction from my stand- 
 point, I assume that the relation between the strength of stimulation 
 and the tendency to bloclving- is quite the same in the polarisation- 
 current as in other currents ; and that therefore, according* to 
 Pfliiger's view, with a certain strength of this current, the stimu- 
 lation overpowers the tendency to blocking, while with a greater 
 strength the latter may become sufficiently great to prevent the 
 excitation from reaching the muscle notwithstanding its greater 
 strength. 
 
 The currents described as weak in Pfliiger's law of contraction 
 are, therefore, naturally those which are not strong enough to evoke 
 a sufficient degree of polarisation in the nerve, and which are conse- 
 quently unaccompanied by a break-contraction. 
 
 If the strength of the polarising current is above this limit, then 
 on breaking a polarisation-current is produced of sufficient intensity 
 to liberate a muscular contraction. It is true that when the 
 polarising current is descending the excitation produced by the 
 polarisation-current has of course to pass its own positive pole, 
 which exerts a blocking action. This blocking action, however, is 
 
THEORY OF THE BEEAK- CONTRACTION. 67 
 
 as yet too weak to prevent the excitation being- propag-ated to the 
 muscle^ just as with an ascending- polarising current of weak or 
 moderate strength the closing excitation at the negative pole is not 
 prevented from reaching the muscle. 
 
 Strong currents are characterised by this, that the blocking 
 which takes place at the positive pole overpowers the excitation 
 proceeding from the negative pole. When the current is descending 
 this is of no consequence ; but if, on the other hand^ the current is 
 ascending, the blocking at the positive pole is sufficient to stop the 
 excitation occurring at the negative pole. 
 
 If the polarising current is ascending, the polarisation -current is 
 descending. However strong it may be made by strengthening 
 the polarising current, it must always evoke a contraction^ since 
 there is nothing to hinder the excitation occurring at its negative 
 pole from being propagated to the muscle. 
 
 It is otherwise if the polarising current is descending. In that 
 case the polarisation-current is ascending, and the excitation has 
 to pass the positive pole. The stronger the polarising current is, 
 the stronger also do the polarisation-current and the block at its 
 positive pole become. Yet, in order completely to prevent the 
 excitation from reaching the muscle, the strength of polarisation 
 must be considerable. Experience, moreover, teaches that the 
 descending break-contraction is far from disappearing with a 
 strength of current which is sufficient completely to do away with 
 the ascending make-contraction. It may even happen, as Pfliiger 
 expressly stated in his original scheme of the law of contraction, 
 that the descending break-contraction does not disappear at all. 
 The reason, of course, is that the polarisation-current is not 
 sufficiently strong. 
 
 The polarisation remaining over from previous stimulations has 
 also in all likelihood much to do with the disappearance of the 
 descending break-contraction. In experiments on Pfliiger's law of 
 contraction, as is well-known, we soon find that the ascending make- 
 contraction disappears. It only remains then to bring about the 
 disappearance of the descending break-contraction to demonstrate 
 the whole law. If we now gradually increase the strength of the 
 polarising current, we need no longer keep changing its direction, 
 but may content ourselves with testing its action when descending, 
 since we have already become acquainted with the whole law of 
 contraction so far as concerns an ascending current. Every such 
 stimulation, however, with a descending cui'rent leaves behind it a 
 
 F 2 
 
68 CONTKIBUTION TO THE 
 
 polarisation-current flowino^ in the opposite direction ; tlie sum of 
 all these residual polarisation-currents make in the end a polarisa- 
 tion-current sufficiently" strong to block the descending break- 
 excitation. 
 
 My theory explains therefore, in a very simple way, all the facts 
 formulated in the law of contraction, and makes it directly intel- 
 ligible why Hhe disappearance of anelectrotonus is a much weaker 
 stimulus than the appearance of hatelectrotonus^ since the polarisation- 
 current is always weaker than the polarising, and both stimulate 
 the nerve in a like manner, namely, at their negative pole. 
 
 If we bear in mind that the break-contraction arises through the 
 passage of a current (the polarisation-current), we can quite simply 
 formulate the law of contraction. " We require in fact only two 
 strengths of current : — 
 
 1. Weak currents, in which the excitation at the negative pole 
 overpowers the blocking at the positive, and 
 
 2. Strong currents, in which the excitation at the negative pole 
 is overpowered by the blocking at the positive. 
 
 These two cases comprise the whole law of contraction, if we only 
 remember at the same time that every polarising current produces 
 polarisation, and also, that every current, to liberate an excitation, 
 must be of a certain strength. 
 
 During the closure of a current the excitability of the nerve is 
 lowered in the I'egion of the positive pole, both on the extrapolar 
 and on the intrapolar side, and it is raised in the region of the 
 negative pole. 
 
 Immediately after breaking, that is while the polarisation-current 
 is still of sufficient strength, the excitability is seen to be diminished 
 at the negative pole (the positive pole of the polarisation-current) 
 and strengthened at the positive pole (the negative pole of the 
 polarisation-current). Like the polarisation-current itself, the 
 changes in excitability on breaking the polarising current are very 
 evanescent. The agreement of the facts with the theory could not 
 be more striking than it is. 
 
 After the stage of diminished excitability at the positive pole of 
 the polarisation-current has passed off, a rise of excitability shows 
 itself for a certain time throughout the whole extent of the nerve ; 
 and this phenomenon must be regarded as the real and only after- 
 effect of the electric current. 
 
 Moreover, the circumstance that, in experiments on living animals 
 and on nerves unseparated from their connections, only make-con- 
 
THEORY OF THE BREAK-CONTRACTION. 69 
 
 tractions occur, is easily explained by my theory. Since the current 
 can with the greatest ease branch off into all possible paths, its 
 density in the nerve is so small that the polarisation cannot attain 
 sufficient strength to liberate a contraction. This is possible only 
 if very strong currents are used (Brenner ^). 
 
 The experiments of Claude Bernard, Schiff, and Valentin, which 
 are often cited in proof of the absence of a break-contraction on 
 stimulation of living nerves, will not bear close examination. It is 
 impossible to obtain exact results if, like Valentin, we insert 
 electrodes directly into the thigh of an animal without dissection ; 
 the branch of the current which reaches the nerve with such a pro- 
 ceeding must always be very small 2. The strength of the current 
 employed by Claude Bernard, or rather by Uousseau, who performed 
 the experiments in question, is nowhere given ^. Chauveau confirms 
 Claude Bernard's results, but adds that, when stronger currents are 
 used, the break-contraction makes its appearance. 
 
 As to the strength of his currents Chauveau also says nothing ; 
 he merely assures us that the break-contraction appears with currents 
 which do not alter the normal properties of nerve*; how the absence 
 of the break-contraction in uninjured nerves can be deduced from 
 this, I cannot conceive. Schiff too says nothing as to the arrange- 
 ment of his experiments and the strengths of current which he 
 used. 
 
 His assertions, therefore, cannot stand as evidence for the absence 
 of break-contractions with uninjured nerves, the more so as he says 
 himself that his law in general holds ' if the duration of the current 
 is not too prolonged ; ' he subsequently admits that ' with consider- 
 ably increased strengths of current ' the second or third degree of 
 contraction may be ' evoked with living nerves ^.' 
 
 ' Living nerves,' according to Schiff, are therefore not quite in- 
 sensible to the break-stimulus. The contrast between ' living ' and 
 
 1 Hermann, Handbuch der Physiologie, i. p. 63, 1879. 
 
 * Valentin, Die Zuckungsgesetze des lebenden Nerven und Muskels, Leipzig und 
 Heidelberg, 1863. 
 
 ^ Claude Bernard, Le9ons sur la Physiologie et la Pathologie du Systfeme nerveux, 
 i. Paris, 1858, p. 185. 
 
 * Chauveau, Journal de la Physiologie, No. x, April, i860, p. 283. 'It is certain 
 that the passage of currents through motor nerves in the normal condition provokes 
 contractions only at the moment of closing, whether the currents are ascending or 
 descending. But it must be understood that we are using batteries sufficiently weak ; 
 otherwise the double contraction manifests itself even with the currents which are far 
 from being sufficiently strong to alter the properties of the nerve and when the physio- 
 logical conditions leave nothing to be desii'ed.' 
 
 * Schiff, Lehrbuch der Physiologie des Menschen.i. pp. 80, Si, 1858-9. 
 
70 CONTEIBUTION TO THE 
 
 injured nerves maintained by the enquirers named with reference to 
 the break-contraction cannot therefore be regarded as established 
 by faultless experiments. 
 
 In the numerous experiments relative to the law of contraction 
 which I have made on nerves still connected with the spinal cord, 
 I have never observed the break -contraction absent when its 
 presence was to be expected, and when the experiment was rightly 
 made. 
 
 My theory is as little capable of explaining opening tetanus as 
 the laws of du Bois-Reymond and Pfliiger are able to account for 
 closing tetanus without a further assumption. The two phenomena 
 are attributable to the same cause. The most likely theory, in my 
 opinion, is that of Engelmann ^, according to which both closing 
 and opening tetanus are dependent on a latent state of stimulation 
 of the nerve from the influence of temperature, evaporation, or other 
 causes. Of course this stimulation by itself is not sufficient to elicit 
 tetanus, but it becomes so in parts of the nerve in which the ex- 
 citability is increased ; that is, in the neighbourhood of the negative 
 pole after closure, and in that of the positive pole (the negative 
 pole of the polarisation-current) after opening. This explanation 
 may or may not be sufficiently complete ; the chief point, in any 
 case, is, that the opening tetanus is to be regarded as the equivalent 
 of the closing tetanus ; and I shall content myself in this paper 
 with establishing a parallel between the phenomena occurring when 
 a current is opened and those which present themselves when it is 
 closed. In the breaking tetanus the excitation starts from the 
 negative pole of the polarisation-current, which is what my theory 
 requires. 
 
 The phenomenon of the ' voltaic alternatives ' can be very simply 
 explained by my theory. We know that, if a weak current is 
 passed in the same direction with another current already traversing 
 the nerve, its action is very considerably increased, since the exciting 
 point now falls on a pai't of the nerve of which the excitability is 
 increased. On the other hand, the excitation-effect by the closure 
 of a weak current in the opposite direction is diminished, since the 
 exciting point then falls on a part of the nerve where the excita- 
 bility is lowered 2. We shall see how the voltaic alternatives may 
 be explained from this point of view. 
 
 We will suppose that the polarising current is ascending, because 
 
 ^ Engelmann, Archiv fiir die ges. Physiologic, iii. p. 411, 1870. 
 
 ^ Compare Griitzner, Archiv fiir die ges. Physiologic, vol. 28, p. 143, 1882. 
 
THEORY OF THE BREAK-CONTRACTION. 71 
 
 this direction favours the occurrence of a break-contraction, which 
 again finds its explanation in this, that the polarisation-current is 
 descending, and the excitation therefore encounters no obstacle in 
 its progress to the muscle. If, now, the circumstances indicated by 
 Engelmann are in operation, a tetanus occurs in place of a simple 
 muscular contraction, when the current is broken. Supposing the 
 polarising current is again passed in the same direction before the 
 tetanus has ceased, the polarisation-current is arrested by the 
 polarising current ; the tetanus vanishes while the current is now 
 passing the second time, it generates a new polarisation-current, 
 which on interruption again produces tetanus. If now the direction 
 of the current is reversed, and the current again closed, there is 
 added to the polarisation-current another flowing in the same 
 direction^ so that their negative poles coincide ; the tetanus must 
 increase. The voltaic alternatives are thus necessary consequences 
 of my theory. 
 
 I have thus proved the agreement between all the most important 
 phenomena occurring when a polarising current is opened and the 
 corresponding phenomena arising when it is closed. In order 
 to subject the theory to a final test it still remains to examine 
 whether it is in accordance with the recent researches of Biedermann 
 and Griitzner. 
 
 In a very thorough investigation Griitzner ^ has shown the great 
 importance of the nerve-current proper in relation to the result of 
 electrical stimulation of nerves. All the phenomena with which 
 Budge, PflUger, Fleischl, and Hermann have made us acquainted 
 relative to the excitability of different parts of a nerve, he explains 
 by the action of the nerve-current, having proved that the current 
 of an ordinary nerve-preparation takes a different direction at dif- 
 ferent spots, and that therefore the exciting action of a current is 
 either strengthened or weakened, according as its direction is 
 opposed to or coincides with that of the nerve-current. 
 
 In connection with these investigations Griitzner also succeeded 
 in showing the very great importance of the nerve-current as deter- 
 mining the opening contraction under certain circumstances. Let 
 us suppose that by means of an ascending current we stimulate a 
 divided nerve near its cut surface. The stimulating current is now 
 flowing in the opposite direction to the nerve-current, and three 
 things are possible ; the stimulation-current may be weaker than 
 the nerve-current, of the same strength, or stronger. If it is weaker 
 
 *' Griitzner, loc. cit. pp. 130^178. 
 
72 CONTRIBUTION TO THE 
 
 than the nerve-current, or of the same strength, there will be no make- 
 contraction, but this will appear whenever the stimulation-current 
 is the stronger of the two. If the exciting current is weaker than the 
 nerve-current the result is the same as if a weak ascending current 
 flowed through the nerve during the whole period of closure ; at the 
 moment the exciting current is opened the nerve-current suddenly 
 attains its full strength. Its negative pole now coincides with a part 
 of the nerve which is in a state of heightened excitability due to the 
 previous descending current. Through this rise of excitability the 
 break-excitation becomes sufficiently strong to evoke a muscular 
 contraction. If the stimulating current is of the same strength 
 as the nerve-current, the latter is neutralized; when the stimulating 
 current is broken the nerve- current appears in its full strength. 
 Now, however, there is no part of the nerve endowed with a 
 heightened excitability, and the break-stimulus by itself is too weak 
 to generate a contraction. If the polarising current is still further 
 strengthened, it overpowers the nerve-current and produces a 
 polarisation, which, in conjunction with the nerve-current flowing 
 in the same direction, calls forth a break-contraction. 
 
 Griitzner's idea that the break-contraction which appears first is 
 due to the nerve-current seems completely pi'oved by the pheno- 
 menon of the ' gap ' (Liicke) which he observed in his series of 
 break-contractions. That the break-contractions occurring after 
 the ' gap,' as to the origin of which Griitzner has not expressed an 
 opinion, do not owe their appearance to the nerve-current, is evident 
 without discussion if we only bear in mind the actual conditions. 
 But it is very easy to see why, after the gap, the break-contractions 
 become stronger and stronger, for they are due to the polarisation- 
 current, the strength of which increases with that of the exciting 
 current. 
 
 Let us now suppose that a descending current is led through a 
 divided nerve in the neighbourhood of the cut surface. This current 
 must evoke a contraction even when of very weak strength, since 
 it is added to the nerve-current flowing in the same direction. (For 
 proofs of this see Griitzner's papers above quoted, pp. 16^, 167.) If 
 the strength of the stimulating current is gradually increased, the 
 contractions increase correspondingly to a maximum, and, as might 
 be anticipated, no gap presents itself. When the current is broken, 
 the polarisation-current flows in a direction opposed to that of the 
 nerve-current. We can in this case, therefore, obtain a break- 
 contraction only when the polarisation-current has become stronger 
 
THEORY OF THE BREAK-CONTRACTION. 73 
 
 than the nerve-curreut. That this is actually the case may be seen 
 from Griitzner's experiments^ p. 163. 
 
 Griitzner further found that in the lower part of the sciatic nerve 
 of the frog an ascending- nerve-current exists. When therefore this 
 part of the nerve is stimulated, the ascending break-contraction 
 must show itself later than the descending, since in the latter case 
 the nerve-current is added to the polarisation-current, while^ in the 
 first case, the two currents flow in opposite directions. At this 
 result Griitzner arrived experimentally. He stimulated the lowest 
 part of the nerve and found a descending break - contraction 
 when the rheochord slider stood at distances of 60, 45, and 690 ™™, 
 while he obtained an ascending break-contraction in the first case 
 with a distance of 360°^, and in the two other instances failed to find 
 it when the slider stood at 1000™™ (see I. c. p. 164). Just as, when 
 the nerve was excited in the neighbourhood of the cut end, where 
 of course the nerve-current is descending, so he found that on 
 exciting the nerve in the neighbourhood of the muscle, where the 
 nerve-current is ascending, there was a gap in the series of opening 
 contractions (see p. 173). 
 
 Where the nerve is free from any current, i. e. at its middle, no 
 difference is to be expected in the strength of current necessary for 
 the appearance of make- and break-contractions respectively, — 
 whatever may be the direction of the stimulating current. That 
 this is actually the case is proved by Griitzner's experiments 
 (p. 164, 165). 
 
 The results found by Griitzner are therefore not at variance with 
 my theory of the break-contraction, but rather furnish new proofs 
 of its correctness. Moreover, they afford indications of a new 
 cause of the break-contraction under certain conditions, which has 
 hitherto received from enquirers only a cursory attention. In proving 
 more thoroughly than has before been done that the nerve-current 
 is sometimes of sufficient strength to liberate a contraction, his 
 results have supplied an additional support to my theory, since they 
 have once more shown of how little consequence for the appearance 
 of a contraction is the absolute strength of the current in com- 
 parison with several other circumstances, such as the abruptness of 
 the increase of current strength, the condition of excitability, and 
 so on. 
 
 I have already mentioned that some facts discovered by Bieder- 
 mann agree very closely with my theory. His paper gives, besides, 
 a number of new results which are in great measure explained by 
 
74 CONTRIBUTION TO THE THEORY OF THE BREAK-CONTRACTION. 
 
 Griitzner's conclusions. Here we have only to enquire liow far 
 Biedermaun's researches on the break-tetanus, with its associated 
 phenomena, favour the theory that the break-contraction is deter- 
 mined by polarisation. 
 
 One of Biedermann's most remarkable results is the retardation 
 of the break-contraction, which he obtained on treating- the nerve 
 with an alcoholic solution of common salt and with salts of potas- 
 sium, and which he regarded as standing" in the closest relationship 
 to the break-tetanus. The reasons which he has adduced for this 
 opinion appear absolutely conclusive, and furnish a new argument 
 for Engelmann's well-known explanation. There is no direct con- 
 nection between this retarded break-contraction and the ordinary 
 one, as Biedermann himself states. The chief point is that, in this 
 case also, the excitation proceeds from the positive pole (the nega- 
 tive pole of the polarisation-current), and that the susceptibility of 
 the nerve to polarisation increases after treatment with alcohol. 
 
 Hitherto I have dealt exclusively with the contraction appearing 
 when the current is completely opened. There is, however, another 
 phenomenon described as a break-contraction, which shows itself 
 when the polarising current is very suddenly weakened without 
 being annulled. This contraction I have not studied at all^ and 
 the investigations made on the subject up to this time are too scanty 
 to permit of a theory being formed with regard to it. We do not 
 know with any certainty whence the excitation-eflfect comes, whether 
 from the positive or from the negative pole. As Griitzner in his 
 paper (p. 143) states that he is engaged on the subject, I have not 
 thought it well to commence any investigations upon it. 
 
 If, therefore, we leave out of account the case of the break-con- 
 traction evoked by a sudden weakening of the current, all the well- 
 established facts, as I have endeavoured to show, point to the con- 
 clusion that the hreah-contraction and all the phenomena occurring on 
 breaJcing a polarising current are attributable to the polarisation-current, 
 with which, in certain exceptional instances, the nerve-current co- 
 operates. 
 
III. 
 
 ON THE 
 
 INTERNAL POLAEISATION OF NERVES. 
 
 By De. EGBERT TIGERSTEDT. 
 Commtinicated to the Swedish Royal Academy of Sciences, September 13, 1882. 
 
III. 
 
 After Peltier had demonstrated polarisation in muscle ^, du Bois- 
 Reymond in the year 1856 showed that internal polarisability is 
 encountered as a widespread property of animal tissues, including' 
 nerve and muscle 2. Later (1867) he proved that this polarisability 
 is incomparably stronger in nerve and muscle than in other bodies 
 moistened with electrolytes, as well as that the strength of the 
 polarisation increases with the duration of the current, up to a 
 limit. This polarisation might arise in two ways. It might be 
 external, at the junction of the muscle with the clay of the elec- 
 trodes ; and it might be in the substance of the muscle itself. 
 He thought it unlikely that the first cause was here capable of an 
 appreciable action. On the other hand, he demonstrated internal 
 polarisability by so placing the electrodes of a galvanometer be- 
 tween those of the polarising current that the galvanometer 
 was not acted on by the muscle while the polarising circuit was 
 closed. After the polarising current had flowed through the 
 muscle a sufficient length of time, with the galvanometer circuit 
 open, the polarising circuit was opened and that of the galvano- 
 meter closed by a switch. A deflection ensued in the reverse 
 direction to that corresponding to the polarising current in the 
 muscle. Du Bois-Reymond obtained similar experimental results 
 with nerves ^. 
 
 ^ Du Bois-Reymond, Untersuchungen uber thierische Elektricitat, i. 1848, p. 376 
 foil. 
 
 '^ Ibid. Gesammelte Abhandlungen, i. p. 19. 
 ^ Ibid. ii. pp. 191-193. 
 
78 ON THE INTERNAL POLARISATION OF NERVES. 
 
 Meanwhile Matteucci had made similar observations. In several 
 papers communicated to the Academic des Sciences of Paris from 
 i860 to 1867, he gave an account of his investigations on this 
 subjects His experimental method was faulty, and could by no 
 possibility give exact results. He thus describes it, ' Lorsque 
 le passage du courant a dure un certain temps, depuis quelques 
 secondes jusqu'a 25-30 minutes, on enleve le nerf avec un support 
 forme d'une lame de gutta-percha, et on le porte en contact des 
 coussinets du galvanometre, dont I'homogeneite a ete reconnue 
 d'avance ' -. Besides demonstrating polarisation, and showing that 
 it increases mth the strength and duration of the current ac- 
 cording to a law which he did not investigate more closely, he 
 claims to have found that the electromotive force of polarisation 
 is much stronger in the neighbourhood of the positive than in that 
 of the negative pole. In proof of this he gives the following 
 experiment. He passed the current (8-10 Daniells) an indefinite 
 time (25-30 minutes and upwards) through a preparation con- 
 sisting of two nerves. Afterwards the nerves were cut off and 
 successively brought into the galvanometer circuit, arranged, how- 
 ever, in opposite directions. There then showed itself as difference 
 of polarisation in the two nerves a current indicating that polarisa- 
 tion at the positive pole was stronger than elsewhere ^. That an 
 experiment so arranged proves nothing is evident without discussion. 
 In contradiction to Matteucci's conclusions, du Bois-Heymond had 
 found that, if the electrodes of the galvanometer are pushed along 
 the conductor (that is the nerve) under investigation, without alter- 
 ing their distance from each other, equal deflections are obtained at 
 all parts of the conductor *. 
 
 In connection with his researches on electrotonus, Hermann 
 made new investigations on the internal polarisation of nerves. 
 To the results obtained by du Bois-Reymond he added observa- 
 tions on the duration and course of polarisation, which showed 
 
 ^ Matteucci, Comptes Eendus de I'Academie des Sciences, torn. 52, pp. 231-235, 
 1861 ; torn. 56, pp. 760-764, 1863 ; torn. 6^, pp. 151-156, 1867. 
 
 ^ Ibid. torn. 52, p. 232, 1861. 'When the passage of the current has lasted for a 
 certain time, from a few seconds up to 25-30 minutes, the nerve is removed on a 
 support consisting of a plate of gutta-percha, and placed in contact with the elec- 
 trodes of the galvanometer, it having been previously ascertained that they are 
 homogeneous.' 
 
 3 Ibid. torn. 65, p. 154, 1867. 
 
 * Du Eois-Reymond, Gesammelte Abhandlungen, i. p. 15. (The original is of the 
 year 1856.) 
 
ON THE INTEKNAL POLARISATION OF NERVES. 79 
 
 that it continues for several minutes with constantly diminishing 
 streng>uh, and almost never entirely disappears, so that for a new 
 experiment with the same tract of nerve, compensation must 
 usually be employed to bring* the needle to zero ^. Hermann's con- 
 clusions relative to the electrotonic after-effect occurring- outside 
 the electrodes we need not consider in the following investiga- 
 tion. 
 
 Since then, no enquirer, so far as I know, has given a thorough 
 study to the subject of internal polarisation in nerves. Yet a property 
 which is so markedly conspicuous in nerve and muscle, as compared 
 with other moist conductors, would seem to deserve investigation. 
 Perhaps one might arrive at new and important points of view 
 with regard to the way in which these tissues react to the electric 
 current. 
 
 The investigations here communicated concern only nerves, partly 
 because these appear to offer greater preliminary interest, partly 
 because in muscle the contraction and the change of position which 
 it entails exert a disturbing action on the investigation. In all 
 probability the same laws hold good for the internal polarisation 
 of muscle as for that of nerve. A future investigation will throw 
 further light on this question. 
 
 In my experiments the polarising current was conducted to the 
 nerves by the same electrodes as were used to lead the polarisation- 
 current to the galvanometer. By this proceeding I hoped to be 
 able to apply directly the results obtained to the phenomena which 
 occur in excitation experiments with a constant current. 
 
 The experiments were arranged as follows. From the battery B-^^ 
 (fig. 3), the polarising current passes through a rheochord Bh. 
 introduced as a derivation-circuit, to a Pohl's reverser B^, and 
 from there through a mercury key K and a Morse key T to the 
 nerve NN, from which it returns to the reverser, and so on. 
 Through the opposite arm of the Morse the current is taken to 
 the galvanometer G. The ' rest-current ' of the nerve is compen- 
 sated in the usual manner by means of a du Bois' compensator CC 
 (represented in the diagram as a straight line). 
 
 A spring s keeps one arm t-^ of the Morse pressed against the 
 plate a, so that the contact t^-i leading to the galvanometer is 
 open. If contact is made at this point by pressing on the other 
 
 ' Hermann, Untersuchungen zur Physiologic der Muskeln und Nerven, iii. pp. 71-74 ; 
 Berlin, 1868. 
 
80 
 
 ON THE INTEKNAL POLARISATION OF NERVES. 
 
 arm t^ of the Morse, the contact t-^-a is broken. The two con- 
 tacts are carefully insulated by a plate of ebonite 14'"™ thick. The 
 base-plate of the Morse key is cut in two, and a is isolated from h 
 by a layer of pitch. In the diagram the continuous lines and 
 arrows indicate the path and direction of the polarising current ; 
 
 Fig. 3. 
 
 the dotted lines and arrows show the path and direction of the 
 current going to the galvanometer. 
 
 There is therefore no possibility of the polarising current break- 
 ing into the circuit of the galvanometer. Du Bois-Reymond found, 
 however, that such an insulation is not sufficient when strong 
 currents are employed, inasmuch as, in spite of it, especially if the 
 atmosphere is moist, there may be a slight escape of the polarising 
 current to the galvanometer^. This drawback, however, did not in- 
 terfere with the investigation. In the first place I used only pretty 
 weak currents ; the strongest being produced by 3 Meidinger's 
 elements with 20,000 units in the rheochord. In the second place I 
 was favoured in my work by dry weather, in consequence of which, 
 for a long time, no action on the galvanometer from this source 
 of error could be perceived. Even when the atmosphere became 
 more highly charged with moisture and in consequence my gal- 
 
 ' Du Bois-Keymond, Untersuchungen tiber thierische Elektricitat, ii. Th. i Abth. 
 p. 496, 1848. 
 
ON THE INTERNAL POLARISATION OF NERVES. 81 
 
 vanometer was acted on when the circuit was open, the deflec- 
 tion even with 3 Meiding-ers and 20,000 rheochord units, did not 
 exceed -3 or at most -3 divisions of the scale. With 3 Mei- 
 ding-ers and 10,000 units it was from -i to -2 div. ; with 4000 
 rheochord units only • i div. ; with 2000 no effect could be per- 
 ceived. The errors which could possibly arise from this source lie 
 therefore almost entirely within the common limits of errors of 
 observation, as an examination of the experimental results shows. 
 
 As electrodes I used in most instances the larg-e unpolarisable 
 ones of du Bois-Reymond. 
 
 The polarising" current was determined by means of a ' damped ' 
 and a periodic g-alvanometer. The distance of the telescope from 
 the mirror of the instrument was 4400 ™™. 
 
 To obtain a greater strength of current I used for every experi- 
 ment the two sciatic nerves of a frog {ra?ia temj^oraria) placed side 
 by side. In all the experiments a battery of 3 Meidinger's ele- 
 ments was employed. 
 
 The experiments were conducted in the following way : — 
 
 The nerves are prepared and laid on the electrodes, while the 
 key K is raised and the current thus shut off from the nerve. The 
 ' nerve current ' is compensated as exactly as possible. This done, 
 and the Morse being depressed at b, the key K is closed. At a 
 given moment the Morse is let go, and the spring now presses the 
 arm t-^^ against a so as to close the polarising current. After the 
 polarising current has acted on the nerve for a given time, the arm 
 t^ is pressed against b so as to break the polarising current, and the 
 polarisation current passes through the galvanometer. 
 
 To determine the time during which the polarising current 
 flowed through the nerve I employed a metronome beating half 
 seconds. After a few experiments I had got sufficient practice to 
 carry out the necessary manipulations with the Morse ; the follow- 
 ing experiments may be cited as examples. 
 
 Experiment I. — 3 Meidingers, 1000 rheochord units ; duration of 
 closure of current 5'' ; distance between the electrodes 1 1 ™'^. 
 No. Deflection. 
 
 Mean variation -13. 
 
 2-0 
 
 2.4 
 
 2.4 
 
 2.4 
 
 2-4 
 
 Mean ... ... 2-32 
 
82 
 
 ON THE INTERNAL POLARISATION OF NERVES. 
 
 Experiment II. — 3 Meidiag-ers, 1000 rheochord xiuits ; duration 
 of closure 5" ; distance between the electrodes 1 1 ^'^. 
 
 No. 
 1 
 2 
 3 
 4 
 5 
 
 Mean 
 
 Deviation. 
 2.7 
 2.95 
 
 2-8 
 
 2.6 
 
 2-75 
 
 2.76 
 
 Mean variation -092. 
 
 Experiment III. — 3 Meiding-ers, 1000 rheochord units ; duration 
 of closure 5" ; distance between the electrodes 10'"'". 
 
 No. 
 1 
 2 
 3 
 4 
 5 
 
 Mean 
 
 Deviation. 
 3-95 
 3-9 
 3-6 
 3-9 
 3-95 
 
 3.86 
 
 Mean variation -104. 
 
 The time which elapsed between the opening of the polarising 
 current and closing of the galvanometer circuit I tried to deter- 
 mine by means of a Depretz' signal. The following table contains 
 the times thus arrived at : — 
 
 
 Time in 
 
 
 No. 
 
 Seconds. 
 
 No 
 
 1 . 
 
 029 
 
 12 
 
 2 
 
 -019 
 
 13 
 
 3 
 
 .018 
 
 14 
 
 4 . 
 
 -025 
 
 15 
 
 5 
 
 -024 
 
 16 
 
 6 
 
 .027 
 
 17 
 
 7 
 
 -032 
 
 18 
 
 8 
 
 -032 
 
 19 
 
 9 
 
 -032 
 
 20 
 
 10 
 
 -025 
 
 21 
 
 11 
 
 -025 
 
 22 
 
 Time in 
 Seconds. 
 .032 
 
 ■033 
 •040 
 •031 
 •041 
 •032 
 •037 
 •037 
 •032 
 
 •030 
 .036 
 
 
 Time in 
 
 No. 
 
 Seconds. 
 
 23 ... 
 
 .. -027 
 
 24 ... 
 
 .. -023 
 
 25 ... 
 
 .. -021 
 
 26 ... 
 
 .. -021 
 
 27 ... 
 
 .. -0 21 
 
 28 ... 
 
 .. -021 
 
 29 ... 
 
 .. -022 
 
 30 ... 
 
 .. -023 
 
 31 ... 
 
 .. -023 
 
 32 ... 
 
 .. -023 
 
 33 ... 
 
 .. -023 
 
 The mean of these observations is -028 seconds ; the mean varia- 
 tion amounts to -005 seconds. 
 
ON THE INTERNAL POLARISATION OF NERVES. 83 
 
 In the Depretz signal, according" to Marey^, magnetization 
 requires a period of 2^0'^ demagnetization g^o''. As a little 
 consideration will show, the first of these numbers must be sub- 
 tracted from, and the second added to, 0-038. We may regard these 
 times as counterbalancing each other, and, this being the case, we 
 have found that the time which elapses between the opening of the 
 polarising circuit and the closing of the galvanometer circuit 
 amounts to -028 seconds with a mean variation of -005 seconds. 
 The extremes are -018"' and •041". In general, therefore, we may 
 say that the time in question ranges at highest between 'Oi" and 
 •04'"j a variation which for my purpose must be looked upon as 
 quite immaterial. 
 
 My investigation of the laws of internal polarisation in nerves 
 dealt mainly with the three following principal questions : — 
 
 (i) The dependence of polarisation on the strength of the polar- 
 ising current. 
 
 (2) The dependence of polarisation on the time of closure of the 
 
 polarising current. 
 
 (3) The durational course of polarisation. 
 
 In the descriptions of experiments Hh. denotes the number of 
 rheoehord units ; ^, the distance between the electrodes ; T, the 
 duration of closure of the current ; Sc, the deflection of the gal- 
 vanometer in divisions of the scale. 
 
 1. The dependence of polarisation on the strength of the 
 polarising current. 
 
 In each of the experiments relating to this subject I led the 
 current through the nerve for the same length of time, raising it 
 gradually from a minimum (3 Meidinger, Mh. = 100) up to a maxi- 
 mum (3 Meidinger, Hit. = 20,000), which I did so gradually that no 
 sudden alteration in the strength of polarisation could occur. The 
 strengths of current which I used were expressed in terms of the 
 number of rheoehord units in the derivation-circuit, 100, 200, 300, 
 400, 500, 600, 700, 800, 900, 1000, 1500, 2000, 2500, 3000, 350O5 
 4000, 4500, 5000, 6000, 7000, 8000, 9000, 10000, 12000, 14000, 
 
 ^ Marey, La mdthode graphique dans les sciences expe'rimentales, p. 476 ; Paris, 
 1878. 
 
 G 2 
 
84 ON THE INTEKNAL POLARISATION OF NERVES. 
 
 16000, 18000, 20000. The mutual relations between these 
 strengths of current I determined by the deflection of the gal- 
 vanometer, when the principal current went to the nerve and a 
 branch of it was led off to the galvanometer by means of a 
 Siemens' rheostat. In order to exclude, as far as possible, the 
 efiect of polarisation of the nerve, I passed the current through 
 the nerve in both directions at each determination, and took the 
 mean of the deflections obtained. The results are recorded in the 
 followino" table : — 
 
 
 RJieochord units in 
 
 
 Current strength 
 
 ^'o. 
 
 the derivation circuit. 
 
 Sc. 
 
 Rh. 100 = 1. 
 
 1 
 
 100 
 
 i-6s .. 
 
 1 
 
 2 
 
 200 
 
 2.65 .. 
 
 1-6 
 
 3 
 
 300 
 
 3-65 .. 
 
 2-2 
 
 4 
 
 400 
 
 4-65 
 
 2-8 
 
 5 
 
 500 
 
 5-4 
 
 3-3 
 
 6 
 
 600 
 
 6.25 .. 
 
 . ... 3-8 
 
 7 
 
 700 
 
 6.9 
 
 4-2 
 
 8 
 
 800 
 
 7-65 .. 
 
 . ... 4-6 
 
 9 
 
 900 
 
 8.25 
 
 5 
 
 10 
 
 1,000 
 
 8.S5 .. 
 
 5-4 
 
 11 
 
 1,500 
 
 11-35 
 
 . ... 6-9 
 
 12 
 
 2,000 
 
 13-3 
 
 ... • 8.1 
 
 13 
 
 2,500 
 
 15 
 
 9-1 
 
 14 
 
 3,000 
 
 . 16.3 
 
 9-9 
 
 15 
 
 3.500 
 
 17-35 
 
 IO-5 
 
 16 
 
 4,000 , 
 
 18-2 
 
 11 
 
 17 
 
 4.500 
 
 19-05 
 
 11-5 
 
 18 
 
 5.000 
 
 19-7 
 
 11-9 
 
 19 
 
 6,000 
 
 20-85 
 
 ... 12-6 
 
 20 
 
 7,000 
 
 21.75 
 
 13-2 
 
 21 
 
 8,000 
 
 22-35 
 
 13-5 
 
 22 
 
 9,000 
 
 22-7 
 
 ... 13-8 
 
 23 
 
 10,000 
 
 23-15 
 
 14 
 
 24 
 
 12,000 
 
 24 
 
 14-5 
 
 25 
 
 14,000 
 
 24-85 ... 
 
 ... I5-I 
 
 26 
 
 16,000 
 
 25-65 
 
 -- 15-5 
 
 27 
 
 iS,ooo 
 
 26-3 
 
 ... 15-9 
 
 28 
 
 20,000 
 
 26-6 
 
 161 
 
 These values, ^ith the corresponding rheochord units, are used 
 in the records of experiment which follow. The following table 
 comprises the results of a series of experiments relating to the 
 subject : — 
 
ON THE INTERNAL POLARISATION OF NERVES. 
 
 85 
 
 TABLE I. 
 
 Ifo. 
 
 B. 
 
 1 
 
 lOO 
 
 2 
 
 200 
 
 3 
 
 300 
 
 4 
 
 400 
 
 5 
 6 
 
 500 
 600 
 
 7 
 
 8 
 
 700 
 
 Soo 
 
 9 
 10 
 
 900 
 1,000 
 
 11 
 
 I, soo 
 
 12 
 
 2,000 
 
 13 
 
 2,500 
 
 14 
 
 3,000 
 
 15 
 16 
 
 17 
 18 
 
 3,500 
 
 4,000 
 
 4,500 
 
 5,000 
 
 19 
 
 6,000 
 
 20 
 
 7,000 
 
 21 
 
 8,000 
 
 22 
 
 9,000 
 
 23 
 
 10,000 
 
 24 
 
 12,000 
 
 25 
 
 14,000 
 
 26 
 
 16,000 
 
 27 
 
 18,000 
 
 28 
 
 20,000 
 
 
 Numbers of experiments and readings of 
 galvanometer. 
 
 1-6 
 
 2-2 
 2-8 
 
 3-3 
 
 3-8 
 
 4-2 
 
 4.6 
 
 5 
 
 5-4 
 
 6.9 
 
 8-1 
 
 9.1 
 
 9.9 
 
 IO-5 
 
 II 
 
 1 1-5 
 1 1.9 
 12.6 
 13-2 
 
 I "3-5 
 13.8 
 
 14-5 
 
 15-5 
 
 Th-9 
 
 i6-i 
 
 1.25 
 
 1-45 
 1.6 
 1.85 
 1.9 
 
 2-1 
 
 2-85 
 3-25 
 
 3-7 
 
 4-3 
 
 4-45 
 
 4-7 
 
 4-85 
 
 5-1 
 
 5-5 
 
 5-85 
 
 6 
 
 6.4 
 
 6.4 
 
 6-7 
 
 6-8 
 
 7 
 
 7-1 
 
 7-3 
 
 •5 
 •75 
 • 8 
 
 I -1 5 
 1-45 
 1-62 
 1.9 
 
 2-1 
 
 2-15 
 2-2 
 
 3-1 
 
 3-9 
 4.4 
 
 4-55 
 
 5 
 
 5-25 
 
 5-5 
 
 5-7 
 
 6-1 
 
 6.4 
 665 
 6.8 
 6.8 
 
 7-1 
 7-1 
 7-3 
 
 7-5 
 7.8 
 
 •45 
 .8 
 
 •9 
 1.2 
 
 1-55 
 
 1-7 
 
 1.85 
 
 2.15 
 
 2-3 
 
 2^45 
 
 3-2 
 
 3-95 
 
 4-65 
 
 4.8 
 
 5-1 
 
 5-55 
 
 5-7 
 
 6.1 
 
 6.4 
 
 6.9 
 
 7 
 
 7-35 
 
 7'5 
 
 7.6 
 
 7-9 
 8 
 
 8.4 
 8.4 
 
 •3 
 
 •75 
 •8 
 
 I-Q: 
 
 1-3 
 1.4 
 1.6 
 1.9 
 ^•95 
 
 2-1 
 
 3-1 
 
 3-65 
 4.1 
 
 4-55 
 
 4-95 
 
 5^2 
 
 5-45 
 
 5-75 
 
 6.2 
 
 6-6 
 
 6-75 
 
 7 
 
 7-3 
 
 7"5 
 
 7-5 
 
 7-95 
 
 8.4 
 
 8.6 
 
 •45 
 
 1^75 
 
 1.9 
 
 2.1 
 
 2-45 
 
 2-55 
 
 2-7.^ 
 
 3^7 
 
 4-45 
 
 5-2 
 
 5-5 
 
 .^•95 
 
 6.4 
 
 6.7 
 
 7 
 
 7-45 
 7-65 
 8 
 
 8-3 
 8-45 
 8-95 
 9.2 
 
 9 -.^5 
 
 9-5 
 
 9-7 
 
 •5 
 .8 
 
 • 
 
 1-25 
 1-6 
 
 I- 
 
 1-95 
 
 
 2.15 
 
 
 2-4 
 2.65 
 
 2- 
 
 2.9 
 
 
 3-2 
 
 3- 
 
 4.2 
 
 
 5-' 
 5-6 
 6.2 
 
 5- 
 6.' 
 
 6.7 
 
 
 7^25 
 
 
 7-7 
 7-9 
 
 8." 
 
 8-45 
 
 
 9 
 
 
 9-15 
 
 9.8 
 
 9.8 
 
 lo- 
 
 10.2 
 
 
 10.4 
 
 
 10.8 
 
 
 II. I 
 
 
 11.4 
 
 ii.( 
 
 The distance between the electrodes was in 
 
 jriment 1 
 
 30mm 
 
 2 
 
 35 » 
 
 3 
 
 25,, 
 
 4 
 
 30,, 
 
 5 
 
 30 >, 
 
 6 
 
 25,, 
 
 7 
 
 25,, 
 
 In all the experiments T was = 5' 
 
 In order to make the irregularities of individual experiments 
 compensate each other and to bring together the results of all the 
 
86 
 
 ON THE INTERNAL POLARISATION OF NERVES. 
 
 thus obtained are 
 
 . calculated the 
 
 mean 
 
 of the first 
 
 six ^. The numbers 
 
 are 
 
 
 
 
 
 No. 
 
 Strength of the 
 
 Current. 
 
 Sc. 
 
 1 
 
 
 I 
 
 
 •45 
 
 2 
 
 
 1.6 
 
 
 
 •77 
 
 3 
 
 
 2-2 
 
 
 
 .96 
 
 4 
 
 
 2.8 
 
 
 
 1.25 
 
 5 
 
 
 3-3 
 
 
 
 1-54 
 
 6 
 
 
 3-8 
 
 
 
 1-7 
 
 7 
 
 
 4-2 
 
 
 
 1.9 1 
 
 8 
 
 
 4-6 
 
 
 
 2-18 
 
 9 
 
 
 5 
 
 
 
 2-29 
 
 10 
 
 
 .V4 
 
 
 
 2-47 
 
 11 
 
 
 6-9 
 
 
 
 3-36 
 
 12 
 
 
 8-1 
 
 
 
 4-05 
 
 13 
 
 
 9.1 
 
 
 
 4.61 
 
 14 
 
 
 9.9 
 
 
 
 4-97 
 
 15 
 
 
 10.5 
 
 
 
 5-36 
 
 16 
 
 
 II 
 
 
 
 5^72 
 
 17 
 
 
 "•5 
 
 
 
 5-99 
 
 18 
 
 
 11.9 
 
 
 
 6.26 
 
 19 
 
 
 12.6 
 
 
 
 6.69 
 
 20 
 
 
 13-2 
 
 
 
 7.07 
 
 21 
 
 
 13-5 
 
 
 
 7.26 
 
 22 
 
 
 13.S 
 
 
 
 7.61 
 
 23 
 
 
 14 
 
 
 
 7-71 
 
 24 
 
 
 14-5 
 
 
 
 8.02 
 
 25 
 
 
 I5-I 
 
 
 
 8-15 
 
 26 
 
 
 15-5 
 
 
 
 8.4 
 
 27 
 
 
 15-9 
 
 
 
 8-67 
 
 28 
 
 
 i6-i 
 
 
 
 8-87 
 
 From these determinations it follows that within the strengths of 
 current employed in my experiments., the polarisation of the nerve is, as 
 near as may be, directly pjroportional to the strength of the polarising 
 current. 
 
 2. The dependence of polarisation on the duration of 
 closure of the polarising current. 
 
 To investigate this question I used a current of the same strength 
 in all the experiments, and passed it through the nerve for unequal 
 lengths of time, namely for periods of i, 3, 5, 10, &c. seconds, and in 
 some experiments of 240 seconds. At the end of each period the 
 
 ' Experiment 7 I have not used in these calculations, as in it the individual deter- 
 minations were too few. 
 
ON THE INTEKNAL POLARISATION OF NERVES. 
 
 87 
 
 g'alvanometer circuit was closed. A number of experimental records 
 are exhibited in Table II. 
 
 TABLE II. 
 
 No. 
 
 T. 
 
 Numlers of experiments and readings of galvanometer scale. 
 
 8 
 
 9 
 
 10 
 
 II 
 
 12 
 
 13 
 
 H 
 
 1 
 2 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 I 
 
 3 
 
 5 
 
 lo 
 
 20 
 30 
 
 45 
 
 60 
 
 90 
 
 120 
 
 180 
 
 240 
 
 1-3 
 1-9 
 
 2-15 
 
 2-5 
 
 2-95 
 
 3-t5 
 
 3-45 
 
 3-65 
 
 3-95 
 
 4-1 
 
 4.8 
 
 1-2 
 
 2.05 
 
 2-55 
 
 2-9 
 
 3-6 
 3-9 
 
 4-05 
 4-65 
 
 4-85 ■ 
 5-1 
 
 5-2 
 
 1-45 
 
 2-2 
 2.4 
 2.65 
 
 3-2 
 
 3-5 
 
 3-7 
 
 3-85 
 
 4.4 
 
 4-75 
 
 1.2 
 
 2-1 
 
 2-5 
 
 3 
 
 3-7 
 
 4-15 
 
 4-5 
 
 4-75 
 
 5-3 
 
 5-75 
 
 .^•9 
 
 6.2 
 
 1.6 
 
 3 
 
 3-5 
 
 4-35 
 
 5-2 
 
 6 
 
 6.6 
 
 6-6 
 
 7 
 
 7-35 
 
 7-7 
 
 7-9 
 
 1-5 
 
 2.7 
 
 3-.S 
 4-6 
 
 5-7 
 6.1 
 
 6-5 
 6.9 
 
 7-2 
 7-5 
 8-35 
 9 
 
 1-5 
 2.6 
 
 3-3 
 
 4-3 
 5-05 
 6.15 
 6.6 
 
 "r 
 
 7-35 
 
 7-65 
 
 7-7 
 
 Note. — The distance between the electrodes and the strength of 
 the current were in 
 
 Experiment 8 
 
 
 9 
 
 E. = 25™™- 
 
 10 
 
 ... E. = 15,, 
 
 11 
 
 E. = 30,, 
 
 12 
 
 E. = 30,, 
 
 13 
 
 E. - 20,, 
 
 14 
 
 E. = 23,, 
 
 Rh. = 
 
 1000 
 1000 
 1 00c 
 1000 
 
 2C0O 
 1000 
 1000 
 
 The means of the seven experiments (mean deflections for each 
 duration of the closure) were as follows : — 
 
 No. 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 10 
 11 
 
 T. 
 
 Sc. 
 
 I 
 
 1-39 
 
 3 
 
 2-37 
 
 5 
 
 2.84 
 
 10 
 
 3-47 
 
 20 
 
 4-2 
 
 30 
 
 471 
 
 45 
 
 5-06 
 
 60 
 
 5-34 
 
 90 
 
 5-72 
 
 120 
 
 6.03 
 
 180 
 
 6.61 
 
 ' The values for a duration of the current amounting to 240" are not counted in the 
 mean, as they only enter into thi-ee experiments. 
 
88 ON THE INTERNAL POLMilSATION OF NERVES. 
 
 From all these experiments it appears : that if the polarising 
 current acts on the nerve for tmequal and increasing lengths of time, 
 other conditions being the same, the polarisation of the nerve increases, 
 more rapidly at first, afterwards more and more slowly, finallij ap- 
 proaching its maximum with extreme slowness. 
 
 3. The duration and course of polarisation. 
 
 After the polarising current is opened, the polarisation remains 
 at its height for only a short time, and then very rapidly declines. 
 It lasts, however, as Hermann found, for a long time afterwards, 
 though with very much diminished strength, so that it is always 
 necessary to compensate the residual polarisation between every 
 two experiments, as it would otherwise be impossible to bring an 
 experiment to a conclusion. 
 
 To obtain a closer acquaintance with the duration and course 
 of polarisation I proceeded as follows, I passed a current of 
 definite strength through the nerve for a given time. Then I 
 opened the polarising current at K (see diagram) and closed the 
 galvanometer-circuit after intervals which differed in different 
 observations. In this way I made myself acquainted with the 
 residual polarisation. I performed the experiments in such a 
 'way, as, first of all, to close the galvanometer-circuit instantly 
 after the polarising current was opened, just as in the experi- 
 ments previously described. Then I closed the galvanometer- 
 circuit I, 2, 3, 4, 5, 6, 7, 8, 9, lo seconds and so on after the 
 polarising-current was opened. In these experiments, then, the 
 current passed along the nerve always for the same length of time, 
 but the galvanometer-circuit was closed at different intervals after 
 the current was broken. 
 
 These experiments are more difficult to carry out than the pre- 
 ceding ones with the simple appliances at my disposal, because 
 the strength of polarisation falls off very rapidly, especially in the 
 first seconds after interruption of the current, while, at the same 
 time, the galvanometer-circuit has to be completed at the right 
 moment with perfect accuracy. Still, the experiments recorded in 
 Table III give an amply sufficient idea of the particular way in 
 which polarisation falls off in nerves. 
 
ON THE INTERNAL POLARISATION OP NERVES. 
 
 89 
 
 TABLE III. 
 
 
 !~ •" . 
 
 
 1 
 
 
 ^ ^ '^ 
 
 
 
 
 
 Numbers of experiments and readings of galvanometer scale. | 
 
 No. 
 
 2^ S s=^ 
 
 
 1 
 
 
 s ^ o 
 
 15 
 
 16 
 
 17 
 
 18 
 
 19 
 
 20 
 
 21 
 
 1 
 
 o 
 
 5-15 
 
 6-55 
 
 7 
 
 7-8 
 
 7^15 
 
 7-4 
 
 11.7 
 
 2 
 
 I 
 
 3 
 
 3-8 
 
 4-1 
 
 4-25 
 
 4.4 
 
 5 
 
 9-5 
 
 3 
 
 2 
 
 2-3 
 
 3-1 
 
 2-6 
 
 2-2 
 
 2-3 
 
 4-3 
 
 7-5 
 
 4 
 
 3 
 
 1.8 
 
 2.4 
 
 2-35 
 
 1-75 
 
 1.65 
 
 2-3 
 
 4.4 
 
 5 
 
 4 
 
 1-7 
 
 1.9 
 
 2 
 
 1-25 
 
 1.4 
 
 1.8 
 
 3-5 
 
 6 
 
 5 
 
 1-55 
 
 1-5 
 
 1.8 
 
 •7 
 
 I-I5 
 
 1-5 
 
 3-05 
 
 7 
 
 6 
 
 I-I5 
 
 1-5 
 
 1-5 
 
 •65 
 
 I 
 
 I-I5 
 
 2.65 
 
 8 
 
 7 
 
 •75 
 
 1-3 
 
 1-35 
 
 •55 
 
 ■75 
 
 
 
 9 
 
 8 
 
 .6 
 
 1.05 
 
 
 
 
 1.05 
 
 ^•7 
 
 10 
 
 9 
 
 •6 
 
 10 
 
 
 
 
 
 
 
 11 
 
 lO 
 
 •6 
 
 •9 
 
 i-i 
 
 
 
 
 09 
 
 i'.'6 
 
 12 
 
 12 
 
 
 
 
 
 
 
 
 1-5 
 
 13 
 
 15 
 
 
 
 
 
 
 
 
 1.05 
 
 14 
 
 20 
 
 
 
 
 
 
 
 
 0.7 
 
 15 
 
 30 
 
 
 
 
 
 
 
 ... 
 
 0.6 
 
 Note. — In all the experiments the polarising- current was closed 
 for 10 seconds. The distance between the electrodes, and the 
 strength of the current used were in 
 
 Exper: 
 
 ment 15 
 16 
 17 
 18 
 19 
 20 
 21 
 
 E. = 3om>». 
 
 Rh. 
 
 = 2000 
 
 E. = 30 „ 
 
 
 = 4000 
 
 E. = 25 „ 
 
 
 = 3000 
 
 E. = 25,, 
 
 
 = 4000 
 
 E. = 20 „ 
 
 
 = 4000 
 
 E. = 30 » 
 
 
 = 2000 
 
 E. = 30 » 
 
 
 = 3000 
 
 The means of the deflections observed in experiments 15-20 
 for each duration of the interval between the opening of the 
 polarising current and the closing of the galvanometer-circuit 
 were as follows : 
 
 No. 
 
 1 
 
 2 
 3 
 4 
 5 
 
 Time after opening of 
 the polarising current. 
 
 3 
 4 
 5 
 6 
 10 
 
 Sc. 
 
 6.84 
 4-09 
 2-80 
 2-04 
 1-68 
 
 1-37 
 1. 16 
 0.88 
 
90 ON THE INTERNAL POLARISATION OF NERVES, 
 
 From all these experiments it appears that when the polarising 
 current is opened, polarisation reaches its highest value instantane- 
 ously, and after that declines continuouslij . This decline piroceeds 
 at first very rapidly, aftenvards more and more slowly, so that the 
 polarisation lasts a long time after the current is shut off, and ap- 
 proaches the zero point only asymptotically. 
 
IV. 
 
 THE NATUEE OF THE ELECTRICAL 
 BEEAK-EXCITATION. 
 
 ' By Dr. P. GEUTZNER. 
 
 Pfliiger's Archiv, Vol. xxxii. p. 357. 
 
 fa 
 
IV. 
 
 INTRODUCTION. 
 
 In a research carried on in conjvinction with P. Moschner, 
 student of medicine \ I found that, in a large number of cases, the 
 excitatory effect of break of a current passing* through a nerve or 
 muscle was in reality due, not to the disappearance, but to the 
 appearance of the current : that there is thus a series of apparent 
 break-excitations which are in reality make-excitations. These 
 cases occurred when nerve-currents were present in the piece of 
 nerve experimented on. As is well known, nerve-currents are of 
 considerable magnitude in the neighbourhood of the cross-section, 
 and have a most important influence on the effects of excitation. 
 If, for instance, the nerve-current and the exciting current were in 
 the same direction — if, that is to say, I excited with so-called 
 ' like ' currents — exceedingly weak currents sufficed to produce 
 very great effects. If, on the other hand, I excited with ' counter- 
 currents,' it was necessary to use very strong currents to elicit any 
 response at all. When there was any response, it was far smaller 
 other things being equal, than in the first case. 
 
 Excitation with counter-currents led us to a peculiar phenomenon 
 hitherto unknown — the 'hiatus' in the series of apparent break- 
 contractions. If one electrode was at the cross-section and the 
 other a few mm. (5 to 7) away from it, and the nerve was excited 
 with an ascending current, whose strength was gradually raised 
 from zero, the following phenomena were observed. The break- 
 contractions first increased in size: as the current strength was 
 
 * Pfliiger's Archiv, vol. xxviii. p. 130. 
 
94 INTKODUCTION. 
 
 raised further, they diminish ed, and finally disappeared : with still 
 further increase of current strength they reappeared, at first small, 
 but sooner or later attainino* to a constant mao-nitude. 
 
 I explained the phenomenon as follows. If currents were passed 
 in a direction opposite to that of the nerve-current, evidently three 
 cases were possible. The counter-current was either smaller than, 
 equal to, or greater than the nerve-current, and therefore either 
 weakened it, reduced it to zero, or reversed it. The effect of 
 opening this counter-current was that the nerve-current suddenly 
 regained its original strength, in the first case from a diminished 
 positive strength, in the second case from zero, and in the third 
 case from a negative strength. In the first case, there was an 
 excitation of the nerve; in the second, that is when the rise of 
 current strength was from zero or near zero, no excitation^ — this 
 was the period of the hiatus ; in the third case, again an excitation. 
 
 All these excitations observed at break, whether before or after 
 the hiatus, were, however, in reality make-excitations, the excita- 
 tion being always due to the rise of the nerve-current, either from a 
 definite positive strength, an increase of current, or from a negative 
 strength, a reversal of current. All these excitations I shall call 
 neuro-electric or nerve-current excitations. Since they occur at 
 break, and a new current is added to a previously existent current 
 of the same or of opposite direction, I shall also refer to them as 
 'additive' in the same, or in the opposite direction. The break- 
 excitations before the hiatus are therefore additive in the same 
 direction, those after the hiatus are additive in the opposite 
 direction ; all being of course neuro-electric, and in reality break- 
 excitations. 
 
 I may mention that about the same time as I made the investi- 
 gations referred to above, Biedermann^ and Hering^ came to similar 
 conclusions, they also finding that the excitation at break of an 
 exciting current is, under certain circumstances, simply a make- 
 excitation of the nerve-current. Hering has thus the credit of 
 having first appreciated, in the case of muscle, the exciting effect of 
 the nerve-current. 
 
 In speaking of compensation of the nerve-current by counter- 
 currents, I do not, as mentioned in my previous paper, p. 170, refer 
 to compensation of the nerve-current in the ordinary sense : not, 
 that is to say, to currents which, as first employed by du Bois- 
 
 * [These observations are more fully referred to at p. 71.] 
 
 ^ Wiener Sitzungsberichte, Ixxxv. Part 3, p. 144. ^ Ibid. p. 237. 
 
INTRODUCTION. < 95 
 
 Reymond in his investigations, reduce to zero the part of the nerve- 
 current led off by the electrodes, so that no current is indicated by 
 a galvanometer inserted in the nerve-circuit, through which the 
 two opposed currents pass. 
 
 I always employed, on the contrary, far stronger exciting cur- 
 rents. If, for instance, the nerve-current was compensated in the 
 ordinary way by a derivation-circuit of 5 cm., the first break- 
 contractions began with a derivation-circuit of 10 to 20 cm. 
 
 In this respect my experiments differ from those of Hering, who 
 employed far weaker currents and very excitable preparations. 
 Hering says that it is necessary for the success of his experiments 
 that the frog's limb employed should be excited even by closure and 
 opening of the nerve-current. In my first series of experiments I 
 never used such preparations, and in my second, referred to in a 
 preliminary communication ^, I did not particularly study their 
 behaviour, since Hering had already thoroughly done so. 
 
 To prevent misunderstanding, or rather to remove those which 
 have already arisen ^, I must further explain that I do not suppose 
 that by compensating the nerve-current in the ordinary way I can 
 reduce the currents present in a nerve to zero. On the contrary, a 
 nerve, whose current is compensated in the ordinary way, behaves, 
 as Hermann has clearly explained ^, like a nerve without any external 
 circuit. Opening of a circuit through which a current exactly com- 
 pensating the nerve- or muscle-current passes cannot, as was found 
 by Hering and Biedermann and emphasised by Hermann, elicit any 
 excitation in the nerve or muscle, for no electric change whatever 
 is thereby occasioned in its fibres. 
 
 But the conditions are, I think, quite different if the currents 
 used are not of the strength of ordinary compensation-currents, but 
 far stronger. In this case it is really possible, by means of so-called 
 over-compensation, to make the nerve currentless, if not in its whole 
 length, yet assuredly in a certain part. With the exceedingly com- 
 plicated conditions under consideration, it will, I think, be best to 
 consider the nerve trunk in four sections, which are affected as 
 follows by the strong counter - current. In the first, perhaps 
 smallest, section, the nerve-currents are unaltered, their lines of 
 flow differing too much from those of the counter-current for any 
 alteration to be produced. In a second section, however, the nerve- 
 
 * Breslauer arztliclie Zeitschrift, 1882, No. 23. 
 
 ^ Virchow's Jahresbericlit iiber Physiologie, by Gad, 1882, p. 193. 
 
 ^ Pfluger's Archiv, xxx. p. 14. 
 
96 INTRODUCTION. 
 
 current will be weakened by the counter current ; in a third 
 reduced to zero ; and in a fourth reversed. To these assumptions 
 no one, I anticipate, will make any objection. I shall assume 
 nothing" more ; for they are amply sufficient to rfender the pheno- 
 mena which I observed, and my explanation of them intellig'ible. 
 
 Two additional jDoints seem worthy of separate consideration. It 
 might perhaps be objected that, since so many currents of different 
 directions and strengths and taking different paths are passing in 
 the nerve, it is impossible to say anything definite at all as to their 
 effects. It would be necessary to suppose that the effects of the 
 incompletely, the completely, and the over-compensated currents 
 would manifest themselves simultaneously. This would be the case 
 if all the fibres in the sciatic nerve were motor fibres of the gastro- 
 cnemius in which we are studying the effects of excitation. The 
 motor fibres however form, if the sensory and other fibres in the 
 sciatic are taken into consideration, perhaps a twelfth of all the 
 fibres, according to a rough calculation. It is only, therefore, what 
 occurs in that twelfth and its immediate neighbourhood that is of 
 importance to us. The current-changes occurring in the remain- 
 ing eleven-twelfths have no significance for us, jot practically 
 none. When, therefore, in the one above-mentioned twelfth of a 
 nerve trunk the diminished nerve-current mounts at break of the 
 counter-current to its original height, we have the break-eon- 
 tractions before the hiatus ; when the nerve-current is exactly com- 
 pensated, the hiatus ; when it is over-compensated, the contractions 
 after the hiatus. 
 
 Hermann^ says, in connection with the subject of compensation 
 of nerve-currents : ' It might be thought possible to do away with 
 the demarcation-current altogether by over-compensating. This 
 is, however, entirely impossible, since the lines of flow of the 
 counter-current in the nerve run quite differently from those of the 
 demarcation-current. Internal compensation is thus inconceivable, 
 and the less so the nearer the proper electro-motive surface. The 
 enormously strong currents in the immediate neighbourhood of the 
 latter will be little affected by external currents ever so strong 
 although these produce a special polarisation in the nerve.' I have 
 nothing to object to this, nor does it contradict what I have said ; 
 for, as already mentioned, it is not necessary to my explanation of 
 the hiatus that the demarcation-current should be entirely abolished.. 
 
 ' Pfliiger's Arcliiv, xxx. p. 15. 
 
INTRODUCTION. 97 
 
 Indeed, I can communicate several experiments which. I performed 
 in the summer session of 1882, in conjunction with Fraulein L. 
 Nemerowsky, and which appear to me to confirm Hermann's 
 assertion that internal compensation is the less conceivable the 
 nearer one gets to the proper electro-motive surface. As is set 
 forth in Fraulein Nemerowsky 's dissertation^, we have investigated 
 the strength of the nerve-current, that is to say its electro-motive 
 forcCj according as the electrodes, of which the one was always at 
 the cross-section, were more or less separated from one another. If 
 the electrodes are placed as near as possible, the electro-motive force 
 of the current led off into the galvanometer is very small. It 
 increases with the distance between the electrodes, and finally 
 reaches a maximum when the electrodes are 5 to 7 mm. apart. 
 With a yet greater separation of the electrodes it diminishes again, 
 evidently because currents of opposite direction (ascending) now 
 begin to be included^, which diminish the strength of the descending 
 current. If now, with a separation of electrodes of 5 to 7 mm., by 
 far the greatest amount of current is led off" by the external cir- 
 cuit, it is intelligible that in these circumstances the nerve-current 
 will be most aflfected by over-compensation, since the electrodes, 
 so to speak, include most current, and the lines of flow of the 
 two currents differ from one another least. We have found as 
 a fact that, since with electrodes close together it is impossible 
 to abolish or weaken the nerve-current even by means of strong 
 counter-currents, no hiatus appears, but the contractions first 
 appear with much stronger exciting currents. The hiatus, how- 
 ever, is most marked when the nerve-currents are most influenced 
 by over-compensation, that is, with a separation of electrodes of 
 5 to 7 mm. 
 
 By these explanations I believe that I have satisfied those who 
 took exception to my theory of the hiatus ^. 
 
 As already mentioned above, I had, in studying with Moschner 
 the phenomenon of the hiatus, opened and closed only in the main 
 circuit. If a compensator (du Bois-Reymond's straight compen- 
 sator) is used for regulating the strength of the exciting current, 
 
 ^ IJber das Phanomen der Liicke, by L. Nemerowsky. Inaugural dissertation, 
 Berne, 1883. 
 
 ^ See my paper on the above subject, Pfltiger's Archiv, xxviii. p. 140. 
 
 ^ See also Biedermann (Wiener Sitzungsberichte, Ixxxv. Abth. 3. p. 160), who 
 has arrived at similar results in the case of muscle as ours in that of nerve, and has 
 drawn similar conclusions. 
 
 H 
 
98 INTRODUCTION. 
 
 the current may be led in to the nerve by closing the exciting 
 current in the main or battery circuit, that is to say by placing a 
 mercury key between element and compensator. But it is also 
 possible to close the nerve-circuit after the main circuit by em- 
 ploying a key placed between electrodes and compensator. 
 
 Du Bois-Reymond^ experimented with regard to this point with 
 polarisable electrodes, and found marked differences in the effects 
 of stimulation according as he employed the one or the other 
 method. ' When I opened and closed the main circuit,' he says, 
 ' contraction resulted with a length - of even a few cm., while often 
 the whole length of the derivation- circuit wire (of about o, m. long) 
 was insufficient to evoke contraction on opening and closing the 
 nerve-circuit. Immediate excitation of muscle showed the same 
 difference, but between limits whose absolute values were higher.' 
 
 Du Bois-Reymond's explanation is as follows : ' When the main 
 circuit is closed after the nerve-circuit, the one division of the 
 current enters the nerve-circuit with an intensity conditioned by 
 the relationship of the resistances. When the main circuit is 
 opened, the current which is abolished in the nerve-circuit has 
 been previously reduced to a minimum by polarisation. But the 
 latter current, since a path is left open through the derivation- 
 circuit, is immediately followed by the polarisation-current in the 
 opposite direction, whose strength is perhaps equal at first to that 
 of the primary current, and which is therefore well suited to elicit 
 contractions. The charges are for the most part got rid of in- 
 stantaneously in the polarisation-current, so that, when the key in 
 the main circuit is again closed, the same series of events is re- 
 peated. If, however, the nerve-circuit is closed after the main 
 circuit, the current strength will be the same at the first closure as 
 at closing the main circuit after the nerve-circuit. Since, how- 
 ever, opening the nerve-circuit prevents the charges which im- 
 mediately develop from discharging themselves, opening of this 
 circuit is comparatively inoperative, and, secondly, the current 
 strength at closure after a not too long interval will not be so 
 great as before.' 
 
 In the more or less complete leading off of the external polarisa- 
 tion-current arising between electrodes and nerve-tissue, du Bois- 
 Reymond thus finds the explanation of this striking difi'erence. 
 
 ^ Ges. Abhandluncjen, i. p, 196, 1875. 
 
 ^ That is, the wire of the compensator or rheochord used. See diagram on a subse- 
 quent page. 
 
INTRODUCTION. 99 
 
 When, however, unpolarisable electrodes are employed, there is no 
 longer the same difference. 
 
 More recently Biedermann and Hering^, the former with 
 parallel-fibred curarised muscles, the latter with exceedingly ex- 
 citable nerve-muscle preparations (both using unpolarisable elec- 
 trodes), made similar experiments, and convinced themselves that 
 excitation of a muscle or a nerve by its own current can, for in- 
 stance, occur if a good external arc of connection is offered to these 
 currents, that is on closure of an arc whose ends are at the cross- 
 section and at the surface, or, what comes to the same thing, on 
 breaking the counter-current in the battery circuit in the case of 
 a nerve- or muscle-current compensated in the ordinary way. 
 Similarly, contractions are observed if the arc of connection is 
 broken, or, what comes to the same, if the compensating counter- 
 current is closed in the battery-circuit. But there is no excitation 
 if, with the nerve- or muscle-current compensated, the key in the 
 nerve-circuit is opened, or if, with the nerve-circuit previously 
 opened, a compensating counter- current is directed through the 
 nerve by closure of the nerve-circuit. 
 
 When the current is stronger than necessary for compensation in 
 the ordinary sense, the conditions become more complicated, but 
 most readily intelligible if what was said at p. 95 on this subject 
 is borne in mind. Hering gives the following explanation of the 
 facts in this case, simultaneously observed by him and by me, namely, 
 that excitation with counter-currents has more effect if the current 
 is made and broken in the main circuit than if in the nerve-circuit, 
 that is, that in the first case weaker currents are able to elicit con- 
 tractions. ' When the counter-current is somewhat stronger than 
 necessary for compensation, after closure of both circuits the current 
 in the nerve is actually ascending, although only traversed by what 
 may be called the remainder of the battery-current. Thus if, with 
 the main circuit already closed, the nerve-circuit is closed, there is 
 no contraction unless the division of the battery-current passing 
 through the nerve-circuit is of very considerable strength. If, on 
 the other hand, the main circuit is closed with the nerve-circuit 
 already closed, the effect of opening the nerve-current is added to the 
 weak and, by itself, insufficient effect of closure of the division of 
 the battery-current, and a contraction occurs. On opening the 
 nerve-circuit with the main circuit previously closed, no contraction 
 occurs, always provided that the division of the battery-circuit 
 1 Wiener Sitzungsberichte, Ixxxv. Part 3. p. 144 ; ibid. p. 237. 
 H a 
 
100 INTRODUCTION. 
 
 ascending in the nerve is not so strong as to produce a break- 
 excitation by itself, in spite of its partial compensation by the 
 nerve-current. On opening the main circuit with the nerve-circuit 
 closed, a new path is afforded to the nerve-current, and a con- 
 traction occurs, which is here further reinforced by the influence of 
 the Voltaic alternatives. For, if the influence of the latter depends 
 on the fact that a point on a nerve which had previously been the 
 point of entrance of a current has become more sensitive to excita- 
 tion by a current leaving the nerve at the same point, the efiect of 
 closing the nerve-current will be greater. I shall here entirely 
 neglect any effects which polarisation-currents might have, since 
 these currents would find a path for discharge by the nerve-circuit 
 at break of the main circuit.^ 
 
 While refraining from all further criticism of Hering's explana- 
 tions, I yet think that they are at least insufficient. It appears to 
 me that the circumstance which Hering puts aside altogether, the 
 formation namely of polarisation-currents, is of far greater import- 
 ance than the other facts. I believe that all these phenomena may 
 be explained in the most simple manner as follows. If a battery 
 of inconstant strength is closed through an arc of connection of 
 small resistance, the electromotive force of the battery, as is well 
 known, quickly falls off from the formation of a polarisation-cur- 
 rent, and any one who has once used such a battery for galvano- 
 caustic experiments knows how quickly the wire ceases to glow 
 when the current has been passed for some time, and, on the 
 other hand, how quickly and certainly it again becomes red-hot on 
 closure after the battery has been left open for a short time ; 
 further, how little the electromotive force diminishes if there is a 
 circuit of great resistance. 
 
 Now a nerve, or at any rate certain parts of it, when an 
 arc of connection of small resistance is applied to it, is analogous 
 to a battery whose current diminishes very rapidly; while a nerve 
 without any such arc of connection is analogous to a battery closed 
 by a circuit of very great resistance. 
 
 Hence if the current is broken in the battery circuit, thus leaving 
 for the nerve an external arc of connection always closed, the 
 counter-current has at closure only to contend with a weak current 
 in the portion of nerve concerned, and will therefore produce an 
 effect sooner ; hence the earlier appearance of excitation on closing 
 the counter-current in the main circuit. The opposite of course 
 occurs on closure in the nerve-circuit. On opening in the battery 
 
INTRODUCTION. J 01 
 
 circuit the nerve-current which appears in the piece of nerve con- 
 cerned passes more easily along- the constantly closed external arc 
 of connection than through the nerve substance ; hence the earlier 
 occurrence of what appear to be break-excitations on openino* 
 the battery circuit. 
 
 A similar explanation, it may be remarked, holds for the fact, 
 also discovered by Hering and me, that with a 'like' excitino* 
 current excitation appears earlier when the current is closed and 
 opened in the nerve-circuit. For in this case also we have a nerve 
 with a stronger current in the parts concerned, since there is no 
 opportunity for the current to discharge itself through an external 
 are of small resistance, and so fall off in strength from internal 
 polarisation. Closure in the nerve-circuit of a like current thus 
 causes alteration from a higher ordinate than closure in the main 
 circuit. The former is therefore more operative, that is to say 
 weaker exciting currents can excite the nerve, because they find it 
 so to speak more strongly electrotonised, and therefore in a more 
 excitable condition. 
 
 I have had little experience of the excitatory effects of like 
 currents, as I observed scarcely any break-excitations at all with 
 the current strengths used by me. I shall therefore refrain from 
 discussing this subject. Hering's explanation is as follows : 
 ' On closure in the main circuit the nerve-current previously 
 passing into the electrode applied to the surface of the nerve is only 
 reinforced by the arrival of the battery-current. But if the nerve- 
 circuit is closed after the main circuity the nerve- and battery- 
 currents are closed at the same moment, and hence the effect of 
 closure will be greater in the latter case. Similarly, the two 
 currents disappear simultaneously on opening in the nerve-circuit.' 
 
 Although, as already said, this subject must be left undiscussed, 
 I nevertheless believe with regard to the effects of closure that my 
 experiments^, which have lately been confirmed and extended by 
 Hermann 2, show that the rise of a current from zero to a consider- 
 able height has far less excitatory effect than even a very slight 
 rise beginning, not at zero, but at a definite positive height. 
 Hence, if Hering's assumption were correct, closure in the nerve- 
 circuit would be less operative than closure in the main circuit, 
 which, however, is not the case. 
 
 It should be noted that all these phenomena appear constantly 
 
 ' Pfluger's Archiv, xxviii. p. 144. * 
 
 * Ibid. XXX. pp. II and 12. 
 
102 ■ INTRODUCTION. 
 
 only when the electrodes are at a moderate distance (5 to 7 mm.) 
 apart. With the electrodes closer there is no difference observable, 
 or there may even be a difference in the opposite direction. With 
 the electrodes at indifferent points we have, as might be expected, 
 never been able to observe a difference in the above sense. Con- 
 tractions begin with exactly the same current strengths, whether 
 the current is opened and closed in the main or in the nerve- 
 circuit. 
 
 THE BREAK-EXCITATION. 
 
 It had thus been shown — and therefore I referred to these 
 experiments again here — that (i) a large class of break-excitations 
 are only so apparently, and (2) that the excitation of a particular 
 set of fibres in a nerve-trunk may be due to the closure of a 
 current in the substance of the nerve, independently of the existence 
 of an external arc of connection. Since, further, I had by earlier 
 experiments convinced myself that (3) in the case of currentless 
 nerve or muscle preparations excitation either does not occur at 
 all at break, or only with very strong currents, there naturally 
 arose a doubt in my mind as to whether the disappearance of an 
 electric current ever excites a nerve or muscle ; whether, to speak 
 with Pfliiger, the disappearance of anelectrotonus does in reality 
 produce an excitatory effect. I now think that this is not the 
 case, and believe, moreover, that there is no hreah-excitaiion, but 
 that every so-called break-excitation is in reality a make-excitation. 
 
 Before, however, I say anj'thing as to my reasons for this belief, 
 I wish to express my pleasure on finding that about the same time 
 K. Tigerstedt^ arrived at exactly the same conclusions, from 
 similar experiments and reasoning. This fact points, I think, 
 with a certain probability to our mutual view of the matter being 
 the right one. As Tigerstedt's work is doubtless accessible and 
 known to the readers of this paper, I can pass more rapidly over 
 many points, and refer the reader to that work. 
 
 Apart from the mention of the matter by Peltier^, du Bois- 
 Reymond^, as is well known, was the first to prove by unimpeachable 
 experiments that in a nerve or muscle through which a current has 
 been passed there is afterwards a current in the o^iposite direction. 
 
 ^ See No. Ill of these Translations, of which the original appeared two or three months 
 after the author's preliminary paper (Breslauer arztliche Zeitschrift, No. 23, 1882), 
 * Untersuchungeu iiber thierische Elektricitat, i. p. 376, 184S. 
 ^ Ibid. p. 377. 
 
ELECTRICAL BREAK-EXCITATION. 103 
 
 Although Peltier attributes to this secondary polarisation-current 
 the excitatory effect on breaking- the original, primary current, 
 this hypothesis was discredited by the authority of du Bois- 
 Reymond, and first became known to me on reading through the 
 literature of the subject after I had reached, on the above- 
 mentioned grounds, my conclusions as to the nature of the 
 break-excitation. ' In what way Peltier meant to explain the 
 break-contraction by means of this fact,' says du Bois-E-eymond, 
 'it is not easy to understand, since these charges require to all 
 appearances a closed circuit to produce a current through the 
 nerve ; this condition however is lost at break.' 
 
 Now this main objection is no longer valid ; for I convinced 
 myself that in the above-mentioned experiments breaking the 
 nerve-circuit allowed the nerve-current to come into existence by 
 being closed through certain paths in the nerve, and so caused 
 excitation. Although, so far as I know, there are no other 
 similar observations on the excitatory effect of a current whose 
 circuit lies wholly in the nerve, yet it was generally assumed that 
 nerve-currents partly close themselves within nerves^. It therefore 
 seemed to me quite incomprehensible that no significance with 
 regard to the break-excitation had been attributed to this secondary 
 polarisation-current, and the more so that this polarisation-current 
 had been recently investigated by various observers under the 
 name of secondary electromotive phenomena^. I have lately 
 turned my attention to this polarisation-current, since in it is to 
 be found, as it seems to me, the real cause of the break-excitation. 
 The results of my investigations may be summed up in few words. 
 
 Unpolarisable electrodes were applied to a nerve or muscle 
 (curarised sartorius) at indifferent points 6-10 mm. apart, and so 
 connected with a Pohl's reverser without cross-wires that in the 
 one position of the reverser the polarising current, of from a given 
 fraction of a Daniell up to three Daniells at the highest, flowed 
 through the nerve, while in the other position the secondary 
 current generated in the nerve flowed through a Wiedemann's 
 galvanometer. Turning over the reverser occupied a fifth of a 
 second on an average. 
 
 I observed, in full agreement with Tigerstedt, that (i) the 
 secondary polarisation-current increases correspondingly to, and 
 approximately in proportion to the strength of the polarising 
 
 ' Cf. Hermann, Handbuch der Physiologic, ii. p. 180. 
 ^ Ibid. ii. p. 164. 
 
104 ■ THE NATURE OF THE 
 
 current ; that (2) the polarisation-current increases with the 
 duration of the polarising* current. My observations were however 
 only continued over shorter periods than those of Tia^erstedt. 
 Nevertheless in them too the law became evident that the greatest 
 effect is produced in the first seconds during which the polarising 
 current remains closed and that a more prolonged closure of the 
 primary current only increases the secondary current to a slight 
 extent. (3) I investigated in exactly the same way as Tigerstedt 
 the rate at which the polarisation-current disappears from the 
 nerve, and found, although more time elapsed in my experiments 
 than in those of Tigerstedt between the moment of opening the 
 primary current and closing the galvanometer-circuit, that the 
 secondary polarisation-current is sfcongest during the first moments 
 after closure, and then gradually becomes weaker, first quickly, 
 and afterwards slowly, but that it can still be observed for a very- 
 long time — in the case of muscle after the lapse of minutes. 
 
 Quite recently there was published a thorough investigation of 
 this subject by du Bois-Reymond, in which a very important 
 and altogether new fact was brought to light, namely that of 
 j)Ositive polarisation. Du Bois-Reymond found that if a current is 
 passed through a muscle or nerve, and its electrical condition 
 immediately after opening tested by means of a second pair of 
 electrodes placed between those of the polarising current and 
 connected with a galvanometer, under certain conditions positive 
 polarisation appears at first, that is to say there is a current with the 
 direction of the primaiy current, while negative polarisation does 
 not appear until later. Like Tigerstedt ^, I have not observed this 
 positive polarisation in my experiments, doubtless for the simple 
 reason that I used primary currents of insufficient strength, and 
 further that my periods of break — the periods of time, that is to say, 
 elapsing between the breaking of the primary current and making 
 of the galvanometer-circuit — and of make were too long. For du 
 Bois-Reymond observed strongest positive polarisation in muscle 
 with a make of '075 seconds' duration and a current of twenty 
 Groves. If the duration of make in the case of nerve was 
 longer than a second, polarisation was always negative ; and the 
 weaker the primary current, the earlier did negative polarisation 
 appear. Now even if the polarisation indicated by the galvano- 
 meter is to be regarded as the algebraical sum of the two kinds of 
 
 ^ Sitzungsberichte der Akademie der Wissenschaften zu Berlin, Sitzung 5 April, 
 1883. P- 343- 
 
ELECTRICAL BREAK-EXCITATION, 105 
 
 polarisation, yet this resulting* polarisation is alone of importance 
 here, since we only used weak currents (usually of less than one 
 Daniell) in our excitation-experiments, and closed the exciting" 
 current for a period of a second on an average. We therefore 
 consider it safe to assume that in these experiments positive 
 polarisation has never occurred, or, to put it otherwise, has always 
 been more than compensated by far stronger negative polarisation. 
 
 However important, then, this positive polarisation may be in 
 itself, we may treat it as if it had been absent in our experiments. 
 
 The further part of our task, namely to show that the cause of 
 the break-excitation is to be sought, not in the disappearance of a 
 current, but in the appearance of the (negative) polarisation-current, 
 is now very easy. We shall bring together aU the cases in which 
 break-excitations have been observed, and endeavour to determine 
 for each case whether there can be a polarisation-current in the 
 nerve without external connection. 
 
 We shall first consider the law of contraction as it occurs when 
 the unpolarisable electrodes have been applied to a nerve at in- 
 different points. In this case the effect, in so far as shown in the 
 muscle, of closure of descending currents (which closure elicits 
 a response with currents of a certain strength and upwards), 
 increases with increase of the strength of current, but soon 
 reaches a maximum at which it remains. Break-excitations with 
 ascending currents follow exactly the same law, except that they 
 require stronger currents for their first production. 
 
 On the other hand, it is known that make-contractions with 
 ascending currents increase with the strength of current up to 
 a certain current-strength, but decrease with greater current- 
 strengths until at last they disappear altogether. Break-con- 
 tractions with descending currents follow the same law, except 
 that they also require stronger currents for their first produc- 
 tion. 
 
 The fact that the phenomena group themselves into these two 
 series is readibly intelligible on our hypothesis ; for the opening of 
 descending curi'ents is equivalent to the closure of weaker ascending 
 ones, while the opening of ascending currents is equivalent to the 
 closure of descending ones. 
 
 Excitation of a muscle or nerve, as Pfliiger, von Bezold, Engel- 
 mann, and Biedermann have shown, occurs at a pole — on closure at 
 the cathode, and on opening at the anode. The latter excitation is 
 also the weaker. We should say, however, that the excitatory 
 
106 THE NATURE OF THE 
 
 action of a current occui's only on closure and at the cathode ; the 
 excitatory effect of opening* a current begins, no doubt, at its 
 anode, but only because this is the cathode of the polarisation- 
 current at closure. 
 
 In the region of the anode there is diminished excitability, and 
 with strong cuiTents excitations cannot be transmitted from the 
 anode ; hence the diminution and disappearance of the make- 
 contraction with an ascending current, and, we now add, the 
 diminution and disappeai'ance of the break-contraction with a 
 descending cun^ent ; for just as in the previous case there was 
 an obstruction at the anode to the transmission of an excitation 
 to the muscle, so now there is an obstruction at the anode of the 
 polarisation-cuiTcnt. 
 
 On opening a polarising current there is, as is well known, 
 diminished excitability immediately afterwards in the region of 
 the cathode, increased excitability in the region of the anode. 
 The relation has apparently been reversed, but on our hypothesis 
 remains just the same ; for in the one case the cathode of the 
 primary current, while in the other the anode of the primary 
 current, is cathode to the polarisation-current. 
 
 The above, as the reader will admit, is a series of very remarkable 
 agreements between what our hypothesis demands and the actual 
 facts. This agreement, however, holds still further. The well- 
 known 'Voltaic alternatives' consist in the fact that a nerve or 
 muscle through which a cmTcnt has been passed for some time 
 becomes very sensitive to closure of a current of opposite direction, 
 while its sensibility to a current of the same direction is diminished. 
 The explanation is simple — that the second current is, in the one 
 case in the same direction as the polarisation-current, in the other 
 case in the opposite direction. 
 
 If our hypothesis as to the nature of the break-excitation is 
 correct, the stronger the secondary polarisation-current, and the 
 less the resistance it encounters, the earlier will a break-contraction 
 occur, provided always that its anode does not block the way from 
 the point of excitation to the muscle. Now my investigations 
 and Tigerstedt's have proved that within the periods taken for 
 stimulation-experiments the strength of the polarisation-current 
 increases with the duration of closure of the polarising current. 
 Nothing is easier to show than that with equal strengths of 
 current break-contractions occur much more readily if the current 
 is kept closed a little longer. Small fractions of a second make 
 
ELECTRICAL BREAK-EXCITATION. 107 
 
 great differences here, since usually the exciting* current is kept 
 closed for barely a second. 
 
 It is an analog"ous fact that if with a weak current there is no 
 break-contraction, one occurs if a stronger current, at the opening 
 of which there may have been an excitation, be first passed through 
 the nerve in the same direction. For the nerve is strongly polarised 
 by the last-mentioned current; the quickly following opening 
 of a weak current, which previously had no effect, now elicits 
 a contraction, since there is now a stronger polarisation-current 
 than before. This is also the explanation of the fact that the 
 phenomenon of the 'hiatus' discovered by me, did not appear, 
 or at least not usually, when I changed the current from a higher 
 to a lower strength. Biedermann ^ and Tigerstedt ^ recently ob- 
 served similar facts. 
 
 The strength of the polarisation-current increases, moreover, 
 with the strength of the primary current ; hence it is that at 
 indifferent points in a nerve or muscle break-contractions do not 
 occur until strong exciting currents are employed, and increase 
 with the current strength. 
 
 Tigerstedt has also found that the sciatic nerve of a frog is 
 easier to polarise near the point of section, or after dilute alcohol 
 has been applied to it. Both these conditions also favour the 
 occurrence of break-contractions, as was shown by my earlier 
 experiments, and those of Biedermann. 
 
 The polarisation, current can also be increased by the use of 
 
 polarisable, instead of unpolarisable electrodes. In this case there 
 
 is an external polarisation-current between the metal and moist 
 
 nerve, and an internal one between the various constituents of the 
 
 nerve. As, however, the closure through the nerve of the external 
 
 polarisation-current is determined by the same, or nearly the same 
 
 conditions as determine that of the internal one, and as the former 
 
 current is added to the latter, a series of phenomena which depend 
 
 on the closure through the nerve of polarisation-currents are in 
 
 this case unusually well-marked. It is thus very easy to convince 
 
 oneself that if instead of unpolarisable electrodes ordinary metallic 
 
 electrodes are used, and with the same resistance, in order that the 
 
 conditions may be as like as possible, break-contractions at break 
 
 of the main-circuit always appear earlier than when unpolarisable 
 
 electrodes are used. An experiment may be cited to exemplify this. 
 
 ^ Wiener Sitzungsberichte, Part iii. vol. Ixxxiii. p. 328, 1881. 
 * Tigerstedt, p. 55 of this book. 
 
108 THE NATURE OF THE 
 
 Experiment. — A nerve-muscle preparation from a Rana esculenta of moderate size 
 is arranged in a Pfliiger's myograph. The electrodes are placed on the nerve at 
 indifferent points, the proximal one at about 30 mm. from the cut end, which is held by 
 a thread, the distal at about 12 mm. from the muscle and 7 mm. from the proximal 
 one. The current is of one Daniell. Mercury keys are inserted in both the main and 
 the nerve circuit. As derivation-circuit (rheochord) I used in these and other experi- 
 ments three German-silver wires stretched backwards and forwards parallel to one 
 another on a board, and each i metre long and -5 mm. in cross-section. Contact of 
 one of the electrodes, which was movable, with the wire was brought about by 
 means of a pair of forceps, made on the principle of ordinary spring-forceps and 
 well nickel-plated, which rode on the wire. The blades of the forceps were thick 
 and heavy at their ends, so that they kept fast. At their other end the forceps carried 
 a small screw, by which the electrode wire was fixed. As the derivation-circuit wire 
 was carried well clear of the board, the forceps could be very easily applied at any 
 desired point on a wire, or transferred from one wire to the otlier. I can recommend 
 this as a very practical arrangement. It appears to me to be both neater and more 
 certain than the mercury contact in du Bois-Eeymond's rheochord. 
 
 The following experiments were made for comparison. Firstly, the break-con- 
 tractions were compared with one another as to size wlien the current was opened 
 only in the nerve-circuit, (a) with unpolarisable and (6) with polarisable electrodes ; 
 secondly, when the current was opened only in the main circuit. There are thus 
 the following four cases : — 
 
 . _, , , • ,1 . • { ((• Unpolarisable electrodes (i) 
 
 A. Current opened in the nerve-circuit ••• j 7 p i •. ii } \ 
 
 T, .-, , J • ji . . ., ^ a. Unpolarisable electrodes (^) 
 
 B. Current opened in the main circuit ...•!,■,-,,.,, ; ( 
 
 { 0. Folarisable ,, (4) 
 
 We shall first compare cases i and 2, and 3 and 4. Afterwards (see below) cases 
 I and 3, and 2 and 4 will also be compared. 
 
 The following is the record of the experiments : — 
 
 First Break-contraction. 
 Experiment i : — 
 
 In case 1 no contraction up to 300 cm. derivation-circuit (rheochord wire). 
 » 2 „ „ 300 „ „ „ 
 
 „ 3 contraction began at 275 „ „ „ 
 
 » 4 » >y 30 » .. » 
 
 Experiment 2 : — 
 
 In case i no contraction up to 300 cm. derivation-circuit. 
 
 >. 2 „ „ 300 „ „ „ 
 
 „ 3 contraction began at 260 ,, „ „ 
 
 j> 4 »' » 4*^ >> )> jj 
 
 In experiment i the exciting current was ascending, in experiment 2 descending. 
 
 Two additional experiments, in which the conditions were exactly the same as in the 
 
 two first ones, gave the following results : — 
 
 First Break-contraction. 
 Experiment i : — 
 
 In case i contraction began at 285 cm. derivation-circuit. 
 „ 2 no contraction up to 300 „ ,, „ 
 
 „ 3 contraction began at 230 ,, „ ,, 
 
 » 4 » .. 35 >. 
 
 Experiment 2 : — 
 
 In case i no contraction up to 300 cm. derivation-circuit. 
 )i 2 „ „ 300 „ „ „ 
 
 „ 3 contraction began at 260 „ „ ,, 
 
 4 .. .. 40 " » " 
 
ELECTRICAL BREAK-EXCITATION. 109 
 
 In a fourth experiment, fur which a current of two Daniells was used, the first 
 break-contractions occurred with the following values of the derivation-circuit, counting 
 in the same order. With ascending current at 170, 220, 150, 50; with descending 
 current at 190, 280, 160, 60. In order to have the resistance equal in cases i and 2, 
 and in cases 3 and 4, when the metallic electrodes were placed on the nerve the pre- 
 viously used impolarisable electrodes were brought into the nerve- or muscle-circuit, 
 lying end to end. 
 
 Thus when the external polarisation-current is added to the 
 internal the occurrence of break-contractions on opening- the main 
 circuit is favoured, as is evident from the experiments and readily 
 intelligible. (Compare cases 3 v^ith cases 4.) If, however, the 
 current is opened in the nerve-circuit, the rule is that the occur- 
 rence of break-contractions is not only not favoured, but even 
 hindered. (See cases i and 3 in the second series, with ascending 
 current.) This would seem to mean that there is far less resistance 
 to an internal polarisation-circuit lying altogether in the nerve, 
 than to a similar external one. To this point I have not further 
 directed my attention. 
 
 Muscle behaves like nerve according as polarisable or unpolaris- 
 able electrodes are used and the main circuit is opened. 
 
 Experiment. — A sartorius muscle of a medium-sized curarised grass-frog is laid on 
 the electrodes in such a way that no muscle-current passes through them (position of 
 indifference). Two Daniells. There was a break-contraction for both ascending and 
 descending currents in case 3 (polarisable electrodes, main current opened) with 
 190 cm. of wire. In all other cases no break-contraction up to 300 cm. of wire. 
 With a current of six carbon-zinc elements a similar result. Break-contraction 
 only in case 3 with 80 cm., in all other cases no break-contraction up to 300 cm. 
 Arrangement similar to that of the previous experiments. 
 
 We now arrive at the consideration of those cases in which the 
 secondary polarisation-current through the tissue to be excited is of 
 insufficient density, or for any other reason fails. 
 
 A number of both earlier and more recent observers (CI. Bernard, 
 Schiff, Valentin, Tick, Rumpf) have drawn attention to the fact 
 that break-contractions occur less readily with uninjured nerves in 
 the living animal than with those which have been cut across and 
 prepared. There are clearly two causes for this. On the one hand 
 the continuance of the circulation must oppose polarisation by 
 removing its products, on the other hand when a nerve is not 
 insulated and lying on the electrodes the polarisation-current can 
 pass along other paths, so that the current passing through the 
 nerve is weakened. 
 
 There is a fact which is evidently' related in nature, and is easy 
 enough to observe, though, so far as I know, it has not been more 
 
110 THE NATURE OF THE 
 
 closely investigated by physiologists. Ritter, as is well known, 
 first maintained that there is a special law of contraction for the 
 thigli of the frog as a whole, and called attention to a difference 
 with respect to excitation between the flexors and extensors. 
 
 The truth of this much-questioned assertion of difference in 
 excitability between physiologically different groups of muscles was 
 established by Rollet, and is confirmed by recent investigations of 
 my own, which will shortly be published. This difference manifests 
 itself at closure and opening in the fact that the more powerful 
 stimulus affects the less excitable extensors most, while the weaker 
 stimulus has most effect on the more excitable flexors. Thus, if one 
 applies a pair of either polarisable or unpolarisable electrodes to 
 the sciatic nerve of a frog at a point midway between its point of 
 exit from the spinal cord and the point where it disappears in 
 the thigh, care being taken that they are kept in exactly the 
 same position, it will not be observed that as the strength of the 
 currents employed is raised the same group of muscles merely con- 
 tract more and more strongly, but that quite different muscles con- 
 tract with weak and strong currents respectively^. If one carefully 
 observes which muscles first contract at closure of weak currents 
 (usually certain muscles of the. thigh and toes), the interesting fact 
 will be observed that on further increase of the current-strength 
 other muscles are excited, and always more and more strongly at 
 closure, while at opening those muscles alone contract which had 
 previously contracted with weak currents at closure. The break- 
 excitations of these relatively strong currents are now similar in 
 their eff'ect to make-excitations of the weak currents. 
 
 Similar phenomena occur in human muscles, and I am astonished 
 to find no mention of them in handbooks of electro-therapeutics. 
 For instance, if with a suflSciently strong current the electrodes are 
 placed a few centimetres apart in the groove on the inside of the 
 biceps, it will be observed constantly that with certain current- 
 strengths other muscles contract at closure than at opening. I 
 often noticed strong flexion of the hand on closure, and moderate 
 pronation on opening the same current. 
 
 In experiments on man (which I have only made cursorily, and 
 intend to pursue further when I have an opportunity) the conditions 
 are of course not so apparent as in the case of separated frogs' nerves. 
 It can be more particularly objected that if the electrodes are 
 sufficiently far apart the points of excitation are also far apart, and 
 
 ' See the dissertation of L. Nemerowsky, Berne, 18S3. 
 
ELECTRICAL BREAK-EXCITATION. Ill 
 
 hence that on closing" and opening* a current the same nerve is not 
 acted on with different current-strengths^ but different nerves are 
 acted on at closure and opening, which, I believe, may often enoug-h 
 happen. For if the break- stimulation of a current is really nothing- 
 else than the make-stimulation of its polarisation-current, it is 
 conceivable that in the above experiment on man the battery- 
 current, which enters the body at one electrode and leaves it at the 
 other, may flow along- paths different from those of the polarisation- 
 current, whose circuit must be completed through the tissues alone 
 and which may therefore stimulate other nerves. 
 
 This experiment brings us to a new fact, which, though it is 
 connected with experiments previously noticed (p. io8), must be 
 separately considered here. As already mentioned, neuro-electric, 
 additive break-contractions appear earlier at break of the main than 
 of the nerve-circuit. The nerve-current which acts here as stimulus 
 meets with less resistance in its passage along- the derivation- 
 circuit than in the substance of the nerve. What is true of the 
 nerve-current is true also of the current of internal polarisation. 
 It is therefore easy to show that if unpolarisable electrodes are 
 applied to a nerve at indifferent points, i.e. where there is no 
 nerve-current, break-contractions appear earlier if the current is 
 broken in the main circuit than if in the nerve-circuit. This 
 will be evident if cases i are compared with the corresponding 
 cases 3 in the experiments at page 109. In each series, i. e., the 
 length of wire required to elicit the first break-contraction is 
 greater in case i than in case 3 ; thus 300 is g-reater than 275, 
 and so on. In case i the current was made and broken in the 
 nerve-circuit, in case 3 in the main circuit. 
 
 Hermann made similar experiments four years ago, but did not 
 publish them until shortly after the publication of my preliminary 
 communication and of Tigerstedt^s complete paper. He also con- 
 cludes from his experiments, the results of which, interesting to us, 
 he thus summarises, that ' where the break-contraction is strug-gling 
 for existence it appears if there is any difference at all with weaker 
 currents and is stronger if the current is broken in the main 
 circuit than if in the nerve-circuit.' Hence he concludes that the 
 counter-current due to polarisation is concerned in the production 
 of the break-contraction. In my opinion Hermann was not justified 
 in drawing this conclusion from the above-mentioned experiments, 
 for the reason that though his experiments were made a long time 
 ago, he has only now drawn the conclusion from them, or at least 
 
112 THE NATURE OF THE 
 
 published it. Thus there are wanting- in Hermann's investigations 
 the exceedingly important control experiments to determine whether 
 the excitation at break of the primary current is due entirely or 
 principally to the polarisation-current. He mig-ht quite as well have 
 been dealing with apparent, additive break-contractions due to the 
 appearance of the nerve-current. To judge from my own experience, 
 it is exceedingly probable that such might have been the case, since 
 as a rule the electrodes are so applied to a nerve that nerve-currents 
 pass through them. The earlier appearance of break-contractions 
 with descending currents, which Hermann records, appears to me 
 to point to additive break-contractions being also included in these 
 experiments (at least very often), or to the nerve- and polarisa- 
 tion-current having passed simultaneously through the derivation- 
 circuit. 
 
 In addition to this, the break-contractions before may be con- 
 founded with those after the hiatus, if the hiatus is short and does not 
 appear at the same point when the main circuit is broken as when 
 the nerve-circuit is broken. The fact that differences, even of a 
 tenth of a second, in the duration of closure of the primary current 
 affect the appearance of break-contractions very considerably should 
 also be mentioned. I myself was once misled by this source of error 
 through the one mercury key being less easy to manage than the 
 other and so keeping the current closed a little longer. Finally, in 
 comparative experiments the current ought not to be changed from 
 a higher to a lower strength, and the long persistence of the 
 polarisation-current must be kept in mind. All this is of importance 
 for the attainment of accurate results. Although I by no means 
 believe that Hermann neglected these facts and so fell into error, 
 yet I am convinced that in his experiments the nerve- and polarisa- 
 tion-current have been intermingled, or have acted simultaneously. 
 Otherwise he would not, if I may so say, have written so hesi- 
 tatingly : ' When the break-contraction is struggling for exist- 
 ence it appears, if there is any difference at all,' &c. What I 
 described was not a struggle : in a very large number of ex- 
 periments, which I have considerably increased since the appear- 
 ance of Hermann's paper, the break-contraction appeared much 
 sooner after opening the main circuit than after opening the 
 nerve-circuit. 
 
 The difference becomes much greater (as is seen in the experi- 
 ments at page io8) if ordinary metallic electrodes are substituted for 
 unpolarisable electrodes. The internal and external polarisation- 
 
ELECTRICAL BREAK-EXCITATION. 113 
 
 currents are conducted much better throug-h the closed derivation- 
 circuit than through the nerve itself. This experiment, which has 
 been already described by du Bois-Reymond, shows how great an 
 influence the polarisation-current (in this case chiefly external) 
 exerts on the appearance of break-contractions ^. But even with 
 unpolarisable electrodes, and therefore with an internal polarisation- 
 current only, the same thing- occurs, though not to the same extent. 
 Thus the less the resistance to the polarisation-current the more 
 readily do break-contractions occur. These I shall call ' polarisation 
 break-contractions,' in contrast to the 'additive' break-contractions 
 previously referred to. 
 
 In conclusion, I shall shortly refer to a break-contraction which 
 diflPers from that already mentioned in that its appearance is retarded. 
 In my preliminary communication I regarded this phenomenon, 
 which had been observed by Pfliiger and was recently studied more 
 thoroughly by Biedermann, as due to the fact that the anode of the 
 polarisation after-current lay between the point of stimulation and 
 the muscle, and hindered for a shorter or longer time (for the 
 strength of the polarisation- current rapidly diminishes) the excitation 
 from reaching the muscle. This explanation applies of course only 
 to the case of opening descending currents, with which Pfliiger had 
 also exclusively observed this phenomenon. My experiments on this 
 subject are however as yet incomplete, therefore I shall not give an 
 account of them at present. 
 
 Apart from this retarded break-contraction, we have met with 
 three distinct kinds of break-contraction in this investigation : 
 (i) additive in the same direction ; (2) additive in the opposite direc- 
 tion, due to the appearance of the nerve-current ; and (3) polari- 
 sation break-contractions, due to the appearance of the current of 
 internal polarisation. All these break-contractions can, as we have 
 seen, be shown to be caused by, or, to speak more cautiously, to be 
 coincident with the appearance of a new current rather than the 
 disappearance of the original one. They may all therefore be 
 in reality make-contractions. At any rate, we know with absolute 
 certainty that closure of a current passing through a muscle or 
 nerve excites these organs : when we see, on the other hand, that 
 on opening a current a new current comes into existence, we are 
 justified in inferring that it is the coming into existence of the 
 
 ^ This meets the objection in Hermann's paper, p. 105. He inferred from my 
 preliminary communication that I had only used polarisable electrodes in my experi- 
 ments. 
 
114 THE NATURE OF THE 
 
 latter, and not the disappearance of the former, that excites. In 
 accordance with this inference, I should express as follows the 
 law accordino" to which currents excite a nerve or muscle : — 
 
 o 
 
 1. A nerve or a muscle is excited hy the coming into existence of a 
 current through it. The height of current-strength and the rapidity 
 with which this height is attained are of account, in that the excita- 
 tion increases with them up to a certain limit. 
 
 2. A muscle or a nerve is also excited, though to a less extent, hy a 
 current floiving through it at a constant strength ^. For muscle this is 
 known ; for sensory nerves and the dilating vascular nerves of the 
 skin I have demonstrated it^; that it is also true for motor and 
 secretory nerves follows if we assume that all nerves are similar, 
 or at least not very different, even though we may not be able 
 to demonstrate excitatory effects in the end-organs of these nerves : 
 for the reaction of the end-organs depends on conditions inherent 
 in themselves, and on other factors not affected by excitation 
 of a nerve. Moreover, Pfiiiger ^ and von Bezold * loug ago asserted 
 that under certain circumstances motor nerves react to a current of 
 constant strength. 
 
 3. On the other hand, a nerve or muscle is not excited hy the dis- 
 appearance of a current from it (at any rate this has not been 
 proved). The excitations observable on opening exciting cur- 
 rents correspond with the coming into existence of other currents 
 (muscle- or nerve-currents or polarisation after-currents), and it is 
 to this that the excitatory action is due. 
 
 4. The direction of the exciting current, provided only that it be 
 parallel to the axis of curreyitless nerves or muscles^ does not appreciably 
 affect the strength of the excitation of these organs themselves. In the 
 case of a nerve, however, the effect on its end-organ is of course 
 modified if at make of a strong ascending current its anode, or if at 
 break of a descending current the anode of the negative polarisation- 
 current lies in the way and obstructs the passage of the excitation. 
 I agree with Tigerstedt in distinguishing two strengths of stimulus 
 here, namel}^, weak currents, at whose closure the excitation at the 
 
 ^ Strictly speaking, no currents — at least, no currents of short duration, such as are 
 commonly employed for stimulation — flow through a nerve or muscle at a constant 
 strength. For since both organs are polarisable there must occur in the first moments 
 after closure an increase of current strength from positive, and immediately afterwards 
 a decrease from negative polarisation. 
 
 ' Pfliiger's Archiv, xvii. p. 215. 
 
 ^ Elektrotonus, p. 453. 
 
 * Untersuchungen iiber die elektrische Erregung, Leipzig, 1861, p. 325. 
 
ELECTRICAL BREAK-EXCITATION. 115 
 
 negative pole overcomes the obstruction at the positive, and strong 
 current, at whose closure the excitation at the negative pole is 
 entirely blocked by the obstruction at the positive i. 
 
 When muscle- or nerve-currents exist, the direction of the exciting 
 current is not indifferent. In this case, closure of a current in the 
 same direction has far more eflFect than closure of one in an opposite 
 direction. 
 
 It still remains to consider some objections which may be made, 
 and which have partly already been made, to these laws, and in 
 particular to the third. 
 
 Hermann comes forward with the objection that he can see no 
 reason for thinking that of two events like in kind — the coming into 
 existence and the disappearance of a current — only the first should 
 produce an effect. 
 
 But according to Hermann's own conception, the coming into 
 existence and the disappearance of a current, in so far as these act 
 as stimuli, are by no means events like in kind ; for it cannot be 
 assumed that the coming into existence of catelectrotonus is an 
 event like in kind to the disappearance of anelectrotonus. But 
 according to my conception, the coming into existence and dis- 
 appearance of a current are like events as regards the way in which 
 they act as stimuli. For with the disappearance of a current there 
 is always associated the coming into existence of a new current, 
 which causes the stimulus. Here there is in both cases a coming 
 into existence of catelectrotonus, so that the two events are clearly 
 absolutely alike in kind, which cannot be said of the other two. 
 For the movement of a body away from its position of equilibrium 
 is not to be assumed without some reason to be an event like in 
 kind to its movement into that position, even if both movements 
 are in the same direction. But according to my conception, only 
 movement away from the position of equilibrium — the coming into 
 existence of cat.electrotonus — has an excitatory effect, not the 
 return to that position — disappearance of anelectrotonus. 
 
 I cannot at present call to mind any state the disappearance of 
 which acts as a stimulus on our organism ; though a large number 
 occur to me which do so at the moment of their coming into 
 existence and durine: their continuance. The electric current 
 
 '■ It is not necessary to consider the effects of exceedingly strong currents, which 
 Hermann (loc. cit. p. 103) has described under the title of the over-maximal phase 
 of the law of contraction ; for in this case, as Hermann has already said, there is only 
 an apparent contradiction of the acknowledged facts of the law of contraction. 
 
 I a 
 
116 THE NATURE OF THE 
 
 may now be placed side by side with these agents ; it differs from 
 them only apparently, since its disappearance only excites when 
 accompanied by a reappearance of the stimulus. Nerve behaves to 
 a current in some respects like an elastic body ; it is not merely 
 disturbed from its equilibrium on closure, but on opening- overshoots 
 its position of equilibrium in returning", before finally coming- to 
 rest. It is in this sense that I understand the almost prophetic 
 sayings of Volta and Ritter in connection with this subject. 
 
 Volta^ imagined that on breaking the conducting circuit an 
 obstacle is suddenly presented to the electric current, against which 
 it strikes, and rebounding gives rise to a wave hi the opposite 
 direction. Ritter^, on the other hand, said that the make-stimulus 
 is imparted to us, while we ourselves impart the break-stimulus. 
 Ritter^s words are : ' We have said that the phenomena occurring 
 on opening the circuit of a galvanic battery form a subject of 
 particular importance. We have now to justify this statement, 
 and do so by drawing attention to the great fact that these 
 phenomena occur at the moment when the organic body and its 
 parts are withdrawn from the influence of the battery. They can 
 in no way however be due to a direct action of the battery; for 
 how could the battery produce such an effect when it is no longer 
 present ? The organism which was in its circuit must itself pro- 
 duce the effect, and can only do so in virtue of having been in 
 the circuit, for apart from this there would have been no effect 
 produced.' If Pfliiger concludes from these words that ' Ritter 
 imagined that a special state produced by the current disappears, 
 which disappearance is accompanied by a contraction, as is every 
 disturbance of the internal equilibriurn of the nerve,' I cannot infer 
 the same from Ritter's words ; for on the disappearance of the 
 current there might just as well reappear a state which had been 
 kept in abeyance by the passage of the current, and act as a 
 stimulus in reappearing. 
 
 Hermann goes on to make the further objection : ' But what 
 seems to me more important is, that if we accept this conception 
 we abandon the advance of most importance embodied in Pfliiger's 
 law, the acknowledgment, namely, that the excitation is due, not 
 to the current itself, but to a definite change of state produced 
 by it in the nerve-electrotonus, as expressly stated in Pfliiger's 
 enunciation.' 
 
 * Du Bois-Reymond, Untersuchungen iiber thierische Elektricitat, Bd. i. p. 314. 
 ''■ Gilbert's Annalen, 1801, xxiii. p. 30, quoted by Pfliiger, Elektrotonus, p. 74. 
 
ELECTRICAL BREAK-EXCITATION. 117 
 
 To tliis I have a double answer. If by experiments and 
 reflection any one reaches a new conception of a complicated process, 
 he has no time to enquire whether his new view is in agreement 
 with older views of the same subject. He has only to ask whether 
 his view is in agreement with the facts. Though the older view 
 may have brought into harmony a whole series of diverse pheno- 
 mena in the most simple and complete manner, it may yet per- 
 haps be erroneous, and, however dearly won, have to be abandoned 
 and to mate way for the others, as has more than once occurred 
 in the history of science. A new theory, therefore, cannot be 
 rejected for the mere reason that it is not entirely in agreement 
 with old ones. 
 
 But in the second place I do not at all see in what way 
 Pfliiger's fundamental law, that the excitation is due not to the 
 current itself but to a definite change of state produced by it in the 
 nerve — electrotonus — is abandoned on my theory. For, according to 
 my view just as much as according to that of Pfliiger, the appear- 
 ance of catelectrotonus acts as a stimulus ; for Pfliiger it is in 
 addition the stronger stimulus ; while the disappearance of anelec- 
 trotonus is for Pfliiger the weaker, but for my theory no stimulus 
 at all. The extremely important fact that the stimulus occurs at 
 the cathode on closure, at the anode on opening, is however also 
 explained by my theory. Besides, if with this fact we often pro- 
 ceed to associate the assertion that the appearance of catelectro- 
 tonus and disappearance of anelectrotonus act as stimuli^ this 
 conclusion is justified only if it can be shown that nothing occurs 
 on closure except the appearance of catelectrotonus, and nothing 
 on opening except disappearance of anelectrotonus. But if it be 
 shown that on opening catelectrotonus appears in addition to 
 anelectrotonus disappearing, the former being an event which is 
 known to act as a stimulus, while it is not known whether the 
 latter also does so, the assumption that in both cases the same 
 cause produces the efiect is the more probable and logical. Time 
 will decide whether my theory constitutes an advance or not. The 
 conditions are certainly not made to appear more complicated by 
 it; it acknowledges only one part of Pfluger's law of excitation, 
 and, so far at least as I can see, does not essentially contradict it. 
 A nerve is excited only if it is moved from its position of equi- 
 librium in a certain direction ; on returning it is not excited. By 
 this return it changes in the one case from a structure of increased, 
 in the other case from one of diminished excitability to one of 
 
118 THE NATURE OF THE 
 
 normal excitability; but neither of these two events alone acts as 
 a stimulus. 
 
 Further on Hermann refers, quoting experiments by Hering and 
 Biedermann, to a series of phenomena from which he attempts to 
 show that my theory is arbitrary, inasmuch as certain excitations 
 can be regarded either as make- or break-excitations. This I 
 fully admit, but would remark that no conclusion can be drawn 
 from such experiments, either for or against any theory. I shall 
 therefore not consider these experiments further. 
 
 But I must refer more particularly to one point, which is 
 essentially of great significance for my theory, and has been 
 brought forward in objection both by myself and by Hermann. It 
 is well known that an opening excitation occurs, not only when 
 the whole exciting current is opened, but also on disappearance of 
 a portion of it. In the first case the stimulus was caused by 
 the appearance of the polarisation-current ; but can the same 
 explanation be offered if part of the original exciting current re- 
 mains ? The accompanying diagrams will express the alterations 
 in the current which occur in this case. 
 
 Let us suppose that a current of strength S passed along a 
 nerve falls to zero at any moment t. According to the ordinary 
 theory it is this fall which produces the excitation, but according 
 to our theory it is the sudden appearance of the polarisation- 
 current in the opposite direction which at its first appearance 
 may have the same strength as the original current. If now 
 instead of letting this cui'rent fall to zero (or, which is the 
 essential point for us, instead of causing a polarisation-current 
 equal to the original current but opposite in direction to appear) 
 we let it fall to half its original strength, it is easy to see that 
 according to the above assumption the polarisation-current will 
 have, not the strength S, but only half the strength. [In other 
 w^ords, if the polarisation-current produced by the passage of an 
 extraneous current S through a nerve is of such strength that they 
 mutually mask each other, the effect of opening the extraneous 
 current is to unmask a counter-current of which the initial inten- 
 sity is equal to S. If S is not opened but merely diminished, the 
 counter current will be similarly unmasked, but will have an 
 initial intensity equal to the amount by which S has been dimin- 
 ished. In the original this is illustrated by a figure which we 
 omit. — Editor.] 
 
 The first question, accordingly, which presents itself is as to the 
 
ELECTRICAL BREAK-EXCITATION. 119 
 
 strenj^th of the polarisation-current at its first appearance as com- 
 pared with the strength of the original current. Is the strength 
 of the former equal to, smaller^ or greater than that of the latter ? 
 I am unfortunately not in a position to give any definite reply to 
 this question. If, however, we reflect that a nerve is an organ 
 which possesses in a high degree the capability of being polarised, 
 if further we call to mind the relative values which du Bois-Rey- 
 mond^ and others ^ have found for the strengths of the original 
 and polarisation-currents, it will not seem too high a value if we 
 assume that the polarisation-current is equal in strength to the 
 original current. 
 
 If, however, this question must for the present be left un- 
 answered, there is a second question which can be decided by 
 experiment. This second question is : To what fraction of its 
 original strength must an exciting current fall in order that an 
 excitation of a nerve and muscle, a so-called 'incomplete' break- 
 contraction, may occur? The way in which this question may 
 be settled is simple. It is only necessary to arrange a mercury 
 key A (see Fig.) in the course of the 
 wire of the compensator DD' and 
 best close to the point I), where the 
 current divides. A second mercury 
 key C is placed in the main circuit 
 UB'B. If the main circuit is closed at 
 C while the derivation-circuit is open, 
 the whole current from the battery 
 B passes through the nerve ; and if 
 the key A in the derivation-circuit 
 is now closed also, the current falls 
 to a certain fraction of its strength, 
 
 which fraction is of course the Fig. i. Main or battery circuit 
 
 n ,1 n ,, T , 1 BCD'D. Derivations or nerve cir- 
 
 smaller the smaller the distance be- cnitDee D". 
 
 tween B and B". The current 
 
 strength could thus be altered from -5' to any strength between 
 
 and I S. 
 
 The most remarkable fact which we observed was that the 
 direction of the original current has a most important influence 
 on the appearance of imperfect break-contractions. It was found 
 as mean of a large number of experiments that an ascending 
 current of the strength of one Daniell, which was of course led 
 
 ^ Ges. Abhandlungen, i. p. 43. ^ See Wiedemann, Electricitatslehre. 
 
120 THE NATURE OF THE 
 
 to a currentless piece of nerve througli unpolarisable electrodes, 
 must fall to about o*6 of its whole strength in order to produce an 
 imperfect break-contraction. A descending current, on the other 
 hand, must fall to at least 0-3 of its original strength to produce 
 a similar excitation. Often, however, after frequent repetition of 
 the second experiment no contraction at all occurs, even though 
 the descending current is reduced to zero. 
 
 The following experiment will serve as an example : — 
 
 Experiment. — Nerve-muscle preparation of a grass-frog of medium size in the 
 moist chamber of the myograph. Unpolarisable electrodes about 7 mm. apart, the 
 distal one about i cm. from the muscle, at an indifference-point. Very weali ascending 
 and descending currents produce equally strong contractions. With key .4 open key B 
 is closed, and key A is afterwards closed regularly after an interval of a second, 
 followed by opening of B, while A remains closed. The first weak imperfect break- 
 contraction with gradually increasing falls of current-strength occurs with DD" = 
 250 cm., and increases in size regularly until DD" nearly = o. The current was 
 ascending, one Daniell being used. A current with DZ*" = 250 cm. corresponds 
 to 0-62 of the original current. It is of importance that in each experiment, 
 ■whether a contraction is produced or not, the key A should remain closed for a 
 short time, in order to get rid of the polarisation-currents. If this is overlooked the 
 results become uncertain and inconstant, so that an imperfect break-contraction first 
 appears at a different point if the current-strengths are altered in the opposite direc- 
 tion, i.e. if the current is first made to fall to zero, and the succeeding falls are made 
 less and less ; in other words, if the distance DD" is gradually increased from its 
 minimum. 
 
 The following fact is also worth mentioning. With plain metallic electrodes of 
 copper, which give a strong polarisation-current, if key B remain open and key A be 
 repeatedly closed, several successive contractions, diminisliing in size, may often occur. 
 These are caused by the polarisation-current in the nerve, which thus does not dis- 
 appear very rapidly, and passes at each closure of A without entirely exhausting 
 itself. It is therefore of great importance to keep the key A closed for a few seconds 
 after each experiment ; the results then become perfectly constant. 
 
 With an ascending current, on the other hand, the first imperfect break-contraction 
 occurs with_DX>"= 100 cm., or sometimes not till DD" = 40 cm., that is to say, with a 
 reduction of current to about 0-22 or ci of its original strength. On further repeti- 
 tion of the same experiment, or with other nerves, no imperfect break-contraction may 
 occur, even on reduction to zero. 
 
 Let us try to explain these results on Hermann's theory that 
 the stimulation in the case of imperfect break-contractions always 
 occurs at the anode of the exciting current, and that there is no 
 reversal of the direction of the exciting current. In the case of 
 an ascending current the stimulation would occur at the anode 
 next the muscle, and would have to traverse a more or less con- 
 siderable extrapolar tract of anelectrotonic nerve in order to 
 reach the muscle. It would thus reach the muscle diminished in 
 strength according to the prevalent view, which may be considered 
 to be firmly established by the experiments of Pfliiger and by 
 Tigerstedt's recent researches. With a descending current the 
 
ELECTRICAL BREAK-EXCITATION. 121 
 
 stimulation, according' to Hermann, would again occur at the 
 anode, which however is now furthest from the muscle. In this 
 ease the excitation, which may be made of the same strength as in 
 the previous case, has now likewise an anodic (intrapolar) tract to 
 traverse, but soon reaches the catelectronic region, which it does 
 not again leave, even when it has passed the neighbourhood of the 
 cathode. This excitation would therefore reach the muscle, not 
 diminished, but increased in intensity; for it follows from the 
 earlier, but particularly from the exceedingly valuable recent re- 
 rearches of Tigerstedt, that an excitation always increases in 
 intensity in passing- the cathode or the catelectrotonic region of 
 a nerve. With a descending current therefore the excitation at 
 the anode would reach the muscle reinforced, with ascending 
 currents it would do so diminished in strength. In other words, 
 the imperfect break-contraction would appear in the case of a 
 descending current with a small fall of current-strength : with 
 ascending currents, on the other hand, only with a much larger 
 fall ; for the former weak excitation would be reinforced in passing 
 the cathode, while the latter stronger one would be weakened in 
 passing along an anodic tract. 
 
 But we see that exactly the opposite actually occurs of what 
 would occur if Hermann's theory were true. Ascending currents 
 have only to be slightly, but descending currents very greatly 
 diminished in strength to produce an imperfect break-contraction. 
 This result is readily intelligible according to our theory ; for the 
 conditions for an imperfect break-contraction are nearly the same 
 as for a complete break-contraction. That is to say, a reversal of 
 the current occurs, and at opening of descending exciting currents 
 the anode of the polarisation after-current intervenes between the 
 point of stimulation and the muscle, while at opening of ascending 
 exciting currents the cathode (at which excitation occurs in every 
 case) of the polarisation after-current is next the nerve. 
 
 Hermann^ will no doubt object to my argument that he can 
 explain all the above results by means of his theory of polarisation 
 increment. For at opening of ascending currents the excitation 
 would occur at the anode and pass to less positive parts of the 
 nerve, while at opening of descending currents the excitation, 
 starting likewise from a point whose polarisation is strongly 
 positive, would pass into parts negatively polarised. In the first 
 
 1 See Hermann, Handbuch der Physiologic, ii. part i. p. 165, and Pfluger's Archiv, 
 vi. p. 359. 
 
122 THE NATURE OF THE 
 
 ease the excitation would be reinforced, in the second weakened. 
 But I can scarcely believe that Hermann will still maintain this 
 theory of polarisation-increment now that Tig-erstedt^ has clearly 
 shown by unimpeachable methods that an excitation which has to 
 pass the cathode is never weakened, but always reinforced, and that 
 the theory of polarisation-increment is therefore incorrect. 
 
 I come now to an additional point which is of importance in 
 explaining- the imperfect break-excitation : this concerns the time 
 during" which the whole current passes along the nerve before 
 beino: reduced to a fraction of its strength. Since a nerve is an 
 organ which becomes strongly polarised, it is intelligible, and has 
 besides been directly demonstrated by du Bois-Reymond^, that a 
 constant current passing through a nerve is weakened by internal 
 polarisation, and, within certain limits, is the more weakened the 
 longer it passes. Although in the latter case the exciting current 
 has already diminished in strength when it is suddenly diminished 
 to a fraction of its height, the fall is no greater than in the case 
 in which there has been no previous diminution. Consequently 
 the same amount of anelectrotonus would disappear in each case, 
 and if this disappearance of anelectrotonus caused the excitation, the 
 fact that differences in the duration of closure affect the break- 
 contraction, which they do very considerably, would be unintel- 
 ligible^. The duration of closure has just as much influence on 
 the appearance of the incomplete break-contraction as it has on 
 that of the complete break-contraction. The longer the current 
 remains closed, the earlier do they both appear. That is to say, 
 with a long closure the imperfect break-contraction appears with 
 quite a small fall as compared with that which is necessary with 
 a short closure. According* to Hermann's theory this is not intel- 
 ligible, but accoi'ding to ours it is readily so. For with a longer 
 closure the polarisation-current which is unmasked is stronger 
 than with a shorter closure. 
 
 The duration of closure varied in our experiments between I, i, 
 and 3 seconds, and the results in each case were in accordance 
 with theory. 
 
 A few words in conclusion as to the influence of current- 
 strength on the appearance of imperfect break-contractions. It 
 
 * Above, pp. 20, 21. 
 
 ^ Gesammelte Abhandlungen, ii. p. 256. 
 
 ^ This reasoning of course applies also to the case of the complete break -contraction. 
 
 * [In the original this statement is illustrated by a diagram. Ed.] 
 
ELECTRICAL BREAK-EXCITATION. 123 
 
 is easy to observe, for both ascending and descending- currents, that 
 within the current-strengths used by us (y\ Daniell to about 
 7 Daniells) imperfect break-contractions appear the earher the 
 stronger the currents are. Thus with a strong current excitation 
 is caused by a fall of a smaller fraction of its height than is 
 necessary with a weaker one. (I shall not give detailed numbers 
 here, since these facts are only mentioned for the sake of complete- 
 ness. They can be as well explained by Hermann's theory as by 
 ours). For if a current of strength S is allowed to fall a half, it 
 produces the polarisation-current of which the initial strength is 
 half /S*; whereas a current of half the strength similarly allowed to 
 fall a half gives only a current of a quarter S. The first current 
 has not therefore to fall a half in order to produce the same 
 effect as is produced by the second when suddenly diminished in 
 the same proportion. This fact is however evidently also in 
 accordance with Hermann's theory ; for in the one case there is 
 
 a fall of current-strength of — , in the second of For produc- 
 
 2 4 
 
 ing weak electromotive forces we used a Noe's Thermopyle, the 
 flame of which was placed at different distances from its middle. 
 
 As the reader will now see, the results of all these experiments 
 on the imperfect break-contraction are not only intelligible on our 
 theory, but present almost insuperable difficulties to the ordinary 
 explanation. For the present I can see no single objection which 
 can be brought against my own and Tigerstedt's conception of the 
 nature of the break-excitation. Of course I shall not persist in my 
 opinion if any one can prove to me that the mere disappearance of 
 anelectrotonus produces excitation. In any case no one can deny 
 that I have referred about 8o per cent, of all so-called break- 
 excitations to make-excitations, and that the polarisation after- 
 current has great significance as regards the causation of the 
 remaining 20 per cent,, as even Hermann acknowledges. But it 
 is more than probable, and to me it seems certain, that the polari- 
 sation-current is the sole cause. I must, however, add that it 
 would have been desirable if I could have directly proved this by 
 experiment for the incomplete break-excitation. The experiment 
 would require to be arranged so that when the fall of current- 
 strength that first produced an incomplete break-excitation was 
 reached, the current passing through the derivation-circuit 
 DAD"e'e should be immediately led to a galvanometer. This 
 current must then be in the opposite direction to that of the 
 
124 THE NATURE OF THE ELECTRICAL BREAK-EXCITATION. 
 
 original exciting current, provided that the whole of the fibres of a 
 nerve-trunk give as strong a polarisation-current as those whose 
 excitation we observed in the case of the gastrocnemius^ which is 
 not altogether certain. Unfortunately I do not possess the ap- 
 paratus necessary for this experiment, and I must ask the reader to 
 be content for the present with the evidence which I have been 
 able to bring forward. 
 
V. 
 
 ON NEEVE-EXCITATION 
 
 NEEVE-CUERENT. 
 
 Bt ewald seeing, 
 
 'Professor of Physiology in the University of Prague. 
 
 Sitzungsbericlite der kaiserlichen Akademie der Wissenschaften, Vienna, vol. Ixxxv. 
 Part 3, p. 237. 1882. 
 
V. 
 
 At the end of my communication^ on the direct excitation of 
 muscle by the muscle-current, I promised to communicate analogous 
 observations on nerves, and to prove that the excitation of nerves 
 by the nerve-current occurs more frequently and is more easily 
 accomplished than has been hitherto supposed. But in the winter 
 of 1879 the frogs at my disposal suddenly became unfit for such 
 experiments, and I was obliged to interrupt them. 
 
 Since then Kiihne^ has communicated observations on the excita- 
 tion of nerve by its own current, which I shall have occasion to 
 confirm in the following pages. My researches, however, in many 
 respects go beyond those of Kiihne, especially in this, that I have 
 succeeded in eliciting from nerves a true secondary contraction 
 and secondary tetanus, which neither du Bois-Reymond nor Kiihne 
 held to be possible, and I have so definitely learned the condi- 
 tions necessary to the success of these experiments, that they will 
 henceforth be easily and successfully repeated by anyone. 
 
 1. Excitation of the Nerve by its own current. 
 
 Du Bois-Reymond' s Ex2)eriment. 
 
 As is well-known, du Bois-Beymond showed that it is possible 
 to excite, by means of its own current, a nerve which is provided 
 with a transverse section. Placing the sciatic, separated from its 
 centre, and in connection with its muscles, so that transverse and 
 longitudinal sections respectively were in contact with the cushions 
 
 ^ Sitzungsberichte, Ixxix. 3. p. 7. 
 
 - Untersuch. aus d. physiol. Inst, zu Heidelberg, iii. p, 90, 1879. 
 
128 ON NERVE-EXCITATION BY THE NERVE-CDKRENT. 
 
 of his electrodes of the old form covered with films of albumen 
 (which had been previously tested by the multiplier and found to 
 be free from polarity), he made and broke the nerve-current by 
 means of a mercury key. ' The leg contracted at closure, and often 
 at opening- ; in a few cases it contracted at opening only^.' 
 
 Later, du Bois-Reymond simplified the experiment by dispensing 
 with all metallic conductors ; he placed two conducting cushions, 
 separated by an interval of a few mm. on a non-conducting support, 
 then laid the nerve upon the two cushions with transverse section 
 upon one cushion, longitudinal surface on the other; then, by means 
 of a third cushion he completed the circuit, thus closing the nerve- 
 current ^. Since this method allows of a rapid closure of the current 
 by the sudden application of the third cushion, but of a less rapid 
 opening by its removal, it is advisable to arrange the two cushions 
 upon which the nerve rests so as to overhang the edge of a glass 
 plate ; a vessel filled with salt-solution can then be raised quickly, 
 so that the overhanging tips of the cushions suddenly dip into the 
 fluid ; or it can be quickly lowered, so as to brgak the nerve-current 
 suddenly. 
 
 Instead of the cushions, one may use blocks of saline clay, which 
 have the advantage that they may be modelled to any desired shape. 
 This is the method which Kiihne has preferred, and which I also 
 have frequently used. I kneaded the clay with concentrated salt- 
 solution, and used the same material to complete the circuit; in 
 order to protect the nerve from the action of the concentrated 
 salt-solution, I placed upon the spots, where the nerve w^as to lie, 
 a thin layer of clay, made with salt-solution o-6°/o5 or a bit of 
 blotting-paper moistened with the same. If sufficiently excitable 
 nerves are to be had, 0-6 °/o solution may be used both in preparing 
 the blocks, and for closing the circuit. The contractions which 
 one obtains in this manner are, as Kuhne also states, very energetic 
 in favourable cases. The opening contractions are generally weaker, 
 or do not occur at all ; sometimes, however, they are strong, as 
 indeed has already been stated by Kiihne. I may also confirm the 
 statement of du Bois-Reymond to the effect that sometimes an 
 opening contraction alone appears. 
 
 In all cases where it is desired to compare make- with break- 
 contractions, or make-contractions taken with different lengths of 
 the interpolar region, it is advisable to employ du Bois-Reymond's 
 
 1 Unters. fiber thier. Elektr. ii. p. 273, 1849. 
 ' Fortschritte d. Physik. iv. Jahrg. p. 314, 1852. 
 
ON NERVE-EXCITATION BY THE NERVE-CURRENT. 129 
 
 original method by metallic closure of the circuit, and to make 
 use of some arrangement providing for equal rapidity of closure 
 and opening. And with this one must naturally be most careful as 
 to the electrical uniformity of the unpolarisable electrodes, since 
 any nerve that will react to its own current, will also be extra- 
 ordinarily sensitive to the weakest additional current, if the latter 
 should be in the same direction as the former in the interpolar 
 region. Closure, by means of a liquid or by a conducting cushion, 
 cannot, at any rate by hand, be performed with sufficient uniformity, 
 and if the contractions are not very strong, one obtains, in spite of 
 the greatest practicable uniformity of speed in raising and lowering 
 the liquid which completes the circuit, a succession of contractions 
 of very different strengths. 
 
 Excitable preparations gave me strong make- and break -contrac- 
 tions even when the portion of nerve between the blocks of clay was as 
 much as one centimeter long, and in one case I still obtained make- 
 contractions when it reached a length of 25 millimeters. I have not 
 closely examined the conditions of the appearance or predominance 
 of the break-contraction. 
 
 I must notj however, omit to mention that this hreaJc-contractlon, 
 if one admits certain assumj)tio7is concerning the so-called internal 
 closure of the nerve-current, is to he regarded, not as a jprojper oj)ening- 
 contraction, hut as a contraction due to the closing of the nerve-current. 
 
 Since it is possible, in the above-described manner, to obtain on 
 a single preparation a long series of closing- and opening-contrac- 
 tions, all that is necessary in order to tetanise the nerve is to make 
 the closures and openings succeed one another with sufficient rapidity 
 just as in the case of a battery-current. Kiihne used for this 
 purpose a vibrating mercury key. I have considered it advisable 
 to avoid the use of any metal in this experiment, partly in order to 
 dispense with the testing of electrodes by the galvanometer, partly 
 in order to demonstrate a ' tetanus without metals.' To this end I 
 fixed the cushion which serves to complete the circuit in du Bois- 
 Reymond's experiment, to the end of a rod which could be made 
 to vibrate, so that the cushion could, just like the hammer of 
 Heidenhain's tetanomotor, be applied to and removed from the 
 blocks of clay in a rapid rhythm. At first, I used for this purpose 
 a large Neef's hammer, which sometimes gave me a very strong 
 tetanus, but often failed, because the amplitude of movement of the 
 cushion was too small, so that threads of fluid between it and the 
 blocks of clay prevented the circuit from being broken. I eventually 
 
 K 
 
130 ON NERYE-EXCITATION BY THE NERVE-CURRENT. 
 
 used a little apparatus, specially constructed for the purpose, which 
 I propose to call a tetanomotor for the production of tetanus without 
 metals. In this apparatus the rapid raising* and lowering of the 
 cushion is effected as follows : The teeth of a toothed wheel raise 
 a lever, to the free end of which the cushion is attached, and a 
 spring" depresses the lever after each elevation ; by means of a 
 screw the amplitude of movement can be exactly regulated, and 
 by varying the rapidity of revolution of the toothed wheel the 
 frequency of the makes and breaks may be varied. 
 
 Excitation, of a nerve by its fall upon a vioist conductor. 
 
 The easiest way to effect an excitation of nerve by its own current 
 is to let its end fall upon some currentless moist conductor. It is less 
 advisable to let it fall on platinum or amalgamated zinc, partly on 
 account of the instantaneous polarisation, partly on account of the 
 admixture of currents due to electrical differences at the points of 
 metal with which the nerve is in contact. The possibility of 
 mechanical excitation, which can only occm- when the nerve is too dry, 
 is easily excluded by control-experiments with non-conductors. 
 
 The fall of the end of a nerve upon a drop of lymph, serum, or 
 normal salt-solution, is, in general, successful only once, because, on 
 removal of the nerve, the adherent fluid permanently short circuits 
 the current, but the experiment can be repeated several times if 
 the nerve is allowed to fall upon a coagulated drop of blood, or a 
 block of clay made wdth salt-solution -6 p.c. 
 
 It need hardly be explained, after what has been said, that a 
 contraction follows also if the end of nerve is dropped upon a com- 
 pletely currentless muscle or any other currentless organ. If one 
 drops the sciatic nerve left in connection \\ith the leg-muscles upon 
 these muscles in the ordinary way, the consequent contraction is 
 by no means a sufficient proof that the nerve has been excited by 
 a muscle-current, although indeed this will generally be the case. In 
 this favourite but not at all unequivocal experiment, the contraction 
 may be due to any one of three causes if the leg has been stripped 
 of its skin : either (i) that the leg gives an ascending current sufficient 
 for excitation, which traverses the nerve from its point of entrance 
 into the muscle to the point where it is in contact with the muscles 
 of the leg, or (2) that the nerve lies upon electrically different points 
 of muscle, or (3) that the nerve-current is suddenly closed through 
 the muscle. It is easy to understand that in most cases all three 
 factors may concur or conflict with one another. 
 
ON NERVE-EXCITATION BY THE NERVE-CURRENT. 131 
 
 Czermak found ^ 'that a current-testing frog's limb gives a 
 closing contraction, if its nerve is raised on a glass rod and suddenly 
 dropped upon the natural longitudinal section of a rabbit's or 
 pigeon's muscle, which happens to be in partial idio-muscular con- 
 traction, so that it is simultaneously in contact with a contracted 
 and a non-contracted portion of such muscle.' He remarks also 
 that 'very excitable frog's limbs' likewise contract if their nerves 
 fall upon an unaltered natural longitudinal section of the muscle, 
 and he explains this by the assumption that there are w^eak currents 
 between different points of such a surface. But Czermak found also 
 that frogs' legs of the highest degree of excitability gave distinct 
 and even very strong contractions if he let their nerves fall upon a 
 quiescent portion of a rabbit's intestine, or upon the kidney or liver 
 of the same animal. To infer from such contractions that there is 
 current from these tissues is not admissible, until it has been shown 
 by adequate control experiments that the nerve does not react, to 
 the closure of its own current. 
 
 Kiihne" has already called attention to the fact that Bonders was 
 mistaken in regarding, as an effect of currents from the cardiac 
 muscle, the contractions of the frog's leg which may be obtained by 
 dropping its nerve on the pericardium during the cardiac pause, 
 and he has shown that similar contractions may be obtained when 
 the nerve is dropped on to the pericardium after the heart has been 
 removed. 
 
 Interference between a nerve-current and a lattery-ctirrent. 
 
 The effects of electrical excitation of a nerve in the vicinity of its 
 transverse section depend so much upon the nerve-current which is 
 present as to be quite unintelligible if this cm*rent is not taken into 
 account. 
 
 According to the report of the fifty-fourth meeting of German 
 savants at Salzburg, Griitzner has recently called attention to the 
 paramount importance of the nerve-current to the excitability of 
 the nerve, or of any other nerve lying in contact with it — a con- 
 sideration which is already, to some extent, suggested by Pfliiger's 
 ' Researches on Electrotonus^.' 
 
 He says^, 'As conducing to the effect of excitation with constant 
 
 * tJber secundare Zuckung von theilweise gereizten Muskeln aus Sitzungsberichte 
 dieser Akad. 1857, '^'^^ Gesammelte Schriften, i. p. 429, Leipzig, 1879. 
 
 ^ Loc. cit. p. 85. ^ p. 151. * Loc. cif. p. 119. 
 
 K 2 
 
132 ON NERVE-EXCITATION BY THE NERVE-CURRENT. 
 
 or with induction currents, the currents of the nerve itself are of 
 great importance, especially such as can be shown to exist in the 
 neig-hbourhood of the section or of branches, as well as in that of 
 the physiological terminations peripheral and central — such currents 
 adding themselves algebraically to the exciting currents, and so 
 increasing or diminishing them. In this wise are explained the state- 
 ments concerning the differences of excitability at different points 
 in the course of a nerve-trunk, as well as the varying results of 
 excitation with weak currents.' 
 
 Since a more detailed account may be expected from Griitzner at 
 some future time, I will here confine myself to the record of a 
 particular instance which belongs to this class of phenomena, and 
 which is of interest because a contraction actually caused by closure 
 of a derived nerve-current appears as if caused by opening of a 
 weak battery-cmTcnt, and conversely a contraction caused by opening 
 of a branch of a nerve-current appears as if due to closure of a weak 
 battery-current ; a case then in w/nc/i, as a matter of fact, a make- 
 contraction 7nay he mistaken for a hreak-contr action, and vice versa. 
 
 Assuming that a rheochord is used to send by derivation a fraction 
 of a battery-current through the nerve, and that the key is intro- 
 duced between the rheochord and the unpolarisahle electrodes, then, 
 if the nerve is in contact with the electrodes by a transverse and 
 by a longitudinal surface, the key will short-circuit the nerve- 
 current, and possibly a contraction will ensue, even when the 
 rheochord is not in connection with the battery. 
 
 Under similar conditions, but more seldom, there may be a 
 contraction when one opens the nerve-rheochord circuit, or, as I 
 shall term it, 'nerve-circuit.' For this to happen it is necessary to 
 choose a preparation which, under the above conditions, gives break- 
 as well as make-contraction with its nerve-current. 
 
 If now the rheochord is connected with the battery, so that the 
 derived current is ascending in the nerve, this current will be 
 opposed to the nerve-current in the nerve-circuit, and may com- 
 pensate it exactly if of appropriate strength^ which particular case 
 we will now assume. The consequence will be that closure of the 
 nerve-circuit will no longer give any contraction, because, in spite 
 of the closure in the conductors between nerve and rheochord, no 
 
 ' I need hardly remark that 1 speak here of compensation of the nerve-current not 
 in any absolute sense, but only in accordance with the conventional sense of the 
 word. For there is no question here of a true compensation of current in the nerve, 
 but only of a compensation of the branch current in the conductors which connect 
 the nerve and rheochord. 
 
ON NERVE-EXCITATION BY THE NERVE-CURRENT. 133 
 
 new current comes into existence whether as a branch of the battery- 
 current or of the nerve-current, and consequently no electrical change 
 of the nerve. Nor can there be a contraction at opening of the 
 nerve-circuit, although such was present in the absence of the 
 battery-current. 
 
 The preparation will act quite differently if a key is placed 
 between battery and rheochord — in the 'battery-circuit ' as I shall 
 henceforth designate it for the sake of brevity. If the latter is 
 closed subsequently to closure of the nerve-circuit, this will be 
 equivalent, as regards the nerve, to opening of the nerve-current, 
 and the muscle will contract. And now what appears to he the closure 
 effect of a current derived from the lattery, is, in reality, the effect of 
 Ireaking the nerve-current'^. Under such conditions then a contrac- 
 tion is obtained if with previously closed nerve-circuit the battery- 
 circuit is closed; but, on the contrary, no contraction if with 
 previously closed battery-circuit the nerve-circuit is closed. 
 
 If when both circuits are closed the key of the battery-circuit 
 is opened, the compensation of the nerve-current in the nerve- 
 circuit is removed, and the muscle contracts because the nerve- 
 current is at the same time reclosed. Thus, what now appears to 
 be an opening effect of the derived current of the battery, is, in 
 reality, a closure effect of the nerve-current, so that under the 
 conditions in question a contraction is obtained if, after previous 
 closure of both circuits, the battery-circuit is opened ; but, on the 
 contrary, no contraction if the nerve-circuit is opened. 
 
 If the branch current from the battery is too weak to compensate 
 the nerve-current in the nerve-circuit, its effects will manifest them- 
 selves in the same direction as above, but not in the same degree. 
 If, on the other hand, it is somewhat stronger than is required for 
 compensation, the nerve will, with the closure of both circuits, be 
 traversed by an ascending current — a remainder, so to speak, of 
 the branch current of the battery. And if the nerve-circuit is 
 closed after previous closure of the battery-circuit, there is no 
 contraction, provided the branch current of the battery does not 
 exceed a certain strength ; if, on the other hand, the battery-circuit 
 is closed after previous closure of the nerve-circuit, the opening 
 effect of the nerve-current adds itself to the weak, and of itself 
 inadequate, closure-effect of the branch current of the battery, and 
 
 ' That under certain conditions the break -contraction due to the nerve-current 
 must in reality be considered as a new make-contraction by the nerve-current, has 
 already been mentioned on p. 129. 
 
134 ON NERVE-EXCITATION BY THE NERVE-CURRENT. 
 
 a contraction is obtained. If with previously closed battery-circuit 
 the nerve-circuit is opened, there is no contraction, assuming* always 
 that the ascending* battery-current in the nerve is not so strong* as 
 to give an openings contraction in spite of its partial compensation 
 by the nerve-current. If, on the other hand, with previously closed 
 nerve-circuit the battery-circuit is opened, the nerve-current finds by 
 this closure, a new channel of derivation, and a contraction follows, 
 which is, in this case, further strengthened by the influence of the 
 voltaic alternative. Then, if we may assume that this is due to the 
 circumstance that a spot at which a cm-rent enters becomes more 
 excitable to a current which makes its exit by it, the closing* exci- 
 tation of the nerve-current will thereby be rendered all the stronger. 
 As to the influence which may be exerted by polarisation currents, 
 I will here entirely pass it over, since such currents are also closed 
 in the nerve-circuit whenever the battery- circuit is opened. 
 
 All the above-mentioned circumstances must be taken into ac- 
 count, if the preparation does not g-ive any contraction to simple 
 closure or opening* of the nerve-circuit in the absence of any 
 battery-current. For, in this case also, the nerve-current is present, 
 and its influence manifests itself in the same direction. 
 
 One thus obtains, toith the loeahest hrancli curreyits from the battery 
 provided that they have the nerve by its transverse section, an ' opening 
 contraction ' first at opening of the battery -circuit, and only loith con- 
 siderably stronger currents at opening of the nerve-circuit ; and in 
 correspondence with this a ' closure contraction ' shows itself first at 
 closure of the battery -circuit, and only with stronger currents at closure 
 of the nerve-circuit. 
 
 Up to this point we have assumed that the weak branch- 
 current from the battery leaves the nerve by its transverse section. 
 In the opposite case the consequence will be different according* as 
 the battery-circuit is closed after previous closure of the nerve- 
 circuit, or vice versa. For, in the former case, a current — that of 
 the nerve namely — already enters by the electrode which is in 
 contact with the longitudinal surface, and this current will simply 
 be increased by the additional branch current of the battery. But 
 if the nerve-circuit is not closed until after the battery-circuit has 
 been closed, the nerve- and battery-currents add themselves together 
 at the moment of closure, and consequently the effect of such 
 closure will be greater. Similarly, by opening the nerve-circuit 
 both currents are simultaneously broken. 
 
ON NERVE-EXCITATION BY THE NERVE-CURRENT. 135 
 
 As a matter of fact, heginnlng toitJi the toeahest ctirrents entering the 
 nerve hy its transverse section, one first obtains the mahe -contraction at 
 closure of the nerve-circuit, and only with stronger currents at closure 
 of the lattery -circuit as well. And similarly in the case of the breaks 
 contraction. 
 
 Biedermann ^ has recently examined the relative inefficiency of 
 currents which pass out of a nerve at its transverse section ; a 
 phenomenon which he has found especially striking in the case 
 of the nerves of warm-blooded animals. Concerning this matter 
 one must be careful to distinguish, in a current that flows through 
 the nerve, between such branches as have their course in the ex- 
 citable substance of the nerve fibres from those which traverse 
 constituents of the nerve which simply do duty as physical con- 
 ductors, in other words, the physiologically important from the 
 physiologically unessential. The points where any such physio- 
 logically important current enters or leaves the excitable substance 
 of a nerve fibre are to be regarded as its physiological anode or 
 kathode. If now a certain part of the proper nervous substance of a 
 fibre is electromotive, as we may, according to Hermann, assume 
 the part adjoining an artificial transverse section to be, and if the 
 current arising at this spot is opposed in direction to any branch 
 current from the battery, then it is clear that a passage of the 
 latter through the electromotive portion of the fibre can only 
 occur when the nerve-current which already exists at that spot is 
 overcompensated by the current which is led in, this requiring, of 
 course, a relatively strong battery-current. Supposing further that 
 the electrical current excites the nerve only at its point of exit 
 from the excitable substance, then an excitation hy a current leaving 
 the nerve by a transverse section would not he j^ossible tmless this current 
 were sufficient to over compensate the nerve-current at its source. 
 
 As to how far the current, once it has exceeded the local strength 
 or density requisite to this internal compensation of the nerve- 
 current, can act as an excitant to the electromotive portion of the 
 nervous substance by means of its residue, so to speak, must for the 
 present remain unanswered on account of the alteration of the 
 nerve-substance which obtains here. Beyond this point the current 
 no longer finds- any excitable substance. 
 
 Similar considerations may be urged concerning the relative 
 
 ' Beitrage zur allg. Nerven- und Muskelphysiologie, Sitzungsberichte, Vienna, vol. 
 Ixxxiii, Part iii, p. 289, 1 88 1. 
 
136 ON NERVE-EXCITATION BY THE NERVE-CURRENT. 
 
 inefficacy of currents having' their point of exit from muscle by 
 an artificial transverse section — a fact which has also been pointed 
 out by Biedermann. 
 
 I need hardly say that what has been advanced above stands 
 or falls with the theoretical assumptions upon which it rests. 
 
 Concerning the liability of frogs kept in the cold to tetanic excitation. 
 
 The frogs which I used for my experiments had been kept for 
 several months in a cellar, at a temperature of about o°. Since, 
 for each series of experiments, I used them immediately after they 
 had been fetched from the cellar, or, at any rate, after they had 
 been placed between a double window, at the temperature of o°, 
 these frogs showed, with few exceptions, the disposition to enter 
 into tetanus which has already been remarked upon by Pfliiger •^. 
 
 This was especially the case wnth regard to Rana esculenta ; Rana 
 temporaria was much less inclined to fall into tetanus. What 
 I have to mention, therefore, relates in general to water-frogs ; 
 in experiments which were made on land-frogs, the fact is spe- 
 cially, mentioned. 
 
 If I divided the spinal cord of a R. esculenta which had been 
 kept in the cold (or, briefly expressed, a ' cooled frog '), tetanus 
 very rarely supervened in the inferior extremities ; but, on the 
 other hand, it did so almost constantly if the incision through the 
 spine was made so low as to affect the nerve-roots ; and the same 
 thing happened if I cut through the sciatic plexus. The strength 
 of the tetanus was very variable ; sometimes the limbs passed into 
 a prolonged and steady rigidity, which then gradually relaxed 
 through a stage of clonic spasms ; sometimes the spasms were clonic 
 from the first. Its duration was equally various, and occasionally 
 reached as much as four minutes. As a rule, the spasmodic stage 
 immediately followed the contractions caused by the incision, but 
 sometimes several seconds elapsed after the cessation of these 
 contractions — as many as six — before the preparation, which had 
 become quite quiescent after the incision, began to twitch again. 
 This was always a sign of a comparatively slight tendency towards 
 tetanus. 
 
 I have made this remarkable tetanus after section of nerves 
 (which was not mentioned by Pfliiger, though it certainly must 
 have been known to him as to other observers) the subject of 
 
 ' Untersuch. iiber die Physiol, des Elektrotonus, p. 133. 
 
ON NERVE-EXCITATION BY THE NERVE-CURRENT. 137 
 
 special investigation, but I shall here mention only such points as 
 bear on what follows. 
 
 To these belongs, in the first place, the fact that a second section, 
 when the muscles have become quiescent, generally causes a 
 renewal of the spasms, though they are as a rule weaker after 
 each new section. 
 
 In the second place, it is to be remarked that the spasms are 
 usually stronger and more lasting the higher the nerve has been 
 cut. If I cut through simultaneously the sciatic plexus of one side, 
 and on the other the sciatic nerve in the thigh or just above the 
 knee, and then compared the spasms on both sides, after fixing the 
 knee so as to prevent any movements caused by the muscles of the 
 thigh, I found that the spasms were as a rule much weaker and 
 shorter in the last named limb, and that occasionally they were quite 
 absent even when the other side entered into active movements. 
 But if only a few such experiments are made it may perhaps be 
 found that this rule does hold good, for it is liable to many excep- 
 tions, the causes of which can only occasionally be assigned. Thus, for 
 example, the neighbourhood of injured and therefore electromotive 
 muscles may cause modifications. But in the large number of 
 experiments I made, the rule in question came out quite clearly. 
 
 Preparations which gave strong tetanus by section of the nerves, 
 often reacted with a tetanus previously to such section, to closure 
 or opening, even when the battery-currents were only just strong 
 enough to act. Simple twitches were less frequ,ent, and appeared 
 for the most part when the battery-current was closed several times 
 in succession in the same direction ; the first closure, for instance, 
 giving closure tetanus; the second, clonic disturbance ; the third, a 
 simple twitch only. Reversal of the current then restored tetanus. 
 All this appeared in a still more striking manner if I cut the 
 nerve, and applied the electrodes to the transverse and longitudinal 
 sections. With descending currents, even the weakest, I then 
 obtained a strong closure tetanus, and with ascending currents an 
 equally strong opening tetanus. And preparations which did not 
 respond by a tetanus to the section of their nerves, fell into a 
 strong closure tetanus (provided they were taken from cooled 
 frogs) as soon as I made a descending current, or broke an ascend- 
 ing current, having thus in the first case its point of entrance, in 
 the second its point of exit, at the transverse section. In these 
 experiments the unpolarisable electrodes were only from 2 to 3 mm. 
 apart. 
 
138 ON NERVE-EXCITATION BY THE NERVE-CUREENT. 
 
 I have already mentioned that the land-frogs which I used 
 showed much less disposition to tetanus after nerve-section than 
 did water-frogs. I, however, obtained from the former also 
 strong closure or opening tetanus by excitation of the transverse 
 section. 
 
 Pfliiger ^ has attempted to explain the great disposition of frogs 
 brought from a cold to a warm place to fall into a tetanic state, by 
 some 'transposition of atoms' brought about by the influence of 
 warmth. But without any desire to call in question the influence 
 of warmth on a preparation, I must mention that frogs which had 
 been kept in a cold room gave a tetanus of the corresponding Kmb 
 after section of the sciatic plexus, even when the section was made 
 in the same cold room and with cold scissors, so that any warming 
 of the preparation was out of the question. They also reacted 
 tetanically to weak battery-currents. 
 
 The tetanus consequent upon nerve section can be completely 
 arrested, or, at any rate, much diminished, by weak ascending 
 currents, if one electrode is applied to the transverse section, and 
 the other to a neighbom-ing point of the longitudinal section. It 
 recommences, however, when the current is broken. A strong 
 muscle-current suffices to arrest it. If I placed the recently cut 
 nerve between two muscles possessing a transverse section, in such 
 a way that the transverse sections of the nerve and of both muscles 
 lay in the same plane, I observed that the tetanus forthwith 
 diminished or entirely disappeared, reappearing after removal of 
 the muscles. Thus it happens that, even with the most excitable 
 frogs, the tetanus is generally absent if the whole thigh is severed 
 at one blow of the scissors, because the currents of the cut muscles 
 forthwith act upon the cut nerve. 
 
 If the sciatic nerve on one side is laid bare, and divided after 
 careful isolation from the muscles, the leg, if the preparation is 
 a sensitive one, becomes tetanised ; while an incision on the other 
 side, and at the same level, but passing through the nerve and all the 
 muscles, causes a contraction of the leg only during the operation, 
 and is followed by complete relaxation. If after section of the sciatic 
 plexus a strong tetanus should have supervened in the leg of that 
 side, it is at once abolished by sudden amputation through the 
 thigh ; but if thereafter the two branches of the sciatic nerve 
 above the knee are divided, tetanic agitation of the leg reappears. 
 
 ' Untersuch. iiber die Physiol, dea Elektrotonua, p. 133, 1859. 
 
ON NERVE-EXCITATION BY THE NERVE-CURRENT. 139 
 
 Closure tetanus of the nerve hj/ the agency of its own current. 
 
 The great liability of cooled frog-s to tetanic excitation explains 
 the fact that even the derivation or short circuiting of the nerve- 
 current is sufficient in suitable preparations to cause a closure tetanus 
 which in favourable cases may be lasting, and of such steadi- 
 ness and energy as to be comparable to the tetanus produced 
 by induction currents. In most cases, it is true, the tetanus is 
 unsteady or clonic, and the derivation of the nerve-current often 
 brings about only an irregular agitation of the muscles ; in short, 
 there are here manifested all those appearances which one mav 
 observe when the closure of a weak battery-current sent through 
 the nerve causes an enduring excitation of the nerve, and not 
 merely closure contractions. 
 
 Whichever of the above described methods of establishing de- 
 rivation be used, there will always result in certain cases, with 
 very excitable preparations, an enduring excitation of the nerve, 
 sometimes after the first closure only, sometimes after repeated 
 ones. This lasting excitation may show itself by weak, irregular, 
 or even rhythmical movements of single parts, or by stronger clonic 
 spasms, or finally by well marked stretching tetanus. If to the 
 original tendency towards tetanic excitation there be added any 
 drying of the nerve, a permanent closure tetanus appears all the 
 more readily. If the preparation is capable of responding to deriva- 
 tion of its nerve-current by a lasting excitation, it is no longer so 
 necessary to provide for the rapid closure of the derivation-circuit, 
 because, even failing a closure contraction, the persistent excitation 
 is still appreciable. 
 
 An observation belonging here is as follows : I allowed the 
 end of a dependent nerve of a galvanoscopic frog's limb to dip 
 into salt-solution 0'6 °/o. I then made above the level of the 
 fluid, a kind of transverse section by crushing the nerve with fine 
 forceps. I allowed the contractions caused by this proceeding 
 to subside, and when the preparation had become quite quiescent, 
 I gradually raised the vessel containing the salt-solution so that 
 the nerve was immersed in it deeper and deeper. As soon as the 
 surface of the fluid rose above the crushed part a tetanus of the 
 preparation appeared, which became stronger as I continued to 
 raise the vessel, and reached its maximum when the surface of the 
 fluid was about i cm. above the crushed part. If I raised the 
 vessel still higher, the tetanus diminished, without, however, entirely 
 
140 ON NERVE-EXCITATION BY THE NERVE-CURRENT. 
 
 disappearing". If I now lowered the vessel, the tetanus increased, 
 and was again at its maximum when the surface of the fluid was 
 I cm. above the injured part. Further lowering of the vessel 
 caused diminution of the tetanus, and finally its disappearance 
 when the fluid was no longer above the crushed point. I could 
 repeat this observation several times with the same preparation. 
 
 In by far the greater number of trials this experiment did not in- 
 deed give such evident results ; at the moment that the nerve current 
 is closed through the fluid a clonic agitation of the preparation 
 occurs which soon disappears, or else the experiment is successful at 
 the first trial only, or not at all. But in such cases even one may 
 still get a reaction if the fluid is raised as rapidly as possible ; a 
 closure contraction then follows, as is generall}^ the rule whenever 
 the nerve-current of a moderately excitable preparation is suddenly 
 closed externally. 
 
 According to previous observers ^, a reaction of the nerve-muscle 
 preparation cannot be obtained if the transverse section is con- 
 nected with a point of the longitudinal surface as near to it as 
 possible. I have repeatedly succeeded in obtaining contraction 
 under these circumstances, because I used preparations which were 
 disposed to fall into tetanic excitation. If I mjide the experiment 
 as soon as the preparation entirely ceased to contract in conse- 
 quence of the freshly made section, and taking care to avoid any 
 wetting of the nerve, the making contact between transverse and 
 longitudinal sections gave me strong clonic agitation of particular 
 groups of muscles. Moreover, I can point to instances in which 
 there occurred strong closure contraction followed by clonic agita- 
 tion, provided I had been able to establish contact with suflicient 
 rapidity, — rather a difficult matter here. 
 
 The same holds good for an experiment which was made by 
 Galvani, and which du Bois-Reymond designated as the fundamental 
 experiment in the electro-physics of nerve, although he was not 
 very successful in its repetition. ' Galvani arranged the nerve of 
 a galvanoscopic leg as an open loop, and allowed the nerve of a 
 second and otherwise completely isolated leg to fall upon the first 
 nerve in such a way that the transverse section of the first nerve 
 constituted one of the two points of contact. In favourable cases 
 both legs contracted ^.' In speaking of the excitation of muscle 
 
 ' See du Bois-Reymond, Unters. iiber thier. Elektr. ii. p, 272, and Kiihne, 
 loc. cit. 
 
 ' Du Bois-R-eymond, Unters. iiber thier. Elektr. i. p. 273. 
 
ON NERVE-EXCITATION BY THE NERVE-CUEEENT. 141 
 
 by its own current, I have already mentioned that if the transversely 
 cut ends of two Sartorius muscles are allowed to fall one upon the 
 other with sufficient rapidity, and in such a way that the trans- 
 verse sections are in the same plane, so constituting- a single 
 transverse section of both muscles, no contraction ever follows, 
 whereas there may be a strong- contraction of both muscles if one 
 end of muscle falls upon the other so that the one is a prolongation 
 of the other, but with the two ends overlapping-. Now in the 
 first case the two muscle-currents are opposed, and compensate 
 each other ; in the second, they have the same direction in the 
 circuit, formed by the two ends of the muscles. Making- use of 
 this artifice, Kiihne ^ has in fact obtained contractions with the 
 nerve-current as well. I have likewise arrang-ed the experiment 
 on this plan, and found that it succeeds with certainty on sufficiently 
 excitable frogs, provided only care be taken that the two ends of 
 nerve have not too much flviid adhering- to them. As soon as they 
 were let fall one upon the other, the two legs contracted vigorously, 
 and sometimes even passed into a prolonged clonic excitation. 
 But if I dropped one nerve on the other so that the two trans- 
 verse sections were in one plane, neither contraction nor tetanus 
 occurred. 
 
 If the sciatic plexus of a cooled frog has been divided close to the 
 vertebrae, or the nerves ligatured at that point, and then cut through 
 above it, and if the cut bundle of nerves is raised, firstly by means 
 of a fine glass rod, secondly by means of the thread, during the tetanus 
 or clonus of the leg consequent on section of the nerves, it is fre- 
 quently observed that the intensity of the tetanus or the form of 
 the clonus is at once altered ; if then the nerve is put back into its 
 moist and conducting bed, the original type of contraction forthwith 
 returns. One may repeat this with success several times. If the 
 contractions of the leg are allowed to come to an end while the 
 plexus is kept raised, and if the plexus is then replaced upon the 
 ileo-coccygeal muscle, the contractions frequently reappear, although 
 generally to a much less degree. This only occurs provided the 
 ileo-coccygeal muscle has not been injured in the course of prepara- 
 tion. The cause of these phenomena (when muscle-currents do not 
 play a part in them) is to be found simply in this, that when the 
 plexus lies on the posterior wall of the pelvis, the nerve-current 
 is closed externally to the nerve, whereas this is not the case when 
 the plexus is raised. 
 
 ^ Loc. cil. p. 92. 
 
142 ON NERVE-EXCITATION BY THE NERYE-CURRENT. 
 
 The fact that sufficiently excitable nerves fall into persistent 
 excitation if an external closui'e of low resistance is offered to their 
 proper current, sug-g-ests that the above-described tetanic excitation 
 which supervenes in cooled frogs after section of the sciatic nerve 
 or plexus, is also simply due to the current which arises in con- 
 sequence of the section. It must be remembered that this current 
 exists in every single fibre, and that the sheath of each fibre, the 
 fluid present between the fibres, as well as the common sheath of 
 all the fibres, provide channels for external closure to each indi- 
 vidual current. Consequently so soon as the nerve-current is pro- 
 duced by the transverse section, all the fibres fall under the 
 influence of the nerve-current, the direction of which in the fibre 
 is descending, and must excite it at the points of exit. Just as a 
 weak battery-current excites tetanically any very excitable nerve, 
 so, in the same way apparently must the proper current of nerve 
 produce its excitation ; and just as the lasting action of a current 
 led in artificially is a gradually diminishing one, so the excita- 
 tory action of the nerve-current on the fibres of the nerve must 
 be of limited duration. Apparently, therefore, there is nothing 
 opposed to the assumption that the more or less prolonged tetanic or 
 clonic excitation of the muscles of cooled frogs consequent upon 
 section of their nerves, is due to the nerve-current which results from 
 section. 
 
 That with this cause other causes participate, as regards the 
 results of nerve section, is not, however, excluded by the above 
 assumption. Just as any li\'ing substance can adapt itself to any 
 persistent stimulus, so the nervous substance also adapts itself to a 
 weak persistent current, whether this be an artificially applied one, 
 or its own proper current. Provisionally one might, in accord- 
 ance with custom, speak of this as ' fatigue.' In consequence of 
 this adaptation or fatigue, the closure tetanus, or, in our case, 
 the tetanus after nerve section comes to an end. A new ex- 
 ternal closure of the nerve-current effects, however, a sudden 
 redistribution of current in the nerve, and therewith a renewed 
 excitation. 
 
 Since other mechanical, thermic, and chemical agents act just 
 like section in giving rise to the development of current in the 
 nerve, there are many considerations regarding the mode of action 
 of these stimuli and regarding the nature of nerve excitation which 
 are suggested by these phenomena. I will not, however, yield 
 to the temptation to enter upon theoretical considerations, just 
 
ON NERVE-EXCITATION BY THE NEEVE-CURRENT. 143 
 
 as I refrained from doing concerning the excitation of muscle, and 
 will confine myself to the suggestion which has just been made. 
 
 Uxcitafiou of the transverse section of a nerve hi/ fluids. 
 
 Just as a muscle contracts so soon as its transverse section comes 
 in contact with a conducting fluid, so does the nerve of a sufficiently 
 excitable nerve-muscle preparation contract when the end of its 
 nerve is dipped into a conducting fluid ; and the contraction is 
 apparently simultaneous with the contact. One is inclined to 
 regard this contraction as caused merely by the sudden external 
 closure of the nerve-current, and in harmony with this view is 
 the fact that badly conducting fluids, such as alcohol, solution of 
 corrosive sublimate, &c., do not give rise to the contraction at the 
 moment of contact. But the assumption of a simultaneous chemical 
 excitation of the nerve substance that has been laid bare by the 
 incision, and is still perhaps, to some extent, excitable, cannot, in 
 this case, be excluded in so easy and convincing a fashion as in the 
 corresponding experiment on muscle. 
 
 As is well-known, Eckhard^, who was the first to carry out 
 systematic experiments relating to the excitation by fluids of 
 the transverse section of nerves, considers this excitation to be 
 exclusively chemical. He obtained no action by dipping the nerve 
 into water, bisulphide of carbon, solutions of metallic salts (with the 
 exception of nitrate of silver), organic acids such as gallic acid, and 
 volatile oils. 
 
 The fluids which Eckhard found efficacious produced effects of 
 two kinds. The first kind of action showed itself in a contrac- 
 tion at the instant of contact, the second followed if the nerve were 
 dipped deeper, and consisted of contractions which began sooner or 
 later after immersion, spread over a greater or smaller number of 
 muscular fasciculi successively, and lasted for a very variable time, 
 according to the agent employed, &c. 
 
 According to Eckhard the first contraction may be avoided if, 
 instead of cutting the nerve, one ties it and immerses the tied end. 
 With solutions of the fixed alcalies, provided they were not too 
 weak, Eckhard obtained regularly both the contraction immediately 
 following contact of the transverse section, and the subsequent 
 contractions ; mineral acids gave the second kind of contractions, 
 and only very rarely the first. All the other fluids which Eckhard 
 
 ^ Die chemische Eeizung der motorisclien Frosclinerveii. Zeitschr. f. rat. Medic. 
 Neue Folge, i. p. 302, 185 1. 
 
144 ON NEKVE-EXCITATION BY THE NERVE-CURKENT. 
 
 used with success — solutions of the haloid and neutral salts of the 
 alcalies and alcaline earths, alcohol, ether, anhydrous acetic acid, 
 saturated solutions of tartaric acid and of sugar — g-ave only the 
 second kind of contractions. 
 
 From all these facts it appears to me that Eckhard worked with 
 much less excitable preparations than I did. Although it was not 
 my intention to make a special study of the excitatory action of 
 fluids on the transverse sections of nerve, I incidentally made 
 experiments of this nature with various fluids. And I found not 
 only that concentrated solutions of sulphate of copper and sulphate 
 of zinc (which, according to Eckhard, produce no eflect) do produce 
 effects in so far as they give contractions at the moment of their 
 contact with the transverse section, hut also that the same effect is 
 obtained by the use of a concentrated solution of sodium chloride, 
 and of 0-6 % solution of the same, fluids which, according to 
 Eckhard, only give the second kind of contractions. Sulphuric 
 acid, of a strength of lo "/oj gave me the same result. Like 
 Eckhard, I found the fixed alcalies especially active. 
 
 A small drop placed on a glass slip suffices for the production of 
 the contraction by contact with sodium chloride, caustic potash, 
 sodium carbonate, sulphuric acid, sulphate of zinc or of copper. I 
 have seen solutions of sodium chloride, or of caustic potash, produce 
 the effect when I had used for the purpose a tiny drop, sufficient 
 to moisten the transverse section only, or at most the very border 
 of the longitudinal surface in addition. I have not used such small 
 drops of other fluids. 
 
 Now, although on account of what has been said above con- 
 cerning the effect of external closure of the nerve-current, I can- 
 not doubt that it plays an important part in the contractions 
 occurring at the moment of contact, I am not in a position to be 
 able to exclude the possible conjoint action of chemical stimulation. 
 It surprised me that fluids differing but slightly in conductivity 
 appeared nevertheless to behave quite differently as to the certainty 
 with which they provoke contraction by their ajiplication to the 
 transverse section. The great activity of the fixed alcalies may 
 perhaps be explained by this, that the nerve is more rapidly and 
 easily moistened by them than by other fluids, so that a more 
 rapid electrical change is produced in the nerve. But I must leave 
 it unsettled whether this is the reason of their apparently greater 
 efficacy. 
 
 There are two circumstances which notably increase the difficulty 
 
ON NERVE-EXCITATION BY THE NERVE-CUREENT. 145 
 
 of this investig-ation. In the first place the specifically lighter 
 nerve frequently fails to sink into a fluid which is of greater density 
 than it, and rests on the surface, so that the external closure of the 
 nerve-current is relatively slowly and incompletely established. In 
 the second place one rarely succeeds in again obtaining the con- 
 traction by a renewed contact of the same transverse section with 
 a fluid. I have already mentioned above, that the clinging of fluid 
 to a nerve, and the permanent derivation thereby established, cause 
 still more interference in this case than in that of muscle. Eckhard 
 says indeed {loc. cit. p. 3 1 8), ' that one never succeeds in eliciting re- 
 peated contractions by repeated immersion of the surface of section.' 
 I have, however, succeeded several times in obtaining rej^eated con- 
 tractions at the instant of immersion, still this was, on the whole, a 
 rarity. As regards the question merely of demonstrating the action 
 of the nerve-current, it would be of little interest to prosecute 
 these experiments ; but in relation to the question of chemical 
 excitation it would be of considerable importance if one could dis- 
 tinguish between the electrical and the possibly coincident chemical 
 stimulation at immersion of the transverse section. If anyone chose 
 to maintain that the so-called chemical excitation is only electrical, 
 and is efficient only in so far as it gives rise to current as well as 
 chemical alteration, this could hardly be disproved in the present 
 state of our knowledge. 
 
 The same difficulty which is presented by the contraction occur- 
 ring at contact of the transverse section, repeats itself as regards 
 the subsequent irregular contractions. For when the nerve-current 
 was externally closed by other means we saw the same thing occur, 
 provided the preparations were very excitable. The question must 
 eventually arise as regards such preparations, what is to be regarded 
 as due to the electrical, and what to the chemical excitation, when a 
 long-enduring agitation results from dipping the nerve into a fluid. 
 If it is possible (as in the experiment with normal saline, described 
 at p. 139) to repeat, with always the same effect, the immersion 
 of an electromotive portion of nerve, it is allowable to exclude 
 any chemical action. But if the experiment succeeds only at the 
 first immersion, its interpretation remains very doubtful. The 
 experiments of Eckhard, which cannot but awaken the keen interest 
 of any one who repeats them, and especially the remarkable action 
 of the fixed alcalies, were well worth a renewed methodical inves- 
 tigation. 
 
146 ON NERVE-EXCITATION BY THE NEKVE-CURKENT. 
 
 2. On true secondary contraction and secondary tetanus 
 derived from nerve. 
 
 On the lehaviour of 7ierve when li is excited hy induction currents in 
 the vicinity of a transverse section. 
 
 Before passing- to the subject-matter proper of this section, I wall 
 refer to the peculiar behaviour which a nerve manifests when it is 
 excited by induction currents in the neig-hbourhood of its transverse 
 section, for this is of importance to what has to follow. 
 
 If to a freshly cut or tied nerve the electrodes of the secondary 
 coil are applied, separated from each other by a distance of only 
 two to three mm., so that one electrode is at the point of section 
 or of ligature, the very weakest currents g-ive strong contractions 
 if the direction of the break-currents is abterminal in the nerve. 
 With an atterminal direction of these currents the contraction is 
 much weaker, or absent, in spite of unaltered position of the elec- 
 trodes and unaltered current strength. If, the dii'ection of the 
 break-currents being abterminal, the electrodes are pushed further 
 and further from the transverse section, the effects diminish rapidly 
 to zero. If, on the other hand, the direction of the break-currents 
 is atterminal, the effect rapidly increases as the electrodes are moved 
 away from the transverse section, soon reaching a maximum, and 
 finally diminishing and disappearing as the electrodes are moved still 
 further away. All this naturally presupposes that one is working 
 with minimal current strengths and freshly-made transverse sections. 
 
 The rule here given is however liable to exceptions. If there are 
 points of section of branches in the vicinity of the transverse section, 
 if the latter is not quite fresh, if the nerve has been injured in its 
 course, or sometimes from other undemonstrable causes, departures 
 from the rule may present themselves. I am here concerned only 
 with the rule itself which anyone will easily find to hold good. 
 From it one may generally determine, with complete certainty, the 
 direction of the break-current in given cases. 
 
 This behaviour of the nerves is especially valuable in all those 
 cases in which it is desired without further trouble to exclude 
 unipolar effects. For on the one hand the low intensity of the 
 currents used makes these effects very improbable, and on the other 
 they ought to persist or indeed become stronger as the electrodes are 
 moved away from the transverse section, and consequently towards 
 where the nerve produces its action ; whereas, indeed, under the 
 circumstances in question, the opposite is the case. 
 
ON NERVE-EXCITATION BY THE NERVE-CURRENT. 147 
 
 It was during- experiments on excitation of the central end of 
 the vagus that I was first struck by the fact that it is necessary, . 
 in order to obtain an effect, to bring the secondary much closer to 
 the primary coil when the break-induction shocks are atterminal 
 than when they are abterminal. Since then I have noted and 
 found this confirmed in the excitation of various nerves of warm- 
 blooded animals. Nerves of frogs behave similarly. But in these, 
 unless it is specially desired to investigate excitability in the 
 neighbourhood of the transverse section, it is seldom necessary to 
 place an electrode at or near a transverse section or ligatured point, 
 whereas in warm-blooded animals, owing" to shortness of the end of 
 nerve, this frequently has to be done. In such cases, it is easy 
 to convince oneself of the great difference in the effect when the 
 electrodes are shifted, or when the current is reversed. Nor does 
 the Helmholtz modification entirely abolish the difference. 
 
 Although the peculiar conditions of excitability of nerve in the 
 neighbourhood of its transverse section have been frequently inves- 
 tigated since Heidenhain called attention to them, I think that 
 the facts which are here mentioned, and which are, in many 
 respects, imjoortant, have not been sufficiently regarded. To ex- 
 perimenters familiar with the department of nerve physiology, this 
 is perhaps no novel statement. 
 
 On the excitation of nerve conseqiient upon the negative variation 
 of the current of another nerve lying vj)on it. 
 
 If no one has yet succeeded in obtaining true secondary contraction 
 from nerve, this want of success must be attributed to the fact that 
 all the attainable conditions which favour it have not been utilised. 
 When I took advantage of these, I succeeded in eliciting from 
 nerve true secondary contraction and secondary tetanus which 
 were as strong and as lasting as if the nerve of the secondary 
 preparation were itself directly excited electrically. 
 
 In designating these as ' true ' secondary e fleets in contradistinc- 
 tion from the secondary e fleets discovered by du Bois-Reymond, and 
 caused by the electrotonic currents of the primary nerve, I am 
 justified on the one hand, by the convenience of the term, on the 
 other, by the consideration that the secondary effects to be presently 
 described are truly analogous with the secondary contraction from 
 muscle. This, as the long-est known kind of secondary contrac- 
 tion, deserves to be regarded as the prototype of such actions. 
 
 L 2 
 
148 ON NERVE-EXCITATION BY THE NERVE-CURRENT. 
 
 The great excitability exhibited by the nerves of cooled frog's in 
 the vicinity of a freshly-made transverse section, led me to choose 
 the central end of such a nerve as the point of excitation of the 
 secondary preparation. I laid bare the two sciatic nerves from the 
 spinal column to near the knee, and used one of these nerves, 
 divided at both ends, as the primary, the other still connected with 
 the leg as the secondary preparation. The peripheral end of the 
 primary nerve was placed upon the central end of the secondary, 
 so that both nerves lay close together for a space of five or six 
 millimeters, and with their transverse sections in one plane. With 
 this arrangement, the current of one nerve, so to speak, compensates 
 that of the other. Granting now that in consequence of a momen- 
 tary excitation of the primary nerve, the current suddenly vanishes 
 in its peripheral end (negative variation down to zero), then the 
 compensation of the current of the overling* secondary nerve is 
 thereby suddenly removed ; at this moment the end of the primary 
 nerve being suddenly deprived of current acts merely as an external 
 closure to the current of the secondary nerve, and the latter, by this 
 sudden derivation of its proper current, must suffer a weak excita- 
 tion. But if, as Bernstein once supposed^, the current of the 
 excited nerve actually alters its direction, this current will, after 
 its reversal, act upon the secondary nerve as a weak descending" 
 current, which is added to the proper current in that nerve at the 
 moment that this is closed externally. 
 
 Before tetanising the central end of the primary nerve by means 
 of weak induction currents, I provided its peripheral end, and at 
 the same time the central end of the adjacent secondary nerve, with 
 a new transverse section common to both, by a single stroke of the 
 scissors, / tlms actually succeeded in provoking iveak tetanic agita- 
 tion of the secotidary preparation at each tetanisation of the primary 
 nerve. Of escape of current or of unipolar action there could be no 
 question, seeing that the exciting currents acting upon the central 
 end of the primary nerve wei'e so weak as to be only just efficacious 
 when the electrodes were placed near the transverse section, and 
 that all secondary action was absent when I moved the electrodes 
 to a part of the primary nerve nearer to the secondar}^ preparation. 
 
 In so far as one may regard as excluded any excitation of the 
 secondary nerve by electrotonic currents — the distance from the point 
 of excitation to that of contact with the secondary nerve being- 
 
 ^ Untersuchungen iiber den Erregungsvorgang im Nerven und Muskelsystem, § 9, 
 1871. 
 
ON NERVE-EXCITATION BY THE NERVE-CURRENT. 149 
 
 equal to almost the entire length of a sciatic nerve, and the distance 
 of secondary from primary coil being- as much as 40 cm. (i Dan. 
 6000 turns of sec. coil.) — it ivas settled hy this experiment that true 
 secondary action is possible from nerve to nerve. If I placed the two 
 nerves together, so that one was a prolongation of the other, with 
 their two ends however in contact for 5 or 6 mm., all secondary 
 action was absent, as indeed was to be expected on theoretical 
 grounds. 
 
 But still the above-described weak and exceptional results were 
 not sufficient to satisfy me. I saw that everything depended upon 
 the most intimate contact between the two ends of nerve^ and that 
 if it were at all incomplete, or if the nerves had but a very little 
 lymph or salt solution adhering to them, the result was interfered 
 with. And so I came to think that it is very unpractical to take great 
 trouble in applying two nerves to each other, instead of using a 
 preparation in which the two bundles of nerve-fibres, which are to 
 serve as primary and secondary respectively, lie together naturally 
 in the same sheath. 
 
 I decapitated a cooled frog, removed the upper extremities and 
 intestines, laid bare the sciatic nerve above the knee, ligatured with 
 the same thread its two branches, divided them below the ligature, 
 prepared the nerve up to near the point of origin of the branches to 
 the muscles of the thigh, divided the sciatic plexus, waited until the 
 consequent disturbance had subsided, and then, when all the muscles 
 were quiescent, excited the end of the nerve at the knee with weak 
 induction currents. The muscles, the nerves of which were still in 
 connection vjith the jplex%is, entered forthwith into strong secondary 
 tetanus, just as if I had excited these nerves directly. 
 
 Often as I have since repeated this experiment it has never failed, 
 provided only that the preparation was sufficiently excitable to 
 give, in addition to the single contraction, a weak and fugitive mus- 
 cular disturbance in the thigh when the sciatic plexus was divided. 
 But I have also made the experiment with almost constantly success- 
 ful result, on cooled frogs — land frogs in particular — which did not 
 give the tetanic agitation after nerve section, if only I excited the 
 peripheral end of the sciatic very shortly after section. In those 
 cases, it is true, a strong secondary tetanus was not reached ; there 
 was for the most part only a tetanic or clonic agitation of the 
 muscles, just as may be seen wdth nerve-muscle preparations, the 
 nerves of which are excited with excessively weak, that is to say 
 barely efficacious, induction currents. 
 
150 ON NERVE-EXCITATION BY THE NERVE-CURRENT. 
 
 In this experiment the secondary tetanus appears not merel}'^ in 
 the muscles which are innervated from the femoral branches of the 
 sciatic, but also in the muscles which are supplied by the crural 
 nerve, which arises from the plexus itself. If the femoral branch is 
 divided, and the sciatic isolated up to the origin of the crural, exci- 
 tation of the end of the' sciatic at the knee gives rise to secondary 
 tetanus at the muscles innervated by the crural. I have also 
 isolated the sciatic nerve in its whole course down to the branch 
 which goes to the semimembranosus and rectus internus, then 
 divided the sciatic above the knee and removed all the rest of the 
 limb with the exception of the above-named muscles, and finally 
 separated the plexus from the spinal column, so that the whole 
 nerve, with these two muscles in connection with it, was completely 
 isolated. These muscles became secondarily tetanised when I 
 appKed weak induction currents to the peripheral end of the 
 nerve. 
 
 If instead of dividing the plexus I divided the spinal column 
 with its nerves between the seventh and eighth, or between the 
 eighth and ninth vertebrae, I only exceptionally obtained secondary 
 tetanus. But on the other hand, I have recorded one case in which 
 the division had been made between the sixth and seventh vertebrae, 
 and in which excitation of the peripheral end of the sciatic gave 
 rise to very strong secondary tetanus of the muscles of the thigh 
 and of the trunk on the corresponding side of the body. The mode 
 of this tetanus did not allow of its being considered as reflex, 
 although the incision had reached the lowest part of the spinal 
 cord — which moreover, according to Engelhai*d\ is incapable of 
 giving reflex movements. 
 
 If after division of the plexus I severed and excited one only 
 of the two branches into which the sciatic divides above the knee, 
 I obtained often, but not always, secondary effects which manifested 
 themselves, partly in the muscles of the thigh, partly in the leg 
 muscles supplied by the other of the two branches. But still 
 this secondary action rarely reached to a steady enduring tetanus, 
 and generally went no further than a tetanic agitation. Once 
 indeed I obtained secondary action by tetanising the sural nerve 
 which I had divided at the ankle joint and prepared up to the knee: 
 the toes spread out vigorously, and the gastrocnemius fell into 
 clonic contractions. Unfortunately the muscles of the thigh were 
 not observed, and I did not succeed in repeating the experiment 
 1 Jliiller's Arch. f. Anat. u. Physiol. Jalirg. 1841, p. 208. 
 
ON NERVE-EXCITATION BY THE NERVE-CUKRENT. 151 
 
 ■with any satisfactory result. I may mention finally that I have 
 repeatedly obtained secondary tetanus of the muscles supplied by 
 the axillary plexus, by peripheral excitation of the ulnar nerve 
 prepared up to the axilla and divided above the elbow, the axillary 
 plexus having been previously divided. 
 
 It can not be supposed that the above-described secondary effects 
 can be attributed to escape of current or to unipolar excitation. 
 The electrodes— fine platinum wires, 2 or at most 3 mm. apart — were 
 placed quite close to the transverse section, so that the break induc- 
 tion currents were ascending* (abterminal) in the nerve, and the 
 induction coils used with one Daniell were separated by a distance 
 of from 50 to 40 cm. or at most 30 cm. for comparatively less 
 excitable preparations. Such weak induction currents, although 
 acting energetically close to the transverse section, generally lost 
 all effect as soon as I moved the electrodes further away from the 
 transverse section towards the muscles. Whenever this was not 
 the case I did not regard the experiment as completel}^ trustworthy. 
 If I made the induction currents descending (atterminal) in the 
 nerve, I obtained no effect with currents which had acted ener- 
 getically in the opposite direction, so long as the electrodes were 
 very near the transverse section. Effects were produced however, 
 either with the same or with a somewhat increased current strength, 
 as soon as I moved up the electrodes from the end of the nerve 
 at the knee ; generally disappearing again if I brought the elec- 
 trodes still nearer to the muscles. There were however exceptions 
 to this. These facts are sufficient to make the above-named objection 
 inadmissible. But as regards the possibility that my experiments 
 have in them nothing but the paradoxical contraction described by 
 du Bois-Ueymond, I shall presentl}^ discuss this matter for the 
 satisfaction of those who may still have doubts. 
 
 I have not confined myself to obtaining secondary tetanus by 
 tetanising the primary nerve with induction currents, but I have 
 also used with the same result weak battery-currents, rendered 
 tetanising by means of a rotating commutator. 
 
 Further, I have obtained strong secondary single contractions 
 in great number by means of single induction shocks and by 
 closure of weak battery-currents. What weak currents suffice for 
 this purpose is shown, for instance, by an experiment in which, 
 with one Daniell and only 4*5 cm. of the double wire of du Bois- 
 Heymond's rheochord as resistance, I obtained a strong secondary 
 contraction at each closure, when I made this weak current 
 
152 ON NERVE-EXCITATION BY THE NERVE-CURKENT. 
 
 ascending" (abterminal) in the end of a sciatic nerve which I had 
 di\-ided above the knee. 
 
 Just as it is possible to obtain secondary contraction or secondary 
 tetanus from any muscle which is set in action by excitation of its 
 nerve, when the nerve of the secondary is laid upon the muscle of 
 the primaiy preparation, so it is possible to obtain similar effects 
 in muscles in which contraction is produced by secondary action 
 derived from nerve. If I caused secondary contraction of the thig-h 
 muscles as above described by the application of weak induction 
 currents to the extremity of the sciatic nerve at the knee, then a 
 second nerve-muscle preparation with its nerve Ipng* upon the thigh- 
 muscles of the first frog" gave what I may call 'tertiary' contraction, 
 and it passed into strong or ' tertiaiy ' tetanus even when the currents 
 applied to the first nerve were only just sufficient to tetanise it. 
 
 I have already pointed out above that the secondary contractions 
 from nerve are most easily obtained just after the plexus has been 
 di^-ided. It follows that with the repetition of the experiment 
 upon otherwise quite unaltered preparations, the secondary action 
 diminishes more and more and finally ceases. This is not at all 
 due to fatigue alone, but chiefly to the decline of the nerve excita- 
 bility which had been pre\'iously enhanced owing to the section. 
 For even if one waits a little after di\dsion of the nerve without 
 subjecting it to excitation, and then excites, the secondary action is 
 weak or absent. W ith comparatively sluggish preparations a delay 
 of a few seconds is usually sufficient for this to be shown, but with 
 good preparations several minutes on an average are required. But 
 if a new transverse section is now made at the plexus, or if it is liga- 
 tured or crushed below the point where it was first divided, excitation 
 of the peripheral end of the nerve, when the muscles have come to 
 rest, gives, with good preparations, renewed and strong secondary 
 effects. In this way, and without altering the seat of peripheral , 
 excitation, I have reproduced the secondary effects with the same 
 preparation after they had vanished, as many as six times. It how- 
 ever often happens Tvath less excitable preparations, that a secondary 
 effect is obtained only after the first severance of the plexus. This 
 rapid disappearance of the heightened excitability consequent upon 
 section which is necessary to the success of the experiment, is the 
 real reason why these have escaped the attention of pre\aous investi- 
 gators and especially of du Bois-Reymond, who must otherwise 
 certainly have met with them in the course of his experiments on 
 paradoxical contraction. 
 
ON NERVE-EXCITATION BY THE NERVE- CURRENT. 153 
 
 If the sciatic nerve is cut a few millimeters above the origin of 
 its femoral branches, excitation of the nerve at the knee with the 
 above-mentioned weak currents usually g-ives a very weak secondary 
 effect or none at all. A similar result follows if the sciatic is cut 
 above its division into tibial and peroneal nerves, and either of 
 these is divided as far down as possible and excited at its peripheral 
 end. Perhaps this is due to the fact that at these points the 
 fibres of the branches in question have already separated from 
 the remaining- fibres so as to be gathered together in separate 
 bundles, although they are contained in the same common sheath 
 with these remaining fibres. In the plexus itself the fibres which 
 are submitted to direct excitation are aj^parently mingled with 
 those which are to be indirectly excited, a disposition which must 
 be very favourable to secondary excitation. But besides this it 
 is possible that a greater excitability of the fibres within the 
 plexus is to be taken into account. I have sought in vain to obtain 
 secondary contraction by dipping into hot water the foot of the leg 
 of which the plexus had been just previously divided ; and I have 
 in vain suddenly frozen, or touched with a hot glass rod, the peri- 
 pheral end of a sciatic nerv^e which had been laid bare and cut 
 above the knee ; secondary contraction was never obtained. But 
 with crushing of the nerve I have seen a very weak and partial 
 contraction of a thigh-muscle twice and three times in succession 
 at each crushing, in two cases out of three preparations. I should 
 not like however to lay much stress upon this, seeing the large num- 
 ber of unsuccessful experiments I have made. 
 
 These negative results need not be surprising if one bears in 
 mind how difficult it is to produce a strong negative variation of 
 the nerve-current by non- electrical excitation, and how much more 
 certainly the secondary contraction derived from muscle itself is 
 produced, when the nerve of the primary preparation is excited 
 electrically, than when other stimuli are employed. Indeed a steady 
 and enduring tetanus can be obtained only by electrical excitation 
 of the primary nerve. 
 
 Distinction between the true secondary contraction and the paradoxical 
 contraction of du Bois-Heymond. 
 
 As is well known, du Bois-Reymond observed that when a 
 sciatic nerve which had been severed from the spinal cord was 
 excited electrically, muscles supplied by branches which left the 
 nerve above the point of excitation, contracted, and that he called 
 
154 ON NERVE-EXCITATION BY THE NERVE-CURRENT. 
 
 these contractions 'paradoxical.' He observed contraction of the 
 muscles supplied by the peroneal nerve when the tibial nerve was 
 excited, and vice versa, and Avith excitation of the sciatic below the 
 orig-in of it^ femoral branches he saw contraction of the muscles sup- 
 plied by these nerves. He showed that the contractions in question 
 are caused by electrotonic currents which pass from excited to 
 adjacent fibres. And in point of fact the conditions of his experi- 
 ments were such that there can be no doubt concerning the correct- 
 ness of this explanation. For he sent the whole current of a 
 Grove's cell through the nerve by means of platinum electrodes, 
 and observed that the paradoxical contraction aj^peared with greater 
 certainty and streng-th as the electrodes were brought closer to the 
 point where the indirectly excited nerve branched off. According; 
 to a figure which he gives of the arrangement of an experiment, the 
 point where the nerve branches and the nearest electrode are distant 
 from each other by not more than about 7 nam. 
 
 I have experimented under quite different conditions. In using 
 the battery-current, for instance, the weak currents which are 
 obtained with one Daniell and only i to 2 mm. of the du Bois- 
 Reymond rheochord consisting in a double wire of platinum, were 
 sufficient. The platinum electrodes were distant from each other 
 about 2 mm., and between the femoral branches the muscles of 
 which were excited to secondary action and the nearest electrode, 
 there was a length of nerve of at least 30 mm. If I used un- 
 polarisable electrodes, no more than 25 mm. of the rheochord 
 wire were required to give the same result. With induction 
 currents, using only i Dan., it was often sufficient to place the 
 secondary coil at a distance of 50 cm. in order to obtain strong 
 secondary effects in the thigh when the sciatic nerve was tetanised 
 at the knee. I have also repeatedly divided the femoral branches, 
 so that none remained attached to the sciatic nerve between the 
 electrodes and the origin of the crural nerve from the plexus 40 to 
 42 mm. higher up. Under these circumstances the muscles 
 supplied by the crural nerve were excited to strong secondary 
 tetanus when I tetanised the sciatic at the knee with the coil 
 standing at a distance of about 40 cm. Whereas du Bois- 
 Reymond obtained stronger secondary effects the nearer he moved 
 the electrodes to the branch, I obtained progressively weaker 
 effects which soon disappeared, if I moved the electrodes further 
 from the section, i.e. nearer to the branches that were to be 
 secondarily excited, especially with ascending (abterminal) currents. 
 
ON NEKVE-EXCITATION BY THE NEKVE-CUREENT. 155 
 
 This indeed only held good with very weak currents. If I made 
 use of considerably stronger currents, I could then obtain effects 
 by placing the electrodes at various parts of the nerve, and if I 
 placed them very near the origin of the femoral branches, the 
 secondary action was sometimes suddenly and strongly increased 
 owing to the presence of electrotonic currents. From all this it 
 appears that the effects observed by me are above all a function of 
 the excitability of the part of the nerve to which the electrical 
 stimulus is applied, while those which du Bois-Reymond has 
 recorded are rather a function of current-strength and of the dis- 
 tance of the point of excitation from the secondarily excited nerve. 
 
 In point of fact these experiments of mine and those of du 
 Bois-Reymond, in spite of their superficial analogy, are concerned 
 with two fundamentally distinct classes of phenomena. The true 
 secondary contraction is dependent upon the excitation propagated 
 to the central end of the divided nerve and consequent variation of 
 the nerve-current. If, instead of secondarily excited fibres in con- 
 tiguity with the directly excited fibres in the central end of the 
 nerve, we imagine a muscle in natural connection with the directly 
 excited fibres, we may lay down the following rule as regards the 
 result of our experiments : Tf this mitsde contracts strongly, the 
 mjiscle of a secondarilij excited nerve will also contract. This rule 
 signifies, then, that the true secondary contraction must follow the 
 law of contraction, just as would be the case for a muscle in connec- 
 tion with the central end of the primarly excited nerve. AjDparent 
 departures from this rule are simply due to the fact that the excita- 
 bility of the secondary nerve does not remain so constant as that of 
 a muscle, but rapidly diminishes after the cross section is made. 
 
 The paradoxical contractions of du Bois-Reymond follow quite 
 another rule. An example will show this best. The sciatic nerve 
 is divided above the knee, isolated up to its femoral branches and 
 placed upon unpolarisable electrodes, so that the interpolar region 
 is not more than a few millimeters long, and the transverse section 
 is but a few millimeters distant from the nearest electrode. The 
 rheochord is in conjunction with 3 or 4 Daniells, and one pole of 
 the battery is joined to the removable binding-screw of the rheo- 
 chord, so that a weak branch current or the total battery current 
 can be alternately sent through the nerve. All the plugs are put 
 in, the slider of the rheochord is placed at zero, and the direction 
 of current is such that the derived current is descending (atter- 
 minal) in the sciatic nerve. The sciatic plexus is now divided, and 
 
156 ON NEKVE-EXCITATION BY THE NERVE-CURRENT. 
 
 by moving the slider the derived current is increased until there is 
 a secondary contraction at closure. This requires, according to the 
 excitability of the preparation, either an extremely small distance 
 of the rheochord slider, or with comparatively poor preparations, 
 a distance of at most a few centimeters. At opening of this weak 
 current there is no secondary contraction. The above-mentioned 
 rheochord binding-screw is now withdrawn, so that the entire 
 current of the battery is sent through the nerve : there is no longer 
 any secondanj contraction at closure-, because the electrotonic altera- 
 tion of the nerve at the anode blocks the excitation proceeding 
 from the kathode. Opening of this current gives strong secondary 
 contraction with sufficiently excitable preparations ; with such as are 
 less so, the contraction fails in this case also. If now the weak 
 derived current is restored, its closure gives a contraction, and its 
 opening has no effect, whereas renewed establishment of the strong 
 current has no effect at closure, but gives an opening contraction. 
 In this manner the alternate excitation with weak and strong 
 currents can be carried on so long as the preparation remains excit- 
 able ; the weak current giving only a make-contraction, the strong 
 current only a break-contraction, or, if the excitability is already 
 depressed, no contraction at all, at a time when the weak current 
 can still give rise to secondary contraction at make. 
 
 But if in the case of the strong current the electrodes are brought 
 so near to the femoral branches of the nerve that electrotonic 
 currents act upon the fibres of these branches which are contiguous 
 with excited fibres, the paradoxical contractions of du Bois-Reymond 
 are produced, and now the strong descending {atterminaT) battery -current 
 gives a make-contraction ; but the weal current applied to the same 
 point generally gives no contraction, for its strength was such that it 
 was only just adequate to evoke secondary contraction when it was 
 applied to the lower transverse section, i. e. to the most excitable 
 part of the nerve. 
 
 It appears thus, that, as regards the true secondary contraction, 
 everything depends upon whether a sufficiently strong excitation 
 wave is propagated to the central end of the primary nerve-fibres, 
 there to give rise to excitation of adjacent fibres by means of the 
 negative variation of the nerve-current. Very weak currents are 
 sufficient for this, provided they are applied to the most excitable 
 portion of the primary nerve. The distance between the excited 
 point and the origin of the fibres to be excited secondarily is in the 
 main indifferent. But in the case of du Bois-Reymond's secondary 
 
ON NEKVE-EXCITATION BY THE NERVE-CURRENT. 157 
 
 contraction, the problem is to excite by means of sufficiently strong" 
 electrotonic currents in directly excited fibres, other fibres adjacent 
 to these, and to excite them near the point where they part company 
 from the directly excited fibres. To this end are required strong- 
 currents and a small distance from the excited point to the point of 
 branching-. 
 
 If the transverse section of a nerve trunk, of which one branch is 
 to act as the primary and the other as the secondary nerve, is in 
 the vicinity of the point of branching-, the increased excitability of 
 the secondary nerve fibres in the neig-hbourhood of the transverse 
 section may favour secondary excitation by means of the electrotonic 
 currents. As a matter of fact du Bois-Reymond has already found 
 that when the peroneal or tibial nerve is excited, a fresh transverse 
 section of the sciatic favours the appearance of the paradoxical con- 
 traction in the muscles supplied by the branch which is not directly 
 excited. And in this case the presence or absence of a paradoxical 
 contraction at make or at break is notably affected by the direction 
 of the electrotonising- current, since the part of the secondary nerve 
 which is to be excited by the electrotonic currents lies very close to 
 the transverse section of that nerve. 
 
 It is easy to understand further, that when strong- currents are 
 applied near the origin of a nerve which is to be excited indirectly, 
 the effects of the electrotonic currents may combine with the effect 
 of the negative variation, if the preparation is excitable enoug-h, and 
 especially if the transverse section is still fresh. It is my intention 
 to consider these complex cases after discussing- more completely 
 the paradoxical contraction of du Bois-Reymond. If the above 
 directions are observed, there will be no danger of the admixture 
 of electrotonic action. The current ought to be so weak that 
 when it is ascending (abterminal) it acts only when quite close 
 to the transverse section, or at any rate much more strong-ly here 
 than at any other part of the nerve. With descending (atterminal) 
 currents even when of excessive strength, the true secondary con- 
 traction may be distinguished from the other by the fact that the 
 former disappears when the current is made so strong- that the 
 nerve is blocked in the reg-ion of the anode (3 to 4 Daniells), whereas 
 du Bois-Reymond's secondary contraction increases with increasing- 
 current-strength. 
 
Part II. 
 
 RESEARCHES 
 
 RELATING TO 
 
 SECONDAEY ELECTEOMOTIVE PHENOMENA. 
 
VI. 
 
 ON 
 
 SECONDAET ELECTEOMOTIVE PHENOMENA 
 
 IN 
 
 MUSCLES, NERVES, AND ELECTRICAL ORGANS. 
 By E. DU BOIS-EEYMOND. 
 
 Sitzungsberichte der koniglich-preussischen Academie der Wissenschaften 
 zu Berlin, vol. xvi. p. 343 (1883). 
 
 M 
 
CONTENTS. 
 
 PAGE 
 
 1. Introduction 163 
 
 2. Historical ............ 164 
 
 3. Preliminary Studies on Internal Polarisation of Moist Porous Bodies . 165 
 
 4. Continuation of Preliminary Studies. On Polarisation at the surface of 
 
 contact of dissimilar Electrolytes . . . . . . .167 
 
 5. External and Internal Polarisation of Moist Porous Bodies . . .168 
 
 6. Preliminary description of Apparatus and Methods . . . .170 
 
 7. Secondary Electromotive Action of Muscles in regard to its dependence 
 
 on the density and duration of the Primary Current . . . -173 
 
 8. Graphic representation of the Polarisation curves in relation to the time 
 
 of closure; Discussion of the same . . . . . . -177 
 
 9. On the curves of Polarisation in relation to the time after opening 
 
 the Primary Current 181 
 
 10. Influence of the direction of the Primary Current on the Secondary 
 
 Electromotive Action of Muscles . . . . . . . .185 
 
 11. Influence of various circumstances on the Polarisation of Muscles . . 190 
 
 12. Positive internal Polarisation of Muscles in conflict with Tetanus . .193 
 
 13. On Secondary Electromotive Phenomena of Nerves .... 195 
 
 14. Matteucci's Experiments on Nerve Polarisation ..... 197 
 
 15. Hermann's Researches on Nerve Polarisation ..... 199 
 
 16. Tigerstedt's Experiments on Nerve Polarisation 203 
 
 17. Influence of the direction of the Primary Current on the Secondary 
 
 Electromotive Actions of Nerves . , 204 
 
 18. Influence of various conditions on the Polarisation of Nerves . . . 209 
 
 19. Positive internal Polarisation in Nerves in conflict with Tetanus . .210 
 
 20. On the Secondary Electromotive Phenomena of Electrical Organs . 211 
 
 21. Theoretical. Closing Remarks , 217 
 
VI. 
 
 1. Introduction. 
 
 I THINK it is now time to break the silence which I have hitherto 
 maintained regarding certain experiments on animal electricity 
 with which I have been occupied for nearly forty years, and to 
 which I attach great importance. I should have been glad to 
 complete this research as far as I was able before publishing the 
 results, but recently electro-physiologists have approached from 
 several sides the province in which I was for so long the only 
 worker, so that by holding back any longer I incur the danger of 
 forfeiting my share of the fruit of a work, respecting which I have 
 more than quadrupled the nonum prematur in annum. I shall there- 
 fore now state the chief views at which I have arrived. In future 
 papers I shall develop more fully what I have said here. 
 
 Long ago I divided the electromotive phenomena of muscle and 
 nerve into three classes ^. The first are perceptible without the co- 
 operation of an extraneous current in a condition of rest or action. To 
 these belongs the current of muscle or nerve at rest as well as its 
 negative variation, the case in which the latter is induced electrically 
 being included in so far as the tetanising current is replaceable by 
 any other excitation. If we wish to extend the classification to 
 electrical organs, we must also include in the same class the weak 
 electromotive efiects observed in thern when at rest, which I call 
 *the organ current^,' and the shock of the organ in whatever way 
 induced. 
 
 In the second class I placed the extra polar electrotonic currents 
 which appear in nerves during the continuance of an extraneous 
 current and to which there exists nothing analogous in muscle or 
 in electrical organs ^. 
 
 1 Untersuchungen iiber thierische Elektricitat, vol. ii. part ii. p. 377. 
 
 ^ Dr. Carl Sachs' Untersuchungen am Zitteraal, Gymnotus electricus, nach. 
 seinem Tode bearbeitet von E. du Bois-Eeymond, Leipzig, 1881, p. 171 fi". 
 
 = Eckhard experimented on the organ of the 'Torpedo,' Sachs on that of the 
 electrical eel, in order to obtain electromotive action of the electrical organs by a 
 
 M 2 
 
164 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 To these two long- known classes I added a third — that of 
 secondary electromotive phenomena ; by which term I understand 
 those electromotive phenomena which an extraneous current pro- 
 duces in the portion of muscle it flows through, and which can gene- 
 rally only be observed owing* to their lasting longer than the 
 extraneous current. These are the facts, to a great extent new, 
 about which I intend to treat. As I shall presently mention, 
 however, it will be found that a distinction can hardly be main- 
 tained between the two last classes. 
 
 Secondary electromotive phenomena appear for the most part 
 as polarisation-currents. I prefer the former of these two expres- 
 sions, which is .quite applicable to the ordinary phenomena of 
 polarisation, and could be used to designate the apparent polar- 
 isation of muscles and nerves, even though it should prove to be 
 different from galvanic polarisation. I do not, however, insist 
 pedantically on the name, but occasionally call the secondary 
 current, generated by the secondary electromotive force, the polarisa- 
 tion-current, and also the after-current ; while the primary current 
 which induces it is styled the polarising current. I also find it 
 sometimes convenient to speak of the body to be polarised as 
 ' object of polarisation.' By negative I mean a polarisation-current 
 in an opposite direction to the primary current, and by positive 
 one in the same direction. 
 
 2. Historical. 
 
 At the commencement of my researches in science I encountered 
 a phenomenon which belongs to this subject. Peltier described in 
 1836 a negative polarisation in a frog's limb through which a 
 current was passing, which he compared to the polarisation of the 
 metals in Ritter's secondary battery. According to him the seat 
 of the polarisation was in the part of the limb immersed in the 
 
 persistent passage of the current ; Eckhard unsuccessfully, Sachs with doubtful 
 results (Unterauchungen am Zitteraal u. s. w., p. 188 et seq.). Even if these experi- 
 ments had succeeded one could not speak of electrotonus of electrical organs. In 
 muscles Hermann thinks he has now found a trace of that electrotonus current which 
 formerly neither he nor I could prove, and which his theory so urgently required 
 (Handbuch der Physiologic, vol. ii. Leipzig, 1878, p. 168) — Die Ergebnisse neuerer 
 TJntersuchungen auf dem Gebiete der thierischen Elektricitat. Separat-Abdruck 
 au8 der Vierteljahrschrift der naturforschenden Gesellschaft in Zurich, 1878, pp. 21, 
 22. He does not consider the possibility that the observed action may proceed from 
 the intramuscular nerves. As early as in 1 849 I warned against this fallacy (Unter- 
 Buchungen, vol. ii. part i. p. 330). \ 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 165 
 
 water which conveyed the current. When he applied a galva- 
 nometer to the parts intermediate between the submerged ones, 
 through which the current had passed, no action whatever was 
 observed. 
 
 In consequence of this Peltier considered the cause of the polar- 
 isation to be the separation of hydrogen and oxygen in the sub- 
 merged parts, which he assumed from the analogy of metallic 
 electrodes, without however demonstrating anything of the 
 sort ^. 
 
 Peltier's observation was the starting-point of my own work 
 on this subject. On repeating his experiment with a solution of 
 ordinary salt instead of water, I actually found the surface where the 
 current entered the preparation giving an alkaline reaction, and 
 where it emerged giving an acid reaction ^. Though this appeared 
 favourable to Peltier's explanation, yet, on the other hand, I 
 found, in opposition to his assertion, that every part of the pre- 
 paration through which the current had passed possessed, like a 
 secondary battery, a negative electromotive action. A portion, 
 then, of the negative polarisation possibly depends on the ions 
 collected at the points where the current enters and leaves the 
 preparation. The largest and by far the most important part 
 has its seat in the interior of the tissues, no doubt principally 
 in the muscles. 
 
 3. Preliminary Studies on tlie internal Polarisation 
 of moist porous bodies. 
 
 I was thus led to conceive the idea of an internal polarisation 
 of muscle. I soon, however, found other moist porous bodies 
 susceptible of internal polarisation, and then I first traced the new 
 phenomenon from a purely physical point of view through a long 
 series of moist porous bodies. These were either of an inorganic or 
 
 1 TJntersuchungen uber thierische Elektricitat, vol. i. part i. p. 377; vol. ii. 
 
 part ii. p. 378. . . 
 
 =" Humphry Davy saw acid and alkali set free in muscular flesh, pieces of livmg 
 plants, and even in the fingers to which distilled water had conducted the current. 
 (Philosophical Transactions for the Year 1807, Pt. I. pp. 52, 53 ; of. Simon in Gilbert's 
 Annalen, 1801, vol. viii. p. 28 ; Eitter loc. cit., vol. ix. p. 329. I also saw acid and 
 alkali set free in a water-pad covered with violet-blue litmus paper between saline 
 pads. (Gesammelte Abhandlungen zur allgemeinen Muskel- und Nervenphysik, vol. i. 
 Leipzig 1875, p. II.) When I said here that this was contrary to Davy's statement 
 that test-paper only changes colour at the metal electrodes through ions, the above 
 experiments of Davy's were not before me. 
 
166 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 an org-anic nature, and in the case of the latter either organised or 
 not organised. In the latter case, again, they might be either 
 dead like wood, or living like fresh tissue of plants ^. It may 
 be stated generally that every porous body which is not too 
 bad a conductor and is soaked with an electrolyte which, in com- 
 parison with the porous body, does not conduct too well, is sus- 
 ceptible of negative internal polarisation ; i. e. after the passage of 
 the current each transverse section lying between two isoelectric 
 surfaces has an electromotive action in a direction opposite to that 
 of the polarising current. This can be explained by the suppo- 
 sition that the current divides itself between the soaking fluid and 
 the porous framework, and that the latter becomes polarised just 
 like a metallic septum by the ions set free on its surface. Each 
 one of the innumerable minute septa has an electromotive action 
 in a direction opposite to that through which the current passed. 
 The partial currents thus generated take their way through two 
 paths, one presented by the internally polarised body itself, the other 
 by the galvanometer-circuit applied to the body. Their strength 
 in the latter depends on the dimensions of the polarisation- 
 object, on the position of the little septa in its interior, on the 
 nature of the material composing the porous framework, and on 
 the soaking fluid. The total current in the galvanometer-circuit 
 results from the superposition of all the partial currents, whose 
 strength is naturally an unkno%\Ti and complicated function of the 
 strength and duration of the polarising current, and the length 
 of time since its opening. All that can be said is that it must 
 increase with the increase of the two first variables up to a certain 
 point, and must diminish with that of the third. 
 
 We have a clear example of internal polarisation in a cylinder 
 of thoroughly burned wood-charcoal soaked with weak sulphuric 
 acid, and through which a current is passed axially. As a material 
 it conducts sufficiently well, so that a portion of the current finds 
 its way through it notwithstanding the good derivation which the 
 dilute acid affords. Each tract of the charcoal cylinder of equal 
 length has an equally strong secondary electromotive action. The 
 action of the whole cylinder equals the sum of all the individual 
 tracts. Wood is also capable of strong internal polarisation. It 
 is necessary, however, as the substance of the wood conducts badly, 
 
 * Untersuchungen tiber thierische Elektricitiit, vol. i. part i. p. 3S0. Gesammelte 
 AbhancUungen, loc. cit. 13-28: ' tJber innere Polarisation poroser, mit Elektrolyten 
 getriinkter Halbleiter' (August 4, 1856). 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 167 
 
 that the soaking" fluid should not belong* to the class of better con- 
 ductors. Only a weak action is obtained with solution of common 
 salt and of su]j)hate of copper. The best liquid to use is distilled 
 water. I studied on prisms of wood the effect of the length and 
 sectional size of bodies susceptible of internal polarisation, upon the 
 strength of the polarisation-current under circumstances in other 
 respects the same. In the first place, it is apparent that in internal 
 polarisation it is not the strength bat the density of the current 
 which comes into consideration. In accordance herewith it became 
 further apparent that the strength of the internal polarisation is a 
 function of the resistance of the moist porous body, in so far as this 
 resistance is determined by the dimensions. This function possesses 
 a maximum, which, in a given circuit, the length of the internally 
 polarisable body remaining constant while the transverse section 
 increases, is arrived at the later, the more imperfectly the soaking 
 fluid conducts ^. 
 
 4. Continuation of Preliminary Studies. On Polarisation at 
 tlie surface of contact of dissimilar Electrolytes. 
 
 While studying' internal polarisation I met with facts showing a 
 second kind of polarisation in moist conductors, i.e. the polarisation 
 at the surface of contact of dissimilar electrolytes. In contrast with 
 internal polarisation it possesses the peculiarity that it can also be 
 positive. If there is at the same time in the circuit an internally 
 polarisable body, action in opposite directions (doppelsinnige Wir- 
 kung) is often obtained, i.e. a deflection first in one direction and 
 then in the other ^. If, for instance, a pad of blotting-paper soaked 
 with water placed between two pads of the same sort soaked with 
 a solution of common salt has a current passed through it, there 
 follows on the opening of the polarising circuit and the closing of 
 the galvanometer-circuit, first a negative preliminary deflection 
 (Vorschlag), which is immediately followed by a positive deflection 
 of longer duration. The preliminary negative deflection arises from 
 internal polarisation of the water-pad, the positive from the sum of 
 the positive polarisations at the contact of salt solution with water, 
 
 ^ Gesammelte Abhandlungen, vol. i. pp. 29-41 : 'Uber den Einflnss welchen die 
 Dimensionen innerlich polarisirbarer Korper auf die Grosse der secundtir Elektro- 
 motorischen Wirkung fiben ' (January 31, 1859), 
 
 ^ Op. cit. vol. ii. Leipzig, 1877, p. 407. 
 
168 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 and water with salt solution. The internal polarisation in this case 
 is stronger but more evanescent in character, the external weaker 
 but more lasting", which explains the slight negative first deflection, 
 followed by a principal deflection which is positive. 
 
 All the fresh animal tissues which I have examined behave 
 between two saline pads like a water-pad. A piece of rib, costal 
 cartilagCj sinew, elastic tissue, frog's skin, human skin, pieces of 
 lung, liver, spleen, rabbit's kidneys, and lastly, muscles and nerves — 
 they all assume positive external polarisation which sums itself 
 algebraically with internal negative polarisation. Under circum- 
 stances otherwise the same, the latter is more strongly developed 
 the greater the length of the portion of tissue through which 
 the current is passed ^. If animal tissues are exposed to a currenifc 
 between the usual pads soaked with a solution of sulphate of zinc, 
 and covered with a layer of clay, kneaded with a ' physiological ' 
 solution of common salt (i.e. o"6 per cent.), the external polarisation 
 is very little marked ^. But it is advisable, in order to avoid errors 
 in investigation of the internal polarisation of muscles and nerves, 
 not to lead off" the secondary electromotive polarisation current by 
 means of the same pads, clay tips, &c. by which the primary 
 current has been led through it, or at least not to lose sight of the 
 fact that if this arrangement becomes necessary under certain cir- 
 cumstances, polarisation at the boundaries of the dissimilar electro- 
 lytes may come into play. 
 
 5. Further Continuation of Preliminary Studies. External 
 and Internal Polarisation of moist porous bodies. 
 
 When I reached this point I thought I knew enough of what 
 takes place during the passage of a current through a series of moist 
 porous bodies, to enable me safely to continue my researches on the 
 secondary electromotive phenomena of muscles and nerves. But I 
 found myself greatly mistaken. After I had made numerous experi- 
 ments of this kind I discovered a new source of error, by which my 
 work up to this time was rendered, if not useless, at least inade- 
 quate. This was what I have called ' secondary resistance ' of 
 moist porous bodies, which I now made the subject of a long 
 
 ^ Op. cit. vol. i. pp. 1-12: 'tJber Polarisation an der Grenze ungleichartiger 
 Elektrolyte' (July 17, 1856). 
 * Op. cit. vol. ii. pp. 189 fP. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 169 
 
 and wearisome course of experiments^, I understand by this a 
 resistance which is created by the current itself; which on ces- 
 sation of the current gradually decreases ; and which on its 
 reversal gradually disappears. The secondary resistance which is 
 developed in coagulated albumen between zinc pads is capable of 
 weakening the current from a Grove's battery of twenty elements 
 to such an extent, that only a tenth of the original strength 
 remains. Consequently, the secondary resistance may greatly 
 surpass the increase of conducting power consequent on the warm- 
 ing of a moist porous body by the passage of the current through 
 it. Reversal of the current in moist porous bodies in which 
 secondary resistance prevails, does not, like reversal in polarised 
 conductors, give rise to a sudden positive increase in the strength 
 of the current, but it presents the remarkable phenomenon of a 
 gradual increase, which goes on for some minutes, until the current 
 almost reaches its original height. From this point the decrease 
 begins, to be again exchanged for slow increase if the current is 
 once more reversed. 
 
 There are two kinds of secondary resistance, i. e. an external and 
 an internal secondary resistance, corresponding to the external 
 and internal polarisation of moist porous bodies. The seat of the 
 first is the point where the current enters, and although much 
 remains obscure, it may be explained by the cataphoric action of 
 the current driving the fluid in the circuit onwards from the 
 anode to the kathode. Where a zinc pad conducts the current 
 to animal tissues a very strong external secondary resistance is 
 developed, but it is easy to prevent it from coming into existence 
 by the methods which I give. The internal secondary resistance 
 which has its seat everywhere in moist porous bodies has up to this 
 time only been observed in any great degree in the tissues of living 
 plants, and therefore comes less into consideration in the experi- 
 ments of which we are treating ^. 
 
 The investigation of external and internal polarisation, of external 
 and internal secondary resistance of moist porous bodies which 
 engaged me for many years, was only one link in the series of 
 
 ^ Gesammelte Abhandlungen, vol. i. pp. 80-130: ' Uber den secundaren Wider- 
 stand, ein durch den Strom bewirktes Widerstandsphiinomen an feuchten porosen 
 Korpern' (Dec. 20, i860). 
 
 * For further information on Secondary Resistance, see H. Munk, Uber die 
 kataphorischen Veranderungen der feucliten porosen Korper, Archiv fur Anatomie, 
 Physiologie, &c., 1873, pp. 241 fF. 
 
170 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 researches by which I sought to elucidate the secondary electro- 
 motive phenomena in muscles and nerves. Though the pursuit of 
 my chief object was excessively hindered by these preliminary 
 studies, yet my impatience was moderated, not only by the evident 
 necessity of laying a firm foundation to work upon in this depart- 
 ment, but also by the consideration that this foundation is equally 
 indispensable for all electro-physiological investigations^. 
 
 I was unfortunately prevented by other problems for nearly 
 twenty years from making use of what I had ascertained. Almost 
 at the same time that I became possessed of unpolarisable elec- 
 trodes, I had discovered in saline clay an invaluable means of 
 leading currents in or off in electro-physiological experiments, I had 
 given Pogg-endorff's method of compensation the right form, and 
 somewhat later I added to these advances in methods of research, 
 the invention of the aperiodic galvanometer. It thus became 
 necessary for me to repeat almost all my former experiments with 
 these new and much better appliances, in which measurement of the 
 electromotive force now took the place of mere estimation of the 
 strength of the current. The investigation of the living specimens 
 of Malapterurus from West Africa which were kindly consigned to 
 me by Professor Goodsir of Edinburgh, also took me away from this 
 subject ", and, more recently, it has taken me nearly two years to 
 work up the posthumous papers of Dr. Sachs on Gymnotus. The 
 latter interruptions had, however, the important result that I gained 
 some information regarding secondary electromotive phenomena in 
 electrical organs. 
 
 6. Preliminary description of Apparatus and Methods. 
 
 I defer the detailed description of the apparatus and methods 
 which I now use until I have further communications to make. 
 It is, however, desirable to say sufficient on the subject to render 
 what follows intelligible. In the experiments on muscles the 
 gracilis and semi-membranosus muscles of the frog were commonly 
 used. They were left in their natural connection, and extended 
 in the ' muscle-stretcher ^,' in the way I have employed whenever I 
 desired to work with muscles in an immobilised state. Close 
 to the ivory plates of the stretcher the edges of the wedge-shaped 
 
 ^ Cf. Gesammelte Abhandlungen, vol. ii. pp. 191 ff. 
 
 ' Untersuchungen, vol. ii. part i. p. 86. Also Gesammelte Abhandlungen, vol. i. 
 pp. 118, 119, and vol. ii. p. 313. 
 
MUSCLES, NERVES, AND ELECTEICAL ORGANS. 171 
 
 pads ^ were applied to the internal (that is, femoral) side of the 
 muscles ; throug-h these pads the polarising" current entered the 
 muscles. Between their edges, the edges of the pads which led off 
 the polarisation current to the galvanometer were in contact with 
 the external side. On account of the resistance of the tendons, 
 which are apt to get dry, it is not feasible to conduct the 
 current to the muscles through the fragments of bone outside of 
 the ivory plates, which would have been in many respects a far 
 better mode of proceeding. Both pairs of electrodes, which in 
 the following pages will be called for shortness sake battery- 
 pads and galvanometer-pads, are covered with clay, which is often 
 renewed. 
 
 In' consequence of the very small duration of most of the ex- 
 periments, errors consequent on external secondary resistance are 
 scarcely to be apprehended ; at all events, if they exist, they can be 
 readily recognised. The muscle-current was compensated, with the 
 aid of the round compensator, by a derivation-current from a 
 Raoult's battery 2. 
 
 In the experiments on nerves the two sciatic nerves of the same 
 frog were commonly used as one. They were secured to the upper 
 surfaces of two pieces of cork with insect needles, and gently ex- 
 tended by shoving the corks along a horizontal glass rod ; the two 
 pairs of electrodes were applied in the same way as in the case 
 of the muscle. In the experiments on the roots of the spinal nerves, 
 unpolarisable tube electrodes with clay tips [Stiefelelektroden) were 
 sometimes used, at other times special contrivances ^. 
 
 A condition essential to certainty of results in these experiments 
 which is difficult to secure, is that of complete insulation of the 
 
 * Gesammelte Abhandlungen, vol. i. pp. 88, 89. 
 
 ^ Untersuchungen am Zitteraal, p. I41. 
 
 ^ Since my last publication of an experimental character, a phenomenon connected 
 with the Thonstiefeln has come under my notice, which not only deserves considera- 
 tion in using this form of non-polarisable electrodes, but is on other grounds interesting. 
 If you have a pair of such tubes, the freedom of which from polarisation you have 
 ascertained by bringing the tips into contact, and then touch the thick upper part 
 {Wulst) of the one with the tip of the other, a current is found to exist of which the 
 direction is from the tip to the ' Wulst.'' This action is less marked in fresh tubes 
 than in tubes that have been prepared for some time. This may be explained on the 
 assumption that there is an electromotive action between moist and less moist clay, 
 though in a series of layers of clay of different degrees of moisture there is no grada- 
 tion of tension. Moist clay brought into contact with less moist between zinc pads 
 gives an e. m.f. amounting to 0-014 Raoult. The relation of the electromotive action 
 between clays of different degrees of moisture to the clay thermal currents of Nobili 
 has not yet been explained. (Untersuchungen, vol. ii. part i. pp. 201-203.) 
 
172 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 battery-circuit from the galvanometer-circuit, so as to ensure that 
 even when very large electromotive forces are used, such as that of 
 a Grove's battery of fifty elements, the galvanometer remains un- 
 affected when both circuits are open. I have already given shortly 
 the means by which this can be accomplished in the description of 
 my experiments on external polarisation in moist conductors ^ It 
 consists in breaking both circuits in two places. The contrivance 
 by which this is effected fulfils at the same time another purpose. 
 It enables the observer to determine the time (' closing time ') 
 during which the polarising current is sent through the object of 
 polarisation, I possess mechanical means by which I can alter this 
 by continuous gradation from xoW ^^ jV ^^ ^ second, and then go 
 on at intervals of 0*3 second up to 20 seconds. Beyond this the 
 variations may be made by the watch. With the same apparatus 
 I am able to determine the interval between the opening of the 
 battery-current and the closing of the galvanometer-current. This 
 I c^ll the ' transmission time ' (tJbertragungszeit). 
 
 The galvanic apparatus used for polarisation was either a Daniell's 
 battery, of which the elements were of the usual drinking-glass 
 size, and the specific gravity of the dilute acid was 1-030 at 19-4° C, 
 or a Grove's battery of the small elements I use, of which I possess 
 fifty ^. In order to be independent of the variable resistance of the 
 pads, muscles, &c., the good condition of the battery was controlled 
 by means of a metallic closure. When other electromotors were 
 used, it is specially stated. 
 
 The secondary electromotive actions were observed with the aid 
 of the aperiodic galvanometer. As these actions are often in two 
 opposite directions one after the other, the aperiodicity of the instru- 
 ment has special value. They render it unnecessary to use con- 
 trivances such as I formerly employed, by which the galvanometer- 
 circuit was closed after a longer or shorter transmission time, instead 
 of, as now, immediately after the opening of the primary circuit. 
 Finally, it can be readily understood that in experiments on polari- 
 sation, as in those relating to electrotonus, it is necessary constantly 
 to watch the condition of the primary current, by means of a special 
 galvanometer. This would only be possible with the Wiedemann 
 
 * Gesammelte Abhatidlungen, vol. i. p. 2. 
 
 ' In the year 1849, the Academy, to which I had not then the honour to belong, 
 ■was good enough to place at my disposal a Daniell's battery made by Siemens and 
 Halske, of 100 elements, for this research. I soon, however, recurred to the small 
 Groves, which are alike more effectual and more convenient. 
 
MUSCLES, NEKVES, AND ELECTKICAL OKGANS. 173 
 
 galvanometer, it being required to compare in experiments which 
 followed each other immediately, the perhaps permanent deflection 
 due to a current of thirty Groves, with the ten thousand times feebler 
 movement produced by a current of a single Daniell lasting a few 
 thousandths of a second. I lost much time and trouble in the 
 attempt to read the two galvanometers, viz. the one for the 
 polarising current, and that for the secondary current, through the 
 same telescope. Besides the technical diSiculties which this in- 
 volves, the observer is overburdened by having to do too many 
 things at the same time. I therefore recurred to the more simple 
 method of having the second galvanometer read by an assistant. 
 
 7. Secondary Electromotive action of muscles in regard to its 
 dependence on the density and duration of the primary 
 current. 
 
 Let us suppose the group of muscles, gracUis and semi-membra- 
 nosus, extended in the muscle -stretcher so firmly that they do not 
 perceptibly move in contraction, and the two pairs of electrodes 
 applied in the manner described, and the muscle-current compen- 
 sated. If the electrodes are applied symmetrically the muscle 
 current ascends; its electromotive force, although often insig- 
 nificant, frequently reaches 0-017 Raoult. 
 
 If a current be now applied to the muscle for a longer or a shorter 
 time, the galvanometer circuit being ' doubly ' open, and if the 
 galvanometer circuit be shut immediately after the battery circuit 
 has been doubly opened, any secondary electromotive action that 
 may have been excited in the muscle equalises ^ itself through the 
 galvanometer circuit, and is truly expressed by the deflection of the 
 galvanometer, provided the compensation remains undisturbed. If 
 the battery circuit is long closed, if the tension of the muscle is 
 diminished, or if strong contraction occurs, this condition is not 
 always certainly fulfilled ; and besides, if a strong muscle-current is 
 to be compensated, the contraction leaves an after-effect. The 
 secondary electromotive actions are however generally too marked 
 to be mistaken for such disturbances. They are due to the internal 
 
 * [When any spot a on the surface of a moist conductor is negative to any other 
 spot b, the current which exists from a to 5 is said sich dbzugleichen by currents out- 
 side of it which are directed from h to a. The same expression is used of a current 
 which flows from h to a through a galvanometer of which the terminal electrodes are 
 applied to these spots. Whenever possible, the word ' dbgleichen ' is Englished by 
 the word ' equalise.' — Ed.] 
 
174 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 polarisation of the muscles alone ; and even if external polarisation 
 occurs to an observable degree in the electrodes of the battery, it 
 cannot equalise itself through the galvanometer circuit. More- 
 over, one may readily convince oneself that every part of the 
 muscles has a secondary electromotive action in the same direction. 
 Consequently, if the resistance in the galvanometer circuit is suffi- 
 cient, the strength of the action increases with the distance apart of 
 the galvanometer pads. If all parts of the muscle were similar, 
 they must all have an equally strong secondary electromotive 
 action. Apart from the impossibility of applying the electrodes 
 twice successively in the same way, this condition is not fulfilled 
 exactly in our group of muscles, because the semi-membranosus 
 becomes gradually smaller towards its lower part. AVe shall find, 
 however, another reason why the two halves of a muscle do not 
 act with the same secondary electromotive strength^. 
 
 If currents of different strengths are passed through the muscle 
 for different periods of time, the secondary electromotive actions 
 appear at first very confused. Sometimes negative deflections occur, 
 as if from ordinary moist porous bodies susceptible of internal 
 polarisation, sometimes, on the other hand, positive deflections, as 
 if external polarisation was mixed up with it. As already remarked, 
 this last is impossible with the arrangement used. It therefore 
 appears that the positive secondary electromotive action depends 
 upon positive internal polarisation, a condition of which the series 
 of bodies which I tested as regards their internal polarisability ^ 
 gave no example. This positive internal polarisation of muscular 
 tissue is therefore the point of novelty and interest, and it is most 
 important to establish the conditions under which it appears. 
 
 There is no other way of accomplishing this but by making a 
 table with double entry, in one column of which, let us say the 
 horizontal one, the time of closure is entered, and in the vertical 
 column the number of battery-cells. In the space corresponding to 
 a given time of closure and a given density of current, must be 
 entered, ist, the deflections which the polarising current creates in 
 
 * Gesammelte Abhandlungen, vol. ii. pp. i6i, 315, 575. 
 
 * I first gave information about it in my ' Untersuchungen,' vol. i. 1848, p. 240, 
 and vol. ii. part i. 1849, p. 331. In 1852 I made a detailed communication upon it 
 to the British Association in Belfast (Report of the twenty-second Meeting of the 
 British Association, &c., held at Belfast in September, 1852, London 1853, Notices 
 and Abstracts, p. 78). Later I alluded to it in the Report on the ' Zitterwelse ' from 
 West Africa (Monatsberichte der Akademie, 1858, p. 106), and in the L^ntersuchungen 
 am Zitteraal, p. 206. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 175 
 
 the galvanometer (P) ; andly, the resulting secondary electromotive 
 action shown by the galvanometer (S), and both effects in the 
 ascending as well as in the descending direction of the polarising 
 current. It is a most tedious business to compile such a table. 
 The secondary actions created by the stronger and longer currents 
 are generally so persistent that one almost requires a fresh prepara- 
 tion for each experiment. The only exceptions are the cases where 
 quite weak currents last only for a small fraction of a second, as 
 then the polarisation is transitory enough to allow of other ex- 
 periments, especially with stronger and longer currents to follow 
 them immediately. On the other hand, those cases are also ex- 
 ceptions, in which a current of long duration creates so persistent 
 a polarisation, that the battery-circuit can be opened and the galva- 
 nometer-circuit closed for a short time without perceptibly altering 
 the condition of polarisation. In the other cases where a polarisa- 
 tion, sinking at first rapidly and then slowly, remains behind, it 
 would no doubt be easy to compensate afresh and so produce equi- 
 librium in the galvanometer-circuit. But in rej)eated experiments 
 in the same preparations one would not be sure of getting the 
 same secondary action as in fresh muscle. If one waits till the 
 polarisation disappears, too much time elapses for it to be assumed 
 that the muscle will behave like a fresh one. On account of the 
 necessity of leaving the pelvis attached to each group of muscles, 
 each frog furnishes only one group, so that every experiment which 
 does not come under these two exceptions requires a new frog. 
 But if the results are to be comparable, it is necessary that, firstly, 
 the frogs should not differ too much in size ; and, secondly, that 
 they should be as much as possible in the same condition. If it 
 is remembered that in order to arrive at a fairly complete know- 
 ledge of the phenomena it is necessary to try at least ten times 
 of closure and as many current densities in both directions, which 
 correspond to two hundred experiments ; that these two hundred 
 experiments must be made successively on almost as many frogs, 
 which must be as far as possible equally fresh and large ; and if 
 in addition to this the unavoidable failures and repetitions are taken 
 into consideration, it is seen that a table such as is required is 
 the work of several weeks in which a great number of favourable 
 circumstances must be present, so that no one will be surprised that 
 in so long a time I only attempted it twice. 
 
 The first time, in the autumn of 1855) I possessed none of the 
 new apparatus or methods of experiment, and I knew as 5'et 
 
176 
 
 SECONDAEY ELECTKOMOTIVE PHENOMENA IN 
 
 nothing about secondary resistance. It was indeed on this occasion 
 that I discovered it, though unfortunately too late, and the whole 
 work, which would however now be behind the time, had to be 
 thrown aside. I succeeded for the first time last summer, 1882, in 
 working" out a new table. Meanwhile the difficulties had not de- 
 creased, for in proportion as the means of assistance were perfected 
 the demand for accuracy had increased also. I took as the normal 
 measurement of a frog 23 centim. from the tip of the nose to the 
 end of the longest toe. I never allowed a greater variation than 
 X a cm. I employed 14 times of closure from o"-oo6 to 25', and 
 9 strengths of current from i Daniell up to 40 Groves' cells. Not 
 counting the faults and repetitions, the number of experiments was 
 198. The present paper would not be a suitable place in which to 
 reproduce the entire table. I will only give one example of it. 
 
 T 
 
 o"'oo6 
 
 o"-o76 
 
 0"-320 
 
 l"-I22 
 
 4".626 
 
 I0"-02 2 
 
 i4"-754 
 
 2o"-oi8 
 
 l' 
 
 5' 
 
 s 
 p 
 
 X 
 
 igsc 
 
 + 133 
 109 
 
 + 109 
 400 
 
 + 84 
 
 305 
 
 + 73 
 1250 
 
 -66 
 
 020 
 
 -79 
 2100 
 
 -125 
 
 2216 
 
 -170 
 
 1669 
 to 
 
 1824 
 
 -119 
 
 1824 
 
 to 
 1962 
 
 s 
 
 p 
 
 1 
 
 + 79 
 14 
 
 + 111 
 
 178 
 
 + 160 
 229 
 
 + 164 
 865 
 
 {-15 
 
 1+ 4 
 
 2129 
 
 J- 9 
 1 + 34 
 
 1337 
 
 (-87 
 \— 112 
 
 2412 
 
 -62 
 
 1777 
 
 — 100 
 
 1657 
 
 to 
 
 i860 
 
 -24 
 IS60 
 
 to 
 
 2I5I 
 
 It is hardly necessary to say that the headings in the top row 
 show the time of closure, under one second before the first and 
 under one minute before the second of the double perpendicular lines. 
 The X means that the polarising-current is produced by 10 Groves. 
 The arrows indicate the direction of the current in the muscle. 
 The horizontal rows S contain the deflections caused by the secondary 
 current. The plus mark indicates positive, the minus mark nega- 
 tive polarisation. The sensitiveness of the galvanometer {S) was 
 diminished, only one bobbin at a distance of 3 cm. being em- 
 ployed. Two numbers bracketed represent polarisation in two 
 directions. In the horizontal rows P, are found the deflections 
 due to the primary current. As 4"-626 is greater than t max. \ 
 the mirror of the galvanometer (P) as deflected by the primaiy 
 current, reaches, whenever the time of closure is equal to 4"-626 or 
 exceeds it, the deflection at which it would have been held by a 
 continuous current. The magnitude of the foregoing numbers 
 
 * Gesammelte Abhandlungen, vol. i. p. 302. 
 
MUSCLES, NEKVES, AND ELECTRICAL ORGANS. 177 
 
 depends on the fact that the deflections are reduced. The galvano- 
 meter is so graduated that the relative rotating moment of the 
 bobbin was known for each 5 mm. of distance. In proportion to 
 the strength of the current the galvanometer bobbin was placed 
 at such distance from the mirror that it produced a suitable de- 
 flection A. Suppose m to be the relative moment for 20 mm. 
 distance of bobbin, m' the moment for the distance chosen each 
 
 Til 
 
 time, the reduced deflection A^ = A -,. With the strongest cur- 
 
 m 
 
 rents were obtained reduced deflections of nearly 10,000 divisions 
 of the scale (see Sect. 6). In the numbers in row P of the table 
 the great imperfection attaching to these experiments notwithstand- 
 ing all my trouble, is seen. From the time of closure ^"-626 
 onwards the collective deflections through the primary current 
 should have been the same. They vary in a ratio of 100: 180, 
 partly from the impossibility of making the resistance of the 
 muscles and the wedge-shaped pads with their clay tips the same 
 size in several consecutive experiments, and (for this explanation 
 is hardly sufficient) partly on other unknown grounds. The 
 increase of the strength of the current, when the closure lasts one 
 minute or more, arises from the warming of the pads, of their edges, 
 and of the muscles themselves. With greater strength and longer 
 continuance of the current, this action is more than balanced by 
 external secondary resistance at the point where the current enters 
 the muscle. The polarisation after 1' closure was so constant that 
 after a short time of opening the battery circuit the galvanometer 
 circuit could be closed without perceptibly altering the condition of 
 polarisation, so that it was unnecessary to use a new preparation 
 (p. 169). 
 
 8. Graphic representation of the Polarisation curves in relation 
 to the time of closure ; Discussion of the same. 
 
 However incomplete these experiments may be, a number of im- 
 portant conclusions may be drawn from them. The subjoined Fig. 5 
 shows shortly the general course of the phenomenon. Let T be the 
 time of closure, A the density of the primary current, + s the 
 strength of the secondary electromotive action, then the plane seen 
 in perspective like a boarded floor, is the T-A-plane. At the points 
 of this plane corresponding to the given times of closure and the given 
 densities of current, the secondary actions are set off" parallel to the 
 
 N 
 
178 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 to 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 179 
 
 xS-axis, the positive above and the negative below the plane. The 
 ordinates s themselves are not drawn, but the curves connecting 
 them corresponding- to a definite density of current are drawn. 
 These therefore represent the strength of the secondary action due 
 to this density of current as a function of the time of closure. 
 Although these curves have similar courses they alter continually 
 with increasing density of current. They agree in the following 
 general characters. The zero time of closure corresponds naturally 
 to the zero of secondary action, consequently the curves themselves 
 start from the A-axis. Each curve rises quickly to a positive 
 maximum, and then sinks less rapidly but still very fast to the 
 I'-A-plane. These curves are drawn in the figure up to the point 
 where they intersect this plane as unbroken lines, beyond this 
 in their negative course they are represented by dotted lines. 
 Here appears a negative maximum which is generally less clearly 
 marked, and is at a greater distance from the point of intersection 
 than the positive maximum, though this distance diminishes with 
 increasing density of the current. Thus it occurs with i Daniell 
 only after 15', with i Grove after 10', with 2 Groves after 5', 
 with 5 Groves after i\ &c. It lies at i' in the part of the table 
 shown, which was obtained with 10 Groves, and it is obvious that it 
 cannot depend at all on the weakening of the primary current by 
 secondary resistance. I have not followed the polarisation-curves 
 further than this negative maximum. However, they probably 
 present nothing particular, but approach very gradually the axis of 
 the times of closure, till from the excessive duration of the experiment 
 the results are obscured by all kinds of collateral actions. 
 
 With a current density of less than 2 Groves and with a very short 
 time of closure there is generally no polarisation appreciable by the 
 galvanometer. The first traces observed with i Daniell and i" time 
 of closure, are negative ; internal polarisation of the muscle is also 
 obtained as I have long ago described, by passing a current com- 
 parable with its own for a long time through it ^. The first positive 
 traces appear for the first time with 2 Groves and about o"-3 time of 
 closure. 
 
 The time of closure at which the positive polarisation passes 
 into negative may be called the critical time. In the figure, 
 the points at which the polarisation-curves described on the time 
 of closure intersect the jT-A-plane, are connected together by a 
 curve drawn with an interrupted line. This curve is therefore 
 ^ Gesammelte Abhandlungen, vol. ii. pp. 191, 192. 
 N 2 
 
180 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 that of the critical times of closure Avith the axis of current 
 densities as abscissal' axis. It is obvious that the curve of the 
 critical times of closure has a maximum ; in my experiments 
 this maximum lay somewhere between the density of the current 
 generated in 2 Groves, and the critical time of closure then reached 
 was about 5'". With 20 Groves it was only about 1". If the 
 positive maxima of the individual polarisation-curves are connected 
 by a curve, this curve also shows a maximum with from 20 to 30 
 Groves ; and similarly a curve which unites the negative maxima 
 exhibits a maximum, but does so with i Grove only. 
 
 It remains to give an idea of the amount of these maxima as 
 compared with that of the muscle-current. If I pass through a 
 galvanometer-circuit, in which the group of muscles have been 
 placed as if for a polarisation experiment, an electromotive force 
 of about 0-045 ^o ^''^55 Raoult corresponding to the force of the 
 whole ^ artificial transverse section of the graciKs or semi-mem- 
 branosus, there follows a deflection of 235-285^°, the sensitiveness 
 of the galvanometer having been diminished. I noticed the 
 strongest negative polarisation after closure of a current of i Grove 
 for 10', and it gave 423^°. I obtained the strongest positive 
 polarisation by closure of 20 Groves during o"-o75 ; it amounted 
 to 239^°, and therefore appeared to have less electromotive force 
 than the muscle-current. We shall soon see, however, that this 
 is a deceptive estimate. What the relation is between the 
 secondary electromotive actions in muscles relatively to their 
 size, and those of other moist porous bodies, has not yet been 
 investigated. 
 
 So far as the above numbers for the position of the maxima, &c., 
 depend on the density of the primary current, they are only applic- 
 able to the particular group of muscles in frogs 22 cm. long, with 
 our method of conduction. Moreover Fig, 5 must not be understood 
 to represent the relation between the numerical values. The avail- 
 able space rendered this impossible. As, for instance, the critical 
 time of closure in our experiments reached at most 5", and the 
 maximum of negative polarisation first appears only at 10-15', ^^^ 
 figure would require to be from 3 to 4 metres long in order to show 
 this maximum at the right place with the same proportional measure- 
 ment as that adopted for the first few seconds. The current 
 densities would require at least five times as broad a space, &c. 
 
 * Gesammelte Abhandlungen, vol. ii. pp. 193, 243. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 181 
 
 Closely connected with experiments on galvanic currents of 
 variable strength and duration, are experiments made with electro- 
 motors, which from their nature produce short sudden shocks. 
 These produce positive polarisation at once. I have not yet found 
 time to pursue this kind of experiment with new means and appa- 
 ratus, and must content myself with citing older results, which are 
 in the main correct. 
 
 In December, 1 846, I obtained to all appearance positive polari- 
 sation in the legs of a living frog with a Saxton's machine, the 
 shocks of which could be sent in the same direction by a commu- 
 tator [Sfromwender). As this machine ^, constructed for Dove by 
 Oertling, passed into the possession of the Physiological Insti- 
 tute, I am in a position to control those experiments which 
 were performed with very imperfect knowledge and experimental 
 methods. Under somewhat better conditions, in November, 1855, 
 I observed positive polarisation in the group of muscles with 
 opening shocks of the induction coil. Lastly, as early as December, 
 1846, I observed positive polarisation in frogs which I killed with 
 a heavily-charged Leyden battery having a surface of about 0-31 
 square metres. The muscles looked blood-stained {llutrilnstig), and 
 only gave traces of reaction on the application of other shocks 2. 
 
 9. On the curves of Polarisation in relation to the time 
 after opening the Primary Current. 
 
 The polarisation of muscles can still be studied in relation to the 
 course it takes after the opening of the primary current, and this 
 course may be represented by a curve drawn relatively to the time 
 which has since elapsed. This period of time may be called the 
 ' opening time.' In general the positive as well as the negative 
 polarisation appears very persistent. In Sect. 7 I have already 
 alluded to the difficulties which arise each time that one wishes to 
 employ the same muscle for several consecutive experiments. 
 After a muscle has been strongly polarised in a positive direction 
 it may require twenty minutes or more to elapse before it has so far 
 regained its original condition that the alteration may be disregarded. 
 As, besides, without having been polarised, the muscle does not 
 during this period retain its electromotive condition unaltered, it is 
 
 * TJntersuchungen fiber thierische Elektricitat, vol. ii. part i. pp. 398 ff. ; Wiede- 
 mann, Die Lehre vom Galvanismu su. Elektromagnetismus, vol. ii. 2 ed. Braun- 
 schweig, 1874, p. 236. 
 
 ^ Untersuchungen, loc. cit. pp. 181, 182. 
 
182 
 
 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 183 
 
 never possible to say if or when the restoration is complete. The 
 same thing" occurs with negative polarisation. Polarisation curves, 
 however, drawn relatively to the opening' time, present other 
 important peculiarities. If the opening occurs at the critical time 
 there often follow deflections in two opposite directions, corre- 
 sponding first to negative and then to positive polarisation. The 
 transition from purely positive polarisation to purely negative, 
 through such double action, may be explained in the manner expressed 
 by Fig. 6. It must not be imagined that from the opening* of the 
 primary circuit to the change of sign of the polarisation, the action 
 is simply positive and then simply negative. It is much more 
 likely that both are present from the moment of closure, and 
 increase in accordance with a difierent law, by which the negative 
 polarisation increases more in proportion to the time of closui'e, 
 while the positive rises first quickly and then slowly. In the figure 
 this is represented on the vertical plane, which is seen in perspec- 
 tive stretching from left to right backwards. In this plane the 
 abseissal axis oT is the increasing time of closure, the broken curve 
 from to + P is that of the positive polarisation for a given den- 
 sity of current, the dotted one from o to — P that of the negative 
 polarisation for the same density drawn relatively to the time of 
 closure. If the ordinates of these component curves ^ are summed 
 up algebraically, and the result inscribed according- to its positive 
 or negative value above or below the abseissal axis, you have the 
 resulting curve o (+«^) Tj^ ( — ^^0' which at Tj^, the critical time of 
 closure, intersects the abseissal axis o T, and in fact is nothing but 
 one of the polarisation-curves drawn with reference to the time of 
 closure such as we had in Fig. ^. The negative maximum, with a 
 longer time of closure, is explained by the supposition that the 
 curve of negative polarisation either begins to sink, or rises more 
 slowly than the positive, or that the latter begins to rise more 
 quickly (Sect, it and 3i). 
 
 On vertical planes which meet at a right angle the /S'-T'- plane of the 
 polarisation-curves, drawn relatively to the time of closure, the com- 
 ponent polarisation-curves are represented relatively to the time of 
 opening ; the abseissal axes ^j z^, t^ z^, t^ z^ represent in each case the 
 increasing time of opening. The polarisation-curves drawn rela- 
 tively to the time of opening are treated in a corresponding way to 
 those of the time of closure, from which they proceed. The com- 
 ponent positive ones are shown by interrupted lines, the component 
 
 * Cf. Untersuchungen am Zitfceraal, p. 215, 
 
184 
 
 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 neg-ative by dotted lines, the resultants by continuous lines. In 
 this representation the assumption is made that positive polar- 
 isation declines more slowly in time, and negative more quickly. 
 A glance now at the figure shows that if the primary current 
 is opened at the time ti the polarisation will always remain 
 positive. On opening* it at the time Tj^, if one could hit this 
 time, the polarisation would be nil for the first moment and then 
 positive. If, however, it is opened at the time t^ a double action 
 follows, first negative and then positive. Further on at t^ the 
 negative polarisation, which increases more in proportion to the 
 time of closure, has risen so high above the positive, which has re- 
 mained nearly at the same height as at first, that notwithstanding 
 the more abrupt descent of the negative curve the curves no longer 
 intersect each other, and purely negative polarisation appears. 
 
 Besides the actions in two opposite directions which can be so 
 satisfactorily explained, there occur in the successive phases of the 
 polarisation-currents peculiarities, which, inasmuch as they do not 
 occur regularly under definite conditions, are difficult to indicate 
 with certainty. In a few cases, polarisation in opposite directions 
 occurs with a first positive and second or principal negative deflec- 
 tion. Often the polarisation increases very slowly, which probably 
 depends on the disappearance of the opposite kind. Frequently, 
 
 S 
 
 indeed, the actions are in one direction only, but double in so 
 far that the deflection first reaches a maximum then sinks to a 
 minimum, and then once again rises to a maximum which often 
 even exceeds the first. A case of the sort is seen in the preceding 
 table, with a descending current, and 14"- 754 time of closure. 
 This is explained by the circumstance that the two component 
 curves, drawn relatively to the time of opening, take such a course 
 that the one is completely above the other, but is more convex 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 185 
 
 towards the abscissa, so that it first approaches the lower curve 
 and then separates from it, as shown in Fig. 7. It is impossible 
 here to follow further these particulars which are still very im- 
 perfectly investigated. The most important result from them con- 
 sists in the necessity of assuming the coexistence of both polarisa- 
 tions in muscle, from which it follows that the apparent magnitude 
 of the resulting positive or negative polarisation teaches us nothing 
 of the true magnitude of each of the component polarisations, for 
 the result we obtain represents the difference between the two 
 components, either of which components may be greater than the 
 difference itself. 
 
 10. On the Influence of the Direction of the Primary Current 
 on the Secondary Electromotive Action of Muscles. 
 I pass over the casual observations and the long list of preparatory 
 experiments which led me at last to the view that in a regular 
 muscle, for instance our group of muscles, the upper half exhibits 
 stronger positive polarisation in an ascending, and the under in a 
 descending direction. The best means of proving this consists in 
 sending the primary current alternately in both directions, first 
 through one half of the muscle and then through the other, allowing 
 exactly the same period of time to elapse between each momentary 
 current, in which case polarisation disappears with the exception 
 of a small residue which is then compensated. If its complete 
 disappearance is waited for, so long a pause becomes unavoidable 
 after each experiment, that great inconveniences arise. If you 
 wait on each occasion, without reference to time, till the polarisa- 
 tion reaches a certain or absolute value, that is to say till it 
 either sinks to a fraction of its original value or to a certain 
 number of divisions of the galvanometer scale, or ' degrees of the 
 compensator/ all kinds of uncertainties are met with. With a short 
 time of closure, and a strong current, the first method of experiment 
 always gives the best result. In any case it is to be preferred to 
 the statistical method which consists in obtaining a mean value 
 for the positive and negative polarisation in both halves, by experi- 
 ments on constantly renewed muscles, so that the current is only 
 once passed in one direction through each half of the muscle. The 
 variations in the strength of the primary current due to irregu- 
 larities in conduction are, as we have already seen (Sect, y), so 
 considerable in spite of every care, that for their compensation an 
 enormous number of experiments would be necessary. 
 
186 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 But even when the battery-electrodes are applied to one and the 
 same half of the group of muscles, we cannot be sure that the 
 primary current is equally strong in both directions. As is well- 
 known, there is no such law of reciprocity for electric currents as 
 there is for rays of light. The same electromotive force, acting in 
 opposite directions through the same series of conductors, often 
 generates currents of very different strengths. The necessity of 
 seeing clearly into the conditions of the irreciprocal conduction of 
 electric currents gave occasion to some experiments of von Beetz 
 and myself, which again were the germ of Professor Christiani's 
 thorough investigation of this subject^. I have convinced myself 
 that a perfectly sufficient reciprocity of conduction occurs through a 
 clay rod through which the current is led as through a muscle. 
 
 This general experience must not, however, be depended upon, 
 but care must be taken to note the strength of the primary cm-rent 
 along with that of the secondary. Professor Georg Quincke, 
 then still in Berlin, had the great kindness to help me with the 
 older observations of this sort, while latterly Professor Christian! 
 aided me, to whom I owe my warmest thanks for his assistance 
 in the present organization of my experiments. 
 
 The following table gives an idea of the great differences which 
 exist between the two halves of the muscle. It is to be under- 
 stood that one of the galvanometer electrodes is applied close to 
 the equator of the group of muscles, and the other invariably either 
 over the under, or under the upper battery-electrodes. 
 
 Ten Groves, — TiiiE of Closuee o"-32o. 
 
 1. Upper Half. 
 Muscle current: 189^, 21'=^ j . 
 
 S t +113"= I + 93 T +145 I + 81 T +159 I + 64 
 P I 171^= i 169 I 153 i 159 I iSo 1 190 
 
 2. Under Half. 
 Muscle current : 21^^'^, 22-^'^^^ \ . 
 
 s I +294 I +199 I +285 j{;^j^3 I +255 ||- ^9 
 pj 200 I 160 i 160 I 169 i 185 I 180 
 
 3. Upper Half repeated. 
 
 Muscle current : 264", 27"=e'^|. 
 
 I S I + 34 r + 87 
 P \ 210 I 220 
 
 ' Arthur Christiani, Beitrage zur Elektricitatslehre ; Uber irreciproke Leitung 
 elektrischer Strome, Berlin, 1876. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 187 
 
 4. Under Half repeated. 
 Muscle current : 285=", zo'^^^ \. 
 
 P I 190 i 190 
 The total of the ascending positive polarisation in the upper 
 half of the muscle is 504, that of the descending- 273. For the 
 lower half the corresponding numbers are 976 and 465. The 
 numbers in themselves are, I may incidentally remark, not to be 
 compared with those mentioned in the previous table, as on this 
 occasion the bobbin of the galvanometer was close up. 
 
 One glance at the table shows that the excess of positive polari- 
 sation cannot be supposed to be attributable to oscillations in the 
 strength of the primary current. These oscillations, which arise in 
 part from defects in the apparatus for regulating the time of 
 closure, are sometimes not apparent, sometimes they have the 
 opposite direction to that which would arouse suspicion, and even 
 when it is otherwise they are much too small to give any grounds 
 for suspicion. Moreover, such an explanation is opposed by the 
 cases of which the table offers three, in which, though the direction 
 of the polarising-current is less favourable to positive polarisation, 
 a preliminary negative deflection appears, and further the not 
 infrequent cases in which the polarisation is purely negative ; for 
 provided the time of closure is equal, either action in two opposite 
 directions, or negative polarisation, imply the existence of a stronger 
 polarising-current. These cases exclude the idea that we have to 
 do with a phenomenon of resistance. 
 
 Another conjecture which offers itself here deserves, however, 
 serious consideration. The whole phenomenon might rest on a 
 deception. As we would expect from the law of the muscle- 
 current, and as is evident from the table, as regards the foregoing 
 case a descending muscle-current prevails in the upper, an as- 
 cending one in the lower half of the muscle. The negative 
 variation and its after-effect are manifest in the upper half as 
 ascending, and in the lower half as descending electromotive action. 
 Consequently the after-effect in the upper half of the muscle, with 
 an ascending polarising-current, will reinforce the positive polarisa- 
 tion, and with a descending one will diminish it. In the under 
 half mutatis mutandis the same will happen. Now since the 
 muscle contracts at the passage of the polarising-current, every- 
 thing might thus be explained on old well-known grounds without 
 having recourse to anything new. 
 
188 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 I believe I have demonstrated in two ways that this explanation, 
 although conceivable, is not correct. I first formed an idea of the 
 magnitude which the after-effect would here attain. Instead of 
 the polarising battery, I allowed the mechanism, which regulates the 
 time of closure and of the ' transmission time ' (Ubertragjuigszeii), 
 (Sect. 6), to close the primary current of the inductorium. I left 
 the nerve still attached to the group of muscles and laid it 
 on the platinum plates of my electrodes {sfromzvfilhrende Vorrich- 
 tung)^. By the action of this mechanism, the muscle was thus, 
 instead of being polarised, tetanised for the fraction of a second, 
 and between the end of the tetanus and closing of the multiplier 
 circuit, the same transmission time intervened as between the end 
 of the polarising curi'ent and the closing of the multiplier circuit. 
 The result of several experiments which, according to the standard 
 of that time were perfect, was that when this method of experi- 
 menting was used, only traces of after-effect were visible, and were 
 by no means sufficient to explain the observed differences of polari- 
 sation. Indeed this might have been predicted, for the after-effect 
 increases with the duration of the tetanus which here only lasted 
 a very short time. 
 
 Secondly, I cut the muscles at the equator about half through 
 their thickness so that there was a gaping wound which behaved 
 as an artificial transverse section. The wedge-pad, which was 
 usually applied to the equator, was now applied to the wound. 
 The muscle-current now had the reverse direction in both halves, 
 and consequently the negative variation was also reversed. Yet 
 now, as formerly, the ascending positive polarisation predominated 
 in the upper half of the muscle, and the descending one in the 
 under half. 
 
 If I consequently considered it proved that the positive polarisation 
 in both halves of the muscle is stronger in the direction from the 
 equator to the ends than in the other direction, this is equivalent 
 merely to a statement of the phenomenon actually observed, a 
 phenomenon which admits of more than one interpretation. If 
 we call the positive and negative polarisation generated by the 
 ascending current respectively Pt, ITT, the corresponding polari- 
 sation from the descending current likewise Pi, 111, then the state 
 of matters is this, that not Pf — nT = Pl — m, but that in the 
 
 * Untersuchungen iiber tliierische ElektricitJit, vol. i. p. 450. It was in March, 
 1857, before ' physiological ' salt clay and unpolarisable electrodes were in use. The 
 muscle current was observed by the nerve-multiplier. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 189 
 
 upper half of the muscle the left, and in the lower half the right 
 term is the greatest. Naturally this can occur from many com- 
 binations of positive and neg-ative alterations of the four terms, 
 upon two of which, however, we must especially fix our attention. 
 It is either Pt^Pt or n^int ; the upper sign applies to the upper, 
 the under one to the lower half of the muscle. What is actually 
 the case will be decided further on with some amount of pro- 
 bability. 
 
 A remarkable fact which is also expressed in the table, is the 
 greater streng-th of all secondary actions in the lower half of the 
 muscle. Neglecting" the negative preliminary deflections, the sum 
 of these actions in the upper half amounts to 776^°, that in the 
 lower half to 1341®°. The corresponding numbers for the primary 
 current are 1453 and 1434^°. The difference of the secondary 
 actions is not explained by the downward tapering form of the 
 semi-membranosus, for I have observed it also in the symmetrically- 
 formed gracilis. With the electrodes of the galvanometer applied 
 symmetrically to the group of muscles, one would expect to obtain 
 with alternately ascending and descending polarising-current, a 
 stronger positive polarisation in a descending than in an ascending 
 direction. This I in fact observed in January, 1857, with less 
 perfect apparatus indeed, but with sufficient distinctness. 
 
 The observation is important, that in the series of experiments 
 here described, when they lasted long their result became uncertain, 
 and finally often became opposite ; so that, for instance, in the 
 upper half of the muscle not only is the descending positive 
 polarisation stronger, but negative polarisation appears even with 
 an ascending primary current. Finally, one circumstance deserves 
 to be mentioned which comes out so distinctly in the table that I 
 must consider it as unquestionable in spite of its paradoxical 
 nature. With very strong descending currents the polarisation in 
 the whole group of muscles becomes positive again after the critical 
 time of closure, which is as usual marked by deflections in oppo- 
 site directions. With 20 Groves there was only at first a trace of 
 this condition ; with 30 Groves it was stronger ; with 40, however, 
 so distinctly marked that with an ascending current and times of 
 closure respectively, id\ 1', 5', I obtained deflections of —125, 
 — 112, — 170^° ; with a descending current, +165, + 166, +ii7^°. 
 Whether this is connected with different polarisabiKty of the 
 two halves of the muscle in both directions, and how it is so, is 
 still quite obscure. In order to embrace the relation in Fig. 5, 
 
190 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 the curves should have been drawn from the ( — A) axis beyond 
 the T-SLxis. This was the less worth while that it is now apparent 
 that such a double figure would be necessary for each half of the 
 muscle. 
 
 Polarisation experiments, with cross direction of the primary 
 current in muscle, are made with great difficulty, and have not yet 
 been satisfactorily carried out. 
 
 11. On the Influence of various circumstances on the 
 Polarisation of Muscles. 
 
 The fact chiefly to be insisted upon is that internal positive 
 polarisation of muscles is only found in the living condition. 
 
 I had formerly stated that internal negative polarisation still 
 existed in boiled muscles ^. This was an error of which I became 
 conscious afterwards, when I experimented on single muscles instead 
 of entire limbs. The internal polarisability of frog's muscle is 
 destroyed by scalding, and completely annulled by boiling. 
 Nevertheless, strong internal negative polarisation is found in an 
 entire boiled frog's leg as I had already rightly observed. On closer 
 investigation, however, it was seen that while the muscle itself was 
 entirely inactive, this polarisation had its seat in the knee joint, 
 that is in the bones or ligaments, or both. The ankle joint of a 
 boiled leg was also found to be strongly polarisable. In the case 
 of the entire limb, the muscles of the limb serve only as conductors 
 to the joint. 
 
 Muscles which have been killed in other ways, — beef from the 
 butcher, frog's muscles which have lain forty-eight hours in a moist 
 chamber, or in water, or which have been dried over chloride of 
 calcium and then been softened again — only indicate on the galva- 
 nometer slight traces of internal negative polarisability with a 
 current of fifty Groves. 
 
 Scalding and boiling therefore exercise a remarkably destructive 
 influence on internal polarisability of the muscles. At the same 
 time, I found in September, 1855, that boiling materially lessened 
 the peculiar resistance of the muscle, which fact has been thoroughly 
 investigated since by Joh. Ranke^. Whether this fact and the 
 destruction of internal polarisability of muscles by boiling are con- 
 nected with each other we shall see further on (Sect. 21). 
 
 I think I have remarked that the long continued passage of a 
 
 ^ Untersuchungen iiber thierische Elektricifat, vol. i. pp. 378, 379. 
 
 ' Joh. Ranke, Tetanus. Eine physiologische Studie, Leipzig, 1865, pp. 10, 19 ff. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 191 
 
 very strong current destroys polarisability. This would explain 
 why a maximum of negative polarisability appears the earlier the 
 greater the density of the current. However this may be, I have 
 in my note-book cases where, in spite of the destruction of polarisa- 
 bility by a strong current, the muscle still contracted. From the 
 incompleteness of my former experiments, I am, however, not sure 
 of these things, and I simply mention them now to show how 
 much has yet to be investigated. 
 
 Between saline solution and muscle, positive polarisation is de- 
 veloped in the same manner as between this solution and other 
 animal tissue (Sect. 4). A slice of beef in which the current only 
 reaches a moderate density gives between saline pads only positive 
 polarisation, but a long thin strip gives polarisation in successively 
 opposite directions, or slight purely negative polarisation, for then 
 the internal negative polarisation makes its appearance along with 
 the external positive polarisation. 
 
 We have seen above (Sect. 8) that the limbs of a living frog 
 are capable of internal positive polarisation. I early tried similar 
 experiments on living men. In the years 1845 and 1846 I pro- 
 cured a powerful voltaic pile of the very earliest fashion consisting 
 of 150 zinc-copper double plates. Of these, 100 pairs were circular 
 having a diameter of i|'', 50 were square being 22" long each .side. 
 When the pile was used, disks of paper pulp steeped in a warm 
 tolerably strong solution of chloride of ammonium, were employed 
 as moist conductors to lay between the plates. With the last zinc 
 plate a copper plate was connected, and with the last copper 
 plate a zinc one, each of which was dipped in a basin with 
 saturated solution of salt. The equality of the two forefingers 
 with regard to their electrical condition was first tested by dipping 
 them in the conducting vessels of the multiplier. Of course I met 
 here with the same difficulties as in the experiments on negative 
 variation through voluntary tetanus ^ — difiiculties which have since 
 been overcome by the method of compensation. The index fingers 
 were now dipped in the basins connected with the battery in order 
 to receive the shock, and immediately transferred to the conducting 
 vessels in order to observe the secondary electromotive action. The 
 shocks of the battery were no trifle. When received with the whole 
 hand they were felt up to the shoulder, and during the time 
 
 ^ Monatsberichte der Akademie, 1852, p. iii ; Moleschott's Untersuchungen 
 zur Naturlehre des Menschen u. der Thiere, 1857, vol. ii. p, 247; Untersuchungen 
 liber thierische Elektricifat, vol. ii. part ii. pp. 186 fF, 
 
192 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 the current lasted those sensations of heat and cold were experienced 
 which J. W. Ritter made the subject of his speculations in the 
 beginning of the century ^. The friends of my youth, Wilhelm 
 Beetz, Ernst Briicke, Karl von Erlach, J. G. Halske, Gustav 
 Karsten, shared then with me the difficulties of these experi- 
 ments. I am happy to be able here, after nearly forty years, to 
 render them my thanks. The result was at first an irregular 
 confusion of positive and negative deflections, but it was soon 
 possible to control the phenomena. With a short closing of the 
 battery, there followed a positive polarisation, and with a long 
 one (15" '^as all that could be endured) a negative polarisation. 
 With a moderately long closure and rapid transference of the 
 fingers from the battery basins into the conducting vessels, 
 negative polarisation followed, and when the transference was slow, 
 positive polarisation. 
 
 The actions were strong enough to try if they would not make 
 themselves apparent in a rheoscopic thigh. This was in reality the 
 case. According to Humboldt's method-, the nerve had two pieces 
 of beef laid on it as conductors ; if I touched these with my fingers 
 after I had polarised myself in the battery circuit, the leg con- 
 tracted most vigorously. As one cannot succeed in making the 
 frog's leg contract by means of voluntary tetanus it is rather 
 interesting that it can be done by secondary electromotive 
 action. Experiments in polarising the human body by means 
 of the shocks of a Leyden battery were ^vithout result. On 
 the whole these facts seem to be in complete accordance with 
 those in frog's muscles and in li^^ng frogs. Unfortunately so much 
 uncertainty attaches to them that it for the present deprives 
 them of value. I was not then aware of the polarisation at 
 the surface of contact of electroljiies. I only looked to see whether 
 the fingers, after the closure of the battery, had an acid or alkaline 
 reaction like the ends of a frog through which a current was passed, 
 and I found once after long closure, traces of acid reaction on one 
 finger, though they appeared to me to be too slight to lay further 
 stress upon them, more especially as intentional soiling of the fingers 
 with dilute nitric acid and liquor potassae produced no actions com- 
 parable to those which we have to explain. I cannot now, however, 
 understand why I did not alter the experiment so as, for instance, to 
 
 '■ Untersuchungen, vol. i. pp. 356, 357. 
 
 ' Versuche iiber die gereizte Muskel und Nervenfaser, Posen uud Berlin, 1797, 
 vol. i. pp. 35 ff. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 193 
 
 take the shock with the forefinger while the secondary electro- 
 motive action was led off by the middle fing-er. Until the ex- 
 periment has been successful in this or some similar form, it is 
 open to the suspicion that we have here to do with external 
 positive polarisation in the skin, and not with internal polari- 
 sation of the muscles ; on the other hand there is no apparent 
 reason to doubt that the negative deflections observed in these 
 experiments depend upon true internal polarisation, and this new 
 action of the current in the human body appears to me to 
 be in some degree worth the attention of electro-therapeutists. 
 I cannot yet say whether the contraction of the rheoscopic frog 
 produced by polarised human limbs arose from positive or negative 
 polarisation. 
 
 12. Positive internal Polarisation of Muscles in Conflict with 
 
 Tetanus. 
 
 One of the most remarkable relations shown by the internal 
 positive polarisation of muscle is, that it is influenced by the state 
 of activity of the muscle, and in reality a tetanised muscle is 
 capable of less strong polarisation than one at rest. 
 
 The experiment is a difficult one on account of the negative 
 variation which is mixed up with it, which cannot always be 
 estimated with certainty. The galvanometer electrodes must be 
 applied to the group of muscles to which the nerve is left attached 
 in such a manner that the muscle-current is as weak as possible ; 
 then the negative variation is also weak. On account of the 
 asymmetrical construction of the semi-membranosus, a more or 
 less strong ascending current is present when the position of the 
 electrodes is symmetrical (Sect. 7). Hence, we obtain the desired 
 equality by applying the lower electrodes higher up on the muscle. 
 The group of muscles must be so firmly stretched that they do not 
 perceptibly shift on being tetanised. The amount and direction of 
 the remaining variation is noted. We then send a polarising shock 
 of short duration through the group of muscles every minute and 
 a half to two minutes, thus leaving it alternately at rest and 
 tetanised. We regulate the direction of the current so that the 
 negative variation is added to the secondary electromotive action. 
 In spite of this, the action is always less during tetanus than 
 during the repose of the muscle. 
 
 
 
194 
 
 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 The following table gives an example of this behaviour : — 
 
 September ii, 1855. 
 Muscle Multiplier, Platinum Electrodes in saturated solution of common salt, 
 Common Salt wedge-pads with albuminous films. Twenty Groves. — Time of closure 
 about \". A minute and a-half between the experiments. 
 
 No. 
 
 Tefanus 
 alone. 
 
 Ascending polarising current. 
 
 Alone. 
 
 With Tetanus. 
 
 A 
 
 I 
 2 
 
 4 
 
 5 
 6 
 
 7 
 8 
 
 5^T 
 
 1-5 r 
 
 + 53°-5 
 + 36.0 
 
 + 12-5 
 
 + 35°-o 
 
 + 16.5 
 
 + 9-5 
 
 — I-O 
 
 + 19-5 
 -3.0 
 
 The strength of the primary current was not noted, but the 
 resistance of the immobilised tetanised muscle, according to my 
 experiments, diminishes a little ^. 
 
 My note books mention no precautions to ensure that tetanus 
 should always last the same length of time, and that the current 
 should commence an equal time after the beginning of the tetanus. 
 With the knowledge and means we now possess it would naturally 
 be easy to obtain much completer numbers than the above. As 
 they stand tliey appear to leave no doubt regarding the thesis I 
 have propounded. 
 
 I have made similar experiments with negative polarisation. On 
 certain grounds I thought that this polarisation would not be altered 
 by the state of activity of the muscle. On account of the weakness 
 of the positive polarisation subtracted from it, it should therefore 
 appear stronger. Unfortunately, in testing whether it be so or not, 
 we meet with the difficulty that it requires a time of closure of at 
 least 10" to obtain adequate negative polarisation. As tetanus 
 begins earlier and must last longer than the polarising current, the 
 muscle is affected to such an extent that one generally succeeds in 
 seeing nothing but a rapid subsidence of all action. However, I 
 
 ' Untersuchungen liber thierische Elektricitat, vol. ii. part i, 1849, pp. 74 ff. 
 When this sheet of my work was printed, Helmholtz' researches, ' Uber die Warmeent- 
 wickelung bei der Muskelaction' (in Joh. Miiller's Archiv fiir Anatomic, &c. 1848, pp. 
 144 ff.) had not yet appeared. I should otherwise have had grounds for considering 
 the possibility that the diminution of resistance which I had observed depended on 
 the action of warming the muscle. I discovered the acidification of muscle by con- 
 traction much later. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 195 
 
 think I have before me in my rows of figures traces of what was 
 anticipated which can hardly depend on mere chance. 
 
 Finally, it is to be noticed that it is not accurately understood 
 in what respect an ordinary experiment on positive polarisation and 
 an experiment on positive polarisation in conflict with tetanus difier 
 from each other. For in experiments without tetanus the muscle is 
 also in contraction during* the passage of a short current. The time 
 has not yet arrived to examine this question. Perhaps the apparently 
 smaller positive polarisability of the tetanised muscle depends upon 
 the fact that the tetanus, which lasts longer than the closure of the 
 battery partly arrests the positive polarisation. 
 
 13. On Secondary Electromotive Phenomena of Nerves. 
 
 When I made a communication to the British Association in 
 Belfast, in the autumn of 1852, on secondary electromotive actions 
 of muscles and nerves (Sect. 7), I had not yet succeeded in obtain- 
 ing positive internal polarisation in nerves. In connection with 
 my hypothesis regarding electrotonus, and considering the fact 
 that muscles exhibit no extrapolar electrotonus, I then compared 
 muscle to hard steel and nerves to soft iron^. The middle tract of 
 a steel rod, surrounded by a coil through which a current is pass- 
 ing, becomes magnetic, and remains so after the cessation of the 
 current ; but it is only by action at a distance that the magnetism 
 spreads beyond the tract immediately involved. If the rod is 
 composed of soft iron it is magnetized throughout its length, 
 although with diminishing strength, from the coil to the ends ; 
 but after the cessation of the current the magnetism entirely dis- 
 appears. In the same way muscles and nerves then seemed to me 
 to behave differently in relation to the polarisation of their electro- 
 motive molecules. But I had theorised too soon. When in the 
 winter of 1852-53 I returned with improved apparatus to the in- 
 vestigation of secondary electromotive actions in nerves and muscles, 
 I found at once that nerves also possess positive internal polari- 
 sability or power of coercion [Coercitivkraft), in accordance with the 
 comparison just made, only that their polarisability is more difficult 
 to seize upon than that of muscles, for reasons that are easily 
 understood. 
 
 From the great importance which this phenomenon appeared 
 to have in relation to the theory of electrotonus, I have since 
 
 ^ Untersuchungen, vol. ii. part i. p. 326. 
 O 3 
 
196 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 repeatedly taken up the investigation as my views and methods 
 improved, and innumerable experiments on polarisation of nerves 
 have been Ij^ng* for years in my note-books. 
 
 Among other things I had already, in the years 1857 and 1858, 
 completed a table which exhibited the secondary electromotive 
 actions in nerves for a series of times of closure and current densi- 
 ties in both directions. The experiments were made with the nerve- 
 multiplier, the two sciatic nerves of a frog laid together being 
 used. A new pair of nerves was employed for each time of closure, 
 and each current density in each direction. Pfliiger, who was 
 then in Berlin, had the great goodness to assist me in these 
 experiments. The table for nerves suffered in the same way as 
 that traced out for muscles two years before, from all the faults 
 which were attached to my results before the discovery of unpolaris- 
 able electrodes, of physiological salt clay, of the method of compen- 
 sation and of the aperiodic galvanometer, besides which the strength 
 of the primary current was not regularly nor closely enough 
 watched. In spite of this I am obliged to make use of the older 
 table, as it has not yet been possible for me, as in the case of 
 muscles, to replace it with one less faulty. As, however, the 
 results of the older table of nerves resemble very much those of the 
 new table of muscles, perhaps the evil is not so great. Indeed the 
 graphic representation given in Figs. 5 and 6 of the polarisation- 
 curves, drawn relatively to the times of closure and opening, apply 
 nearly as well to nerves as to muscles. Current densities less than 
 a certain amount only give negative polarisation. With greater 
 current densities of from 5 to 50 Groves, and very short time 
 of closure, purely positive polarisation appears. With a somewhat 
 longer time of closure, which amounted to perhaps 0-2 second, 
 the polarisation in nerves was already in opposite directions, that is 
 to say, a positive deflection immediately succeeded a negative one. 
 If the time of closure amounted to more than a second, the polar- 
 isation was purely negative. I obtained the strongest positive 
 polarisation with a very short closure, from 25 to 30 Groves ; as 
 in the case of muscle, the force remained apparently below that of 
 the current of the nerve at rest. The strongest negative action 
 resulted, exactly as in muscle, from the passage of relatively 
 weaker currents for a longer time. After 45' five Groves, after 
 15' seven Groves, after 5' ten Groves swung the needle against 
 the stop of the multiplier, that is, a stronger action was produced 
 than by the nerve-current. Here also, as in muscle, the observ- 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 197 
 
 ation holds g-ood, that the resulting- polarisation, which is the 
 difference between two component polarisations, is no measure of 
 their strength, for this may be many times as great as the observed 
 diflFerence. For, obviously, the various kinds of phenomena of 
 nerve-polarisation lead unavoidably to the assumption that in 
 nerves as well as in muscles there are present at the same time 
 two polarisations, — a positive polarisation which at once attains 
 almost its full amount a] though it continues to increase slowly and 
 is very persistent, and a negative polarisation which increases with 
 the time of closure, and rises more and more above the former, 
 but disappears more quickly after the opening of the battery- 
 circuit. Opening currents of the ordinary inductorium of an 
 adequate strength produce purely positive polarisation. 
 
 In regard to the finer details of the phenomenon, the various 
 maxima of polarisation-curves recognised in muscles, and the influence 
 on the position of these maxima of current-density, time of closure, 
 and time of opening, it is desirable to await further investigations, 
 although even in this respect the old nerve-table, with all its 
 incompleteness, has a certain resemblance to the new table of 
 muscle. We know nothing as yet about the proportional strength 
 and duration of polarisation in muscle and in nerve with an equal 
 current-density, &c., and it would be very difficult to form a 
 reliable conception of it. 
 
 14. Matteucci's Experiments on Nerve Polarisation. 
 
 I first gave an account of internal negative polarisation of nerves 
 in 1856, in my paper upon this phenomenon in moist porous bodies 
 generally^. In 1867 I again returned to it as a circumstance 
 which is opposed to the constancy of currents in circuits containing' 
 nerves 2. That nerves possess, in addition, internal positive polari- 
 sation, although perhaps in a manner only apparent to persons well 
 informed on the subject •\ I have often pointed out when speaking 
 of the secondary electromotive actions of electrical organs. 
 
 Although my papers on polarisation of electrolytes and moist 
 porous bodies found their way into French and Italian journals*, 
 
 1 Monatsberichte etc., August 4, 1856, p. 457 ; G-esammelte Abhandlungen etc., 
 voL i. p. 19. 
 
 =" Arcliiv fiir Anatomie, Physiologie, etc., 1S67, p. 262 ; Gesammelte Abhandlungen 
 etc., vol. ii. p. 192. 
 
 3 Monatsberichte etc., 185S, p. 106; Untersuchungen am Zitteraal etc., p. 206. 
 
 * II nuovo Cimento ec, vol. v. Maggie e Giugno, Pubbl. il 9 Luglio, 1857, p. 338 ; 
 Annales de Chimie et de Physique, 3">« S^rie, 1S60, vol. Iviii. pp. 314, 318. 
 
198 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 Matteucci in i860, four years later than I and without mentioning 
 me, described similar phenomena as if he had discovered them, and 
 laid claim on the same occasion to the discovery of internal negative 
 polarisation of nerves also. 
 
 Like me, he terms the actions 'secondary-electromotive.' His 
 methods of investigation certainly do not resemble mine. He lays 
 the two ends of a nerve first on the platinum electrode of a battery of 
 from two to eight cells composed of zinc, carbon, and salt-water ele- 
 ments, passes the current through, from two seconds to three minutes, 
 and then transfers the nerve to the multiplier pads. Negative polar- 
 isation is then found in the intrapolar tract, but positive in the 
 two extrapolar tracts, and stronger in the one bordering on the 
 kathode. These experiments were made not only with nerves 
 of frogs, but also and by preference with those of warm-blooded 
 animals, sheep, rabbits, and fowls. Negative polarisability was 
 found to persist for several hours after the loss of vital properties. 
 It was destroyed only by boiling temperature and compression. 
 Matteucci attributes secondary electromotive action to the aci-ds 
 and alkalies set free at the platinum electrodes^. 
 
 In a second communication, however, he alters his expressions 
 in many ways in a manner peculiar to himself, without saying 
 whether the first were mistaken or not. Now, the nerves were 
 ' in most cases ' not brought directly into contact with platinum 
 electrodes, but passed through strips of woollen stuff steeped in 
 spring water. The batteries were of 8 to 10 Groves, and the 
 time of closure amounted to 25-30 minutes. He no longer speaks 
 of polarisation of the extrapolar tracts, but now states that 
 the part of the intrapolar tract adjoining the anode has a greater 
 negative secondary electromotive force than that adjoining the 
 kathode, and that the difference between them is greater when the 
 current ascends than when it descends in the nerve. He professes 
 to have convinced himself of this by opposing the two sections to 
 one another in the same circuit. This kind of action he ascribes to 
 contamination of the nerves with ions. Between the terminal 
 portions of the nerves having an electro-chemical action of this 
 kind, he seems to distinguish a middle portion, everj^ point of 
 which has a negative secondary electromotive action even after 
 very short closure of the current ^. 
 
 ' Comptes rendus etc., 27 Fevrier, i860, vol. 1. p. 412; Archives des Sciences 
 physiques et naturelles, Fevrier, 1S60, vol. vii. p. 173. 
 
 * Comptes rendus etc., vol. lii. 1861, 11 Fevrier, p. 231 ; 13 Mai, p. 954. 
 
MUSCLES, NEBVES, AND ELECTRICAL ORGANS. 199 
 
 In a third communication Matteucci returns to the extrapolar 
 electrotonic currents, and there, as usual, repeats his statements 
 on this subject many times with all kinds of variations^, through 
 which it is useless to follow him. On the polarisability of nerves, 
 however, he founded his conception of extrapolar electrotonus cur- 
 rents as being- escape-currents, which spread from the anode along 
 the anelectrotonic tract externally, and reach the kathelectrotonic 
 tract by the axis cylinder, along the outside of which they return 
 to the kathode : all of this he imitated by a platinum wire covered 
 by a moist coating. If such a wire is laid on a pair of electrodes, 
 extrapolar diffusion of current occurs according to the law of 
 electrotonus, but is absent when zinc wire covered by a coating 
 steeped in sulphate of zinc has been employed. As is well known, 
 this is the theory which has since been taken up by Hermann, to 
 whose experiments on internal polarisation of nerves we now come. 
 
 15, Hermann's Researches on Nerve Polarisation. 
 
 In his paper of the year 1867, in which Hermann first advanced 
 his necrosis hypothesis, he gave a philosophical^ explanation of 
 electrotonus in connection with this hypothesis, according to which 
 a current opposed to the polarising current predominates in the 
 intrapolar tract, which is comparable to Peltier's counter current 
 in a thermopile through which an external current is passing. In 
 this supposed coincidence Hermann saw a striking confirmation of 
 his speculation and a new support of his view^. 
 
 In my confutation I contented myself with exposing a flaw 
 in the series of conclusions by which Hermann believed that he 
 could connect his conception with the law of the conservation of 
 energy ^. I had reasons for not withdrawing the veil from the whole 
 subject of the secondary electromotive actions of nerves with which 
 only I and a few initiated people, such as Pfliiger, were acquainted. 
 
 ^ Comptes rendus etc., 16 September, 1861, vol. liii. p. 503; 20 Avril, 1863, voL 
 Ivi. p. 760 ; 22 Juillet, 1S67, vol. Ixv. pp. 131, 195 ; Annales de Cliimie et de Physique, 
 3™° S^rie, vol. Hx. p. 385 ; Electro-Physiological Researches, Eleventh Series, Philo- 
 sophical Transactions, etc., 1861, p. 363; Proceedings of the Royal Society, vol. xi. 
 p. 384; Philosophical Magazine, etc., 4th Series, vol. xxiv, October, 1862, p. 311; 
 Corso di Elettro-Fisiologia in sei Lezioni, date in Torino ec, Torino, 1861, 
 
 P-59- 
 
 ^ [The word used is ' naturphilosophisch,' which cannot be rendered in English. (Tr.)] • 
 ^ Weitere Untersuchungen zur Physiologie der Muskeln und Nerven, Berlin, 1867, 
 
 p. 40. 
 
 * Gesammelte Abhandlungen, vol. ii. p. 341. 
 
200 SECONDA.RY ELECTROMOTIVE PHENOMENA IN 
 
 In the g-cneral interest it would indeed have been better if I had 
 at once told Hermann that positive polarisation prevailed in the in- 
 trapolar tract. The course of science in this direction would have 
 been different, and perhaps more fruitful. 
 
 A year later Hermann broug-ht forward his hypothesis anew, 
 somewhat more carefully worked out ; after having- made use of 
 some of the hints contained in my confutation ; he this time made 
 a point of supporting" his electrotonus theory by experiments ^. ' The 
 battery consisted of from 3 to 6 very small Daniells ; the nerve- 
 current was always accurately compensated to zero before the passag-e 
 of the current. The passag-e of the current lasted, on an averag-e, 
 for a minute.' Under these circumstances Hermann naturally suc- 
 ceeded in seeing nothing* but negative polarisation in the intrapolar 
 tract. As this result appeared to agree with his hypothesis that 
 ' the splitting of the nerve-molecules is quickened by the kathode 
 and retarded by the anode,' and as it appeared to contradict my 
 molecular theory, he was satisfied to hold by it. 
 
 Hermann's experiments also embraced the extrapolar tracts with 
 which Adolph Fick had occupied himself after Matteucci-. He 
 and Hermann arrived finally at the same conclusion, though the 
 latter really has the priority, inasmuch as Fick's first communications 
 did not correctly represent the facts. These are in Hermann's 
 words, ' Both the extrapolar tracts act for a short time after 
 the current is broken, and both are directed away from the tract 
 in which the current flows ; ' but the anelectrotonic after-current 
 which is opposed to the polarising current is the stronger of the 
 two^. 
 
 In the meantime, Hermann allowed his hypothesis upon the 
 difl'erent ' rapidity of splitting ' (^SjjaUimgsgescJiw'mdigkeit) as the 
 cause of the electrotonic current, to drop, and took up in its place 
 Matteucci's above-mentioned explanation of the extrapolar elec- 
 trotonic current by diffusion, the complete verification of which 
 he made his chief task. His experiments on intrapolar negative 
 polarisation served him now as a proof of Matteucci's hypothesis *, 
 
 ' Untersuchungen zur Physiologie der Muskeln und Nerven, Drittes Heft, Berlin, 
 i868, pp. 71 ff. 
 
 " Centralblatt fiir die medicinischen Wissenscliaften, 1867, p. 436; Untersuchungen 
 aus dem physiologisclien Laboratorium der Ziiriclier Hochschule, parti, Wien, 1869, 
 p. 129. 
 
 ^ Handbuch der Physiologie, vol. ii. p. 164. 
 
 * Pfliiger's Archiv fiir die Geaammte Physiologie, 1872, vol. vi. p. 357 ; Handbuch 
 der Physiologie, vol. ii. pp. 164, 165. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 201 
 
 as they had done formerly for his own. Matteucci's hypothesis 
 is undeniably to be preferred, inasmuch as something- may be made 
 of it. It is a g-ood discussable mode of conception, of which I 
 am all the more ready to take account that the fundamental idea 
 is my own. As early as in my researches in 1849 I considered 
 expressly, and tested with all the apparatus which then existed, the 
 possibility of exj^laining- the extrapolar electrotonic current by the 
 assumption then much in vogue, that the axis cylinder conducted, 
 and the medullary sheath insulated. In that case a branch current 
 would flow from the anode along the perineurium to the next cross 
 section, along- the axis cylinder to the section nearest to the anode, 
 and back through the perineurium to the kathode. In this form, 
 however, the hypothesis did not stand the test^. 
 
 Whether it would do so in the form which Hermann gave it, 
 on the ground of the inferior transverse conductivity of nerves 
 discovered by him, I rather hesitate to believe on his word, and 
 I have not yet had time to convince myself about it by experi- 
 ments of my own. If I succeed in doing this I shall freely 
 acknowledge it. In the meantime it is certain that the intrapolar 
 tract not only possesses the negative polarisation which is rather 
 physical, and on which the extrapolar currents may depend, but 
 also that positive polarisation of a more physiological nature which 
 my hypothesis requires. It is certain that even if the extrapolar 
 electrotonic currents depend merely on diffusion of current, the 
 nature of electrotonus as consisting in positive polarisation is 
 not thereby affected. We should only be rid of the doubtful 
 necessity of explaining its spread beyond the electrodes. It is 
 further certain that while I discovered the positive polarisation of 
 the intrapolar tract at a time when I possessed none of the appa- 
 ratus of the present day, Hermann, who found this apparatus ready, 
 has from 1867 till now overlooked the fact which is fundamental in 
 the field upon which he appeared as a reformer. Finally, it is certain 
 that all that he thought out regarding electrotonus, and which he 
 propounded with so much confidence, is reduced to mere chafi" in 
 presence of this fact, and that the investigation of electrotonus 
 in general must begin from this fact. Hermann stated that the 
 comparison of a living nerve and one killed by exposure to steam, 
 showed no apparent difference as a conductor in favour of the 
 former, as must be the case if a current set free electromotive forces 
 in the same direction as itself in the living nerve ; and he regards 
 
 * Loc. cit. vol. ii. part i. pp. 229, 275-282, 347-350. 
 
202 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 this experiment, which was moreov^er first suggested by me^, though 
 on account of theoretical difficulties it was not carried out, as an 
 ex2oenmentum crucis against the molecular theory of electrotonus-. 
 
 To Hermann's estimate of the electromotive force acting in a 
 nerve whose dipolar molecules collectively turn their similar poles 
 to the same side, I am the less able to object in principle that 
 I explain the functions of the electrical organs by similar esti- 
 mates^. That no force comparable with the calculated one appears 
 in nerves has been vindicated by the assumption made by me 
 already in the ' Untersuchungen,' that when the molecules are 
 thrown out of the true direction in relation to the polarising 
 current their rotation amounts only to a small angle, and not 
 to 1 80°.* 
 
 The difficulty which Hermann's experiment offers disappears 
 before the known facts of nerve polarisation. With a long time of 
 closure, negative polarisation exceeds positive, and as boiling heat 
 destroys both, the living nerve ought apparently to conduct worse 
 than the dead one, not better^ as Hermann expected. In order to 
 find the living nerve a better conductor than the dead one the 
 experiment must be made with a very short time of closure. Even 
 then positive polarisation predominates over the negative only when 
 the strength of the primarv^ current is so considerable^ that there 
 is little hope of observing the slight excess of positive over negative 
 polarisation when this is added to the primary force. There is, 
 however, one experiment which might be explained in this sense. 
 If a nerve closes the secondary circuit of an inductorium, the alter- 
 nating currents of which would otherwise, according to well-known 
 rules, compensate one another in the galvanometer, the quicker and 
 shorter opening currents predominate, according to Fleischl. A 
 single opening shock also passed through the nerve causes a 
 stronger deflection than a closing shock ^. Inasmuch as the opening 
 shock is more exciting than the closing one, Hermann sees in this 
 nothing more than a special case of his 'theory of polarisation 
 
 ' Untersuchungen uber thierische Elektricitat, vol. ii. part i. p. 328. 
 
 ^ Untersuchungen zur Physiologie der Muskeln u. Nerven, part iii, 1868, p. 67; 
 Pfliiger's Archiv, 1872, vol. vi. p. 328 ; Die Ergebnisse neuerer Untersuchungen auf 
 dem Gebiete der thierischen Elektricitat, Sep. Abdr. aus der Vierteljahrschrift der 
 naturforschenden Gesellschaft in Zurich, 1 878, part i. p. 17; Handbuch der Phy- 
 siologie, vol. ii. 1S78, p. 172. 
 
 ^ Untersuchungen am Zitteraal, p. 275. 
 
 * Untersuchungen Uber thierische Elektricitat, vol. ii. part i. p. 325. 
 
 ^ Sitzungsberichte der Wiener Akademie der Wissenschaften, 1878, part iii. 
 vol. Ixxvii. p. 159. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 203 
 
 increment' (compare Sect. 19)^ After what I have just said about 
 nerve polarisation, however, positive polarisation in Fleischl's experi- 
 ment must exceed negative to a considerably greater degree in the 
 case of the opening than in the case of the closing shock, and the 
 possibility that this participated in the result is at least to be borne 
 in mind. 
 
 If the negative polarisation does not spread beyond the poles 
 (which I do not as yet consider proved), while positive polarisation 
 does so (which I, for my own part, never regarded as demonstrated), 
 we might perhaps believe that the extrapolar positive polarisa- 
 tion exceeded the similar intrapolar to some considerable extent. 
 But I have no intention of entering now on a discussion of these 
 obscure and complicated questions. 
 
 In so far as the extrapolar electrotonus currents outlast the 
 opening of the polarising current, the intrapolarisation being 
 however necessarily present during the closing of the battery, one 
 sees that the distinction between the second and third of the 
 different classes of electromotive phenomena in muscle and nerve 
 referred to in Sect. 2, disappears, and it would be better to distin- 
 guish only two classes, that of the independent or primary, and 
 that of the secondary phenomena, by including under the latter, 
 without reference to the time of their appearance, all the phenomena 
 an extraneous current generates as a current, and not merely as 
 an irritant in muscle and nerve. 
 
 16. Tigerstedt's Experiments on Nerve Polarisation. 
 
 Last year Tigerstedt of Stockholm described some experiments 
 on internal polarisation of nerves ^. For certain reasons he led off 
 the after-current through the same electrodes which had led in the 
 polarising current. He used no stronger battery than three 
 Meidingers. He regulated the time of closure with the hand, 
 according to the metronome : the shortest amounted to i'', the 
 longest to 240", The transmission time was determined by means 
 of the Marcel Deprez's electromagnetic marker and varied between 
 o"-o2 and o"'04. 
 
 Tigerstedt also obtained only negative polarisation. He formu- 
 lated his results as follows: ' i. With a current up to three Mei- 
 dinger cells the polarisation is almost directly proportional to the 
 
 ' Pfliiger's Archiv, 1879, ^°^- ^^^- P- i^^- 
 
 '■' See Tigerstedt's paper on Internal Polarisation, p. 77 of this volume. 
 
204 SECONDAKY ELECTROMOTIVE PHENOMENA IN 
 
 strength of the polarising- current. 2. If the polarising current 
 acts during an unequal length of time on the nerves under other- 
 wise unaltered circumstances, the polarisation increases ; it rises at 
 first faster, and later more slowly, finally reaching its maximum 
 extremely slowly. 3. If the polarising current is opened, the 
 polarisation instantly reaches its highest amount, and after that 
 sinks continuously. This decrease is at first very rapid, later on 
 always slower, so that the polarisation persists a long time after 
 the opening of the polarisation-current, and only approaches the 
 zero point asymptotically.' Tigerstedt would certainly have met 
 wdth the internal positive polarisation of nerves in his carefully- 
 conducted experiments, if he had employed greater current den- 
 sities and a shorter time of closure. As it was, however, he 
 remained, like his predecessors, on the further side of the critical 
 time of closure. Neither Hermann, Fick, nor Tigerstedt had a gal- 
 vanometer in the battery circuit. I do not think that this was of 
 any importance in their experiments, hut I would caution any one 
 from engaging in a series of experiments with strong currents and 
 long times of closure without following this rule. (Sect. 6.) 
 
 17. On the Influence of the Direction of the Primary Current 
 on the Secondary Electromotive Actions of Nerves. 
 
 It was the desire and hope of communicating something definite 
 about the dependence of internal positive polarisation of nerves on 
 the direction of the polarising current, that chiefly withheld me 
 from publishing my investigations in their present condition. The 
 remarkable fact that the greater strength of positive polarisation of 
 muscle in the direction from the equator, in or near which the hilus 
 is situated, towards the ends (Sect. 9), led me for a time to suppose 
 that the intra-muscular nerves were concerned, and that they 
 underwent a stronger positive polarisation in a centrifugal direc- 
 tion. This led me to the idea of experimenting as to how the 
 motor and sensory roots of the spinal nerves would behave, in 
 relation to internal polarisation. 
 
 Polarisation experiments on the roots are naturally made much 
 more difficult by their shortness. The first requisite is to procure 
 the largest possible frogs. In 1857 I measured a K.. esculenta of 
 298 mm. and one of 305 mm. from the point of the nose to the 
 longest toe. I succeeded, however, in applying four wedge-shaped 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 205 
 
 pads, covered with films of albumen, even to root-bundles of 
 smaller animals. Thoug-h not absolutely necessary, it is very 
 much to be desired that the experimenter in these researches 
 should be able to g-ive his undivided attention to the manipula- 
 tion of the apparatus, and that he should be relieved of the 
 trouble of preparation. Here again I remember, with heartfelt 
 thanks, Pfliig-er's self-sacrificing- assistance. During many a July 
 afternoon in the tropical summer of 1857, while I kept everything* 
 in readiness at the nerve-multiplier, he prepared bundles, first 
 of anterior and then of posterior roots, which, as it must be done 
 as quickly as possible without pinching- or displacing- the roots, is 
 not so easy as might perhaps be imagined. 
 
 In these experiments the battery consisted of only a few Groves ; 
 the time of closure amounted to about o'^-z. Any nerve-current 
 present was compensated by means of a primitive arrangement, out 
 of which the round compensator was subsequently developed ^. The 
 roots were placed, along with the four wedge-shaped pads, on a 
 small support, the ' root carrier,' which could be rotated on a per- 
 pendicular axis. This allowed them to be reversed between the 
 'conducting vessels^' without risk of displacing them on the pads. 
 The object of this was to obviate as much as possible the electrical 
 inequalities of these vessels and any irreciprocity of conduction in 
 the circuit (Sect. 10). The current was sent through the roots 
 alternately, first in one direction and then in the other, the direc- 
 tion being changed every minute and a half. After every four 
 experiments the roots were turned round. Such series as the 
 following were thus obtained : — 
 
 Three Groves. Time op Closure about o"-2. 
 Motor Boots. 
 
 " '-13 
 
 |r_io°tr-i2 |f-ii tr-i: 
 n + i6° I 1 + 10.5 it + 16.5 I t+ic 
 
 Hoots reversed. 
 
 tf-12 |f-ii Ti-i2|r-i2 
 
 I 1 + I2-51 L + ii-5 I L + i3i 1 + 13 
 Boots reversed. 
 
 |r-i2Tf-i3 |r-i4Tr-i2. 
 
 I 1 + 12 I 1 + 12-5 1 1 + 12 I 1+12 
 
 Sensory Boots. 
 |r_ii|r_i7 I r_ii 1 r_i7.5^ 
 I 1 + 1411 + 12.5 I 1 + 13 jl+I2 
 
 ' Gesammelte Abhandlungen, voL i. pp. 176, 177. 
 
 ^ [Du Bois' ' Zuleitungsgefasse,' not so familiar now as formerly. Ed.] 
 
206 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 Boots reversed. 
 
 I r-34rr-i6| r-4irr-i 
 i 1+14 1 1+311 1+12 1 1+3 
 
 Jioofs reversed. 
 + 16 
 
 tj_is|f-28 Tr-2i ir-27 
 
 il + 16.5 I 1 + I6.5U + I7- 
 
 In order to judg-e of these figures correctly it must be remembered 
 that the currents were still led off to the nerve multiplier throug-h 
 platinum in saline solution, in the old way. After a momentary 
 current the returning* needle would swing back over the zero point 
 into the other quadrant, and here it would be perhaps carried 
 beyond its deflection in the first quadrant in consequence of the 
 charging- of the platinum plates. The polarisation- current of nerves 
 is not however comparable to an instantaneous current of this kind. 
 Even the neg-ative polarisation, althoug-h the more transitory of 
 the two, sinks more or less gradually. The middle point of oscilla- 
 tion for the undamped needle is therefore temporarily transferred 
 to the negative quadrant, as we will call it. In its backward smng- 
 the needle may indeed cross the zero point, but its positive deflection 
 will not exceed its negative. If this should happen, we may conclude 
 that action successively in opposite directions is present. Con- 
 sequently where in the above table a larger positive number follows 
 a smaller negative one, action in two directions occurred, that is to 
 say, a negative after-current changed suddenly into a positive one, 
 and the difierence of the numbers gives an approximate measure for 
 the comparative strength of the latter. Accordingly in both tables 
 up to the mark || we have the expected result. The subsequent ap- 
 pearance of the opposite result reminds us of the reversal of the normal 
 relation in the two halves of a regular muscle (Sect. 10). Almost 
 without exception the series began normally in the roots, but fre- 
 quently the reversal took place not later than the second or third 
 momentary current. This was especially the case in the motor roots, as 
 in the examples given the sensory roots exhibited the condition in 
 question longer than the motor ones. If we consider that in these 
 experiments both sides of the separated roots were brought into 
 contact with white of egg and injured with electric shocks, one can- 
 not avoid the impression that apparently, in spite of all imperfec- 
 tions, they render the thesis probable that, — In motor roots positive 
 polarisation predominates in a descending direction, and in sensory 
 roots in an ascending direction ; the direction in both cases being that 
 of the physiological innervation wave. Certainly until more is known, 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 207 
 
 the same observation holds good as for the two halves of a muscle. 
 For strongs positive polarisation weak negative polarisation may be 
 substituted in the thesis ; thus we may say that in the motor roots 
 the ascending- negative polarisation predominates, in the sensory the 
 descending. When discussing muscles I referred to reasons to be 
 afterwards given, for taking the same view of them which I do of 
 nerves. 
 
 Various circumstances, amongst others the arrival of the West 
 African Malapterurus, compelled me to break off these experiments, 
 and I was only able to resume them twenty-two years later, in the 
 autumn of 1879. I was then furnished with all the apparatus in- 
 vented during the interval, and this time kindly aided by Professor 
 Gustav Fritsch, whose anatomical skill was the more welcome, that 
 in a consignment of Hungarian frogs for which I was indebted to 
 Professor Jendrassik of Pesth, we found no such giants as the waters 
 of the Brandenburg Mark sometimes supplied formerly. This series 
 of experiments also was not completed. It did not seem advisable to 
 continue them during the winter, after the larger frogs had been 
 used. After that I was under the necessity of devoting myself to 
 the editing of Sachs' results. These new observations, however, 
 yielded much that was of importance. First of all, I tested the 
 behaviour of a mixed nerve-trunk by the method of experiment to 
 which the roots were to be subjected. Here no definite difference 
 was apparent between the actions observed when the current was 
 ascending and when it was descending, as the following example 
 shows : — 
 
 Five Groves. Time of Closure o"-oi*j. 
 
 Portion of the Sciatic Nerve between the Vertebral Column and the origin of the 
 branches to the Muscles of the Thigh. 
 
 I r- 5-Tf- 2 I r_ 7t r- 4| J- 3 etc 
 n + 33'"= I 1 + 36-5 i t + 23 I 1 + 26 i \ + 2o ^'''• 
 
 Time op Closure o'^-osi. — Another similar portion. 
 
 I r-o.5Tr-o-5 I r-o-sT f- I I J-o-5T(- I I r-o-5 gt^ 
 
 II + 23 |L+i5il + 2i |l + i7il + ^7 11 + 1711+21 
 
 Central portion of the nerve. 
 
 T- 5||- 6rf- 6 I f- 8 T- 5 I J-i°etc 
 I + 18 i 1 + 16 I 1+ 19-5 i 1+15-5 I +i8.5i\+ 8 ^'°- 
 
 Time of Closure o^'-o'jS. 
 Lower portion with the peroneal and tibial nerves. 
 
 IJ-4Tf-3|/-2 TJ- 3 i;-2-5Tr-2-5 ,,, 
 n + 6|t+i5it + 2i.5|t+i4.5it + i4 11+15 ^*'=- 
 
208 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 The actions are always in two directions, as is at once evident on 
 tlie g-alvanometer, but the first negative deflection is sometimes 
 limited to a scarcely perceptible jerk. The inequalities in the several 
 series depend undoubtedly on the imperfection of the apparatus for 
 regulating the times of closure and transmission. Taking together 
 all the ex2:)eriments I have up to this time made on mixed nerves, 
 there appears a slight preponderance of ascending positive polarisa- 
 tion. In descending it seems to decrease, and indeed even to 
 become reversed. However, the experiments are still too imperfect 
 to permit of such a generalisation, at all events no discussion of the 
 question whether this is connected with the proportion in which the 
 two kinds of fibres are contained in the transverse section of the 
 nerve, can be entered on. On comparison of the above series with 
 those on sensory roots a striking difference appears in favour of the 
 suggested law. 
 
 Three Groves. Time of Clositre o"-oz'j. 
 Sensory Boots. 
 
 Boots reversed. 
 
 tf- 5lJ-Sr[- 5l/-9Tf- 6|/-9Tf- 5. 
 I 1 + I3I 1 + 7 I i + '3il + 9l l+"i L + 6| 1 + ^3 
 
 It is remarkable, however, that the motor roots here again 
 yielded a less favourable result. 
 
 Five Groves. Time of Closure o''-oi5. 
 Motor Boots. 
 
 1 J- 3TJ- 2 I f- 3t f- 2 I r_ 3tj- 3l J- 4 II U-2-5 1 J-3^8_ 
 i 1 + 23 I 1 + 17 1 1 + 22 1 1 + 20.5 i 1+17 1 l + i2i 1 + 16 II I 1+17 i 1 + 18 
 
 Up to the mark || the suggested law is pretty clearly expressed ; 
 but for the most part even to this extent such was not the case. 
 I was no more successful with the nerves of the thigh. 
 
 If consequently the law cannot be considered to be demonstrated, 
 on the other hand it is not yet determined that it does not hold. 
 It is possible that a combination of current density and time of 
 closure which I did not hit upon in my experiments is required, in 
 order to demonstrate it on the motor roots. In order to make 
 further progress we should require a consignment of bull -frogs 
 (R. mugiens or Catesbyana) from North America. So far my 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 209 
 
 endeavours to procure them have been fruitless. The optic nerve of 
 a large osseous fish ought to he an excellent object for the fibres 
 which conduct centripetally, but this I have not yet found time to 
 try. For fibres having a physiological centrifugal action there seems 
 to be nothing better than the electric nerves of a Torpedo. I will 
 not say anything of an abundance of other experiments which are 
 here suggested. Even the experiments already communicated seem 
 well deserving of the attention of physiologists. They present the 
 first example of an electromotive difference between centrifugally 
 and centripetally acting fibres. One might well in this case em- 
 ploy the expression of which Paul Erman made use with reference 
 to the law of contraction : ' The mere suspicion of such a difierence 
 inspires awe ^.' One must therefore be all the more cautious. Even 
 if the supposed law were demonstrated, it remains to be con- 
 sidered whether it must be regarded as an action of the physiological 
 innervation wave, or a provision for facilitating its propagation in a 
 definite direction. I once doubted its existence or even its possi- 
 bility on theoretical grounds^. 
 
 18. On the influence of various conditions on the Polarisation 
 
 of Nerves. 
 
 Regarding the question of the conditions which influence the 
 polarisability of nerves, I have scarcely any results which are 
 methodically arranged. Matteucci already knew that nerves 
 which have been exposed to boiling-heat do not any longer 
 show a negative after-current (Sect. 14). If, however, the nerves 
 are protected from drying and allowed gradually to die at a low 
 temperature, they retain their secondary electromotive action for 
 a long time but with decreasing strength. Even after one-and- 
 twenty hours a trace of polarisation could be seen. For example, 
 after this length of time, sensory roots which had lain undisturbed 
 in the moist chamber gave with 2 Groves and o^'-o^i time of 
 closure : — 
 
 11+ 3il '"I l + 3-5il "^l 1+ 2 
 
 The experiments belong to the same series as those last com- 
 municated. After twenty-seven hours, however, there followed 
 only purely negative polarisation : — 
 
 t-9i -9t-9i-9; 
 
 ^ Untersuchungen iiber thierische Elektricitat, vol. i. p. 334. 
 - Ibid. vol. ii. part i. pp. 574, 575. 
 P 
 
210 SECONDARY ELECTROMOTIVE PHENOMENA IN . 
 
 and there was no longer any difference between ascending and 
 descending polarisation. After fifty-five and a-half hours there 
 was still a trace of negative polarisation to be seen amounting to 
 about one degree of the scale. When the negative polarisation 
 alone remains, it is the same in both directions in the sensory 
 roots. Consequently the difference which living roots exhibit wath 
 regard to polarisation by the ascending and descending currents, is 
 to be attributed to positive polarisation ; in both kinds of fibres, 
 though less markedly in the motor fibres, the negative polarisa- 
 tion is as strong and the positive not weaker in the direction of 
 the physiological wave of innervation than in the opposite direction. 
 The question referred to in Sect. lo has thus been decided as regards 
 nerves as was anticipated, and the method I chose of representing 
 it is justified. 
 
 Without having made the experiment on the two halves of the 
 muscle, I do not doubt that it would yield a similar confirmation. 
 We shall meet with another fact pointing in the same direction 
 further on. 
 
 It would be interesting to make polarisation experiments with a 
 bundle of nerve fibres without perineurium, such as one gets in 
 the manner described by Harless \ 
 
 19. Positive internal Polarisation in Nerves in conflict with 
 
 Tetanus. 
 
 The idea suggested itself of transferring the experiment de- 
 scribed in Sect. 12 as made on muscles to nerves, and to examine 
 what influence the state of activity of nerves would have on their 
 internal positive polarisation. Although this plan of research has 
 been among my agenda since 1857, I have not succeeded in carry- 
 ing it out. In the meantime another fact became known from a 
 different source, which did not indeed directly answer the question 
 here stated, but which made it possible to foresee the answer with 
 some certainty. This was the fact discovered by Griinhagen, that the 
 strength of a current passed through a nerve increases on tetanising 
 the nerve. Griinhagen explains this increase by diminution of 
 
 ^ Moleculare Vorgiinge in der Nervensubstanz, ii. Abhandlung, Voruntersuch- 
 ungen. Aus den Abhandlungen der K. bayer. Akademie, vol. viii. part ii. 
 Miinchen, 1.858, p. 538 if.; cf. Charles E. Morgan, Archiv fiir Anatomie, Phy- 
 siologic, etc. 1863, p. 340; Electro-Physiology and Therapeutics, etc., New York, 
 1868, p. 464. 
 
MUSCLES, NERVES, AND ELEOTRICAL ORGANS. 211 
 
 resistance in the nerve ^. It g-ives a measure of the improvement 
 of methods that Griinhagen obtained a result for which I had 
 sought in vain in October 1844, althoug-h my arrangements were 
 faultless according to the knowledge we then possessed ^. 
 
 Hermann, who several years later made an independent discovery 
 of Griinhagen's facts, showed that it was not on account of 
 diminished resistance, but of an electromotive force aroused in the 
 intrapolar tract, which he calls the ' polarisation Increment ^.' 
 
 This is hardly the occasion for testing Hermann^'s theory of the 
 phenomenon, for the positive polarisation of the intrapolar tract 
 will render it necessary for him to make some changes in his con- 
 ceptions. The fact in itself has for us this significance, that if the 
 positive polarisation during the duration of the polarising current 
 in a nerve is strengthened by tetanus, this will probably also be the 
 case with the positive after-current. There would then exist 
 between the secondary electromotive actions of muscles and those 
 of nerves this difference, that the condition of activity weakens 
 the positive after-current in muscles and strengthens it in nerves. 
 
 20. On the Secondary Electromotive Phenomena of 
 Electrical Organs. 
 
 I have already given an account of the secondary electromotive 
 phenomena in the electrical organs of the Malapterurus in my 
 ' Experimentalkritik der Entladungshypothese *.' "What I have 
 already said shows how I was led to these experiments. The 
 agreement between the polarisation-currents of muscle and nerve 
 and those of the electrical organs, which I pointed out in my lecture 
 on the West African Malapterurus in January^ 1858^, is now 
 recognised. 
 
 A strip of the fish cut in the direction of its length exhibits ^ a 
 secondary electromotive action like muscle or nerve. With medium 
 
 * Henle's u. Pfeufer's Zeitschrift fiir rationelle Medicin. 3 series, 1869, vol. xxxvi. 
 p. 140. 
 
 ^ Untersuchungen iiber thierische Elektricitat, vol. ii. part i. 1849, P- 444- 
 ^ Pfliiger's Archiv, 1872, vol. vi. p. 561 ; 1873, vol. vii. p. 323 ff . ; 1874, vol. viii. 
 p. 264; 1875, vol. X. p. 215 ; 1876, vol. xii. p. 151 ; 1879, '^ol- ^'x. p. 416; 1881, 
 vol. xxiv. p. 246; Handbuch der Physiologie, vol. ii. 1878, p. 165; Die Ergebnisse 
 neuerer Untersuchungen, etc., Sep. Abdr. aus der Ziiricher Vierteljahrsschrift, 1878, 
 p. 32. 
 
 * Gesammelte Abhandlungen, vol. ii. p. 718. 
 ^ Monatsberichte, 1858, p. 106. 
 
 * [Here the Multiplier was the instrument used. Ed.] 
 
 P 2 
 
213 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 density of current, or greater density but longer time of closure, 
 negative action follows in both directions, though weaker in the 
 direction of the shock. With greater current density and shorter 
 time of closure positive action appears, but is stronger in the 
 direction of the shock. As the shock in the electrical organ is 
 directed from the head to the tail we may describe the polarising 
 current in that direction as a descending one, and the one in the 
 opposite direction as ascending. Thus a purely negative after- 
 current often follows a momentary ascending' current, though the 
 descending one generates a strong positive after-current. The 
 latter is distinguished by its longer duration from a shock of the 
 organ, however induced. While the shock deflects the needle in 
 the same way as a momentary volta-electric current, the positive 
 after-current actually holds it pressed against the stop for some 
 time. At the same time the primary current is also stronger in 
 the direction of the shock than in the opposite direction. From the 
 great electromotive force which the primary current must possess in 
 order to obtain positive polarisation, the electromotive force of this 
 polarisation is relatively considerable. As it will probably be a 
 long time before any one finds an opportunity of making such 
 experiments again, I will communicate a few particulars and a few 
 numbers to show what magnitudes we have to do with. 
 
 I had already made polarisation experiments on the smallest of 
 the three fishes, sent to me by Goodsir, and which were killed on 
 the 23rd November, 1857, in which however the primary current had 
 not been observed. But I convinced myself, by their means, of the 
 agreement of the secondary electromotive behaviour of the organ 
 with that of muscle and nerve. For further experiments I used 
 strips of the organ of the largest specimen, which was 233 mm. 
 long. On the morning of the 1 2th of January the animal was found 
 dead, yet the organ still gave reflex shocks on irritation of the skin 
 with a forceps, so that a frog's leg, the nerve of which was applied 
 to the fish, contracted. As its death was unexpected, the necessary 
 galvanometrical apparatus was unfortunately not at hand. The 
 only available galvanometer had to be kept for the primary 
 current. The nerve multiplier was employed for the secondary 
 actions. It soon proved to be too sensitive ; but after the experi- 
 ment was begun with it, it could not be exchanged for another in- 
 strument without sacrificing the comparableness of the result ^. Had 
 I then known the astonishing tenacity of life which the organ 
 
 * Gesammelte Abhandlungen, vol. ii. p. 718. 
 
MUSCLES, NERVES, AND ELECTEICAL ORGANS. 213 
 
 possessed, and which I then discovered for the first time ^, I should 
 not have allowed myself to be deterred from at least beg-inning a new 
 series of experiments with the muscle multiplier. As I did not 
 know whether each action we observed might not be the last, 
 naturally nothing- could be obtained but a rough general outline 
 of the phenomena. Filling up a table systematically with double 
 entries, such as I already possessed for muscle and nerve, was not 
 to be thought of. 
 
 The strip separated from the organ with scissors was about 
 30mm. long and from 5 to 8 mm. wide. The thickness of the 
 organ might have been 5i^^-j so that supposing the conducting 
 power to be the same, the strip possessed about the same resistance 
 as our group of muscles. The plan of experiment was the same as 
 for muscles, except that, as the organ did not twitch, the stretcher 
 and the wedge-shaped pads for conducting the polarising current 
 were unnecessary. To the two ends of the strip were generally 
 applied ordinary pads soaked in a solution of sulphate of copper 
 covered with a film of albumen^. The wedge-shaped pads for 
 leading off the current were steeped in saline solution, and also 
 covered with films of albumen, and were continuous with the ordi- 
 nary old form of conducting vessels with platinum plates in saline 
 solution. Lastly, the alternate closing of the primary and secondary 
 currents, the isolation of both circuits from each other, the control 
 of the time of closure and so on, were effected by means of the 
 same apparatus which I still use for muscle and nerve. With, 
 the exception that the external surface of the skin was slightly 
 positive in relation to all the other surfaces of a strip of the 
 organ, there was no trace of electromotive action in the organ 
 at rest^. 
 
 The following tabular statement, coming after all the others, will 
 be comprehensible without further explanation. The roman num- ■ 
 bers are Groves, the times are the times of closure, +90° + a; 
 indicates that the needle was deflected until the stop was struck. 
 The deflections of the galvanometer, with ascending and descending 
 currents, are strictly only comparable within the limits of one 
 experiment, as the distance of the bobbin was often altered to make 
 the reading of the deflection easier. 
 
 ^ Cf. UntersucliuTigen am Zitteraal, p. 188. 
 
 '^ It is stated in the Experimentalkritik, through an oversight, that the conducting 
 pads were also wedge-shaped pads {loc. cit. p. 717). 
 
 ^ Loc. cit. p. 718 ; XJntersuchungen am Zitteraal, p. 172. 
 
214 
 
 SECONDARY ELECTROMOTIVE PHENOMEXA IN 
 
 
 X. 
 
 XX. 
 
 XXX. 
 
 Fresh. 
 
 Tired. 
 
 Exhausted. 
 
 IS 
 
 1 p 
 
 °"- 1 s 
 ip 
 
 -90° 
 
 -90° 
 15 
 
 -90 
 25 
 
 -90 
 25 
 
 - 90 
 118 
 + 90 + .T 
 132 
 
 -90 
 44 
 
 + 70 
 44 
 
 -90 
 
 57 
 -.50 
 
 62 
 
 ip 
 
 
 -90 
 
 80-5 (steady deflection) 
 -90 
 81 (steady deflection) 
 
 rs 
 ip 
 
 -90 
 -90 
 
 
 It is seen that the liminal strength for the appearance of positive 
 polarisation is very high, between 20 and 30 Groves, and the 
 primary descending current appears stronger than the ascending 
 one in the proportion of ii3 : 100. The table also shows that in 
 the same strip, in proportion as its conducting power is exhausted 
 after repeated experiments, positive polarisation recedes and nega- 
 tive reappears. In the same proportion the difference in the 
 strength of the ascending and descending polarising current dis- 
 appears. 
 
 The polarisability of the organ is reduced still more quickly and 
 completely to a mere trace, by boiling temperature, which at the 
 same time, as in muscle, lowers the resistance of the organ in a 
 remarkable degree, as, e.g., in an experiment not sj^ecially directed to 
 the observation of polarisation, in the proportion of 100 : 43. 
 
 
 XXX. 
 
 Fresh. 
 
 Boiled. 
 
 IS 
 
 °-ns 
 
 ip 
 
 -co" (bis) 
 
 45 ; 48=<= 
 
 + 90 + X (bis) 
 
 61; 80 
 
 — 24; —20 
 
 176; 176 
 
 -34; -26.5 
 
 186; 186 
 
 As stronger negative polarisation followed immediately, the 
 greater strength of the descending primary current in the boiled 
 piece had nothing to do with the preponderance which is observed 
 in the same direction in fresh pieces, but for some reason there was 
 a disturbance of reciprocity (see Sect. 10). That this preponderance 
 
MUSCLES, NERVES, AND ELECTEICAL ORGANS. 
 
 215 
 
 arises from positive polarisation, and that the preponderance of this 
 polarisation in the direction of the shock is not merely accidental, 
 follows further from the fact that when the current passes trans- 
 versely nothing of the kind is observed. 
 
 Strip from the hack cut transversely. 20 Groves. 
 
 + 90 
 
 -90 
 
 92 ; 104 
 
 + 90 
 
 76 
 
 -90 
 
 r, I indicate right and left. Through the thickness of the organ 
 from tha outer skin to the internal fascia 40 Groves with 0"-% time 
 of closure only produced 25° in both directions. The so-called Haut- 
 sc/nvarfe produced with 30 Groves and the same time of closure, only 
 — 3° in both directions. 
 
 From a piece of the organ I removed the skin and fascia ; 
 it continued to give secondary actions, though in a somewhat 
 irregular manner, which was not surprising considering that it had 
 lost all regularity of structure. No secondary electromotive action 
 could be elicited from the fin, either with 30 Groves and a time 
 of closure of o-"3, or with 30 Groves and 20", or with 20 Groves 
 and 15". Certainly this was the day after that on which the fish 
 was found dead ; but this would not account for the inactivity of 
 the fin, if it were polarisable like the electrical organ, seeing that 
 the latter still showed vigorous electromotive action. In fact, when 
 kept in a damp chamber at a temperature of little over o°C. the 
 organ retained, like a dying nerve (Sect. 18), a small quantity of 
 polarisability for an astonishingly long time, as is shown in the 
 following table, in which the primary action is recorded only 
 exceptionally, as a second observer was not always at hand. 
 
 XXX. 
 
 Jan. 
 
 T ^3, 
 
 14. 
 
 15. 
 
 16, 
 
 17, 
 
 18, 
 
 S -90° 
 
 — 90+a; 
 
 — 
 
 -90 
 
 -45 
 
 -S 
 
 p — 
 
 
 
 
 
 
 — 
 
 41^'= 
 
 S +90 
 
 1 -r^ 
 
 + 60 
 
 — 
 
 r-14 
 1 + 48 
 
 [ + 12 
 
 {;i 
 
 i P — 
 
 
 
 
 
 
 
 — 
 
 .^8 
 
 A strip perpendicular to the lateral line gave, with a current passed 
 transversely through it on the i6th of Jan., — ^^° in one direction, 
 and in the other —59°. 
 
 Thus on the sixth day after the death of the fish traces of 
 polarisation were still present, according to the same law as for fresh 
 organs. They finally disappeared with the properties of life, the 
 
216 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 org-an beginning' to emit a peculiar stale odour. From a given 
 moment the actions generated by the descending current are in two 
 oi5j)osite directions. This is important, as it teaches that also in the 
 electrical organ these actions are the algebraic sums of two polari- 
 sations, a negative more transitory polarisation, and a positive which 
 subsides more slowly. 
 
 At my suggestion, the late Dr. Sachs began some experiments in 
 Venezuela on the polarisability of the Gymnotus, the results of which 
 I have already published so fully from his note book, as well as 
 discussed in the ' Untersuchungen am Zitteraal,' that I have nothing 
 to add to what has there been said^. I there showed that these 
 results may be satisfactorily deduced on the supposition that in the 
 organ of the Gymnotus, as well as in that of the Malapterurus, two 
 polarisations exist together, a positive and a negative, of which 
 the algebraic sum presents itself to observation ; and that there 
 also, the positive polarisation curve drawn relatively to the time of 
 opening is the less abrupt, while the negative is the more abrupt. 
 If Dr. Sachs never saw purely positive deflections, this is explained 
 only too certainly by a defect in his outfit and instructions, for 
 which I myself am to blame. He possessed only twenty small 
 Groves, of which three arrived broken, an accident on which I should 
 have calculated. Then I had not sufficiently impressed upon him, 
 because it had not been enough brought home to myself, that he 
 should use the smallest animals for these experiments and should 
 cut the thinnest possible strips from the organ. Considering the 
 exceptionally high liminal intensity which the electrical organ 
 possesses for positive polarisation, and which perhaps lies still 
 higher in the Gymnotus than in the Malapterurus, it is no wonder 
 that Dr. Sachs with seventeen Groves and a piece of the organ of 
 6-'] cm. transverse section, obtained no purely positive deflections. 
 
 In the meantime, the discussion of Dr. Sachs' ' empirically 
 resulting ' curves supplied in some degree the lack of direct observ- 
 ation, as it left no reasonable doubt that both polarisations also 
 co-exist in the organ of the Gymnotus. Dr. Sachs also convinced 
 himself that boiling heat destroyed the polarisability of the organ. 
 Thus everything appears to take the same course as in the Malap- 
 terurus. Unfortunately, however, there is an important difference 
 between my experiments and Dr. Sachs' on another point. Accord- 
 ing to him, in the organ of the Gymnotus negative polarisation 
 
 ^ Loc. cit. i^p. 205-221 . The curves in Table II. of the work are, as will be easily 
 recognised, in our present mode of expression, polarisation curves drawn relatively to 
 the time of opening. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 217 
 
 is stronger in the direction of the shock, while in the organ of 
 the Malapternrus positive polarisation is the stronger. The 
 difference as regards the relation between the nerve-ending and the 
 direction of the shock in the two fishes might be referred to in ex- 
 planation ^ But in my work on Gymnotus^ I explained the reasons 
 why I did not consider Dr. Sachs' proof sufficient for his statement. 
 Dr. Sachs had evidently no idea of the complete opposition to me 
 into which he fell, and he contented himself with a single series of 
 experiments on this difficult subject, during which he did not 
 once observe the primary current and therefore was satisfied without 
 guarding against irreciprocity of conduction. Under these circum- 
 stances, I must consider the question to be still an open one, which 
 polarisation is the strongest in the organ of the Gymnotus in the 
 direction of the shock ; whether the positive, as in the organ of the 
 Malapterurus, or the negative, as Dr. Sachs asserts. 
 
 Nothing is known as yet about the polarisability of the organ of 
 the Torpedo, except that I discovered an experiment of Configliachi's 
 of the year 1 805 in which it appears that several organs of Torpedos 
 laid upon each other were charged like a Ritter's secondary battery^, 
 i.e. were polarised negatively. 
 
 It is unnecessary to say that in the secondary electromotive actions 
 of the electrical organs we have to do with the sum of the actions of 
 the individual electrical plates. The view might be taken that on 
 polarising the organ only the electrical nerves of the organ were 
 the seat of polarisation. In the Malapterurus, however, the mass of 
 these nerves is so small in proportion to that of the electrical 
 structures, that this idea appears to be quite untenable. 
 
 I have already repeatedly treated of the teleology of the polarisa- 
 tion of the electrical organ — that is to say, of the part it possibly 
 plays in the mechanism of the shock *. 
 
 21. Theoretical. Closing Remarks. 
 
 The sum of the observations I have communicated, proves that 
 muscles, nerves, and electrical organs are polarisable according to a 
 common law, up to this time only observed in themselves. In ad- 
 dition to internal negative polarisation, which though at first sight 
 indistinguishable from that of other moist porous bodies, yet is 
 
 * Cf. Gesammelte Abhandlungen, vol. ii. p. 618. 
 ^ Loc. cit. pp. 218, 219. 
 
 ^ Gesammelte Abhandlungen, vol. ii. p. 719, Anm. 
 
 * Gesammelte Abhandlungen, p. 722; Untersuchungen am Zitteraal, p. 220. 
 
218 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 intimately connected with their vital properties, they are also 
 capable of internal positive polarisation, which has never been 
 observed elsewhere ; and in all three, the positive polarisation appears 
 more persistent, the negative more transitory, whence actions in 
 two opposite directions, first neg-ative and then positive, arise. It 
 must however be remembered that I did not experiment with many 
 nor the most important objects, such as glands, to see if they might 
 perhaps show positive polarisation with greater current density and 
 shorter time of closure. 
 
 My investigation of secondary electromotive actions is still in 
 quite an unfinished condition, in which I am not in the habit of 
 publishing results. I hardly think, however, I could have obtained 
 the desired completeness by continuing to work at it longer without 
 publication. Any one who has followed the preceding' exposition 
 must perceive that he has before him an overpowering mass of new 
 facts and relations, so that although the best points of view may have 
 been opened out and the right methods indicated, the exhaustive treat- 
 ment of the subject w^ould still require the lifetime of a new observer. 
 
 If, however, positive internal polarisation should be found in 
 other moist porous bodies, that of muscles, nerves, and electrical 
 organs will still remain distinguished by its peculiar relation to the 
 vital activity of the structures. In regular ' monomeric ^ ' muscles 
 positive polarisation in the direction from the equator to the two ends 
 is stronger than in the opposite direction. Unfortunately, in order 
 to draw conclusions on substantial bases from this fact, accurate 
 knowledge of the distribution of the nerve-endings in these muscles 
 is wanting. I will not take up the question again as I have already 
 discussed it ^, but will content myself with the remark that it is 
 much the same whether each muscle fibre, as W. Krause states, 
 receives its innervation only at a single point, or as Kiihne says ^. at 
 
 * Gesammelte Abhandlungen, vol. ii. p. 570. 
 
 * Ibid. pp. 568 ff. 
 
 ^ I am not quite sure what KuLne's later views are on this point. He now dis- 
 cards, as based on imperfect examination, his statement that there are six or eight 
 nerve-endings in the fibres of the sartorius (Uber die peripherischen Endorgane der 
 motorischen Nerven, Leipsig, 1862, 4". Plate iii. Fig. xiv. F), and, as a proof of the 
 innervation of muscular fibre at several points, he figures single nerve-endings of 
 amphibia in which the nerve-fibre divides into several parallel terminal branches, 
 which however in his theory form together only one point of innervation (Unter- 
 suchungen aus dem physiologischen Institute der Universitiit Heidelberg. Sonderab- 
 druck, Heidelberg, 1879, pp. 115, 129; Untersuchungen am Zitteraal, pp. 416, 417). 
 It is hard to see what this has to do with the question whether each muscular 
 fibre receives its nervous supply at several points distinct from one another to the 
 naked eye, or only at a single point. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 219 
 
 several points, for on the whole a contraction wave will always run 
 from the middle of the muscle to its ends. In the sensory, and less 
 certainly in the motor nerve-fibres, positive polarisation predominates 
 in the direction of physiological innervation. In the electrical plates 
 of the Malapterurus it predominates in the direction of the shock. 
 
 Here it may be remarked, by the way, that there is no doubt that 
 the positive polarisation is the stronger, and not that the negative is 
 the weaker. If the polarising current density is but small, the 
 negative polarisation which is then alone present appears equal in 
 both directions. So also when exposure to a boiling temperature has 
 left, instead of both kinds of polarisation, only a small residue of nega- 
 tive polarisation in the organ, this is the same in both directions 
 just as it is in dying sensory roots. As it may therefore be con- 
 sidered proved that in electrical plates and in nerve roots it is the 
 positive polarisation which becomes stronger in one direction, not the 
 negative which recedes, we confidently apply the same explanation by 
 analogy, to the behaviour of both polarisations in the upper and under 
 halves of regular monomeric muscles, notwithstanding that a direct 
 proof for this explanation is still lacking. In all three structures 
 then — muscles, nerves, and electrical plates — positive polarisation 
 appears more strongly in the direction in which the physiological 
 process belonging to them — whether contraction-wave, innervation- 
 wave, or electrical shock — is propagated in them. This view, if further 
 verified, cannot be considered unimportant. For the possibility 
 already discussed (Sect. i8), that the preponderance of ascending 
 positive polarisation in the sensory roots may be only an action of 
 the physiological innervation-wave which is freqviently propagated 
 in this direction, is rendered improbable by the necessity which it 
 involves of attributing the same action to three perfectly different 
 occurrences, the contraction-wave, the innervation-wave, and the 
 electric shock. 
 
 In the endeavour to picture to oneself what takes place in each 
 of these three structures during polarisation, one is naturally inclined 
 to rank the negative polarisation in them with the internal polari- 
 sation of moist porous bodies. The idea at once suggests itself of 
 connecting the destruction of the internal polarisability of muscles 
 and of the electrical organ by boiling temperature, with the dimi- 
 nished resistance of the ' interstitial fluid ^ ', and to connect its 
 
 * ' Binnenfliissigkeit.' This is the name which Munk has proposed for the inter- 
 stitial fluid of moist porous bodies, so far as they indicate secondary resistance ; it 
 may also be applied to them as objects of polarisation (Archiv fur Anatomic, Physi- 
 ologic, etc., 1873, p. 254). 
 
220 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 gradual disappearance in the process of dying* with the acidification, 
 which, according" to Ranke, lowers the resistance of muscle 2-4 times 
 more than boiling temperature ^. In agreement with this also is 
 the observ^ation of Ranke that the connective tissues which do not 
 lose their internal polarisability by exposure to a boiling temperature 
 (Sect. 11) undergo through it no diminution of their resistance. On 
 the other hand, it is not clear why a boiling temperature is more 
 injurious to polarisability than death is, since the latter reduces the 
 resistance considerably more than boiling temperature ; and the 
 general question is whether such a small reduction in the resistance 
 of the interstitial fluid as occurs on boiling the muscle, is capable 
 of an action for which the conducting capacity of vinegar, ammonia, 
 sulphate of copper and zinc solution is insufiicient in other porous 
 bodies, such as blotting-paper 2. 
 
 Ranke places nerves among the tissues, the resistance of which 
 undergoes no alteration through boiling heat or death ; this creates 
 further difficulties ^. In analogy with the white matter of the brain, 
 it may be inferred that the nerve-trunks are rendered acid by a 
 boiling temperature *. This would explain the destruction of inter- 
 nal polarisability at a boiling temperature, were it possible to under- 
 stand why the resistance of boiled nerves does not diminish, and 
 why dying nerves which do not become acid, lose their polarisability ^. 
 
 There is still another consideration in opposition to the view that 
 negative polarisation of miiscle and nerve is a simple physical occur- 
 rence like the internal polarisation of wood, leather, etc. The de- 
 pendence of this negative polarisation on current density and time 
 of closure agrees well at first sight with such a view. Negative 
 polarisation seems to increase at first to some extent proportion- 
 ally to the product of these variables. It then slowly approaches a 
 limit. This is exactly the behaviour to which we are accustomed 
 in polarisation-currents generally, and which is easily explained by 
 the nature of polarisation. In muscles, however, and if I may trust 
 my old observations, also in nerves, a maximum of negative polari- 
 sation makes its appearance with an increasing time of closure. 
 
 * Eegarding the acidification of boiled and dying electrical organs, see the Gesam- 
 melte Abhandlungen, vol. ii. p. 646 ; Untersuchungen am Zitteraal, p. 70. 
 
 ^ Tetanus; Eine physiologische Studie, Leipzig, 1865, pp. 35 fF. 
 ' Gesammelte Abhandlungen, vol. i. pp. 25, 26; vol. ii. p. 190. 
 
 * Loc. cit. pp. 33, 38. 
 
 5 Gscheidlen in Pfluger's Archiv, 1874, '^^l- "^"- P- ^7^- Oddly enough, there 
 seem to be no experiments on the reaction of boiled nerve-trunks themselves. Comp. 
 Hermann in his Handbook, p. 139. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 221 
 
 This does not agree well with the idea that we have here to do with 
 ordinary internal polarisation. That this maximum points to an 
 increase of positive polarisation which appears at this time is the 
 more difScult to admits inasmuch as the maximum appears rather to 
 be coincident with the destruction of polarisability hy prolonged 
 action of very strong currents (Sect. ii). 
 
 The help which the comparison of the negative polarisation of 
 muscles, nerves, and electrical organs with ordinary internal polari- 
 sation gives in explaining it, is entirely wanting as regards positive 
 polarisation. If negative polarisation could he identified wath ordi- 
 nary internal polarisation, it would be covered by the physical process 
 in the hypothetical septa referred to in Sect. 3. Of the occurrence of 
 negative and positive polarisation in the same electrodes there is only 
 one example observed by me, viz. that of iron, and zinc containing iron 
 in a solution of zinc sulphate^. Although this double polarisation 
 depends on the time of closing and opening, in the same way as that 
 with which we are now occupied, we can scarcely have recourse to 
 it for an explanation. In order to refer the polarisation of muscles 
 and nerves and electrical organs to known physical facts, it is there- 
 fore necessary to assume another frequently recurring discontinuity 
 in the direction of the current, which, like the surfaces of certain 
 electrolytes, would be the seat of a positive polarisation ^. If such 
 a discontinuity could be assumed in muscle, or even in nerve, it 
 would still be entirely wanting in the electrical plates of the malap- 
 terurus. It would also be very hazardous to suppose the existence 
 in three such different structures of any such alternation of materials 
 as could produce secondary electromotive actions. Besides, the 
 positive polarisation in the electrical organs is much too strong to 
 be thus explained. 
 
 In addition to this there is the peculiar manner in which positive 
 polarisation depends on current density, and on times of closure or 
 opening. Without reference to the maximum in prolonged time of 
 closure, the dependence of negative polarisation on these conditions 
 is at all events similar in form to that of the internal polarisation of 
 moist porous bodies. This is not the ease with positive polarisation, 
 in respect of which there is a liminal intensity of the current density, 
 below which it does not appear at all. It then appears suddenly with 
 a force which increases indeed with current density, but slightly if 
 at all with the time of closure. The limen lies higher for the electrical 
 
 * Gesammelte Abhandlungen, vol. i. pp. 57-60. 
 ^ Ibid. vol. i. p. 6. 
 
222 SECOXDARY ELECTROMOTIVE PHENOMENA IN 
 
 plate than for muscle and nerve. While neg-ative polarisation quickly 
 disappears with increasing- time of opening", like the internal polarisa- 
 tion of porous bodies, positive polarisation shows a persistence which 
 points to another kind of mechanism. It should not however remain 
 unnoticed that negative polarisation also persists a long time at the 
 junctions of saturated solution of sulj^hate of zinc, and physiolog-ical 
 solution of salt ^ 
 
 Althoug-h, then, traces of positive polarisation can be detected 
 for a long time in nerves and electrical plates which are dying" at a 
 low temperature, it is more dependent on the vitality of the struc- 
 tures than negative polarisation. Boiling- heat destroys it at once 
 and absolutely, althoug-h still leaving a remnant of negative polari- 
 sation ; and the destructive action of strong- currents on polarisability 
 in general attacks principally positive polarisation. In addition to 
 this, the importance and bearing- of the phenomenon may be inferred 
 from what has already been said as to its relation to the direction 
 of contraction, to that of innervation, and to that of the shock in 
 electrical fishes, as well as to its not less impoi-tant relation to the 
 state of activity of the tissue. 
 
 I do not think that any one trained in physics and who keeps 
 these facts before his eyes in their totality and in an unprejudiced 
 manner, will arrive at any other conclusion to-day than that to which 
 I was compelled to come many years ago. It is the conclusion not 
 that in positively polarisable structures electromotive forces having 
 the same direction as the primary current are generated, but that the 
 conducting molecules [Trdger) of electromotive forces already existing 
 take the direction of the primary current. If to this it is added that 
 in the natural condition of the structures the electromotive molecules 
 are maintained by certain forces in a position, from which, although 
 they can be displaced by the weakest currents, they can be removed 
 permanently only by currents of a certain density, this simple and 
 natural assumption reconciles the known Hminal intensity of positive 
 polarisation with the electrotonising properties even of the weakest 
 currents. It is unnecessary to mention that the great persistence 
 of positive polarisation agrees ecpially with this view. 
 
 As I became acquainted with positive polarisation of muscle 
 shortly after the publication of my 'Preliminary Sketch' (1843), 
 and long before that of the first volume of the ' Untersuchungen ' 
 (1848), it is easy to see how I arrived at the conception of the 
 
 ' Gesammelte Abhandlungen, vol. ii. p. 190. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 223 
 
 electromotive phenomena of muscle and nerve expressed in that work. 
 The g-eneralisation by which I concluded from the positive polarisation 
 of muscles that there was also polarisation of the intrapolar nerve- 
 tract, had perhaps something- of youthful rashness in it, but nature 
 proved me to be right when a few years later, I discovered in the 
 positive after-current of nerves an indication of the prevalence of 
 positive polarisation during the primary current. In so far as no 
 other reasonable explanation of positive polarisation could be given, 
 excepting that which regards it as due to the directing of already 
 existing forces, the evidence obtained of this polarisation is at the 
 same time serviceable in another way. In my eyes it afforded and 
 will continue to afford proof of the presence of electromotive forces 
 in uninjured nerves, a proof which, in the absence of a natural 
 transverse section of a nerve, was otherwise lacking. The evidence 
 of the existence of such forces in muscle obtained by carefully 
 testing the natural transverse section of muscle, is also strengthened 
 by it. In the same way I hoped to show the electromotive forces 
 of muscles at rest in living men, the observation of which is ren- 
 dered impossible by cutaneous currents, by parelectronomy, and by 
 derivation through the skin. I had this in my mind in the ex- 
 periments described in Sections ii and 12, and in this sense the 
 secondary electromotive phenomena should form the conclusion of 
 the eighth chapter of the ' Untersuchungen ' w^hich treats ' of the 
 muscle-current and its motor phenomena in living uninjured 
 animals ^.' 
 
 Furthermore, as I now saw the possibility of explaining the 
 electromotive forces of the shock in electrical organs by the physio- 
 logical polarisation of molecules {Trdgern) ; as I was fortunate 
 enough here also to prove the existence of positive polarisation on 
 the largest scale and in undoubted connection with the shock of 
 the organ ; lastly, as Delle Chiaie's and Babuchin's statement 
 regarding the preformation of electrical elements in the organs of 
 electrical fishes leaves no doubt of a pile-like strengthening 
 of the shock in the plate, — therefore it need occasion no surprise 
 that I still believe myself on the rig-ht road, and that for the 
 present I contemplate with calmness the polemic which Hermann 
 has directed against my views for sixteen years, and which has 
 hitherto been more fruitful in technical ex^Dressions than in new 
 facts. Repeatedly, and from two quite different hypothetical bases, 
 
 ■ Loc. cit., vol. i. p. 240; vol. ii. part i. p. 331 ; part ii. pp. i, 377. 
 
224 SECONDARY ELECTROMOTIVE PHENOMENA IN 
 
 Hermann has demonstrated that in the intrapolar tract a polari- 
 sation-current opposite to the polarising- current must prevail (Sect. 
 15), I am curious to see to what auxiliary hypothesis he will have 
 recourse for an explanation of the actual polarisation-current which 
 is in the sa7ue direction. 
 
 These 'directahle' carriers of electromotive forces in muscles, nerves, 
 and electrical org-ans, referred to above as molecules. I have, as is 
 well kno^vn, called electromotive molecules ; and without in the 
 first instance saying* anything about their nature, I represented 
 them diagrammatically in the simplest manner ^. The discovery 
 of inclination-currents {1^ eigungstrdme) placed me later in a 
 position to state something definite about the arrangement of pre- 
 existing electromotive forces in muscles. In the place of spheres 
 of which one half was electro-positive, the other electro-negative, 
 there came electromotive discs {FUlchenelemente) ^, while at the 
 same time I completed the representation which had remained 
 unfinished in the ' Untersuchungen,' by representing these elements 
 as foci of chemical change something like that w^hich constitutes 
 tissue resj)iration. 
 
 I know the difficulties that stand in the way of these views as 
 well as any one. Here the question is how to represent the directing 
 of the electromotive molecules by the current. In order to make it 
 clearer, I had referred to Grothuss' theory of electrolysis. Hermann 
 erroneously imputed the idea to me that the current rotated the 
 molecules electrodynamically, and he tried to deduce this rotation 
 from Ampere's fundamental law by substituting for a molecule 
 capable of being directed a current element capable of being rotated, 
 and for the total current a thread of current passing through the 
 middle of the element ^. Later on Hermann remarked indeed that 
 he was presenting me with a theory with which I was unacquainted, 
 but notwithstanding he thinks himself justified in setting forth my 
 electrotonus hypothesis in this form (calling it simply the ' electro- 
 dynamic hypothesis '), on the ground that Grothuss' view is not 
 applicable to electromotive molecules separated by interstitial fluid *. 
 
 I had good grounds for avoiding any electrodynamic theory of 
 the sort. Hermann's view is quite inadequate. The problem of 
 
 * Lioc. cii., vol. ii. part i. p. 323. 
 
 ^ Loc. cit., vol. ii. pp. 122, 291, 671, 672. 
 
 3 Untersuchungen zur Physiologie Ser Muskeln und Nerven, part iii. p. 66. Berlin, 
 1866. 
 
 * Pfluger's Archiv, 1874, vol. viii. p. 268; Haudbuch der Physiologie, 1878, 
 pp. 171, 172. 
 
MUSCLES, NERVES, AND ELECTRICAL ORGANS. 225 
 
 determining the eleetrodynamic interaction between an electrical 
 current and an electromotive molecule certainly raises several knotty 
 points, nevertheless, so far as the eleetrodynamic forces are known 
 with certainty, it may now be affirmed that no rotation of the 
 molecule through these forces is possible. 
 
 So far as Grothuss' theory in its original form presupposes that 
 atoms of oxygen and hydrogen form a kind of molecular chain in 
 the conductors of the second class undergoing electro-chemical 
 decomposition, this representation does not agree strictly with the 
 polarisation of muscles, nerves, and electrical organs by the current. 
 The explanation I gave in the ' Untersuchungen ' even now seems 
 me to sufficient to give to the theory of directable molecules as 
 much plausibility as is usually attainable in dealing with molecular 
 processes of this kind. 
 
 It has indeed since then received some support from the dis- 
 covery of the anaphoric actions of the current by Jiirgensen ^, and 
 their explanation by Quincke ^. If particles floating in a fluid and 
 negatively electrified by contact with it, migrate to the anode before 
 our eyes, the conclusion does not seem too venturous that a molecule 
 of which one half is electropositive, the other half electronegative, 
 not displaceable but capable of rotating freely, would turn its posi- 
 tive pole to the kathode and its negative to the anode. 
 
 But I attach little importance to such speculations. It is quite 
 sufficient for me to know that in muscles, nerves, and electrical 
 organs, carriers of electromotive forces exist which the current 
 somehow directs, which are in relation to the vital activity of 
 these structures, and in electrical organs serve to explain their 
 enigmatical action. 
 
 In the face of the profusion of questions regarding facts which 
 are here awaiting an answer, I consider it lost labour to indulge in 
 further conjectures. One can scarcely hope to penetrate more deeply 
 into the molecular mechanism of electromotive tissues until the 
 region of secondary electromotive phenomena has been explored in 
 every direction. 
 
 * Archiv fiir Anatomie, Physiologie, etc., i860, p. 673. 
 
 ^ Poggendorff's Annalen der Physik iind Chemie, 1861, voL cxiii. p. 565. 
 
VII. 
 
 ON 
 
 DU BO IS-EEYMOND'S EESEAECHES 
 ON SECONDARY ELECTEOMOTIVE PHENOMENA OF lUSCLE. 
 
 By dr. EWALD HEEING, 
 Professor of Physiology in the German Universiti/, Prague. 
 
 Sitzungsberichte der kaiserlichen Akademie der Wissenschaften, Vienna, 
 Vol. Ixxxviii, Part 3, p. 445? iS^S- 
 
 Q3 
 
CONTENTS. 
 
 PAGE 
 
 Introduction 229 
 
 1. Criticism of du Bois-Eeymond's experiments on the gracilis and semimem- 
 
 branosus muscles of the frog. Influence of tendinous intersections on 
 
 * polarisation ' of muscle 231 
 
 2. Experiments on polarisation of a muscle without tendinous intersections 
 
 — the sartorius of the frog ■. . . .241 
 
VII. 
 
 In my twelfth communication ^ I mentioned dn Bois-Reymond's 
 treatise on ' Secondary Electromotive Phenomena in Muscle, Nerve, 
 and in Electrical Organs,' in which he arrives at deductions directly 
 contradicted by facts which I have established. Du Bois-Eeymond 
 concludes from his researches that when an electrical current flows 
 through a muscle, it polarises the whole of the region through 
 which it flows. If this is so, every part of the intrapolar region 
 lying between the galvanometer electrodes should, on breaking 
 the circuit, give an after- or polarisation-current, which, according 
 to the density and duration of the primary current, now opposes, 
 now reinforces it. Equal lengths of the intrapolar region should 
 give equal polarisation-currents, provided the sectional area be the 
 same throughout, so that it would be of no importance from what 
 portion of the intrapolar region the polarisation-current was led ofi", 
 so long as the length of that portion remained the same. 
 
 Provided the resistance in the galvanometer-circuit is sufiicient, 
 the current of polarisation should be correspondingly stronger if the 
 leading ofi" electrodes include a longer tract than if they include a 
 shorter tract, on account of the greater number of electromotive 
 parts between the galvanometer electrodes, and therefore strongest 
 when the whole intrapolar tract is interposed. In this respect the 
 positive or negative polarisation -currents would be quite analogous 
 to the negative polarisation-currents of a charcoal cylinder steeped 
 in weak sulphuric acid, on which account, moreover, du Bois-Rey- 
 mond assumes an ' internal polarisation ' for the whole of the intra- 
 polar tract. 
 
 » tJber Veranderungen des elektromotorischen Verhalten der Muskeln im Folge 
 elektrischer Reizung. Sitzungsberichte, vol. Ixxxvii. p. 415. 
 
230 EESEARCHES ON SECONDARY ELECTROMOTIVE 
 
 In contradiction to this, I have an-ived at the conchision that 
 the polarisation-cun-ents in a muscle consequent on breaking the 
 primary or stimulating current, result solely from alterations which 
 the latter undergoes at any point of the contractile substance of 
 the muscle-fibres at which it enters or leaves that substance ; in 
 short, at the anodes and kathodes. So long as the current flows in 
 a direction which is mathematically parallel with the muscle-fibres, 
 in all probability it does not polarise them at all ; at least there is 
 not a single recorded observation which would lead one to such a 
 conclusion. If polarisation of the muscle -substance occurred in the 
 intrapolar tract (in the strict sense of the term), in any case it 
 would be so trifling that it might be provisionally quite neglected 
 in comparison with the polarisation-currents which accompany 
 changes at the anode and kathode. Without prejudging in any 
 way the essential nature of these changes, they may be regarded 
 as a polarisation which has an analogy with external polarisation, 
 but certainly not with internal. 
 
 This holds at least for negative polarisation, whilst positive, 
 resting as it does on excitation of the living substance, does not 
 admit of comparison with physical polarisation. 
 
 The changes which occur at the anodes of the contractile sub- 
 stance in consequence of stimulation, I call anodic polarisation, 
 those which occur at the kathodes, kathodic polarisation of those 
 spots. In a muscle-fibre through which a current is passing in a 
 longitudinal direction, only a portion where there are no anodes nor 
 kathodes is to be understood as an intrapolar tract. But, whereas 
 in a ' monomerous ' muscle, the fibres of which are parallel and 
 through which a current is passing, the anodic or kathodic spots 
 are for the most part aggregated in the region where the current 
 enters or leaves the whole muscle, the tract between these two 
 spots can with approximate accuracy be termed the intrapolar 
 tract. 
 
 At the same time it must not be forgotten that in point of fact, 
 on grounds which I have already discussed, there are always in- 
 dividual anodic and kathodic spots within this tract even when no 
 muscle-fibre terminates in it. Finally, it must not be forgotten 
 that there are other tissues in a muscle besides the muscle-fibres, 
 which possibly might be appreciably polarised. 
 
 According to my interpretation, the positive polarisation-current 
 consequent on changes at anodic spots of the contractile substance 
 is a so-called action-current, and is generated by a break-excitation 
 
PHENOMENA OF MUSCLE. 231 
 
 orig-inating" in the neighbourhood of the anode — an action-current 
 indeed which behaves in a manner essentially different from the 
 action-currents consequent on instantaneous stimulations. Never- 
 theless, I will provisionally speak of this break action-current as a 
 polarisation-current. 
 
 The facts contained in the preceding- communication gave suf- 
 ficient evidence that the internal polarisation adopted by du Bois- 
 Reymond was not the actual cause of the secondary electromotive 
 phenomena observed by him ; yet I feel it my duty to give a 
 still more thorough refutation of his interpretation of those 
 phenomena. 
 
 I. 
 
 In his researches, du Bois-Reymond made use of two muscles at 
 the same time, i.e. the rectus intermis major (Ecker), which Cuvier 
 calls adductor magmis and du Bois-Reymond gracilis^ and the semi- 
 membranosus (Cuvier, Ecker, du Bois-Reymond). 
 
 The first tapers to a thin tendon at both ends, and is a tolerably 
 symmetrical muscle. A tendinous intersection, {imcriptio tendinea,) 
 however, runs obliquely through the middle third, which com- 
 pletely divides the muscle into an upper and a lower portion. 
 
 The greater portion of the semimembranosus is likewise traversed 
 by an obKque tendinous intersection, hence it only partly consists 
 of long fibres. Besides this it is very unsymmetrical, rising by a 
 broad tendon from the pelvis, and tapering from above downwards, 
 at first gradually, afterwards rather rapidly, to end in a thin tendon. 
 The intersections of both muscles were fully described and depicted 
 by du Bois-Reymond in a former research^. He noticed moreover 
 ' that the semimembranosus in a portion of its length from below 
 upwards receives fresh accessions of fibres, which arise from both 
 sides of a strip of tendon projecting at its outer side.' 
 
 Both muscles were so prepared for the purpose of experiments on 
 polarisation, as to remain attached to each other, and in addition to 
 portions of the pelvis and tibia. By means of these bones they 
 were extended on a ' muscle-frame ' or stretcher and arranged as 
 one single thick muscle. 
 
 ' Close to the ivory plates of the stretcher, the wedge-shaped pads 
 
 On negative variation of the muscle-current; Gesammelte Abh. vol. ii. p. 575. 
 
283 RESEAKCHES ON SECONDARY ELECTROMOTIVE 
 
 of the polarising current were brought into contact with the muscle 
 from the inner (femoral) side ; between these, on the outer side, the 
 electrodes were placed to lead off the polarisation- current. On 
 account of the high resistance of the tendons, which would become 
 heated and dried, it is impossible to lead the current to the muscle 
 by means of ordinary pads applied to the fragments of bone out- 
 side the ivory plates ; a method which otherwise would have 
 had great advantages.' With reference to the last point, I may 
 mention that du Bois-Reymond used up to as many as 50 Groves. 
 I at first never went beyond 8 Daniells, for I only wished to 
 stimulate the muscle, not seriously to injure it at the same time. 
 
 In order to understand the results of du Bois-Reymond' s re- 
 searches, it will next be necessary to indicate how the anodic and 
 kathodic spots are distributed in the muscle. 
 
 The current from the stimulating anode as it enters the natural 
 longitude of the muscle near its tendon cannot, as a rule, enter the 
 contractile substance through the natural transverse section of the 
 individual muscle-fibres ; it must rather in its course from fibres in 
 the neighbourhood of the electrode to fibres at a greater distance 
 from it, pass through the first-named transversely or obliquely. 
 Hence, the lines of current radiating from the electrode become 
 parallel to the muscle-fibres only at some distance from their source. 
 Every fibre through which a portion of current flows in a trans- 
 verse direction has both anodic and kathodic spots. Moreover, 
 as the double muscle taken as a whole tapers off" at the upper and 
 lower ends, a considerable number of its fibres terminate at varying 
 distances from the commencement of the proper tendons. In the 
 case of the semimembranosus there is in addition the circum- 
 stance above mentioned, that it has considerable accessions of fibres 
 in the direction from below upwards. 
 
 If the current entered and left through the tendons, the whole 
 tract of muscle in which fibres take origin or in which they 
 terminate would have a corresponding number of anodic or 
 kathodic spots, and one portion of the muscle would present an 
 anodic tract, the other a kathodic, understanding thereby the tracts 
 of muscle in w^hich the anodic and kathodic spots are respectively 
 aggregated. If the stimulating electrodes are applied to the longi- 
 tudinal surface of the muscle but near the tendons, and yet in 
 the tapering part of the muscle, the current would meet numerous 
 muscle-fibres in its course, and would enter or leave them through 
 their natural ends. More especially would this be the case, as 
 
PHENOMENA OF MUSCLE. 233 
 
 we have already shown, at the lower end of the semimem- 
 branosus. 
 
 The sig-nificance of the circumstance, that the greater portion of 
 muscle-fibres in the neighbourhood of the anode are crossed by 
 transversely or obliquely running lines of current, and hence that 
 every such fibre has anodic spots of somewhat greater density of 
 current on the side turned towards the stimulating electrodes, and 
 kathodic spots of somewhat less density on the other side, we shall 
 not discuss at length here, for in order to do so it would be 
 necessary to go minutely into the so-called transverse polarisation 
 of muscle. In any case, it is apparent from the above-described 
 distribution of anodic and kathodic spots, that from the stimu- 
 lating electrode which lies on the side of the muscle in the 
 neighbourhood of the tendon, to the point where the muscle no 
 longer sensibly increases or diminishes in thickness, anodic or 
 kathodic spots must prevail as the case may be, and hence the 
 whole tract must be regarded as anodic or kathodic. 
 
 Now, when the stimulating electrodes are in the neighbourhood 
 of the tendons of the muscle, and the galvanometer electrodes lie 
 near them, the galvanometer electrodes are within the region of 
 the anodic and kathodic tract, and this is the case at the lower end 
 of the pair of muscles on account of the unsymmetrical structure 
 of the semimembranosus, even if the stimulating electrodes lie at 
 some distance from the tendinous end. 
 
 Even if the muscles had no tendinous intersection, still there 
 would be a polarisation-current in the galvanometer-circuit after 
 every break of the stimulating current, because all the anodic and 
 kathodic spots lying in the anodic and kathodic tracts are the seat 
 of a change caused by the stimulating current, which change must 
 have an electromotive action. 
 
 Du Bois-E-eymond has not stated how far the galvanometer 
 electrodes were from the stimulating electrodes, and these, as 
 appears from the description, were very near the tendons of the 
 pair of muscles. It is very probable that both galvanometer elec- 
 trodes were within the limits of the anodic and kathodic tracts, 
 and this may be assumed in any case of the lower electrode, because, 
 as we have already stated, the semimembranosus receives new fibres 
 very high up. 
 
 But even if the galvanometer electrodes were at such a distance 
 from the stimulating electrodes that they were completely with- 
 drawn from the limit of the anodic and kathodic tracts as above 
 
234 EESEARCHES ON SECONDARY ELECTROMOTIVE 
 
 defined, and that from the electromotive changes taking place in 
 this region set up b}^ the stimulating current no polarisation-current 
 could flow into the galvanometer electrodes, yet the far more 
 important polarisation occurring at the tendinous intersections 
 would not be excluded. 
 
 For between the two galvanometer electrodes lies the tendinous 
 septum, which severs the two muscles in such a way that each 
 consists of two separate muscles lying one behind the other. In 
 front of the intersection the current leaves the fibres of one of 
 the separate muscles, and behind it, enters the fibres of the second 
 separate muscle. On one side, therefore, of the intersection lie 
 innumerable kathodic spots, on the other side as many anodic 
 spots, and the former as well as the latter are the seat of polar 
 changes. The change at the anodic spots gives rise to a polarisation- 
 current, and that at the kathodic spots gives rise to a second such 
 current. As was shown in the first communication, both currents, 
 according to circumstances, flow sometimes in the same direction, 
 sometimes in an opposite direction to the stimulating current. 
 The polarisation-current entering the galvanometer-circuit, re- 
 presents the sum, or the difference, in short, the algebraic sum 
 of both currents, which would appear separate if either the whole 
 of the anodic spots or the whole of the kathodic spots could be 
 suddenly rendered inert at the tendinous intersection. 
 
 Hence du Bois-Reymonds researches prove nothing with regard 
 to an ' internal polarisation,' that is to say, with regard to the fact 
 that the stimulating current polarises the contractile substance 
 in the proper intrapolar tract, where it runs within the contractile 
 substance of all the fibres. 
 
 The polarisation-currents described by du Bois-RejTuond were 
 for the most part the result of polarisation taking place at the 
 tendinous intersection, or the algebraic sums of the two polarisation- 
 currents produced there. 
 
 Besides this, polarisations in the neighbourhood of the upper, 
 and still more in that of the lower electrode, became more con- 
 siderable the nearer the galvanometer electrodes were to the 
 stimulating electrodes. Hence du Bois-Reymond observed the 
 algebraic sum of four synchronous polarisation-currents. 
 
 Another special disadvantage in the experiments of du Bois- 
 Reymond consists in the fact that the tendinous intersection of each 
 of the muscles used by him runs in a very oblique direction to the 
 axis of the muscle, and that the intersections of the two muscles 
 
PHENOMENA OF MUSCLE. 235 
 
 diverge from one another. The tract which extends from the upper 
 end of one of the muscles to the lower end of the other is equal to 
 about half the length of the muscles themselves. This whole 
 tract was, so to speak, anodic and kathodic at the same time ; and 
 it would have been necessary to put the galvanometer electrodes 
 on the end quarters of the muscles in order to escape it. 
 
 If this was done, however, they would get into the region of the 
 anodic and kathodic tracts in the neighbourhood of the stimulating 
 electrodes. Taking into consideration the fact that the direction 
 and strength of the polarisation-currents produced by one stimu- 
 lating current depend on its density at the individual anodic and 
 kathodic spots, and that in every muscle there must be two 
 break-stimulations, which affect sometimes a larger, sometimes 
 a smaller muscle tract (see below), it is easy to see under what 
 extraordinarily complex conditions du Bois-Reymond's experiments 
 were conducted. On this ground, also, it is not possible to deduce 
 the theoretical result of each experiment. 
 
 According to the propositions set forth in my last treatise some 
 points appear explicable, viz. first, that very weak currents in- 
 variably only gave negative polarisation. This was inevitable, 
 for such currents produce negative polarisation at the anode as 
 well as at the kathode, hence in such cases the total result of the 
 combined anodic and kathodic polarisations of du Bois-Reymond's 
 preparation could only be a negative polarisation-current. 
 
 Again, it is clear why very strong currents of short duration 
 only gave positive polarisation-currents. For such currents either 
 affect the anode only, and always ia the direction of positive 
 polarisation, or in any case their positive anodic action is much 
 stronger than their negative kathodic action ; hence the combined 
 result of the polarising actions can only be a positive one. 
 
 With very weak currents, as well as with strong ones of short 
 duration, the relations are much simpler and more easily reviewed, 
 because the persistent break-stimulation is then confined to those 
 portions of the fibres immediately adjacent to the anodic spots. 
 Stronger stimulating currents of longer duration, after the break, 
 eflPect however a persistent stimulation spreading farther in the 
 direction of the stimulating current. As a necessary consequence of 
 this in du Bois-Reymond's preparation, the galvanometer electrode 
 which was next the stimulating anode, came within the limits of 
 the stimulated and therefore relatively negative muscle tract, when 
 the break-excitations were strong. If the galvanometer electrodes 
 
236 EESEARCHES ON SECONDARY ELECTROMOTIVE 
 
 were at such a distance from the exciting- ones as to be beyond 
 the limits of the corresponding- anodic and kathodic tracts, and 
 the exciting- currents were weak or of very short duration, the 
 polarisation-currents produced at the intersection would be, in 
 fact, the only ones which manifested themselves galvanometrically. 
 But even when the exciting currents were stronger or more pro- 
 tracted, and persistent break-excitations were induced, a new factor 
 must have come into play. 
 
 Every stimulation with persistent and stronger currents neces- 
 sarily induces persistent break-excitation in two tracts of the 
 muscle ; in the first place, an excitation which, commencing in the 
 region of the stimulating anode spreads more or less towards the 
 centre of the muscle, and a second which starting from the tendinous 
 intersection spreads more or less towards the stimulating kathode. 
 These two concomitant persistent excitations must, as a rule, be of 
 different strengths and extent, for the conditions under which the 
 current enters the individual fibres of the muscle from the stimu- 
 lating electrode applied laterally, are quite different from those 
 under which it enters the natural cross sections of the fibres of 
 the second half of the muscle from the tendinous intersection. 
 It is at all events plain, that the relations of current density in 
 the anodic spots of the individual fibres must be quite different in 
 the one case from what they are in the other, and on these current 
 densities depends essentially the break-excitation of each fibre. 
 
 Another equally complicating element I have hitherto only in- 
 cidentally mentioned. Every polarisation-current is to be regarded 
 as the algebraic sum of the individual amounts contributed by the 
 muscle-fibres. Supposing, for example, the current entered the 
 thin end of a muscle and passed thence to its thicker centre. The 
 current w^ould in this case decrease in density so long as there was 
 an increase in the transverse section of the muscle. In such a tract 
 the current enters the natural ends of all the added fibres of the 
 muscle, and the density with which it enters is much less to- 
 wards the equator of the muscle than it is towards the end. In 
 this last position the density of the current would perhaps be suffi- 
 cient to produce positive anodic polarisation, whereas in the former 
 position it only shows negative polarisation. The anodic tract will 
 therefore show, after breaking the stimulating current, polarisation 
 in two opposite directions. As the semimembranosus has fresh 
 accessions of fibres far up, the circumstances above mentioned 
 demand consideration. 
 
PHENOMENA OF MUSCLE. 237 
 
 There is always a similar condition when a current enters a muscle 
 composed entirely of parallel fibres, not at its natural end but at the 
 side, even if the current meet with no fresh fibres on its way. The 
 current enters those fibres which are in the immediate neighbour- 
 hood of the stimulating anode with g-reater density than those 
 which are more remote. It is further to be noted that only a 
 portion of the current which enters a particular fibre flows on into 
 it, the rest passing on to fibres more remote from the stimulating 
 anode. It is therefore first necessary to know exactly the laws of 
 transverse polarisation, in order to be able to say to what extent 
 in this case kathodic polarisation can take place along with anodic. 
 In any case, there is a preponderance of anodic polarisation (as 
 would be supposed theoretically, and as can be practically demon- 
 strated), and I have therefore considered it alone in the above 
 pages. Sufiicient has been said to show how difficult or impos- 
 sible it is to unravel the intricacies of the different individual 
 polarisations which must have existed in du Bois-Reymond's ex- 
 periments. 
 
 One other result on which du Bois-Reymond seems to lay great 
 weight remains to be considered. In a special series of experiments 
 he brought one galvanometer electrode to the equator of the pair of 
 muscles, whilst the other was placed either immediately above the 
 lower or below the upper stimulating electrode. The stimulating 
 electrodes, if I understand rightly, were placed near the ends of the 
 muscle. Thus the polarisation-current was led off" either entirely 
 from the upper or entirely from the lower half of the muscle. It 
 now resulted that the upper half showed stronger positive polarisa- 
 tion when the current was ascending than when it was descending, 
 whereas the lower half gave stronger positive polarisation when the 
 stimulating current flowed from above downwards. 
 
 Let us for the moment imagine that the upper and lower 
 halves of the pair of muscles were strictly symmetrical, both 
 tendinous intersections passing exactly through the equator, that 
 the stimulating electrodes were arranged symmetrically, and, finally, 
 that one electrode lay on the equator, the other alternately on 
 symmetrical spots of the upper or of the lower half of the muscle. 
 In this case any one who knows that polarisation phenomena 
 depend not on internal polarisation but on changes occurring at the 
 points where the current enters and leaves the muscle, would readily 
 see that the upper half must necessarily be in the same relation to 
 the ascending current in every respect as the lower half is to the 
 
238 EESEAECHES ON SECONDARY ELECTROMOTIVE 
 
 descending", and vice versa ; for the direction of the current in the 
 first case would have exactly the same significance as regards the 
 first half, as the opposite direction of the current in the second 
 case would have as regards the second half, because being sym- 
 metrical the one half would be the exact reflection of the other. In 
 like manner it would be expected that each individual half would 
 behave in general differently towards the two directions of current, 
 because in each half there woidd be no symmetry of structure in the 
 up and down direction. 
 
 Du Bois-Reymond's muscle-preparation deviates widely from the 
 strictly symmetrical muscle scheme above described, yet there is 
 an essential analogy, in so far as there are tendinous intersections 
 passing through the middle of both muscles, and as the pair of 
 muscles taper away above and below. The result of du Bois-Rey- 
 mond's experiments does not therefore surprise us at all, whilst 
 to him it appeared striking and of great import. 
 
 The same thing* may be said as to the further observation of du 
 Bois-Reymond, that the strength of the secondary electromotive 
 actions in both halves of the pair of muscles (or even in the gracilis 
 alone) was not equal. The simple reason of this is, that the two 
 halves of the muscle are not exactly symmetrical in relation to 
 the equator ; the tendinous intersections do not lie in the equator 
 itself, nor is the configuration of the upper half of the pair of 
 muscles (or of one of them) the exact reflection of the lower. More- 
 over there could scarcely be a muscle adapted to this kind of re- 
 search, of which both the external and internal structure would be 
 strictly symmetrical. We should only have been surprised if du 
 Bois-Reymond had found the strength of the secondary electro- 
 motive phenomena to be always equal in the two halves of the 
 muscle, whereas the opposite result is what might have been antici- 
 pated. 
 
 In answer to the final point emphasised by du Bois-Reymond, 
 that when the galvanometer electrodes are placed symmetrically on 
 the muscle, so that one is below the upper stimulating electrode, 
 the other above the lower one, the polarisation-currents are not 
 equal in both directions, it must also be said that it would be 
 strange if they ever were equal. So long as we start with the 
 assumption of an internal polarisation of the whole tract through 
 which the stimulating current flows, all inequalities of polarisation 
 dependent on the direction of the stimulating current, must con- 
 tinue to afford new and enigmatical problems. 
 
PHENOMENA OF MUSCLE. 239 
 
 Among these difficulties may be mentioned the difference in 
 streng-th of the polarisation-currents according to the direction of 
 the stimulating current in the pair of muscles, and the fact that 
 the relations of the polarisation-currents led off from the upper 
 half of the muscle are in some respects the reverse of those led off 
 from the lower half Another question is, why the positive polar- 
 isations in each half are stronger when the stimulating current 
 flows from the equator towards the extremity, than when it flows 
 in the opposite direction, and why the reverse is not the case ? and 
 why, further, all secondary actions are stronger in the lower half of 
 the muscle than in the upper? why very strong descending currents 
 (30 to 40 Groves) with closing times of %o sees, i min., 5 mins. 
 again give positive variation, ascending currents negative variation, 
 and so on ? For each of these questions I could find an answer based 
 on the experience I have gained in working with the sartorius. 
 That such answers would be correct as regards the pair of muscles in 
 question, could only be proved by the aid of numerous and varied 
 researches on them. I cannot however regard it as my task to 
 institute a tedious series of researches under such complicated con- 
 ditions as those offered by du Bois-K-eymond's preparation. Such 
 researches could at best only lead by a long and circuitous path to 
 such a knowledge of the same laws as is afforded directly by the 
 study of a regularly constructed muscle. 
 
 With the observation just mentioned, that with certain current 
 strengths and closing times the positive polarisation was stronger 
 in each half of the muscle when the current flowed in that half 
 from the equator to the end, du Bois-Reymond connects the suppo- 
 sition that this phenomenon stands in relation to the circumstance, 
 that in natural stimulation of a muscle through the nerve, the 
 excitation spreads from the neighbourhood of the equator where the 
 nerve-filaments enter the muscle-fibres, towards both ends. Hence 
 he infers that the direction in which normally the proper physiolo- 
 gical changes are propagated in muscle-fibres is the most favourable 
 likewise for positive polarisation. 
 
 It is not however clear how the observations of du Bois-Reymond 
 on his muscle-preparation could lead to such a supposition. For 
 each muscle of the pair, on account of the tendinous intersections, 
 consists functionally of two separate muscles. With regard to the 
 rectus internus major or gracilis this is strictly true ; with regard 
 to the semimembranosus, it is at least partly so, for only some 
 of its fibres run the whole length of the muscle, all the others 
 
340 RESEARCHES ON SECONDARY ELECTROMOTIVE 
 
 either end or commence in the tendinous intersection. Hence the 
 stimulus imparted by the nerve would not start from the neig-h- 
 bourhood of the equator of the pair of muscles, i.e. from the region 
 of the intersection, but each half being functionally a muscle of 
 itself would have its own equator. 
 
 It would be contrary to all analogy if the motor end plates were 
 especially crowded together in the vicinity of the intersections ; this 
 would be much more likely to be the case in the neighbourhood of 
 the equator of each individual half of the muscle, with the exception 
 of the long fibres of the semimembranosus. Instead of comparing 
 the positive polarisation in both halves of the muscle, du Bois- 
 Reymond ought to have investigated each quarter of the muscle for 
 comparison. In the first and third quarters, counting from above, 
 the ascending positive polarisation ought to have proved strongest, 
 in the second and fourth the descending, if the experimental 
 evidence was intended to support the suggestion of a connection 
 between functional direction and positive polarisation. 
 
 For the rest in Sect. 21 of his paper (No. VI), du Bois-Reymond 
 makes no further statement about the muscles with which the 
 researches were actually conducted, when he says, ' In regular 
 monomerous muscles (i. e. those consisting of longitudinal fibres only) 
 positive polarisation is stronger in the direction from the equator 
 towards the ends, than in the opposite direction.' No experiments 
 however on monomerous muscles are contained in the whole treatise, 
 but only those already mentioned with the rectus internus and semi- 
 membranosus, which are pre-eminently 'pleiomerous' muscles, i.e. 
 such as have fibres commencing and terminating within their limits. 
 Moreover, had du Bois-Reymond repeated his experiments with 
 monomerous muscles, he would have found the true condition to be 
 the reverse of his statement with regard to them, viz. that with 
 stimulating currents of relatively equal strength (and corresponding 
 density) and of the same short closing time (0-32'^) he would have 
 found positive polarisation in the half of the muscle next the 
 stimulating anode, none or weak negative polarisation in the half 
 next the stimulating kathode, and through which therefore the 
 current flowed in the direction from the equator towards the end. 
 Had du Bois-Reymond instituted such experiments as these, he 
 would at once have recognised that the muscle is not polarised in 
 the same direction in the whole intrapolar tract (see below). 
 
 In du Bois-Reymond's speculations as to the connection between 
 positive polarisation of muscle and the direction in which the 
 
PHENOMENA OF MUSCLE. 241 
 
 wave of contraction liberated from the nerve flows, there appears 
 to me to be a contradiction which is the more inexplicable from 
 the fact that he described the structure of both muscles used by 
 him, and especially mentioned the circumstance that each half 
 of the muscle received its own special nerve supply. 
 
 II. 
 
 Du Bois-Reymond informs us at the end of his treatise that he 
 was acquainted with positive polarisation of muscle long* before the 
 issue of the first volume of his ' Researches' (i 848), and that from it he 
 was led to the conception of the electromotive phenomena of nerve 
 and muscle. From positive polarisation of muscle he was led to 
 a similar conclusion with reg-ard to the intrapolar tract of nerve, and 
 a few years later he actually discovered the intrinsic polarisation of 
 nerve which before he had only assumed theoretically. 
 
 ' In so far as no other reasonable explanation of positive polarisation 
 could be given,' continues du Bois-Reymond, ' excepting that which 
 regards it as due to the directing of already existing forces, the 
 evidence obtained of this polarisation is at the same time serviceable 
 in another way. In my eyes it afforded and will continue to afford 
 proof of the presence of electromotive forces in uninjured nerves — 
 proof which in the absence of a natural transverse section of a nerve 
 was otherwise lacking. The evidence of the existence of such forces 
 in muscle, obtained by carefully testing the natural transverse 
 section of muscle, is also strengthened by it.' Hence, in internal 
 positive polarisation, according to du Bois-Reymond's own assertion, 
 we are expected to see the origin and mainstay of his whole theory 
 of nerve- and muscle-currents. Here again I find myself com- 
 pelled, by means of a series of experimental examples, to show that 
 intrapolar positive polarisation in du Bois-Reymond's sense cannot 
 be demonstrated. 
 
 I begin with an experiment in which the stimulating electrodes 
 were placed on the muscle in the same manner as they were in 
 du Bois-Reymond's experiments, although this method of leading 
 in the current is not generally to be recommended for reasons given 
 above. Instead, however, of a muscle divided by a tendinous inter- 
 section, I used a curarised sartorius, which^ according to Aeby's re- 
 searches, is in most cases though not invariably really monomerous, 
 which matter I discussed in my twelfth communication^ (see p. 229). 
 ' [Not included in the present series of translations. See Preface. Ed.] 
 
 R 
 
242 llESKARCHES OX SECONDAEY ELECTROMOTIVE 
 
 This muscle, stretched by the clamping of the pelvis and tibia, 
 lies horizontally with its borders directed upwards and downwards. 
 As the muscle tapers away by termination of fibres towards its 
 lower attachment, I place the lower electrode at the junction of 
 its middle and lower thirds, the upper one I cm. below the upper 
 end on the wide surface of the muscle. Exactly opposite both these 
 electrodes, on the other wide surface of the muscle, I place the 
 galvanometer electrodes — one opposite each. Midway between 
 these two, on the same surface, a third galvanometer electrode is 
 applied. In rapid succession, the lower and upper electrodes, or 
 the lower and middle, or the middle and upper can l)e thrown 
 into the galvanometer circuit, so that alternately the whole of the 
 intrapolar tracts, or only the lower or only the upper half of the 
 muscle is in circuit. All the electrodes are tube electrodes. I use 
 the whole current of 7 or 8 Daniells for stimulation. The period 
 of stimulation is 5 sees., the ' transmission time ' -026 to -034 sec. 
 A current of this strength, with this period of stimulation with 
 a fresh, well-excitable muscle and electrodes arranged as above, 
 gives strong positive anodic and weak negative kathodic po- 
 larisation. 
 
 If immediately after opening the stimulating current I throw in 
 the lower half of the intrapolar tract next the stimulating anode, 
 I get a strong positive polarisation current ; if I throw in the upper 
 half next the stimulating kathode, I get a weak negative polarisation 
 current. If, finally, after opening the stimulating current I throw the 
 whole of the intrapolar region into the galvanometer circuit, this 
 gives a polarisation current corresponding to the difference between, 
 the positive anodic and, the negative haihodic polarisation, making 
 allowance for the change of resistance caused by the introduction of 
 a longer tract of muscle. Naturally it would be best if after one and 
 the same stimulation the polarisation of all three tracts could be 
 measured separately, but simultaneously. As this is impossible, it 
 is necessary to make several experiments on the same muscle. Each 
 preceding experiment influences that which follows it. Tliis cir- 
 cumstance only changes to some extent the strength, not the 
 direction of the individual polarisation currents. In these experi- 
 ments we have to deal not with subtle but with relatively substantial 
 phenomena. To eliminate the influence of the order in which 
 the three tracts are investigated, either researches are made with 
 several muscles in different order, or a series of experiments is re- 
 peated in reversed order on the same muscle, as was partly the case in 
 
PHENOMENA OF MUSCLE. 
 
 243 
 
 the researches tabulated below ; but the direction of the current 
 must not be chang-ed in the course of an experiment. For, so 
 strong- a current at its point of exit, impairs considerably the 
 susceptibility of the muscle substance for positive anodic polarisation. 
 Thus, when in an experiment the spot of muscle which had acted as 
 the point of entry for the descending current, served as the point 
 of exit for the stimulating- ascending- current, it showed much 
 weaker positive polarisation, and mig-ht even show neg-ative polari- 
 sation, a circumstance of which I made mention in the preceding- 
 communication. An exactly analogous experiment can be made 
 on another muscle with a descending current ; it is best to use the 
 other sartorius of the same frog. 
 
 The following table gives examples of such a series of experiments. 
 The Roman numerals correspond to the order of the individual 
 experiments on the same muscle : — 
 
 Ascending current. 7 Daniells. Closing time 5 sees. Transmission time -026 
 to -034 sec. 
 
 
 Entire intrapolar 
 tract. 
 
 Lower half of 
 intrapolar tract. 
 
 Upper half of 
 intrapolar tract. 
 
 I. 
 
 + 220 
 
 
 
 II. 
 III. 
 
 IV. 
 
 
 ■•-342 
 
 + 287 
 
 -45 
 
 V. 
 
 VI. 
 
 VII. 
 
 + 177 
 + 123 
 
 
 -43 
 
 2, 
 
 Descending current. 8 Daniells. Closing time 5 sees. Transmission time .026 sec. 
 to -034 sec. 
 
 
 Entire intrapolar 
 tract. 
 
 Lower half of 
 intrapolar tract. 
 
 Upper half of 
 intrapolar tract. 
 
 I. 
 
 + 200 
 
 
 
 II. 
 III. 
 
 IV. 
 V. 
 
 + 136 
 
 -35 
 
 + 300 
 + 239 
 
 VI. 
 
 
 -26 
 
 
 R 3 
 
244 
 
 EESEAECHES ON SECONDARY ELECTROMOTIVE 
 
 With weaker currents and tlie same arrangement, negative 
 kathodic polarisation outweig-hs positive anodic, so that the current 
 led off from the whole intrapolar tract is negative, of which an 
 example may be given : — 
 
 Descending current. 4 Darnells. Closing time g sees. Transmission time -026 see. 
 to -034 sec. 
 
 
 Entire intrapolar 
 tract. 
 
 Loioer half of 
 intrapolar tract. 
 
 Upper half of 
 intrapolar tract. 
 
 I. 
 
 II. 
 
 III. 
 
 -30 
 
 -92 
 
 + 60 
 
 After such experimental results no one will be disposed to 
 maintain that the entire intrapolar tract is internally polarised in 
 the same direction. The utmost that could be supposed with 
 regard to that matter, is that one half undergoes internal positive 
 the other internal negative polarisation, or that both are positively 
 and negatively polarised at the same time, but in such a manner 
 that the positive is in excess at the anodic half, the negative at 
 the kathodic half of the intrapolar tract. This would be a totally 
 different kind of polarisation of the intrapolar tract from that 
 assumed by du Bois-Reymond. 
 
 Admitting, however, the existence of an internal polarisation 
 in opposite directions of the two halves of the intrapolar tract, 
 it can be conclusively shown that such a polarisation could not be 
 generated by the current within the intrapolar tract, but must be 
 caused by a change of the muscle substance propagated from the 
 anodic or kathodic spots to the intrapolar tract. For if the anodic 
 spots of the muscle which showed strong positive anodic polarisa- 
 tion be destroyed, without making any other alterations in the 
 arrangement of the apparatus, and if then the same stimulating 
 current be again sent through in the same direction, there is no 
 positive but at the utmost weak negative anodic polarisation on 
 breaking the current. 
 
 I have performed this experiment in various ways, and always 
 with the above result. For instance, the stimulating current was led 
 through the bones to the horizontally extended sartorius, and one 
 
PHENOMENA OF MUSCLE. 245 
 
 electrode was placed on the middle of the muscle, the other on its 
 attachment to the pelvis. After breaking the descending- current the 
 positive polarisation current was measured. The upper end of the 
 muscle was destroyed by means of a heated glass rod, with which 
 it was touched in the region of its natural section. After the 
 muscle was thus destroyed I replaced the upper electrode on the 
 same spot. On heating the muscle it thickened, and now the 
 current no longer flowed into the living muscular substance through 
 the short tendon, but through the dead portion. Thus the 
 density of the current was now less at its point of entry into the 
 living portion of muscle than it had been previously. On this 
 account, in the experiment with the uninjured muscle, I placed on 
 both sides of the tendon, in the groove between the muscle substance 
 and the bone, a piece of blotting-paper folded four times and steeped 
 in normal salt solution. By this means the density of the current 
 was decreased at its ' point of entry into the muscle substance. 
 A current from 4 Daniells was now sent through the muscle for 
 5 sees, in a descending direction, with the result, to quote a par- 
 ticular instance, of positive polarisation to the extent of 129 degrees 
 of the scale ^ After this the upper electrode was removed, the end 
 of the muscle was destroyed, and the electrode again replaced. The 
 strong descending demarcation-current of the muscle now flowed 
 into the galvanometer-circuit. As soon as the muscle had cooled 
 and its current had become constant, this was compensated, and the 
 same current as before was again sent through for 5 sees, in a 
 descending direction : there was now negative polarisation to 
 the amount of 13 scale. 
 
 I also made similar experiments with the current led in from the 
 side. Let it be imagined that the two electrodes are placed on one 
 of the wide surfaces of the muscle at modemte distances from the 
 ends, that one galvanometer electrode is directly opposite the upper 
 stimulating electrode, the second in the middle of the muscle. 
 With this arrangement I obtained, for example, with a descending 
 current of 8 Daniells closed for 5 sees, after the first stimulation, a 
 positive polarisation current amounting to 300 scale; after the 
 third, a similar one still amounting to 239'"'. 
 
 Hereupon the upper stimulating electrode and the upper galva- 
 nometer electrode were removed, the spots of muscle on which they 
 had rested were destroyed on either side with a heated glass rod, then 
 
 1 Two excitations with the ascending current had preceded this, hence positive 
 polarisation was already diminished. 
 
246 RESEARCHES ON SECONDARY ELECTROMOTIVE 
 
 moistened with -6 per cent, salt solution, and the two electrodes re- 
 placed in their respective positions. After the strong- descending* 
 'demarcation' current had become constant, it was compensated, 
 and the muscle was again stimulated for the same period Avith the 
 same descending* cui'rent ; on breaking, negative polarisation resulted 
 to the amount of ^6 divisions of scale. 
 
 Tn destroying the muscle by heat it is necessary to defer 
 exciting anew until the heated portion cools down. The result 
 is the same if the muscle is locally destroyed by crushing with 
 a pair of pincers. After crushing the lower end of the sartorius, I 
 obtained only slight negative anodic polarisation, with ascending 
 currents of the same strength and duration as those with which I 
 had obtained positive anodic polarisation, amounting to far more 
 than lOO divisions of the scale, from the uninjured lower end of the 
 muscle. And even with the whole current from 3 Daniells and 
 with lO sees, closing time, I obtained only weak negative polari- 
 sation to the amount of 20^*^, whereas when the lower end of the 
 muscle is in a normal condition, the whole current is wont to give 
 such a strong positive anodic polarisation current, even wuth short 
 closing time, that the image of the scale shoots out of sight. The 
 difference of action before and after destruction is in this case more 
 prominent, because the current enters the normal lower end of the 
 muscle with greater density than it does the living muscle substance 
 above the crushed end. That the muscle substance lying above the 
 lower destroyed portion, and that lying below the upper destroyed 
 portion as well as the w^hole remainder of the uninjured muscle, has 
 not lost its susceptibility for positive anodic polarisation, can be 
 easily shown by control experiments. 
 
 The fact thus established, that the stimulating current which 
 passes through dead and dying muscle substance to enter muscle 
 which is still excitable and living, is not in a condition to produce 
 positive anodic polarisation in the anodic spots of the latter, 
 as well as the fact shown in Biedermann's researclies that it is 
 incapable of producing the break-contraction, proves that the 
 cause of so-called positive polarisation must be sought for, for the 
 most part if not exclusively, not in changes which the current 
 directly produces in the whole or in some part of the intrapolar tract, 
 but in changes taking place at the anodic spots of the muscle sub- 
 stance, which spread more or less from the latter into the former. 
 T/ie positive polarisalion-curreni is thus dependent upon the integrity 
 of the anodic spots of the excitable substance. 
 
PHENOMENA OF MUSCLE. 247 
 
 That weak positive anodic polarisation can be obtained by passing- 
 excessively strono- currents throug-h a destroyed spot into the 
 muscle, is on obvious grounds not to be doubted ; but weak break- 
 contractions are also obtained under those circumstances. 
 
 Negative kathodic polarisation becomes almost impossible when 
 the point of exit of the current is killed, provided that its strength 
 and duration are not too great. 
 
 Experiments in which the electrodes are applied to the surface of 
 the muscle have the disadvantage, that the strength of the polarisation 
 phenomena, with equal strength of stimulating- current, must de- 
 pend on the extent of surface of contact of the electrodes. For it is 
 this which determines the density with which the current enters or 
 leaves the muscle, and on this density the strength of anodic and 
 kathodic polarisation depends. Even if the density of the current is 
 equal throughout the intrapolar tract, i.e. the tract in which the 
 fibres are parallel, yet it may be different at the spot of entry or 
 exit, and at the same time the strength of the polarisation-current 
 may be very different, whereby the measurement of the intensity 
 of the stimulating current loses much in value. It is also of 
 no little importance, as has already been explained above, that the 
 current enters the fibres cross-wise or obliquely, and that nearly all 
 the fibres are traversed by portions of current, and therefore have 
 both anodic and kathodic spots. 
 
 Tschiriew ^ has of late given an accovmt of experiments made with 
 the sartorius. in the results of which he sees confirmation of du Bois- 
 Reymond's theory. He laid both stimulating electrodes on one sur- 
 face of the muscle, both galvanometer electrodes (Fleischl's brush 
 electrodes) exactly opposite them on the other surface, so that the 
 intrapolar tract amounted to about 15 mm. He obtained, 'with 
 short periods of closure of the polarising current — up to 4 sees. 
 and sometimes more — and with the freshest possible preparations, 
 purely positive excursions of the needle. With a longer closing 
 time, and when from the duration of the experiment the ex- 
 citability of the preparation had decreased, the excursions became 
 diphasic, a negative deflection preceded the positive, and finally 
 a purely negative polarisation remained.' As appears from a table 
 of experiments, Tschiriew changed the direction of the current 
 from one experiment to another in the same series 2. The change 
 in the direction of the current made the spot of the muscle 
 
 » Archiv f. Anat. u. Physiol. 1S83. Festschrift, Suppl. Vol. p. 284. 
 
 ^ The explanation of his results follows naturally from what I have already said. 
 
248 RESEAKCHES ON SECONDARY ELECTROMOTIVE 
 
 which was in contact with onb stimulating* electrode, first the point 
 of entry, then the point of exit of the stimulating- current. Through 
 the kathoclic action of the current the susceptibility of the muscle 
 for positive anodic polarisation was soon diminished in the neig-h- 
 bourhood of both the stimulating electrodes, so that the preponder- 
 ance of negative kathodic polarisation kept increasing, and at the 
 same time there was negative polarisation even at the anode. 
 
 As the whole intrapolar tract was included in the galvanometer 
 circuit, only the algebraic sum of the anodic and kathodic polari- 
 sation-current would find expression in the deflection of the magnet. 
 I adopted substantially Tschiriew's method in my earliest experi- 
 ments on polarisation of muscle. I applied to the under surface of 
 the horizontally extended muscle, across its long axis, two small 
 cylinders of the clay used for electrodes, these cylinders being in 
 contact on one side with the stimulating electrodes, on the other with 
 the galvanometer electrodes. This arrangement has the advantage 
 over that of Tschiriew, that the polarising cm-rents enter and leave 
 the muscle at the same points as the galvanometer-currents. In 
 comparison with du Bois-Reymond's arrangement, Tschiriew's 
 has the advantage that the polarisation currents recorded were 
 always the result of two polarisations, one anodic, the other ka- 
 thodic, whilst in du Bois-Reymond's experiments, besides the two 
 polarisations in the neighbourhood of the stimulating electrodes, the 
 considerable polarisations at the tendinous intersections come into 
 account. 
 
 I also made some experiments with du Bois-Reymond's muscle 
 preparation, as well as with the muscle which he calls gracilis, 
 by itself, introducing the current through the bones. With this 
 arrangement from one and the same muscle, with the same intensity 
 and strength of current, I obtained polarisation currents which 
 were of different strengths and differently directed, according 
 as I applied the electrodes to one or another spot of the muscle, 
 keeping the electrodes in all cases about i cm, apart. This was 
 the case whether I used the same pair of electrodes, and shifted 
 them before each new stimulation, or whether I used several pairs 
 of electrodes, and connected them alternately with the galvanometer, 
 as described above for analogous experiments. The confusion of 
 differently directed or diphasic polarisations was so great that I did 
 not feel tempted to pursue the matter further. If the electrodes are 
 always applied in the same position to the muscle, the polarisation 
 currents must necessarily follow a certain law. 
 
PHENOMENA OF MUSCLE, ' 249 
 
 Du Bois-Reymond called the closing time which was necessary 
 with a definite streng-th of stimulating" current and a fresh prepara- 
 tion, in order to obtain polarisation in alternately opposite direc- 
 tions, the ' critical time.' (See No. VI. Sect. 9.) Shorter closing- 
 times gave with the same stimulating cvirrent positive polarisation, 
 and longer, generally negative. Under special conditions, as when, 
 descending currents of 30 to 40 Groves were used, positive polarisa- 
 tion reappeared. After all that has been said, the determination of 
 this ' critical time ' has no great value. In anodic polarisation, how- 
 ever, a critical closing time might be spoken of, inasmuch as with 
 fresh muscle and weak stimulating currents the polarisation is nega- 
 tive with short, positive with longer closing times. Even if it 
 were possible always to produce equal density of current at the 
 anodic spots in a series of experiments on fresh muscles, this 
 ' critical time ' would depend essentially on excitability of the 
 muscle, which varies greatly even in frogs which have been kept 
 under the same conditions. 
 
 I cannot yet speak of a ' critical time ' for kathodic polarisation 
 as I have not yet succeeded in demonstrating positive kathodic 
 polarisation. By applying the electrodes to the sides of the prepara- 
 tion, in which case the stimulating current traverses the fibres 
 transversely or obliquely, I have repeatedly observed diphasic polar- 
 isation, and once indeed positive kathodic polarisation. As, how- 
 ever, with this method of leading in the stimulating current, the 
 greater number of the fibres of the muscle must contain both anodic 
 and kathodic spots in the neighbourhood of the kathode, and as the 
 density of current is not the same in the former as in the latter, 
 it is not surprising that we get in this case diphasic polarisation. 
 But, on this point, I can at present express no definite opinion, 
 because, as I said, I have seldom employed the lateral method of 
 leading in, which is in general unadapted to the purpose, and 
 therefore I possess no series of experiments as to the polarisation 
 which occurs when this method is employed. The discussion of 
 this question must be the object of special researches on polarisa- 
 tion with transverse and oblique currents. 
 
 Du Bois-Reymond avoided leading in through the natural ends 
 of the muscle, because he feared the secondary resistance pointed out 
 by himself in the thin tendons, considering that as 50 Groves were 
 used, and the closing times amounted to 5 min., this must have 
 been considerable. 
 
 In our experiments, which were made with at the most 
 
250 RESEAECHES ON SECONDARY ELECTROMOTIVE 
 
 8 Daniells, the question of secondary resistance was of the less im- 
 portance, as from the first it proved to be unnecessary to exercise 
 any strict supervision of the streng-th of the stimulating- current. 
 According-ly we did not spend much time in measuring- it. In 
 more minute experiments, however, this would occasionally be 
 necessary. 
 
 It must not be thought that our researches are not comparable 
 with those of du Bois-Reymond because we used relatively much 
 weaker currents. The question here is not of the intensity, 
 but of the density of the current in the muscle, and this is incom- 
 parably g-reater at the natural ends, especially at the lower end of 
 the sartorius, with the same streng-th of current, than at the thick 
 portion of the pair of muscles used by du Bois-Reymond ; and 
 this holds good notwithstanding* that the sartorius, when the current 
 is led in through the bones, offers a much greater resistance than 
 the pair of muscles does when the current is led in laterally. 
 Hence it is only those experiments conducted by du Bois-Reymond 
 with extraordinarily strong currents which cannot be at once 
 brought into comparison with ours. 
 
 If the care and accuracy with which du Bois-Reymond is wont 
 to conduct his experiments, as well as the acuteness with which he 
 designs and interprets them, be taken into consideration, it will 
 appear surprising that so distinguished an observer should have 
 fallen into so fundamental an error, an error which derives its im- 
 portance from the circumstance that he founds on it his whole 
 theory of nerve and muscle currents. I find the cause of the error 
 mainly in the fact that du Bois-Reymond has not devoted the same 
 thorough study to the mechanical effects of excitation of muscle as 
 he did to the electrical effects. Among other indications this appears 
 from the fact that in 1881 in his work on the researches con- 
 ducted by Dr. Sachs on the Gymnotus, he remarks ^ cursorily 
 that ' in direct stimulation ' (with the electric current) ' contrac- 
 tion commences at all points of the muscle at the same mo- 
 ment.' Quite apart from the researches referring to this matter 
 of Bezold and Engelmann, Biedermann in 1879 brought con- 
 clusive evidence of the fact that it is not so, but that the make- 
 contraction commences at the kathodic, the break-contraction at 
 
 ' Dr. Carl Sachs, Untersuchungen am Zitteraal. Leipzig, 1881, p. 226. 
 
PHENOMENA OF MUSCLE. 251 
 
 the anodic spot of the muscle. According'ly contraction can only 
 be supposed to commence simultaneously throug-hout the whole 
 intrapolar tract when the currents are of g-reater density in the 
 muscle than the strong-est currents used by Biedermann. 
 
 Had not du Bois-Reymond adhered to the old view of simul- 
 taneous stimulation in the whole intrapolar tract, had he even 
 regarded as an open question that of the starting-point of excita- 
 tion in direct electrical stimulation, he would scarcely have em- 
 ployed in his experiments on polarisation muscles divided by 
 tendinous intersections. At the same time it was an unfortunate 
 circumstance that the very muscles which he chose on account of 
 certain advantages presented by them should have been traversed 
 by tendinous intersections, for had those intersections not been 
 present, du Bois-Reymond would certainly have recognised the 
 error of his assumption, at all events in those experiments in which 
 he led off the polarisation-current from one or the other half of 
 the intrapolar tract. 
 
 In a future communication I shall have to speak of du Bois- 
 Reymond's researches on secondary electromotive phenomena 
 in nerve. 
 
VIII. 
 
 ON 
 
 THE POSITIVE VAEIATION 
 
 OF THE 
 
 NEME CUEEENT AFTEE ELECTEICAL STIMULATION. 
 
 By dr. EWALD HERING, 
 
 Professor of Physiology in the German University, Prague. 
 
 Sitzungsberichte der kaiserlichen Akademie der Wissenschaften, Vienna, 
 Vol. Ixxxix, Part 3, p. 137, 1884. 
 
CO:N^TE]SrTS. 
 
 PAGE 
 
 1. Discussion of Methods of Experimentation ..... 255 
 
 2. Metliod actually adopted by the Author 260 
 
 3. Tables of Experimental Results 262 
 
 4. Discussion of the same . 267 
 
 5. Modification of Phenomena when the Stimulations were of such 
 
 strength as to give rise to unipolar action ..... 26S 
 
 6. Discussion of du Bois-Eeymond's older Observations in this relation 2^0 
 
VIII. 
 
 The investigation of negative variation of the nerve current, 
 with the improved methods for which we are indebted to du Bois- 
 Reymond, leads to results which modify in many respects the doc- 
 trines taug-ht by its discoverer respecting- it. 
 
 I may premise that in future communications I shall frequently 
 have the opportunity of recurring to the negative variation, and 
 shall confine myself here to a short discussion of the fact that the 
 negative variation of the nerve current caused by electrical tetani- 
 Bation is generally followed by a positive variation, which commences 
 at the close of the stimulation and thus immediately follows the 
 negative variation. 
 
 Du Bois-Reymond's statement, 'that after tetanic stimulation 
 the needle of the multiplier only returns more or less incompletely 
 to its state of rest,' a phenomenon which he attributes to loss of 
 electromotive power of the nerve in consequence of the preceding 
 Btimulation, accordingly aj^pears to be insufficient. My observa- 
 tions were made in December and Januarys, partly with cold, 
 partly with warm frogs (R. esculenta) ; the former term I apply 
 to frogs brought straight from the cellar, the latter to those kept 
 for some days in a room the temperature of which never fell below 
 
 The tract of nerve lying between the stimulating electrodes I 
 call the stimulated region. It always consists of a portion of the 
 sciatic plexus a little below the ligatured upper end, and was 5 mm. 
 in length. The portion lying between the leading off electrodes 
 will be designated the galvanometer tract. It is always at the 
 peripheral end of the nerve. The portion between the lower 
 stimulating electrode and the upper galvanometer electrode will 
 be called the intermediate tract. 
 
256 ON POSITIVE VARIATION OP THE NERVE CURRENT 
 
 Unpolarisable electrodes were almost exclusively employed. The 
 lower ends of these were filled with saline clay, into which was intro- 
 duced a brush steeped in -6 per cent, salt solution. An ordinary 
 camel-hair pencil was used, and this was cut off about i cm. above 
 the lower end of the quill. If there are several such brushes at hand, 
 it is easy to find one to suit any electrode tube, fitting firmly into 
 it. These electrodes are made like the pencil electrodes of Fleischl, 
 with this difference, that modelling" clay serves instead of g-ypsum 
 to close the lower extremity. This material is easily renewed after 
 each experiment. These electrodes are only occasionally completely 
 free from current. I therefore measured this current, before each 
 experiment, by bringing the two points into direct contact. Its 
 strength is given in each case in the experiments hereafter quoted. 
 As we are dealing with nerve currents, the galvanometer used was 
 very sensitive. Consequently the electrode deflection was sometimes 
 considerable ; but it must not be forgotten, however, that after intro- 
 duction of the nerve with its great resistance into the circuit, the 
 current of the electrodes has not nearly so marked an effect on the 
 deflection of the needle caused by the nerve current, as it would have 
 had without that increase of resistance. On the amount of negative 
 or positive variation observed after compensation of the nerve current, 
 the current of the electrodes has no influence ; I therefore took no 
 further pains to secure perfectly unpolarised electrodes, although these 
 are indispensable in other cases. If a larger number of electrodes be 
 prepared on the type above described, two may generally be found 
 which are quite, or nearly devoid of current at least for a time. 
 The stimiilating currents were also introduced by these electrodes. 
 A moist chamber protected the nerve from drying, but unavoidably 
 favoured unipolar action. Very weak stimulating currents were 
 therefore used and numerous control experiments were made, 
 especially those in which the nerve was crushed in the inter- 
 mediate tract. The secondary coil (5000 turns) of the induction 
 apparatus (worked by i Daniell) was scarcely ever brought nearer 
 than 20 cm. to the primary, under which condition unipolar action 
 was not noticeable. In spite of the resistance of the electrodes, the 
 induction current thus obtained proved quite sufficient. On account 
 of the many drawbacks to the ordinary induction apparatus in use 
 in physiological laboratories, the following method was employed to 
 produce induction currents. 
 
 In order to obtain reversed but physiologically equal currents, a 
 few years ago I improAased the following apparatus. A secondary 
 
AFTER ELECTRICAL STIMULATION. 257 
 
 coil of 7 II o turns was mounted so as to revolve on a vertical axis. 
 The ends of the sjiiral terminate in slender plates running- in circular 
 troughs filled with mercury, of which one is placed above, the other 
 below the revolving* coil. These channels communicate with the 
 stimulating electrodes. The coil is made to revolve by means of a 
 puUey. If a strong magnet or the primary coil of an induction 
 apparatus through which a current is passing-, be brought into the 
 neighbourhood of the revolving coil, and if the rate of revolution is 
 high enough, currents in alternately opposite directions are obtained, 
 which are sufficiently strong and physiologically equal. The strength 
 of the currents may be varied on the one hand by changing the 
 number of elements which supply the primary current, on the other 
 they may be adjusted with the utmost delicacy by altering the posi- 
 tion of the primary coil or magnet. With only i Daniell and 20 
 revolutions per second, I got tetanus with sensitive preparations 
 when the axes of the two coils (when parallel) were 9 cm. distant 
 from each other, the most superficial layers of the two coils being 
 25 mm. apart. In the experiments I am about to describe, 3 Daniells 
 gave the primary current, the coils were as near together as possible, 
 and the number of revolutions amounted to about 20 per second, 
 corresponding to twice as many induction currents. 
 
 An induction apparatus was also used in which the primary coil was 
 replaced by a single conducting ring. A flat wooden ring (vulcanite 
 would have been a better material) was made 87 mm. in diameter 
 and 30 mm. in width. Around this 11 20 turns of thin copper wire 
 were wound. Then a second ring was made of the same width but 
 somewhat larger. This encircled the first ring, and a space was left 
 between the two for a thin copper ring 104 mm, in diameter. The 
 copper ring was nearly the same width as the wooden rings (2501111.). 
 A slit was made in it so that it did not form a closed circuit. It 
 consisted of a circularly bent strip of copper, the ends of which 
 were brought into electrical continuity by means of terminals w^hich 
 communicated with wires from the battery. The primary circuit thus 
 constituted required a special interrupter. For this purpose, I used 
 either a wheel provided with twelve copper points which passed in 
 quick succession through a vessel of mercury, or one of Wagner's 
 hammers adapted for the purpose. This was set in action by a 
 special cell, and besides interrupting its own current, effected 
 the regular interruption of the above-mentioned primary circuit. 
 The introduction of the hammer into the primary circuit as in the 
 ordinary induction apparatus is inadmissible here, for if the coil of 
 
 s 
 
258 ON POSITIVE VARIATION OF THE NERVE CURRENT 
 
 the electromagnet were in the primary current, extra currents 
 would be produced which it is desirable to avoid. The turns on the 
 outer ring- tog-ether with those on the inner ring form the secondary- 
 coil. In consequence of the close proximity of all the turns of this 
 double coil to the inducing ring of copper, even i Daniell is quite 
 sufficient to obtain tetanus from a sensitive nerve-muscle preparation. 
 In the experiments which we are going to consider, 3 Daniells 
 wTre used. In order to ensure the possibility of varying the 
 strength of the induction currents, the copper ring is made move- 
 able, and can be partly or entirely removed from the rings of the 
 secondary coil. 
 
 Although this induction apparatus is preferable to the ordinary 
 one in all experiments where it is desirable to obtain an equal 
 duration of the make- and break-currents, yet it shares with it some 
 of the disadvantages connected with the mode of interruption, par- 
 ticularly that the interval between the make and the following 
 break is not quite equal to that between the break and the follow- 
 ing make, unless a special contrivance is used for the purpose. 
 Moreover, exact equality between make- and break-currents was not 
 obtained. 
 
 The apparatus I use is only intended for the stimulation 
 of nerve. The strength of the currents can however be in- 
 creased, with the same number of turns, by increasing the diameter 
 of the rings. I also stimulated with ascending and descending 
 break-currents of an ordinary induction apparatus, the make currents 
 being cut off. Finally, interrupted battery-currents ascending, 
 descending, and alternately in opposite directions were employed 
 to stimulate the nerve. In all cases the positive variation of the 
 nerve current was clearly demonstrated. All experiments were 
 made with a Wiedemann's galvanometer wuth the magnet aperiodic. 
 This instrument is not especially well adapted to these experiments, 
 because the time which tHe magnet takes in coming to rest is too 
 great in proportion to the duration of the positive variation. I shall 
 take some opportunity of repeating the experiments with a galvano- 
 meter more suited to the purpose. So much of the phenomenon 
 as I am now concerned with can be satisfactorily examined with the 
 above-named galvanometer. 
 
AFTER ELECTRICAL STIMULATION. 259 
 
 The positive after- variation may be investigated by tbree distinct 
 methods. 
 
 I. Without a compensator, the nerve current is led off from 
 the lower end of the nerve, and a short period is allowed after the 
 magnet has taken up its new position in order to be assured of the 
 constancy of this current. When this is satisfactory, the nerve is 
 tetanised near the other end. As a result of stimulation the magnet 
 swings back for a certain distance. At the end of excitation, the 
 magnet returns to and even passes beyond the position which it had 
 before stimulation, but as a rule comes back immediately, and slowly 
 regains its original position. The positive after- variation is now 
 over. 
 
 It is here of course understood that the magnet swings aperiodi- 
 cally, otherwise its excursion beyond the original deflection would 
 afford no proof of a transitory increase of the nerve current. I give 
 an example of experiments conducted in this manner : — 
 
 Wann Frog. Du Bois-Reymond coil. Galvanometer tract 5 mm. 
 Intermediate tract 30 mm. 
 
 Deflection caused by current of electrodes . . 500 to 510. 
 
 „ nerve current . 
 
 Duration of stimulation 10 sees. Return of magnet 
 Deflection at close of stimulation 
 Immediate return of magnet .... 
 Positive after-variation therefore amounts to . 
 
 500 „ 621 = + T2I. 
 „ 609 = — 12. 
 „ 625. 
 „ 621. 
 
 + 4. 
 
 2. Experiments conducted with the nerve current compensated are 
 much preferable to the above. The negative variation of the nerve 
 current is now shown by deflection of the magnet in the opposite 
 direction under the influence of the compensation current. At the 
 close of stimulation, the magnet not only returns to its position of 
 equilibrium, but passes beyond it, and then immediately, or after a 
 very short interval comes back and slowly regains its original 
 position. This concludes the positive after-effect. When the object 
 is not to observe the whole course of the positive after- variation, it 
 is desirable to break the circuit of the galvanometer as soon as the 
 positive effect has arrived at its maximum, and the magnet is on 
 the point of retiirning ; for in this way it is more easy to make sure 
 that the magnet has not spontaneously changed position dm-ing the 
 period of the two deflections. 
 
 The following is an example : — 
 
 Cold frog. Du Bois-Reymond coil. Galvanometer tract 10 mm, _ Intermediate 
 tract about 30 mm. 
 
 S 2 
 
260 ON POSITIVE VARIATION OF THE NERVE CURRENT 
 
 The Roman numerals in this and in the following: table refer to 
 the order of the individual stimulations. The number with a minus 
 sig-n gives the diminution of the nerve current during* stimulation, 
 that with a plus sig-n gives the amount of transitory increase of the 
 nerve current at the close of stimulation. The duration of stimula- 
 tion is g-iven in seconds in each case. 
 
 Current of electrodes +40^. Nerve current +154°. 
 
 I. i; sees. < 7 ^ III- 10 sees. •^~ 
 
 =1+2 1+5 
 
 6 
 
 L+ 8-5 
 
 II. 5 sees. < "^ IV. 20 sees. < 
 
 {;■ 
 
 -16 
 
 6-5 
 
 Cold frog. Stimulation with rotation apparatus. Galvanometer tract 8 mm. 
 Intermediate tract 38 mm. 
 
 Current of electrodes + 7. Nerve current + 195. 
 I. Duration of stimulus i ;; sees. < ~ "^ 
 
 "^ 1 + 12 
 
 TT /-38 
 
 II. ,, ,, „ 20 sees. < ^ 
 
 L+ 9-5 
 
 The increase of nerve current after close of stimulation is a transi- 
 tory one. I often noticed that the magnet, when negatively de- 
 flected during the stimulation of the nerve, suddenly moved in an 
 opposite direction at the moment the stimulation ceased, this 
 movement being out of relation with the final amount of the 
 deflection : thus it seemed probable that the positive increase of the 
 nerve current is greater immediately after stimulation than it ap- 
 pears to be from the weak positive deflection which the magnet 
 eventually shows, that is, that before the dilatory magnet as it 
 slowly returns to zero has reached that point, an appreciable amount 
 of the positive increase of nerve current may already have been 
 spent, so that only the remainder conducts the magnet through its 
 position of equilibrium to a position of positive deflection. From 
 this may be explained the fact that occasionally, and especially with 
 feebly excitable or fatigued nerves, the positive deflection is quite 
 lost when this method of observ^ation is adopted. 
 
 3. Hence it is desu-able, after accurate compensation of the nerve 
 current, to open the galvanometer-circuit during stimulation and 
 only to close it again immediately after ceasing to stimulate. This 
 plan has another advantage, \'iz. that a portion of the compensation 
 current corresponding in amount to the negative variation is pre- 
 vented from passing through the nerve during the period of 
 
AFTER ELECTRICAL STIMULATION. 261 
 
 stimulation. This is the third and by far the best method of in- 
 vestigating* the positive after-variation. In this case an aperiodic 
 mag-net is unnecessary. 
 
 With investig-ations carried out in this manner, the correctness of 
 the above supposition was shown, in so far that I got appreciably 
 stronger positive deflections than when the galvanometer-circuit 
 was closed during stimulation. In this investigation I used the 
 same double switch which I used for the researches on polarisation 
 described in my twelfth contribution. 
 
 In one position of the switch the stimulating circuit (in this case 
 generally the circuit of the secondary coil) is broken in two places, 
 while the galvanometer-circuit is closed ; in the other position the 
 galvanometer-circuit is broken in two places, while the stimulat- 
 ing circuit is closed. From this last position the switch is re- 
 versed in such a way that the time of its passage, i.e. the interval 
 between the opening of the stimulating circuit (end of stimulation) 
 and the closing of the galvanometer-circuit was always the same. 
 In our experiments it amounted to between 0-026 and 0-034 of a 
 second ^ 
 
 With this method of investigation a determination of the amount 
 of negative variation is not obtained. For this purpose two ob- 
 servers and two galvanometers, each with a special compensator, 
 would be necessary. The electrodes conducting the nerve current 
 would have to be arranged by means of a switch so as to permit of 
 their being connected now with the one, now with the other galvano- 
 meter. In both galvanometers the nerve current would have to be 
 compensated. On one the negative variation would have to be ob- 
 served, and at the end of stimulation the electrodes would have to be 
 quickly connected with the other galvanometer, on which the positive 
 variation would have to be read. At the same time the above- 
 mentioned drawback is to some extent remedied, if observations are 
 made with the same nerve and with equal strength and duration 
 of stimulation alternately according to the third and second methods. 
 If it is desired to make an observation according to the second method 
 after one has been made according to the third, it is only necessary, 
 
 * It is not to be understood that the transmission time varied between these 
 limits ; I can even venture to say that it was always tolerably nearly the same. 
 The method which I adopted to detei-mine the time in question does not permit of 
 direct measurement of that time, but gives on the one hand a duration which the 
 transmission time cannot possibly exceed, on the other hand a duration of which it 
 cannot possibly fall short. Whether the true transmission time approaches more 
 nearly to the one or the other cannot be decided by this method. 
 
262 ON POSITIVE VARIATION OF THE NERVE CURRENT 
 
 while the galvanometer-circuit is closed by the smtch, to join to- 
 gether the four wires of the switch at both points of interruption of 
 the stimulating current, so as to convert them into two continuous 
 wires. Moreover, as it is desirable not to let the individual stimu- 
 lations of the nerve follow each other too rapidly, there is ample 
 time to use each arrangement alternately, and thus to observe first 
 the positive variation alone, then the negative with the positive 
 following. 
 
 With the last-mentioned method of observation (' without switch') 
 I found without exception a smaller positive variation than with 
 the other. 
 
 A few examples will make this clear. 
 
 Cold frog. Excitation with rotation apparatus. Galvanometer tract 8 mm. 
 Intermediate tract 37-5 mm. Duration of stimulation 20 sees. 
 Current of electrodes —38. Nerve current + 1S8. 
 
 I, Stimulation with switch +21 III. Stimulation with switch +13 
 
 11. „ without switch I ~ 38 IV. „ „ „ +12 
 
 The positive variation of 21 which appeared with the first stimu- 
 lation is the greatest I have ever seen. It was also very lasting, for 
 10" after reaching the maximum, a deflection of + 10 was still 
 shown. I did not await the complete return of the magnet in order 
 to verify the zero point. 
 
 W^arm frog. Stimulation with rotation apparatus. Galvanometer tract 7-5 mm. 
 Intermediate tract 33 mm. Duration of stimulation 20 sees. 
 Current of electrodes —46. Nerve current + 167. 
 
 I. Without switch I ~^^ III. Without switch j"^^ 
 
 1+9 1+2 
 
 II, With switch +13 IV. With switch + 6 
 
 Cold frog. Stimulation with du Bois induction-coil. Galvanometer tract 9 mm. 
 Intermediate region about 30 mm. Duration of stimulation 10 sees. 
 
 Current of electrodes +17. Nerve current + 104. 
 
 I. With switch +8 II. Without switch \~^^ 
 
 L+ 4 
 III, With switch + 5-5 
 
 Cold frog. Stimulation with du Bois induction-coil. Galvanometer tract 8 mm. 
 Intermediate tract 25 mm. Duration of stimulation 12 sees. 
 Cuitent of electrodes —43. Nerve current + 85. 
 
 I. With switch +10 III. With switch + 7-5 
 
 II. Without switch I "" ^ 9 i y. Without switch { ~ ^ \ 
 
 L+ 5-5 I.+ 3 
 
 V. With switch + 3-5 
 
AFTER ELECTRICAL STIMULATION. 263 
 
 While negative variation falls off only very slig-htly or not at all 
 with good nerves when moderate stimulations are frequently re- 
 peated, notwithstanding- the shortness of the intervals between the 
 individual stimulations, positive after-variation, as far as I could 
 see, attained its maximum after the first stimulation, declining- 
 with succeeding- stimulations, at first very quickly, then more 
 slowly, and finally almost imperceptibly. In spite of very fre- 
 quently repeated moderate stimulations of short duration, I never 
 saw the positive variation quite disappear, although in comparison 
 with the negative variation which for a long period remained 
 unaltered, it was very small. In this statement it is assumed, 
 however, that the nerves which demonstrate this point are healthy. 
 
 One example will suffice. 
 
 Cold frog. Both nerves used at the same time. Du Bois induction-coil. Gal- 
 vanometeAract ^ mm. Intermediate tract 26 mm. Duration of stimulation 12 sees. 
 Alternately with and without switch. 
 
 Current of electrodes + 40- Nerve current + 294. 
 
 Without With Without With 
 
 hrs. min. switch. switch. hrs. min. switch, switch. 
 
 I. 5 5.5 +10 XII. 5 37-5 +4 
 
 11. 5 7-5 {-'I XI"- 5 41 {-'I 
 
 III. 5 10 +6 XIV. 5 43-5 +3 
 
 IV. 5 14 {-'I XV. 5 44-5 {^o 
 
 V. 5 17-5 +5 XVI. 5 46 +4 
 
 VI. 5 22 {-'I XVII. 5 49 1^'^ 
 
 VII. 5 25 +5 XVm. 5 50-5 +4 
 
 VIII. 5 26.5 ? XIX. 5 52-5 { + '2 
 
 IX. 5 3-5 {-'I XX. 5 54 ? 
 
 X. 5 34-5 +4-5 XXI. 5 55-5 +4 
 
 XI. 5 36-5 {-'I XXII. 5 57 [^ V 
 
 Positive after-variation increases up to a certain limit with in- 
 creased duration of stimulation. It was already noticeable when the 
 period of stimulation was only a fraction of a second, and after a 
 stimulation of a second's duration it was at times considerable. The 
 increase dependent on duration of stimulation is only conspicuous 
 when the duration is very short, subsequently it only increases very 
 slightly with increased duration; it falls again when duration of 
 stimulation exceeds a certain limit, and finally with further increase 
 quite disappears. With a stimulation which lasted for five minutes, 
 I found in two instances after cessation of stimulation not only 
 
264 ON POSITIVE VAIIIATION OF THE NEEVE CUKKENT 
 
 no trace of positive variation, but an actual weakening* of the nerve 
 current amounting- to several divisions of the scale. In this respect 
 the positive after-effect behaves in an opposite way to the negative 
 variation. The weakening or negative variation of the nerve 
 current which takes place during stimulation, is constant with 
 moderate stimulations which may last for several minutes, and in- 
 creases rather than diminishes with the duration of the stimulation. 
 With stimulations which lasted for five minutes, I repeatedly 
 observed that there was no falling- off in the amount of negative 
 variation, but rather a moderate increase in it. This increase corre- 
 sponded only in isolated cases and then only partially, with a per- 
 sistent weakening of the nerve current, while in other cases the 
 nerve current was quite unchanged at the close of a long stimulation. 
 The law of the dependence of the positive variation on the duration 
 of stimulation could not be accurately deduced without the statistics 
 of numerous experiments ; because with repeated stimulation of a 
 nerve the diminution of the positive after-variation must present 
 itself even when the duration of stimulation remains unaltered. 
 Immediately after a long stimulation, a short stimulation with the 
 same strength of current gives no positive variation. If, however, 
 the nerve has a longer period of rest, a strong positive variation 
 again appears. 
 
 The following tables, together with those already quoted, give an 
 approximate idea of the manner in which the amount of positive 
 variation changes with the duration of stimulus on repeated stimu- 
 lation of the same nerve. The Roman numerals again indicate the 
 order of the individual stimulations, which followed each other at 
 intervals of about 2-3 minutes. 
 
 Warm frog. Du Bois induction-coil. Galvanometer tract lo mm. Inter- 
 mediate tract about 30 mm. With switch. 
 
 Current of electrodes +17. Nerve current + 1 70. 
 
 I. 
 
 Duration of stimulation 
 
 3 sec. 
 
 + 2 
 
 II. 
 
 
 5 „ 
 
 + 3 
 
 III. 
 
 
 10 „ 
 
 + 4-5 
 
 IV. 
 
 
 20 ,, 
 
 + 5 
 
 V. 
 
 
 30 „ 
 
 + 4 
 
 VI. 
 
 
 5 „ 
 
 + 0-5 
 
 Cold frog. Du Bois induction-coil. Galvanometer tract 10 mm. Intermediate 
 tract 35 mm. With switch. Pause between the several stimulations about 2 min. 
 Current of electrodes — 59. Nerve current -f- 81. 
 
 I. Duration of stimulation i sec. + 1 
 11. ,. „ 2 ., ? 
 
AFTER ELECTRICAL STIMULATION. 
 
 265 
 
 III. Duration of stimulation 2 
 
 IV. „ „ 4 
 V. „ „ 8 
 
 VI. „ „ 16 
 
 VIL „ „ 32 
 
 + 2 
 + 2.5 
 + 3 
 + 3-5 
 + 3 
 
 Galvanometer tract 9 
 
 Cold frog. Du Bois induction-coil, 
 tract about 30 mm. With switch. 
 
 Current of electrodes + 40. Nerve current + 1 34. 
 
 I. Duration of stimulation i sec. 
 
 II. 
 
 IIL 
 
 IV. 
 
 V. 
 
 VI. 
 
 VII. 
 
 VIII. 
 
 IX. 
 
 X. 
 
 XI. 
 
 XII. 
 
 XIIL 
 
 XIV. 
 
 XV. 
 
 I 
 
 3 
 
 5 
 
 10 
 
 3 
 10 
 
 20 
 
 I 
 
 5 
 10 
 
 30 
 
 5 
 60 
 10 
 
 + 4 
 
 + 3 
 + 4-5 
 + 6 
 
 + 7 
 + 4-5 
 + 6-5 
 + 6 
 + 1 
 
 + 4 
 + 5 
 + 5 
 + 2.5 
 
 + 5-5 
 + 4 
 
 mm. Intermediate 
 
 Cold frog. Stimulation with induction-coil. Galvanometer tract 9 mm. Inter- 
 mediate tract about 30 mm. With switch. 
 
 Current of electrodes +17. Nerve current + 103. 
 I. Duration of stimulation 
 
 n- 
 in. 
 
 IV. 
 
 VI. 
 VII. 
 
 Cold frog. Stimulation as with preceding. 
 8 mm. Intermediate tract 32 mm. 
 
 I 
 
 sec. 
 
 + 3 
 
 2 
 
 
 + 5-5 
 
 5 
 
 
 + 7 
 
 10 
 
 
 + 8 
 
 20 
 
 
 + 9 
 
 5 
 
 
 + 4 
 
 10 
 
 
 + 6 
 
 S^itl 
 
 I switch. < 
 
 Galvanometer tract 
 
 First nerve. 
 Current of electrodes —40. Nerve current +81. 
 
 I. 
 
 Duration of stimulation 
 
 20 sec. 
 
 + 6 
 
 II. 
 
 >> j> 
 
 30 » 
 
 + 7 
 
 IIL 
 
 >> >> 
 
 40 » 
 
 + 5 
 
 IV. 
 
 3> J> 
 
 50 » 
 
 + 4 
 
 V. 
 
 5> }> 
 
 60 ,, 
 
 + 3 
 
 Second nerve. 
 Current of electrodes —45. Nerve current + 84, 
 I. Diiration of stimulation 60 sec. + 10.5 
 II- „ „ 50 » +7 
 
 III- » „ 40 » +5 
 
266 
 
 ON POSITIVE VARIATION OF THE NEEVE CURRENT 
 
 IV. 
 
 Duration of stimulation 
 
 .^o 
 
 sec. + 4 
 
 V. 
 
 » J) 
 
 20 
 
 ,y +3 
 
 VI. 
 
 >j )' 
 
 lO 
 
 „ +2-5 
 
 VII. 
 
 )) )) 
 
 5 
 
 » +2 
 
 VIII. 
 
 5) 5) 
 
 30 
 
 „ +3-5 
 
 IX. 
 
 ;> >> 
 
 30 
 
 » +3 
 
 The circumstance that it is sufficient to tetanise the nerve for 
 a fraction of a second in order to obtain positive after-variation, 
 suggested that even a single momentary excitation might be suffi- 
 cient to produce this result to a slight extent. My attempts to 
 demonstrate this by means of the repeating method failed from 
 causes which we shall have to discuss in a further communication 
 on the negative variation. 
 
 As negative variation increases with the strength of the stimu- 
 lating current, so likewise does positive variation, but only up to a 
 certain limit. In this case again more accurate data can only be 
 arrived at from a very large number of researches. Naturally the 
 decrease in consequence of repeated stimulation is veiy evident 
 in this case, where we are dealing with increasingly stronger 
 stimulations. 
 
 In the following series of experiments the lapse of time between 
 two successive stimulations was about 3 minutes. 
 
 Cold frog. Du Bois induction coil. Galvanometer tract 9 mm. Intermediate 
 tract 34 mm. With switch. Duration of stimulation 10 sees, in each case. 
 Current of electrodes +30. Nerve current + 133. 
 
 I. Distance apart of coils 50 cm. + 4 
 
 n. 
 
 HI. 
 
 IV. 
 
 V 
 
 VI. 
 
 VII. 
 
 VIII. 
 
 IX- 3 J >> 
 
 X. „ 5, 
 
 XII. 
 
 Cold frog. Stimulation as in foregoing. Galvanometer tract 9 mm. Inter- 
 mediate tract 27 mm. Duration of stimulation 10 sees. With switch. 
 Current of electrodes + 30. Nerve current +151. 
 
 I. Distance apart of coils 50 cm. +0 
 II- ,y „ 40 » +3-5. 
 
 Ill- „ „ 30 „ -^5-5 
 
 IV. „ „ 20 „ +7 
 
 V. ,, „ ^° " 1 + 2 - f ^^^t^°"* sv^tch 
 
 40 
 
 
 + 4 
 
 30 
 
 
 + 6 
 
 20 
 
 
 + 7 
 
 15 
 
 
 ? 
 
 15 
 
 
 + 6 
 
 20 
 
 
 + 5-5 
 
 30 
 
 >J 
 
 + 4-5 
 
 40 
 
 
 + 3 
 
 50 
 
 ;> 
 
 + 1 
 
 20 
 
 
 1 + I without switch 
 
 20 
 
 1) 
 
 + 5 
 
AFTER ELECTRICAL STIMULATION. 267 
 
 In order to get more accurate data of the relationship in amount 
 between neg-ative and positive variation, it would be necessary to make 
 observations with two galvanometers at the same time, as already- 
 mentioned. With a stimulation of short duration, on account of the 
 inertia of the magnet, the deflection of the needle does not cor- 
 respond to the amount of negative variation : before it has reached 
 a point in its deflection which would correspond in measurement to 
 the decrease of the nerve current, the stimulation is already at an end. 
 On this account all values recorded and quoted above of negative 
 variation were far too small when the duration of stimulation was 
 short (less than about lo seconds). This drawback does not hold good 
 for positive variation, because in this case one may always await the 
 maximum deflection of the magnet. Yet even the value so found does 
 not give a correct representation of the amount of positive variation, 
 because this, as has been already stated, rapidly declines, and indeed 
 in all probability too quickly for the sluggish magnet to make an 
 excursion corresponding to the original amount of the variation. 
 Hence it is desirable to repeat these experiments under more favour- 
 able circumstances. 
 
 The relationship in amount of positive and negative variation 
 is best determined by experimenting with stimulations of equal 
 strength and duration alternately with and without a switch, as has 
 already been discussed. In this manner it was shown that the' 
 amount of positive variation not unfrequently reached half the 
 amount of negative variation, seldom exceeded that proportion, but 
 generally fell short of it . 
 
 The duration of positive variation under different circumstances I 
 have not accurately determined. In cases where positive variation 
 was considerable, minutes might pass before it fully died away. In 
 most cases its duration is measured by seconds. 
 
 The variation of the nerve current may assume a totally different 
 aspect if stronger induction currents be employed to tetanise the 
 nerve, for then unipolar actions mix with them, especially if the 
 break-shock is used. With what distance of coils and with what 
 strength of primary current this occurs, must be specially deter- 
 mined for each form of experiment by special control experiments. 
 
 Du Bois-Reymond has already pointed out that, even with the 
 most careful insulation of the circuit of the secondary coil on the one 
 
268 ON POSITIVE VAEIATION OF THE NERVE CURRENT 
 
 hand, and of the multiplier- or galvanometer-circuit on the other 
 hand, it is not possible to exclude unipolar action, especially with 
 break-induction currents. Unless these are extremely weak, elec- 
 tricity flows appreciably throug-h the intermediate tract into the 
 galvanometer-circuit. Now, seeing- that an electrical current excites 
 a nerve especially where it enters or leaves the excitable sub- 
 stance of that nerve, the stimulation is no longer limited to that 
 portion of the nerve to which the stimulating electrodes are applied, 
 but a second direct stimulation occurs in the neighbourhood of 
 the galvanometer electrodes. This unipolar action is greater, the 
 greater the portion of nerve between the stimulating electrodes, 
 and the less it is moistened. I therefore took care in my ex- 
 periments that with stimulations with make and break induction 
 currents, the intermediate tract should not be less than 25 mm., 
 and that the stimulated tract should not be more than 5 mm. ; 
 and although numerous test experiments showed that even with- 
 in these limits, when the coils were more than 13 cm. apart, 
 unipolar action never took place, so long as the primary coil was 
 supplied by i Daniell only, yet I nearly always stimulated with 
 the coils at a distance of 20 cm. apart. If, as an exception, the 
 coils were brought nearer together, a test experiment was always 
 made at the end by crushing the nerve in the intermediate tract. 
 When the stimulated tract was smaller and well-moistened, a much 
 nearer approach of the two coils was possible without noticeable 
 unipolar action. If the end of the nerve is not introduced into the 
 galvanometer-circuit, but is in connection with the well-isolated 
 muscle, and the above-mentioned conditions are fulfilled, there is 
 not the least occasion to fear unipolar action. 
 
 The alterations in the course of the variation of the nerve current, 
 which are brought about in consequence of the application of stronger 
 currents and of the concomitant intermixture of unipolar action, as 
 far as I have yet observed, are in substance the following: — 
 
 The positive after- variation is weaker, and when unipolar action 
 becomes to a certain degree considerable, is quite lost. If unipolar 
 action still increases, the magnet falls further short of its original 
 deflection ; or if it has been compensated, does not return to the zero 
 point after cessation of stimulation. Hence the weakening of 
 nerve current, which du Bois-Reymond regarded as the rule. 
 By degrees, however, this weakening passes away, occupying 
 several minutes in so doing if considerable, but only a few seconds 
 if very small. Again, if excitations are repeated at short intervals 
 
AFTER ELECTRICAL STIMULATION. 269 
 
 and with the same strength and duration of current, negative varia- 
 tion is seen to be less at each successive stimulation. The subse- 
 quent weakening of the nerve current is at the same time greater, 
 and the impression is given of increasing exhaustion of the nerve. 
 If the strength of current be increased by degrees, the negative 
 variation still increases, as does at the same time the subsequent 
 weakening of the nerve current. During stimulation with very- 
 strong currents, the nerve current disappears altogether, and it 
 returns only partially and very slowly after stimulation has ceased. 
 Finally, this may proceed to such an extent that the nerve current 
 may reverse its direction during stimulation. After ceasing to 
 stimulate, this reversed current gradually dies away ; if stimulation 
 is again repeated, the reversed current may become still stronger, and 
 after ceasing to stimulate may continue for a period in the negative 
 direction. At other times a lasting disappearance of nerve current 
 ensues^ the nerve appearing to lose permanently its electromotive force. 
 But these last-named deep-seated alterations in the electrical 
 condition of the nerve are solely a consequence of change in the 
 neighbourhood of the galvanometer electrodes, especially of the upper 
 one. If the altered portion of the nerve be cut away and the elec- 
 trodes be connected, one with the new section, the other with a point 
 proportionately higher up the length of the nerve, a strong nerve 
 current is again obtained. If now the other end of the nerve be 
 stimulated afresh, the whole series of changes will be repeated. 
 
 It cannot therefore be supposed that this weakening or total 
 disappearance of the nerve current depends on a corresponding 
 decrease or total annihilation of electromotive power of the whole 
 nerve ; we are not dealing here with ' complete exhaustion of 
 animal excitability ^.' A nerve is a relatively durable structure, 
 and it is astonishing how long it holds out even against strong 
 stimuli. It is only locally that it is permanently altered by strong 
 currents. The portion which is not directly stimulated, although 
 it may be in a condition of strong propagated excitation, suffers 
 much less even when stimulated for a long period. 
 
 The ease with which a nerve is exhausted has been greatly over- 
 rated. If unipolar action be excluded, the nerve can be repeatedly 
 stimulated, or stimulated for a long period, without any appreciable 
 weakening of the nerve current or its negative variation, provided 
 that the stimulating currents be not too strong. 
 
 ^ Compare on this point N. Wedenskii's recent communication, ' Centralblatt fur 
 die medicinische Wissenschaft ' of Feb. 2, 1884. 
 
270 ON POSITIVE VAKIATION OF THE NERVE CURRENT 
 
 False estimates may easily be formed of the influence of the 
 streng-th of the stimulating" current on the amount of neg-ative 
 variation, unless unipolar action be strictly excluded. Neg-ative 
 variation amounts only to a very small fraction of the current from 
 transverse to longitudinal section so long* as it depends solely on 
 propag-ated excitation, and under these circumstances I have seldom 
 seen it amount to a fourth part of that current. Very weak stimu- 
 lations must be employed at the onset, unless it is desirable to g-et 
 at once almost maximal action. That increase of the streng-th of 
 the stimulating" current beyond a certain very low limit has scarcely 
 any further influence on the amount of neg-ative variation, can 
 naturally only be observed with methods of stimidation which exclude 
 all unipolar action. For as soon as stimulation is started with the 
 ordinary induction apparatus, neg-ative variation can be made so 
 great with moderate currents, that the whole nerve current tempo- 
 rarily disappears, or even chang-es its direction. 
 
 In his researches on negative variation, du Bois-Reymond did 
 not exclude the unipolar actions which are occasioned by the escape 
 of electricity into the circuit of the galvanometer. After he had 
 convinced himself that, when tetanising a nerve with induction- 
 currents, the electricity which flowed over and produced unipolar 
 action did not deflect the needle, he took measures to exclude that 
 action in special cases only. His general purpose was to show that 
 the electromotive force of the nerve decreased durino- excitation. 
 Whether this excitation was produced exclusively by stimulation 
 of the tract between the stimulating electrodes, or whether it was 
 produced at the same time by stimulation of the tract through which 
 electricity was discharging itself unipolarly, appeared to him in 
 general of no consequence, so long as the electricity led into the 
 nerve did not influence the deflection of the needle. 
 
 In order to prove that in du Bois-Reymond's researches he did 
 not as a rule exclude unipolar action when tetanising with induction- 
 currents, I will quote a few passages from his ' Researches,' remark- 
 ing that the words in italics are not so in the original text. 
 
 In vol. i. p. 436 ^, du Bois-Beymond says : — 
 
 ' The question is whether the electricity which with this method 
 of experiment continually spreads through the nerve, considering that 
 the apparatus is not completely isolated, may not have an action 
 on the needle of the multiplier. That in this case, as iti that pro- 
 duced hy unipolar induction-contractions, electricity continually flows 
 
 ^ Untersuchungen uber thierische Elektricitiit. 
 
AFTER ELECTRICAL STIMULATION. 371 
 
 out, is not to he donhted,for the contractions of the animaVs limls can 
 he only very incompletely arrested by ligature of the nerve in the middle 
 of the intermediate tract ; for this ligature to he efficient it must he 
 applied immediateli/ above the entratice of the nerve into the muscle. 
 Witli reg-ard to the action of this outflowing of tension electricity 
 on the needle of the multiplier we need have no anxiety.' 
 
 It only need be added to this remark, that even with the best pos- 
 sible insulation, unipolar action is not excluded under the above circum- 
 stances, if only the conditions already mentioned are favourable to it. 
 
 In vol. ii. p. 51, du Bois-Reymond shows proof that the negative 
 variation of the muscle-current cannot depend on the invasion of the 
 circuit of the multiplier by inductive electricity. For this purpose 
 it is only necessary to ligature the nerve of the gastrocnemius 
 between the stimulated tract and the galvanometer tract, in order 
 to see the needle remain motionless in its place, the muscle also 
 remaining at rest. ' Care must be taken ' in using an induction 
 apparatus of any kind for tetanising, as du Bois-Reymond observes, 
 ' that the ligature be placed close to the point where the nerve 
 enters the muscle. The ground for this rule lies in unipolar 
 induction-contractions. And if it he not adhered to, the inuscle does 
 not remain at rest, as above described, in spite of the most careful and 
 seciire ligature of the nerve ; and simultaneously with the slight 
 tetanus which is still produced, there occurs a correspondingly small 
 action on the needle.' 
 
 Du Bois-Reymond expresses himself in a precisely similar manner 
 when dealing with the negative variation of the nerve current. 
 
 ' Ligature and section,' he says, ' here also prevent all trace of 
 movement of the needle. In these experiments one must not use the 
 induction apparatus for tetanising, on account of unipolar actiotis, hut 
 must employ instead a battery, the current of which is broken and 
 reversed at the same time by means of a Poggendorff " inversor." ' 
 
 In the section which treats of the influence of various circum- 
 stances on the amount of negative variation when tetanising with 
 reversed currents, du Bois-Reymond says (vol. ii. p. 449), ' Further, 
 it is not unfrequently necessary to exchange the induction apparatus 
 for Poggendorfl"s inversor. This is the case in the first place when 
 the amount of negative variation in tetanisation is to be compared 
 with the amount of the negative phase of the electrotonic state, 
 which is itself produced by a current of equal strength with the 
 
 tetanising current In the second place, this necessity arises 
 
 when one is apprehensive of disturbances of the result consequent 
 
272 ON POSITIVE VARIATION OF THE NERVE CURRENT 
 
 on the actions which occasion unipolar induction contractions. In 
 other words, unless the whole secondary circuit, the apparatus for 
 leading the current to the nerve, the leading- off electrodes and the 
 multiplier are completely insulated (a state of things not easy to 
 secure as may he ivell understood^ there is a continual current, now 
 negative, now positive, according to the direction of the tetanising 
 current flowing through the tract between the platinum plates 
 and the pads, and through the latter and the remainder of the 
 apparatus to the ground.' ' On the needle, indeed, this electricity 
 has no action, but it does not fail slightly to tetanise the nerve, 
 consequently it might show an illusory activity where there should 
 have been none, or an exaggeration of actually existing activity. 
 Hence, for example, it appears that ligature or sectioii of the nerve in 
 the tract between platinum plates and pads does not completely arrest 
 negative variation when induction currents are used for tetanising^ 
 
 Du Bois-Reymond discussing, two pages farther on, the in- 
 fluence of the strength of the stimulating current on the amount of 
 negative variation, says expressly, ' There is no drawhack to the use 
 of the induction apparatus in this case ' (that is to say, in comparing 
 the actions of two currents of very different strength). 
 
 A few lines farther he remarks, ' In order to show that here like- 
 wise all depends on the density of the current and not directly on 
 its strength, one can go to work in the manner above described 
 for the electrotonic condition. The source of the stimulating current 
 is left unaltered, but a collateral closure is established between the 
 electrodes ; such as a pad of blotting-paper folded several times and 
 steeped in white of egg. It is however advisable in this case to 
 exchange the induction coil for an inversor, because the lessening of 
 density no longer affects unipolar action, as it does when the source of 
 the current itself is weakened! In like manner du Bois-Reymond 
 rejects the use of the induction coil when dealing with the influence 
 of the length of the stimulated tract, or of the influence of the 
 distance apart of the stimulated and led off tracts. 
 
 The above is sufficient to show that in general it appeared a matter 
 of indifference to du Bois-Reymond, whether the nerve were stimu- 
 lated by induction currents exclusively in the tract between the 
 stimulating electrodes, or whether it were stimulated in the rest of 
 its extent at the same time. Only in particular cases, when it 
 appeared necessary to limit the stimulation to the tract between 
 the electrodes, did he exclude unipolar action by tetanising with 
 batterj'^-eurrents in alternately opposite directions. 
 
AFTER EI-ECTRICAL STIMULATION. 273 
 
 It must be remembered that du Bois-Reymond was then of 
 opinion that the electrical current excited the nerve in an equal 
 manner throug-hout the whole of the tract through which it flowed. 
 Since however Pfliig-er's orig-inal researches on Electrotonus, we know 
 that the current has polar actions which are quite different at the 
 anode from what they are at the kathode. And when du Bois-Rey- 
 mond, in his researches on secondary electromotive phenomena in 
 nerve and muscle, still maintains an equal internal (positive and 
 negative) polarisation for the whole of the intrapolar tract, he is 
 in my opinion in error. The so-called secondary electromotive 
 phenomena are the consequence of polar actions, and like the 
 excitability exhibit a polar difference. Thus it happens that elec- 
 tricity flowing" unipolarly throug-h the g-alvanometer electrodes into 
 the circuit of the multiplier, appreciably alters the nerve in their 
 neighbourhood, and there completely chang-es its electromotive 
 behaviour ; hence the alterations in the nerve current which occur 
 during- and after stimulation are in no wise solely the result of the 
 general alteration of the nerve. 
 
 Where du Bois-Reymond discusses the influence of the strength 
 of the stimulating current on the amount of negative variation, he 
 says (vol, ii. p. 451) expressly : ' In this case the use of the induction- 
 coil has no disadvantages.' And yet it is just here, where strong 
 stimulating currents have to be dealt with, that unipolar action is 
 most to be feared. 
 
 Thus it is sufficiently evident that du Bois-Reymond's state- 
 ments differ in many respects from what we learn from the investi- 
 gation of negative variation when unipolar actions are excluded. 
 I will not here examine how far the above described behaviour of 
 the nerve current on stimulation with strong induction currents, 
 depends on unipolar action and thus on direct stimulation and 
 alteration of the galvanometer tract, or how far it depends only on 
 an increase of excitation propagated from the stimulated to the 
 galvanometer tract. I would only add this, that with vigorous 
 nerves and with methods of excitation which exclude unipolar 
 actions, I never succeeded in observing such behaviour of the nerve 
 current during stimulation as was shown when strong induction 
 currents were used and unipolar actions in consequence obviously 
 came into play. 
 
 I will not enter on a theoretical discussion of the positive after- 
 variation until I have described some other electrical phenomena in 
 stimulated nerves. 
 
 T 
 
IX. 
 
 THE SO-CALLED 
 
 SECONDAEY ELECTEOMOTIYE PHENOMENA 
 OF MUSCLE AND NERYE. 
 
 By L. HERMANN. 
 
 Pfliigev's Archiv f. Physiologie, vol. xxxiii. p. 103. 
 
 T 2 
 
CONTENTS, 
 
 PAGE 
 
 1. Facts observed by du Bois-Reymond 277 
 
 2. The Theoretical and Controversial Observations of du Bois-Eeymond . 278 
 
 3. Investigation of the intrapolar after-currents of Muscles and Nerves . 284 
 
 4. Investigation of extrapolar after-currents in Nerves and Muscles . . 304 
 
 5. Explanation of the intrapolar and extrapolar after-currents of Nerves and 
 
 Muscles, together with further experiments 310 
 
 6. Concluding Remarks 325 
 
IX. 
 
 In a recently published treatise ^, E. du Bois-Reymond discusses 
 the so-called secondary electromotive phenomena ^ as observed by 
 him in muscle, nerve, and electromotive tissues, in order to give fresh 
 support to his molecular hypothesis and his theory of electrotonus, 
 against both of which, to my mind, fatal objections have been 
 raised. 
 
 How important the author regards this support is seen in the 
 opening sentences, ' I think it is now time to break the silence 
 which I have hitherto maintained regarding certain experiments 
 on the electromotive phenomena of tissues, with which I have been 
 occupied nearly forty years, and to which I attach great import- 
 ance,' as well as in the repeated expressions according to which all 
 my ideas upon electrotonus had now been exploded, so that the 
 elucidation of the subject would have to be begun anew^. 
 
 In consideration of the heavy artillery brought against me, it 
 will not be taken amiss if, without waiting for the further communi- 
 cation upon the same subject announced by the author, I go over 
 the facts of the new position. It will then be seen to what extent 
 the expectations and expressions of du Bois-Reymond are justified. 
 
 1. Facts otoserved by du Bois-Reymond. 
 
 The facts, in part new, in part previously stated, which du Bois- 
 Reymond adduces, are briefly as follows : — 
 
 • ' Sitzungsberichte der Konigl. Preuss.' Acad, der Wissenschaften, 1883, p. 404. 
 
 * Du Bois-Reymond describes as secondary electromotive such electrical pheno- 
 mena as are produced by leading a current into an animal tissue, but are not 
 detected in the tract through which the current has passed until after this is broken. 
 He places the electromotive phenomena of animal tissues under three categories : 
 (1) Currents inherent to the tissue at rest or in action ; (2) Extrapolar effects of led- 
 through currents during their passage ; (3) Intrapolar effects of the same, i.e. the 
 secondary electromotive phenomena. In this enumeration are omitted the extra- 
 polar effects following the break of a current, the laws of which I first stated with 
 regard to nerve, as du Bois-Reymond himself has mentioned. Concerning these 
 phenomena in nerve and muscle see later. 
 
 3 See No. VI of this Series, Sect. 15. 
 
378 THE SO-CALLED SECONDARY ELECTROMOTIVE. 
 
 If a current is allowed to flow lengthwise throuo-li a tract of 
 muscle, nerve, or electrical organ, and immediately this is broken 
 a portion of the tract is connected with a galvanometer, ordinary 
 internal polarisation — i. e. a current opposed to the one led through 
 — does not always manifest itself. 
 
 Thus, if the current led into the tissue is very strong (for the 
 thigh muscles at least that of 2 Groves, 5 Groves for the sciatic 
 nei've) and the duration of its closure is very short, an effect 
 appears which is in the same direction instead of in the opposite 
 direction ; or a double effect may show itself, which is first in the 
 opposite and then in the same direction^. 
 
 This after-effect in the same direction as the polarising current is 
 connected more intimately with the living condition of the tissue 
 than the one opposed to it in direction ; moreover, its production 
 is more or less clearly favoured by a direction of current which 
 coincides with that of the natviral excitation (in muscle and nerve) 
 or of the shock (in electrical organs). 
 
 2. The Theoretical and Controversial Observations of 
 du Bois-Reymond. 
 
 Du Bois-Reymond denotes as 'negative polarisation' the after- 
 effect found by Peltier in a led-through tract, and which is present 
 even with weak polarising currents, especially if these are of 
 long duration. There is nothing to be said against this nomen- 
 clature, except that the word ' negative ' is superfluous. That the 
 phenomenon depends upon true galvanic polarisation undoubtedly 
 follows from its whole character, from its direction as compared with 
 that of the current which produces it, its rapid disappearance after 
 the break of this current, the relationship of its magnitude to that 
 of the current and to the duration of closure of the latter, and further 
 the relationship of its amount to the position of the laminae of hetero- 
 geneous substances in the tissues ^ ; a fact discovered by myself, 
 which du Bois-Reymond has omitted to refer to, although such 
 an indication is of much more value than pure speculations 
 
 ' [For convenience of description, the after-effect in the same direction as the polar- 
 ising current, termed by our author ' gleichsinnig,' is denoted in this translation by 
 the sign + , the opposed effect, termed ' gegensinnig,' by the sign — , the signs merely 
 indicating the relation of the direction of the after-effect to that of the polarising 
 current. Tr.] 
 
 ^ See Pfliiger's ' Archiv,' vol. v. p. 232. 
 
PHENOMENA OF MUSCLE AND NERVE. 379 
 
 as to the cause of internal polarisation (du Bois-Reymond, No. VI, 
 Sect. 2i). 
 
 That galvanic polarisation is opposed to the polarising* current, 
 and thus must be, as du Bois-Reymond says, negative, is as much a 
 physical necessity as the principle of the conservation of energy, with 
 which in point of fact the polarisation law is in the most intimate 
 relationship; on this account the prefix ' negative ' might be done away 
 with. But it was necessary for du Bois-Reymond, since he had arrived 
 at the surprising idea of calling the + effect, observed by him, 
 'positive polarisation.' I do not find fault with the confusion 
 which is thus produced, on the ground that for twelve years I 
 have distinguished as positive and negative polarisation the polar- 
 isation of nerve at the positive and negative electrode ; for I am 
 not so presumptuous as to suppose that du Bois-Reymond has any 
 regard for the technical terms introduced by myself, the discovery of 
 which, according to him, my entire work consists of (see Sect. 2i). 
 Nor can I ground my objection on the circumstance that du Bois- 
 Reymond once used these expressions with reference to the + and 
 — electrodes, as Pfliiger indicates in a quotation ^, as 1 have re- 
 marked elsewhere ^, for I cannot find the passage. It is not on this 
 account that the term 'positive polarisation' is a surprising one, but 
 because it represents the phenomenon as a reversal of ordinary 
 polarisation ; as if the galvanic current could produce in its circuit, 
 instead of the customary opposing forces, forces acting in the same 
 direction as itself, and this on a strictly comparable principle. 
 
 That du Bois-Reymond is not far from holding this very opinion, 
 that he really holds, on the ground of his experiments, a ' reversed 
 polarisation ' as possible, follows from his whole treatise. This is 
 evident from the circumstance that he places this side by side with 
 the polarisation, supposed by him to be also reversed, which occurs 
 at the limiting surfaces between salt solution and water (du Bois- 
 Reymond, Sect. 4), or iron and zinc sulphate (Sect, ai, par. 9), quite 
 in accordance with the scheme of true polarisation. 
 
 But if du Bois-Reymond had had the slightest doubt as to this 
 relationship, it would have been his duty, even on the assumption 
 that he was writing entirely for readers schooled in physics, to avoid 
 most carefully this expression. If, on approaching a magnet to a 
 conductor, it were observed that a current appeared in the latter 
 which must from its direction affect the magnet pole attractively, 
 
 ^ ' Untersucliungeii liber die Physiologic des Elektrotonos,' p. 438. 
 ^ See Pfliiger's ' Archiv,' vol. xxxi. p. 100. 
 
280 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 it would certainly not be inferred with truth that under certain 
 circumstances a reversed induction occurs in addition to the ordinary- 
 one, but it would be rather said that there must have been some 
 error, or that the cause of the phenomenon was something wholly 
 unconnected with induction. In like manner, if in such compli- 
 cated structures as those of muscle and nerve, and with currents 
 of from 20-50 Groves', streng-ths of current repudiated for other 
 purposes, an after-effect similar in direction to the current were 
 found to exist, no one would think of describing* this effect as re- 
 versed polarisation. On the contrary, it would be said either there 
 is some source of error, or the phenomenon has nothing to do 
 with polarisation. Should the physics of the future indeed bring 
 to light a reversed (' positive ') polarisation, it will certainly not be 
 by means of experiments with such damaging strengths of current, 
 and made upon the most perishable objects in the world. The 
 cases of iron and zinc sulphate &c. before referred to will, on 
 closer investigation, undoubtedly show themselves to be based upon 
 something quite different from reversal of the law of polarisation. 
 
 The author reproaches me (Sect. 15) that I have not discovered 
 the fundamental fact of the + intrapolar after-effect (the posi- 
 tive polarisation of du Bois-Reymond) in a realm in which I have 
 ' entered as a reformer ;' an entirely new method, by the way, of 
 making an attack upon irrelevant grounds. I might just as well 
 reproach him with not having discovered the extrapolar after-effect, 
 the law of the action current, or that of the secretion current, &c., 
 and particularly certain phenomena of this, his own special subject, 
 which will be brought to light in what follows. And how much 
 more indeed might not be discovered if some one would make up his 
 mind to go at muscle and nerve, not with the 20-50 Groves, which 
 we have hitherto used with apprehension, but with hundreds or 
 thousands of elements ! 
 
 The most noteworthy point in the whole treatise is the manner 
 in which the new fact of the + intrapolar after-effect is made 
 use of. If it is true that my theory of electrotonus ' so urgently 
 needed ' for its support the electrotonus currents in muscle dis- 
 covered by myself (du Bois-Reymond, Sect, i),^ it is still more 
 evident that ' positive polarisation ' is indispensable for du Bois- 
 
 ' I ask myself in vain on what grounds du Bois-Reymond's theory of electrotonus 
 less urgently needs an electrotonus of muscle than my own. The molecular scheme 
 extends to both muscle and nerve, and what is correct for the molecules in nerve, 
 vk. their rotation by the current, must also be right for the molecules of muscle. 
 
PHENOMENA OF MUSCLE AND NERVE, 281 
 
 Reymond's molecular theory of electrotonus. It is indeed nothing' 
 more or less than the clemonstratio ad oculos, that the current produces 
 corresponding rotation of the electromotive molecules in the a^iimal 
 tissues (Sect. 2t) ! 
 
 ' Repeatedly,' says du Bois-Reymond (p. 223), ' and on two quite 
 distinct hypothetical bases, Hermann has demonstrated that in the 
 intrapolar tract a polarisation -current opposed in direction to the po- 
 larising current must prevail. I am curious to see by what auxiliary 
 hypotheses he will deal with the polarisation current in the same 
 direction which actually prevails.' That some of his most zealous 
 molecular adherents, e. g. Bernstein ^ and v. Fleischl ^, have followed 
 me and have been heretical enough to confirm the fact of an effect in 
 the intrapolar region opposed to the current, in contradiction to 
 himself, is a subject upon which he is silent, and he is not curious 
 as to the auxiliary hypotheses of these authors, a circumstance 
 which I regard as most flattering to myself. 
 
 But what is the ' positive polarisation current actually prevailing 
 there ' which my theory and my auxiliary hypotheses will have to 
 satisfy ? Since when is the proof of an after-current on opening, a 
 ground for the assumption that the same current existed durhig 
 closure ? ^ Would physicists have ventured to deduce ordinary 
 polarisation from the polarisation after-current alone, had not the 
 steady decrease of the current led through, furnished a proof of the 
 existence of the former during closure ? And what would du Bois- 
 Reymond say if some one took it into his head to conclude from 
 the fact proved by myself and confirmed by Fick, that the anodal 
 extrapolar region of a nerve shows on opening, an after-current 
 opposed in direction to the polarising current, that electrotonus has 
 also, during closure this direction on the anodal side ? 
 
 Du Bois-Reymond observes in the intrapolar region two conflict- 
 ing after-effects, one opposed to, the other in the same direction 
 
 * Zioc. cit. vol. viii. p. 51. 
 
 ^ ' Sitzungsber. d. Wiener Acad. Part 3, vol. Ixxvii. 
 
 ' This absolutely incomprehensible error of transferring, without question, condi- 
 tions which have been observed after opening, to the time of closure, plays the chief 
 part in an objection raised by du Bois-Reymond against an experiment of mine 
 (p. 200). Fifteen years ago I experimented to see if the intrapolar tract of nerve 
 showed the large increase of current which the du Bois theory requires, and I found 
 no trace of it. Du Bois-Eeymond now thinks that this arose from my long time of 
 closure, in consequence of which the polarisation must have been negative ; with 
 shorter times of closure positive polarisation would have given me the sought-for 
 result. In this suggestion we have an instance of the very transference which is so 
 unreliable. 
 
282 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 with the current led throng-h, the former more transitory, the latter 
 more persistent, and, above all, observable only after short closures 
 of very strong* currents. He concludes, without any justification, 
 that both effects are also present during closure. Now, as 
 two simultaneous ordinary polarisations, a true (' negative ') and a 
 reversed (' positive '), are not conceivable, he assumes that the first 
 only is due to true internal electrolytic polarisation, and that the 
 second is due to the 'set' of the electromotive molecules in the 
 direction of the current. He ' does not believe that any one schooled 
 in physics will come to any other conclusion ' (du Bois-E-eymond, 
 p. 222). And this and no other is the ' positive polarisation cur- 
 rent which actually prevails,' in contrast with which I had opposed 
 a negative ; it was this which was to bring me to despair ! 
 
 As to the question how this rotation of molecules is to be effected 
 by the current, unfortunately nothing is learnt even from the new 
 treatise. With respect to this process, which belongs to the 
 physics of the future, du Bois-Eeymond is still unable to set up any 
 theory. He is therefore obliged to have recourse to pulling the 
 strings of the model of rotating molecules for the satisfaction of his 
 credulous auditory. My request, expressed years ago, that the pro- 
 pounders of the molecular theory would once for all set forth an 
 exact theory of their molecules, remains therefore unfulfilled ^. It 
 is characteristic that the moment an attempt is made to apply the 
 ordinary physical modes of thought to the subject of molecules, one 
 is repelled on the ground that the physics of the future have first to 
 teach the necessary mode of thought^. My objection to the credi- 
 bility of du Bois' original view according to Grothuss' theory, du 
 Bois-Reymond appears to have acknowledged as not unfounded, 
 but he disposes of the subject with half-intelligible evasions and 
 with a facility which is not reassuring. He is not willing to 
 admit electrodynamical explanations, and rightly so, as I believe. 
 But in that case what becomes of the theory ? 
 
 I can only repeat what I said ten years ago, of which I am not 
 able to retract . a word ". * If the molecular schema is used for the 
 purpose of referring the electromotive activity of the nerve 
 cylinder to that of small particles, and definite physical actions 
 are treated of as affecting these particles, we are compelled to de- 
 mand that they should receive a more definite form which allows 
 
 ^ Pfliiger's ' Archiv,' vol. viii. p. 267 ; vol. xx. p. 393. 
 * hoc. cit. vol. viii. p. 267. 
 ^ Loc. cit. vol. viii. p. 267. 
 
PHENOMENA OF MUSCLE AND NERVE. 283 
 
 of physical discussion. The original molecular theory of du Bois is 
 quite an allowable assumption in so far as it explains the phe- 
 nomena of the current of rest, (whether it is necessary is another 
 question). The second assumption of du Bois-Reymond's, that the 
 force of the molecules decreases during- the condition of activity, 
 is also a permissible one, for as physiological conditions produce 
 the force of these molecules, other physiological conditions may be 
 able to diminish it. But with his theory of electrotonus du Bois- 
 Keymond entered another field ; here, from the existence of the 
 molecules, he drew conclusions as to the influence exercised upon 
 them by physical processes which admit of strict definition. Im- 
 mediately the question arises as to the special physical charac- 
 teristics of these molecules, and it is not possible to agree with 
 Bernstein in characterising as ridiculous either the experiments in 
 which such physical characteristics are assumed or the objections 
 founded upon them. If a molecule is in truth such an unapproach- 
 able thing that it is not possible to think about it in a definite 
 physical way, then no assertion is permissible as to the effect of 
 electrical currents upon it ^.' 
 
 The simple theory of electrotonus I have advanced, a theory 
 founded upon proved facts, du Bois-Reymond is naturally unwilling 
 to admit, at all events for the present, for in ten years he has not 
 found time to make any investigations with regard to it (du Bois- 
 Reymond, p. aoi). His own position has changed in a most 
 notable fashion. Formerly he assumed as a pure hypothesis the 
 intrapolar rotation of molecules in the direction of the current, in 
 order to be able to deduce from it the extrapolar effects. Now he 
 believes that he has directly proved the intrapolar rotation (with 
 what amount of truth has been already discussed and will be clearly 
 explained in what follows), that he may be able to give up extrapolar 
 electrotonus. Thus he says (p. 30 1), ' It is certain that even if the 
 extrapolar electrotonic currents depend merely on diffusion of current, 
 the nature of electrotonus as consisting in positive polarisation is 
 not thereby affected. We should only be rid of the doubtful neces- 
 sity of explaining its spread beyond the electrodes.' The bank- 
 mptcy of the molecular hypothesis could not he more distinctly 
 declared. The intrapolar molecular change was originally invented 
 merely to explain the extrapolar effects, and on its proving inade- 
 
 ^ Further remarks upon the value of the molecular hypothesis may be found in 
 Pfliiger's 'Archiv,' vol. xxvi. p. 485 ; to this passage I refer as being more accessible 
 to most readers. 
 
284 THE SO-CALLED SECONDARY ELECTKOMOTIYE 
 
 quate to accomplish this, became the subject of a new and unreliable 
 assumption. 
 
 It was coolly explained as constituting- the ' essence of electro- 
 tonus,' notwithstanding that I have repeatedly drawn attention to 
 the fact that it was incapable of explaining either the contrast 
 between an- and catelectrotonus or the indifference point. But the 
 actually proved polarisation of the nerve-core, which does admirably 
 explain this contrast as well as the extrapolar spread, is not the 
 essence of electrotonus ! 
 
 Excepting possibly in the case of the fanciful speculations upon 
 the shock of the electrical fish which I have recently treated of ^ 
 the abyss of misconception of nature by which the molecular theory 
 is confronted has never been so appallingly displayed. Others 
 may hold as loyal the sacrificmm intellectus which abstains from 
 criticising an hypothesis which claims to be sacred, I place it to 
 my credit that I have brought back into the physiology of this 
 department the principles which in natural science are paramount, 
 strict thought and the rejection of error ; and it is nothing to me 
 that du Bois-Reymond, who to-day still idly talks of currents of 
 rest of uninjured muscles, can only describe the result of my 
 sixteen years' work (p. 402) as a ' polemic, which has brought to 
 light more new terms than facts.' 
 
 Those however who in text-books, lectures, and popular writings 
 blindly go on spreading statements and hypotheses which have 
 been refuted, will by so doing place themselves in a position of 
 which one day they will be ashamed. 
 
 3. Investigation of the intrapolar after- currents of Muscles 
 
 and Nerves. 
 
 We now enter upon the task of investigating the -H after-effect 
 discovered by du Bois-Reymond, in a way which may guide us to 
 its elucidation. I repeated du Bois-Reymond's chief experiment so 
 as to obtain an idea of the j)henomenon in question ; it then 
 appeared desirable to modify in one respect his method. He 
 placed the leading-in electrodes on one lateral surface of the muscle, 
 the leading-off ones on the other. Under certain assumptions a 
 
 * Loc. cit. vol. xxvi. p. 483. 
 
PHENOMENA OF MUSCLE AND NERVE. 285 
 
 + after-effect mig-ht be caused by ordinary polarisation with this 
 arrang-ement, supposing-, for instance, enveloping layers of badly- 
 conducting- polarised portions of tissue to be present. Should the 
 + after-current also appear by leading'-off on the same surface 
 with the leading--in electrodes, then this possibility is removed. 
 As a matter of fact this is the case : no marked difference appears 
 between the results obtained with contacts on the same and on 
 opposite surfaces. 
 
 The simplest arrangements sufficed to show the phenomenon. 
 A Pohl's commutator without cross-wires served to close the 
 galvanometer-circuit immediately after the battery-circuit was 
 opened, the commutator being coupled up by i , 3 with the battery, 
 by 4, 5 with the galvanometer-circuit. In my 
 former experiments upon extrapolar after-currents ^ 
 this method was not satisfactory, owing to insuffi- 
 ciency of insulation; but the effects of insufficient 
 insulation are much less marked in the case of intra- 
 polar than in that of xtrapolar after-currents. Sub- 
 sequently the wooden reverser was replaced by a paraffin one. 
 The switching from battery to galvanometer-circuits was effected 
 as quickly as possible by hand, in the case of long closures by a 
 clock. 
 
 In muscle, after a very short closure, the after-current appeared 
 to be mostly diphasic^ ; a quick, sometimes momentary — being 
 followed by a long-enduring + deflection which mostly only 
 
 ^ ' Untersuchungen zur Physiologie der Muskeln und Nerven,' Part 3, Berlin, 1868 
 p. 71. 
 
 ^ It seems to me that du Bois-Reymond has seen this diphasic {doppelsinnig) effect 
 more seldom than I have, and the cause probably lies in the extreme lightness of the 
 magnet system of my galvanometer ; it weighs only 0-9 gr. (see ' Pfliig. Archiv ' 
 vol. xxi. p. 436). Besides this, very effectual damping gives aperiodicity with a 
 relatively slight amount of astasia : it is thus conceivable that a heavier or more 
 astatic system may not show the very small negative jerk of which my instrument 
 gives plain indications. 
 
 It may also be noticed here that with rapidly vanishing polarisation-currents the 
 velocity of the movement of the magnet must largely influence the amount of the 
 deflection, since the current itself becomes smaller during the deflection. This 
 influence, as to which I have many times convinced myself, is rendered especially 
 predominant through aperiodicity. On this ground, as I have often seen, the deflec- 
 tions caused by rapidly decreasing polarisation-currents are not diminished by with- 
 drawal of the bobbin to the same extent as in the case of constant currents ; the 
 shorter time which the diminished deflection requires enables the maximum of deflec- 
 tion to coincide with a greater strength of current than would be the case if the 
 instrument were rendered more sensitive. 
 
286 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 incompletely disappeared. I can entirely verify the fact that the 
 appearance of the + after-current, and of this only, is favoured by 
 short times of closure and by strong- currents ; that, further, the 
 phenomenon is intimately connected with the \atality and, as 
 I am able to show, with the excitability of the animal-tissue. 
 By fatigue, even when produced by an ordinary tetanus, + after- 
 currents become diphasic, and diphasic ones become entirely — . 
 On the other hand, the — after-current, especially in a nerve, is 
 a phenomenon which, even in summer, can be demonstrated for 
 from 24 to 48 hours, feebly indeed, but quite plainly ; whilst the 
 + after-current appears always with fresh preparations, and these 
 alone. As boiling-water, according to du Bois-Reymond, con- 
 firmed by myself, puts an end also to the — after-current, possibly 
 the most correct expression of the facts would be this, t/ie — after- 
 current is associated icith the maintenance of structure, the -\- with 
 that of life. Electrotonus has a certain resemblance to the — after- 
 effect, inasmuch as it often shows plainly traces of its presence in 
 exhausted and long-kept nerves. 
 
 If then the + after-effect is thus intimately connected with 
 excitability, the first question which arises is, whether it may 
 not possibly be itself an excitation phenomenon. It is incon- 
 ceivable to me how du Bois-Reymond could have overlooked this, 
 and so given himself the trouble of seeking an explanation in the 
 physics of the future. 
 
 My first thought on reading- du Bois-Reymond's communication 
 M'as that this + after-effect was the current of action proceeding- 
 from a prolonged break-excitation. It is well known, as Pfliiger 
 discovered, that both in nerve and muscle the break of a strong- 
 current is followed in the anelectrotonic tract by a violent and pro- 
 longed excitation. We may fairly assume that this excitation is 
 strongest and most prolonged at the anode itself, and decreases 
 towards the indifference point. According- to the laws of action- 
 currents which I have advanced, a point near the anode must, after 
 break, be negative to one further off, and thus an intrapolar after-cur- 
 rent must be produced having- the same direction as the polarising- 
 current ; as, according to Pfliiger, the indifference-point when 
 strong- currents are used lies close to the kathode (at least in 
 nerve), this condition must spread throughout the entire intrapolar 
 region. 
 
 This current is so directly necessary, and without being con- 
 nected with anv theory is in itself so undoubted, that du Bois- 
 
PHENOMENA OF MUSCLE AND NERVE. 287 
 
 Reymond must have thought of it, and he was then bound to 
 investigate to what extent it was connected with the phenomenon 
 discovered by himself, even if there had been some reasons for 
 assuming" that it was not by itself a sufficient explanation of the 
 facts ; but, as will be seen, such reasons are not forthcoming. In- 
 stead of this, such an explanation is not once mentioned, simply 
 because the molecular theory has again asserted its power of 
 paralysing the next step forwards. 
 
 Let us next see how the indications of this action -current should 
 present themselves. When the indifference-point is in the middle, 
 then, after break, the anodic half of the intrapolar region, if ap- 
 propriately led off, would show an action-current in the same 
 direction as the polarising current ; this must also be the case when 
 one leading-off electrode is in the kathodic, the other in the anodic 
 half ; on the other hand, if both leading-off electrodes were situated 
 in the kathodic half, the current w^ould be absent ; but it is con- 
 ceivable that, owing to electrotonic spread of the anodic action- 
 current, it might be present even in this last case. With strong 
 currents the + after-current should be found throughout the intra- 
 polar region. 
 
 This current thus comes in conflict with Peltier's — polarisation 
 after-current in the intrapolar region. But it is essential to polari- 
 sation after-currents, first that they become stronger, the longer 
 within certain limits the polarising current has been closed ; and, 
 secondly, that they rapidly decrease after the break of the latter. 
 The most favourable condition for the development of the action- 
 current (' positive polarisation ' of du Bois) is above all very short 
 closure of polarising current, so that the maximum of polarisation is 
 not obtained with such closure ; but even under such conditions, 
 during the first moment after break, the polarisation — effect is 
 stronger than the + action-current, and the after-current is there- 
 fore diphasic. The chief facts can be thus thoroughly and simply 
 explained without taking into account certain new phenomena to 
 be presently referred to. 
 
 The next step, necessary to prove the validity of this explanation, 
 is to investigate the influence which the position in the intrapolar 
 region of the leading-off electrodes exerts on both — and + after- 
 effects ; a problem which du Bois-Reymond has not investigated. 
 
 Two pairs of leading-off" electrodes were brought into contact 
 with the intrapolar region, and each pair connected with the circuit 
 of a separate galvanometer and compensator. 
 
288 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 Herr A. von Gendre of St. Petersburg* undertook to read one of the 
 
 galvanometers. The two ""alvanometer-circuits were closed at the 
 
 2 3 same moment by means of a commutator, ar- 
 
 I I rang-ed so that one was connected with i and 
 
 : : 6, the other with 4 and 5. The commutator 
 
 " ■ : had a paraffin bed instead of the ordinary 
 
 I I wooden one, this excellent insulating material 
 
 5 6 being" readily workable on the lathe. The two 
 
 cups 2 and 3 were in the battery-circuit, and on the commutator 
 
 being turned to that side were connected by means of a copper arc. 
 
 The connections between the three copper arcs, indicated in the 
 
 figure by dotted lines, were of paraffin. Further, the battery, of 
 
 from 1-18 zinc carbon elements, stood upon paraffin feet. 
 
 For the experiments on nerve, as a rule two sciatics were used, 
 joined together in such a way that the central end of one lay in 
 connection with the peripheral end of the other. This was in order 
 to eliminate as completely as possible the difference in the be- 
 haviour of nerve, stated by du Bois-Reymond to depend on the 
 centrifugal or centripetal direction of a current. For the experi- 
 ments on muscle, the groups of gracilis and semimembranosus were 
 used, the left and right groups being left in their natural attach- 
 ments to the symphysis ; and hooks fastened to ivory pegs by 
 silk thread w^ere attached to the muscle groups and immovably 
 extended them. 
 
 The leading-in contacts were at the two knee ends respectively, 
 and the two pairs of leading-off electrodes were brought as close as 
 possible to these, each pair being thus in connection with the lower 
 division of a muscle group, so that du Bois-Reymond's asserted 
 difference in the behaviour of the upper and lower part of the 
 muscle might be eliminated from the experiment. 
 
 With nerve there occurred no marked difference in the deflections 
 of the two led-off regions attributable to the direction of the polar- 
 ising current ; that is to say, it made no difference as regards the 
 after-current whether the led-off region lay in the neighbourhood of 
 the anode or in that of the kathode. 
 
 Example. Two sciatic nerves placed together in the manner before 
 described. The schema indicates the position of the electrodes (the 
 leading-off electrodes are marked thus \ ). 
 
 
 c 
 
 c' 
 
 
 d 
 
 d' 
 
 1 
 
 ^ 
 
 i 
 
 8 i 
 
 40 
 
 i 8 
 
 i 
 
 5 I 
 
 A ' 
 
 1 — 
 
 ...J 
 
 
 1 
 
 . — ! 
 
 ^B 
 
PHENOMENA OF MUSCLE AND NERVE. 
 
 289 
 
 The numbers denote the intervals in millimeters — polarising current led in at 
 A-B, leading-ofF electrodes at cc' and dd'. 
 
 The battery elements, as in all the following examples, were zinc and carbon, the 
 duration of closure, unless otherwise stated, being extremely short. As no special 
 care was taken to make the closure always of equal length — this being effected by 
 the hand — the inequality in the deflections of successive experiments may be taken 
 as due to this. Comparisons of the amounts of deflection can only therefore be made 
 in the case of observations taken at the same time in cc' and in dd'. This holds good 
 for all the following experiments. 
 
 The comparison of the two deflections in cc' and dd' is naturally further compli- 
 cated through differences in the two galvanometers — differences as well in the delicacy 
 as in the period of oscillation of the magnets. We have taken the necessary read- 
 ings for correcting the deflections so as to give them the value which they would have 
 had if the two instruments had been identical. But this reduction has no practical 
 importance, and is therefore omitted in the experiments given. The first glance 
 shows that the galvanometer in connection with cc' was the more delicate one. A 
 deflection off the scale is denoted thus cc. 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current deflection. 
 
 In cc'. 
 
 In dd'. 
 
 3 
 
 9 
 )> 
 
 i8 
 
 ► 
 
 — -> 
 
 • > 
 
 •< 
 
 ► 
 
 - 75C 
 
 - 32C 
 -158 A 
 
 - 102 A 
 
 -88|c 
 + 43J 
 
 - 981c 
 + 30J 
 
 + 38/^ 
 
 + 30/ ^ 
 
 -2i5lc 
 + 95J 
 
 + 100/^ 
 
 - 30 A 
 
 - 20 A 
 
 - 80C 
 
 - 40C 
 
 - 55l A 
 
 - 40I A 
 
 " ^4 c , 
 
 + 35J ' 
 
 - 50I c 
 + 40J 
 
 ~ 55l A 
 + 65!^ 
 
 -2Sol ^ 
 
 + 80/^ 
 
 The letters A and C denote that the leading-off electrodes lay respectively in the 
 neighbourhood of the anode and kathode. 
 
 Another example. Two sciatic nerves, as in the preceding experi- 
 ment. 
 
 c c' d d' 
 
 35 i 8 i 6 I 
 
 A 
 
 6 1 10 i 
 1 . i_ 
 
 -I B 
 
290 
 
 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 Number 
 of cells. 
 
 Direction and dura- 
 tion of closure of 
 polarising current. 
 
 After-current deflection. 
 
 In cc'. 
 
 In dd'. 
 
 3 
 
 5> 
 
 9 
 
 » 
 >> 
 i8 
 
 ^ short 
 
 ► very short 
 
 < — — short 
 
 ► „ 
 
 * very short 
 
 * JJ 5> 
 
 < longer 
 
 ► short 
 
 ► . longer 
 
 -167 C 
 -138 C 
 
 - 46 A 
 
 — 210 A 
 
 -=^ Ic 
 
 + 47/ 
 
 -4981c 
 + 41J 
 -216I . 
 + 17/^ 
 
 -500I 4 
 + 10/^ 
 
 -3431 c 
 
 + 153J 
 
 -=^ Ic 
 
 + 132I 
 -coj 
 
 + I24J 
 
 - 40 A 
 
 - 34 -A. 
 
 - 15C 
 
 - 70C 
 
 -164U 
 + "J 
 
 -I40I A 
 
 + loj"^ 
 
 - 4°jc 
 + 35J 
 
 -1051 c 
 
 + 20J 
 
 - 4oiA 
 
 + 391 
 -2661^ 
 + 35J 
 
 + 60/^ 
 + 60/^ 
 
 Quite otherwise is the case in muscle. In many experiments the 
 + after-current was so decidedly favoured by the neig-hbourhood of 
 the anode as to leave no question of chance : that is to say, the 
 + phase was, both absolutely and relatively to the — phase, markedly 
 stronger in the led-ofF reg-ion near the entry of the polarising 
 current than in the other regions ; and by certain strengths of 
 polarising current it was generally only found in this region. As 
 this relation remained constant with reversal of current, its cause 
 could not be special to the muscle group investigated. (The 
 direction of the cun-ent was generally changed after two experi- 
 ments.) Although the majority of experiments showed the above 
 relations, there were some dubious ones ; some in which nothing of 
 the kind was observed, and some which appeared to give opposite 
 results. But one single exception is sufficient to invalidate a law, 
 however distinctly the majority of experiments may indicate it. 
 
 In this embarrassment a circumstance happily came to my mind 
 which has very often been overlooked, namely, the inscrijjtio tendinea 
 with which, as is well known, the gracilis and semimembranosus of 
 
PHENOMENA OF MUSCLE AND NERVE. 291 
 
 the frog are provided ^ If a current is led through the entire 
 length of the muscle, then clearly each fibre has an anodic and a 
 kathodic region where it is broken by the hiHcriptio. Supposing 
 one led-off region to lie in the lower half of a gracilis through 
 which an ascending current passes^ and to be symmetrically placed 
 between hiscripiio and the muscle end ; then it is by no means, 
 as might be thought, in the anodic region, but exactly mid- 
 way between an anode and a kathode. That, in spite of this, the 
 gracilis and semimembranosus show for the most part the before- 
 mentioned relations maybe due, first, to tbe oblique direction of their 
 inscriptiones, a truly symmetrical position of leading-ofF contacts 
 between these and the lower ends of the muscles not being pos- 
 sible ; secondly, to the fact that the inscripfio does not completely 
 traverse the semimembranosus ^ ; and thirdly, to the close proximity 
 (as close as possible) of the led-off region to the lower end of the 
 muscle. 
 
 In any case, as the inscripUo must tend to obscure any law, it was 
 advisable to pass on to other muscles, and I chose the sartorius. This 
 rewarded investigation by bringing out, without exception and in 
 the most forcible manner, the following simple law : — 
 
 In muscle the + phase either, as is generally the case, appears only 
 in the led-off portion near the anode ; or, if it comes into view iti the 
 hathodic region {xvhich is the case ichen strong currents are used), is 
 incomparably weaker than in the anodic. 
 
 The — (first) phase, on the other hand, is stronger in the kathodic 
 region ; but this difference is much less marked, and can be ex- 
 plained by the fact that the strong development of the + effect 
 must necessarily prevent the complete accomplishment of the — de- 
 flection. 
 
 Example. Two sartorius muscles, both attached to the symphysis 
 S. Current led in at the knee-ends A and B. Leading-off" contacts 
 in the neighbourhood of the knee-ends at cc' and dd". The follow- 
 ing schema gives the intervals : — 
 
 c c' S d d' 
 
 I 4 4 8 i 25 I 25 1 8 i 4 I 
 
 A L ^-L. 1 -I ' ' ' B 
 
 * See Pfluger's 'Arcliiv,' vol. x. p. 49 ; vol. xv. p. 223. 
 
 "See du Bois-Keymond, ' Arcliiv fur Anat. u. Physiol.' 1876, p. 351 (Ges. Abh. 
 
 ii.p-S73). 
 
 U 3 
 
392 
 
 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 Number 
 of cells. 
 
 Direction of 
 current. 
 
 After-current deflection. 
 
 In cc'. 
 
 In dd'. 
 
 I 
 3 
 
 !J 
 
 9 
 j> 
 
 » 
 
 i8 
 
 ► 
 
 - lo c, C 
 
 - S a, A 
 
 - 20 C, C 
 
 - 45U 
 + 45J 
 
 - 53c,C 
 
 - 3na,A 
 
 + 451 
 
 - 140 c, C 
 
 -Ha . 
 
 + 202J 
 
 -3571 c,C 
 + 36J ' 
 
 "'1^1 a, A 
 + 177J 
 
 - 16 a, A 
 
 - 13 c, c 
 
 + 100}^'^ 
 -107 k, K 
 
 - 76 c, c 
 
 - 290 c, C 
 
 - 'na,A 
 
 -^°nc,c 
 + 32J ' 
 
 — 2021 . 
 
 + 674 "■* 
 
 The readings in cc' were taken by Herr v. Gendre, those in dd' by myself. The 
 delicacy of the two galvanometers was nearly equal ; mine was a little the more 
 delicate. That the deflections in spite of this were relatively smaller in cc' than in 
 dd' was due to the slower swing of the magnet of the first galvanometer, so that 
 during the deflection the after-current had already markedly decreased ; this we 
 ascertained by special experiments*. 
 
 In order to establish the law still more firmly, experiments were 
 now arranged in which the two sartorius muscles were led off at the 
 upper end in the neig-hbourhood of the symphysis. If the law holds 
 good, then the + phase should appear more easily or even alone in 
 that muscle to which the kathode is applied (i. e. the reverse of what 
 was previously the case), since now the leading-off contacts, though 
 on the kathodic muscle, are in the neighbourhood of the physio- 
 logical anode. The prediction was most beautifully fulfilled. 
 
 w ^ The galvanometer used in cc' was an ordinary Wiedemann, that in dd' one con- 
 structed by myself and described in Pflliger's ' Archiv,' vol. xxi. p. 430. The damping 
 arrangement allows aperiodicity with relatively small astasia ; hence its superiority 
 as mentioned before in the text. 
 
PHENOMENA OF MUSCLE AND NERVE. 
 
 293 
 
 In the tables the letters a and c indicate that the deflection to the record of which 
 they are appended belongs to the muscle to which the anode and kathode are respec- 
 tively applied. The capital letters A and C indicate that the region led off was in 
 the neighbourhood of the physiological anode and kathode respectively. In the 
 preceding experiments the physiological and actual anodes and kathodes were identical. 
 This is not the case in the following ones. 
 
 Examjple. Sartorius muscles attaclied to the symphysis, Leading-- 
 
 in electrodes at the two knee-ends A and B. Both muscles led off 
 
 in the neig-hbourhood of the symphysis at cc' and dd'. The schema 
 
 gives the intervals — 
 
 c c' S d d' 
 29 49i4|4i9i 29 
 
 A L- 
 
 -J B 
 
 Number 
 
 Direction of 
 
 After-current deflection. 
 
 of cells. 
 
 current. 
 
 
 
 
 
 In cc'. 
 
 In dd'. 
 
 I 
 
 4 
 
 — 20 c, A 
 
 - 15 a, C 
 
 ;, 
 
 ► 
 
 - 16 a, C 
 
 - 53 c, A 
 
 3 
 
 ' 
 
 ~ 54 ) „ A 
 + 26 P'^ 
 
 — 119 a, C 
 
 >) 
 
 ► 
 
 — 24 a, C 
 
 -.f.\-^ 
 
 J) 
 
 — 
 
 ~ 99I c 1 
 + 13 1 ^"^ 
 
 - 97 a,C 
 
 J) 
 
 ► 
 
 - 26 a, C 
 
 -.%\'^- 
 
 9 
 
 — 
 
 ~ 43 K A 
 + 35 1 "'"^ 
 
 -134 a, C 
 
 j> 
 
 » 
 
 — 69 a, C 
 
 - 28 ) . 
 + 361 1^'^ 
 
 j> 
 
 < 
 
 :^^i"'- 
 
 — 191 a, C 
 
 5> 
 
 ► 
 
 - 88 a, C 
 
 ;3^n«.- 
 
 18 
 
 — 
 
 :Si=.- 
 
 — 269 a, C 
 
 5) 
 
 » 
 
 -i86a, C 
 
 ~"^°^l c A 
 + 408 J •''^ 
 
 5> 
 
 < 
 
 ~ 53 j c,A 
 
 + 45 i ' 
 
 -188 a, C 
 
 5> 
 
 . 
 
 — 220 a, C 
 
 ""M c A 
 
 The galvanometers used were the same as in the previous experiments, and, as is 
 evident, the remarks as to the diflferences between their deflections apply here. (The 
 readings in cc' were taken by Herr v. Gendre with the Wiedemann, those in dd' by 
 myself with my galvanometer,) 
 
294 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 The striking result of this set of observations made me think 
 that the different influence of the anode and kathode respectively 
 mig-ht be more markedly obtained by uniting one of the leading- 
 off electrodes vnth. the anode or the kathode of the polarising 
 current. This method, if literally carried out, would involve the 
 danger of bringing in external polarisation (in the electrode itself), 
 but naturally it is only necessary to make the physiological 
 electrodes identical, i. e. the points of connection of the battery 
 and of the galvanometer with the muscle fibres ; and to do this I 
 proceeded as follows. 
 
 A gastrocnemius was stretched out by means of my extension 
 arrangement, and a polarising current led in at the Achilles tendon 
 and the knee-end. One of the leading-ofF electrodes was also placed 
 on the Achilles tendon somewhat above the leading-in electrode, 
 the other lay on the surface of the muscle at about the junction of 
 the lower and middle thirds of the fleshy part. It is evident that 
 with an ascending current the lower ends of the muscle fibres con- 
 stitute the physiological anode and that it is here that the lower lead- 
 ing-ofl" electrode is in contact with the muscle substance, whilst the 
 vipper leading-off contact lies far removed from the physiological 
 kathode ; with a descending current the physiological kathode and 
 lower leading-off contact are identified. 
 
 The expected result was most beautifully brought out. With de- 
 scending currents, — deflections only were almost always obtained ; 
 
 with ascending currents, diphasic ( h) deflections or + alone. 
 
 Moreover, now even with one zinc-carbon cell the + phase showed 
 itself; whereas with the ordinary method at least two or three cells 
 were needed. It may therefore be stated that leading-off" at the anode 
 itself is the most favourable, at the kathode itself the least favour- 
 able condition for obtaining the + intrapolar after-effect. 
 
 Example i. (Left hand.) Gastrocnemius led in and off as just 
 described. Current of rest T318 scale = 0-0033 D. O^'^ ^^ course 
 of the experiment the current of rest due to the summing up of 
 the residual + phases with ascending polarising currents increased 
 to five times this amount ; this increase showed itself in all the 
 experiments.) 
 
 Example 2. (Right hand.) Gastrocnemius as above. Current of 
 rest T 32 (delicacy diminished) = 0-0065 D. (Increased during experi- 
 ment to 42- times as much.) 
 
PHENOMENA OF MUSCLE AND NERVE, 
 
 295 
 
 Example i. 
 
 Example 2. 
 
 Number 
 
 Direction 
 
 Deflection 
 (lower end 
 of muscle). 
 
 Number 
 
 Direction 
 
 Deflection 
 (lower end 
 of muscle). 
 
 of cells. 
 
 of curi-ent. 
 
 of cells. 
 
 of current. 
 
 I 
 
 
 -167c 
 
 Delicacy of galvanometer 
 
 
 
 - 35 1 „ 
 
 considerably decreased. 
 
 7> 
 
 
 + 122 \ 
 
 I 
 
 
 - 13 c 
 
 }J 
 
 
 — 290 c 
 
 
 
 - ^L 
 
 
 
 +i27.r 
 
 " 
 
 
 + 7r 
 
 " 
 
 
 ,, 
 
 
 - 25 c 
 
 3> 
 
 
 -154 c 
 
 
 
 - ^sL 
 
 
 
 - 89U 
 
 + I25i 
 
 " 
 
 
 + 3r 
 
 " 
 
 
 3 
 
 
 - 47 c 
 
 3 
 
 
 -320 c 
 + 00 p 
 
 )) 
 
 
 - ^9|a 
 
 j> 
 
 
 „ 
 
 
 - 51 c 
 
 Delicacy of galvanometer 
 
 
 
 - ^3 la 
 
 greatly decreased. 
 
 " 
 
 
 + 44i* 
 
 )> 
 
 
 - 6c 
 
 - I ) 
 
 9 
 
 
 - 38 c 
 + 123 a 
 
 j> 
 
 
 + I5i^ 
 
 >) 
 
 
 — 27 c 
 
 >> 
 
 
 — 22 c 
 
 - I ) 
 
 " 
 
 
 — no ) 
 + 81P 
 
 5) 
 
 
 + isp 
 
 18 
 
 
 -237 c 
 
 6 
 
 
 - 13 c 
 
 M 
 
 
 — 40 
 
 )> 
 
 
 - ^4ja 
 + 315 
 
 „ 
 
 
 7%\' 
 
 9 
 
 
 - 9c 
 
 - 'n. 
 
 + 32 i 
 
 ,, 
 
 
 -.is\^ 
 
 " 
 
 
 
 
 
 Delicacy of galvanometer 
 
 
 
 
 slightly increased. 
 
 
 
 
 5> 
 
 I 
 
 — 41 c 
 
 
 
 
 
 r 
 i 
 
 -^47 j a 
 + 93 i 
 - 63 c 
 
 
 
 
 " 
 
 r 
 
 -^^7ja 
 + 43 i 
 
 
 
 
 There is another method by which the physiological electrode of 
 a polarising current can be identified with the physiological leading- 
 off contact, namely, leading a polarising current into a piece of dead 
 muscle, and leading-off to a galvanometer from this in place of the 
 tendon in the experiment just described. The employment of this 
 method did not give those results which might at first sight have 
 
296 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 been expected, but, on the contrary g-ave to a certain extent opposite 
 ones ; these however confirmed our main law in a most striking and 
 instructive way. 
 
 For these experiments I made use of a method of my own for 
 causing heat rigor in a tract of muscle^; which du Bois-Reymond 
 has more recently described as the production of a ' thermal cross- 
 section^.' The experiment naturally allowed the use of a single 
 muscle arranged as in the schema of that just described, a tract 
 in heat rigor taking the place of the tendon. The following 
 arrangement was however still better adapted for this ex- 
 periment. 
 
 The two sartorius muscles were so prepared that they remained 
 attached to the symphysis. They were then placed together, 
 and the symphysis with the muscle attachments was dipped 
 8 mm. deep into hot water ; heat rigor being thus produced. The 
 muscles were then drawn apart and stretched out by fixing the 
 knee-ends with hedgehog spines to a cork board. The polarising 
 current was led in at the knee-ends, the leading-off contacts w^ere 
 placed, one in the region of heat rigor (at the symphysis), the other 
 on the unimpaired surface of the same muscle in its upper half. If 
 the right muscle be thus led off, then the left one serves merely as 
 an indifferent conductor of the polarising current ; the muscle ex- 
 perimented upon is however the right one, in which the upper 
 polarising and upper leading-off electrodes are identified and lie on 
 the thermal cross-section. At the end of an experiment the other 
 muscle may be used by shifting the leading-off electrodes. The full 
 demarcation current is naturally led off and has to be com- 
 pensated. 
 
 ■ The preceding experiments would lead us to expect that the -f- 
 phase of the after-current should only occur when the physiological 
 anode is united with one leading-off contact, or, in other words, 
 when the polarising current is directed from the muscle not led off 
 to the led- off one; and hence is in this ' abterminal.' Instead 
 of this, the + phase never appeared under such circumstances ; 
 but sometimes appeared with the polarising current directed the 
 opposite way. 
 
 The cause of this apparent discrepancy was clear to me. It is known 
 that Biedermann found that the breaking- excitation failed in the case 
 
 * See Pfltiger's 'Archiv,' vol. iv. p. 167. 
 
 ^ ' Archiv f. Anat. u. Physiolog.' 1S73, p. 526 (Ges. Abh. ii. p. 409). 
 
PHENOMENA OP MUSCLE AND NERVE. 297 
 
 of currents whose anode coincided with an artificial cross-section. 
 The failure of the + j)ha8e of the after-current loith a j)olarismg 
 current directed from dead to living tissue, is therefore a neio and 
 striking proof that this after-current is connected in the most intimate 
 manner loith the hrealdng excitatioti. I may remark here incidentally, 
 that these experiments offer many opportunities for verifying- the 
 statements of Biedermann and Eng-elmann (with van Loon) as to 
 the excitatory effects of currents directed respectively from dead 
 to living- and from living to dead tissues, since there are here 
 two muscles in the circuit, in one of which the current is in 
 the first direction 'abmortal,' in the other or second direction 
 ' admortal ^.' 
 
 It has already been noted that the + phase sometimes appeared 
 with atterminal direction of current. As mig-ht be readily pre- 
 dicted, its appearance was more marked the nearer to the knee end 
 the leading--off electrode was placed ; the farther therefore, with 
 livino- to dead direction of current, this electrode was moved into 
 the region of anelectrotonus. 
 
 Examples. Each experiment was made upon two sartorius 
 muscles arranged and led off in the manner just described. The 
 following- schema shows the arrangement : — 
 
 A^ 
 
 The symphysis is marked S as before. AS, BS are the two sartorii ; 
 the shaded parts indicate the portions of the two muscles in heat 
 rigor. The electrodes of the polarising- current are at A and B ; 
 and the positions of the pairs of leading--off electrodes in the several 
 experiments are indicated by cc', cc'^ dd^ 
 
 ^ The nomenclature, as bearing upon the direction of a current with reference to 
 the ends of the fibres (' Archiv,' vol. xvi. p. 193), is an adaptation of that used by 
 Biedermann, Engelmann, Griitzner, and others ; these writers describe as abterminal 
 a current directed from the ends of the fibres into the muscle when the end of the 
 fibre is an artificial cross-section ; such currents I term ' abmortal.' Biedermann's 
 facts make such a distinction desirable. For instance, I can thus shortly express 
 my results — an abterminal current gives under certain conditions a + after-current 
 phase, an abmortal does not. [In the text the terms ' abmortal ' and ' admortal ' 
 used by Hermann are translated ' dead to living,' ' living to dead.' Tr.] 
 
 d' 
 
 d 
 
 s 
 
 c 
 
 c' 
 
 c" 
 
 
 i 
 
 i 
 
 ^^^^^ 
 
 i 
 
 4 
 
 i 
 
 -Jb 
 
298 
 
 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 Example i. 
 
 
 Examp! 
 
 e 2. 
 
 Number 
 of cells. 
 
 Direction 
 of current 
 in AB. 
 
 After- current 
 deflection. 
 
 Number 
 of cells. 
 
 Direction 
 of current 
 in AB. 
 
 After-current 
 deflection. 
 
 Led of 
 
 F at cc', current of rest -> oo 
 
 Led off at cc", current of rest — ► 00 
 
 
 = 0-0311 D. 
 
 
 = 0-046 
 
 5D. 
 
 I 
 
 - 
 
 — 26 living to 
 dead 
 
 I 
 
 - 
 
 — 41 living to 
 dead 
 
 )) 
 
 - 
 
 — 32 dead to 
 living 
 
 >) 
 
 -> 
 
 — 40 dead to 
 living 
 
 3 
 
 *- 
 
 - 64 1. to d. 
 
 3 
 
 *- 
 
 + 30 1. to d. 
 
 
 -► 
 
 — 42 d. to 1. 
 
 )> 
 
 — ► 
 
 — 160 d. to 1. 
 
 6 
 
 _ 
 
 — 42 1. to d. 
 
 - 61 d. to 1. 
 
 6 
 
 - 
 
 -^?3 i.tod. 
 + 98 
 
 9 
 
 - 
 
 ~ M 1. to d. 
 + 20 \ 
 
 — 60 d. to 1. 
 
 9 
 
 — 
 
 — 260 d. to 1. 
 
 i8 
 
 •*— 
 
 — 590 d. to 1. 
 
 18 
 
 < — 
 
 — 630 d. to 1. 
 
 + 659 s 
 
 — 00 d. to 1. 
 
 Number 
 of cells. 
 
 Direction 
 
 of current 
 
 in AB. 
 
 After-current 
 deflection. 
 
 Number 
 of cells. 
 
 Direction 
 
 of current 
 
 in AB. 
 
 After-current 
 deflection. 
 
 Led off 
 
 at dd', curre 
 scale = 0- 
 
 nt of rest <— 565 
 3436 D. 
 
 Ledofl 
 
 ' at dd', curr 
 = 0-04^ 
 
 ent of rest <— 00 
 ^8D. 
 
 I 
 
 - 
 
 — 20 dead to 
 living 
 
 1 
 
 - 
 
 — 70 dead to 
 living 
 
 6 
 
 5) 
 
 9 
 
 - 
 
 — 10 1. tod. 
 -315 d. tol. 
 
 -48Ml.tod. 
 + 65 i 
 
 >> 
 4 
 
 - 
 
 — 30 d to 1. 
 -183 d. tol. 
 
 5» 
 
 - 
 
 7\llV-^^- 
 
 5> 
 
 - 
 
 Tk!'-'""- 
 
 18 
 
 « — 
 
 — 00 d. to 1. 
 
 9 
 
 *- 
 
 — 496 d. to 1. 
 
 )) 
 
 - 
 
 ZusV-^i' 
 
 )> 
 
 — ► 
 
 :^^^|>--^- 
 
 
 
 
 18 
 
 5> 
 
 ■» — 
 
 — 00 d. to 1. 
 
 - °" ! 1. to d. 
 
 + CO \ 
 
PHENOMENA OP MUSCLE AND NERVE. 299 
 
 The striking result of this experiment led to a method by 
 which the ditt'erent effects dependent upon direction of current 
 were still more markedly shown. It is seen above that (at least 
 within the limits of current strength there employed) currents 
 directed from dead to living tissue hardly ever give the + phase, 
 which only appeared with a current directed from living to dead 
 tissue, the arrangement being that one leading-in and one leading- 
 off electrode lie in the region of artificial cross- section ; further, 
 that with currents directed from living to dead tissue the appear- 
 ance of the + phase is favoured by the proximity to the anode of 
 the electrode leading off the living tissue. Could this latter be 
 identified with the anode, the conditions for the production of the 
 phenomena would be the most favourable. 
 
 This is obtained in the gastrocnemius, for instance, when a 
 thermal cross-section is made in the upper end by dipping it into 
 hot water. Upon this one leading-ofi' and one polarising electrode 
 are placed, and then the other leading-ofF and the other polarising 
 electrode are placed upon the Achilles tendon ; both places offer suf- 
 ficient room for two kaolin points. With this arrangement the de- 
 scending (dead to living) current never gives the + phase ; on the 
 other hand, the ascending (living to dead) current gives it even 
 when quite weak. The state of things is evidently this ; the lead- 
 ing off occurs quite at the anode and kathode, independently of the 
 direction of the current. The anode, however, in the case of 
 a descending current, lies upon the artificial cross-section ; ac- 
 cordingly it is only with ascending currents that the + phase is 
 produced. 
 
 But it is possible to go still farther. So far the union of a 
 leading-in and a leading-off electrode has been effected by a part 
 of the muscle which acts as an indifferent conductor, whether 
 this be the tendon or a portion of tissue in heat rigor. There 
 is however no objection to effecting this by a piece of electrode 
 kaolin. This is indeed capable of internal polarisation according 
 to du Bois-Reymond, but only 'negatively;' and at any rate 
 it can scarcely be more polarisable than a piece of tendon or dead 
 muscle. 
 
 When it is desired to lead off from the entire portion of tissue 
 through which the polarising current passes, a very simple arrange- 
 ment is the following ; the kaolin points of the leading-off elec- 
 trodes are laid upon the tissue, and upon these points are placed 
 those of the polarising electrodes. The * external ' polarisation 
 
300 
 
 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 which prevented du Bois-Reymond (p. 348) from using* polarising^ 
 as leading--ofF electrodes is in this way completely avoided ^. 
 
 When one such double electrode lies upon a living", the other 
 upon a dead portion of muscle, then, as might be expected, the 
 -h after-current only appears following such currents as are directed 
 from the former to the latter, i.e. from living to dead tissue. In 
 every way the above arrang'ement presents itself as the most favour- 
 able one for obtaining the + after-effect of muscle, even when both 
 double electrodes lie upon living tissue, since in all cases the leading- 
 off is from the most effective portion of the anodal intrapolar region. 
 
 Exawples. Sartorius muscle, with one end A in heat-rigor. 
 Leading in and off" by means of double electrodes at cc'. 
 
 I c c 
 
 A 
 
 Example X. 
 
 Example 
 
 
 
 After-effect. 
 
 
 
 After-effect. 
 
 Num- 
 ber of 
 cells. 
 
 Direc- 
 tion of 
 current. 
 
 
 
 Num- 
 ber of 
 cells. 
 
 Direc- 
 tion of 
 current. 
 
 
 Actual 
 reading. 
 
 Calculated reading 
 
 with most delicate 
 
 position. 
 
 Actual 
 reading. 
 
 Calculated reading 
 
 with most delicate 
 
 position. 
 
 
 Current o 
 One ( 
 
 ' rest —*■ 
 ;oil only u 
 
 = 0-0189 D. 
 
 sed (1) I 
 
 
 Current of rest ♦- uo = 0-0386 D. 
 Full delicacy of galvanometer. 
 
 I 
 
 - 
 
 -267) 
 + 35 ) 
 
 -334) living 
 + 44 ^ to dead 
 
 I 
 
 - 
 
 — 426 ) 
 
 -H05 \ 
 
 — 426 ) living 
 -M05 ) to dead 
 
 " 
 
 -> 
 
 -298 
 
 ) dead to 
 37^ J living 
 
 » 
 
 -> 
 
 -343 
 
 dead to 
 ^'*^ living 
 
 
 Coil 
 
 farther av 
 
 ^ay (aV)- 
 
 
 Coil farther away (2^). 
 
 4 
 
 - 
 
 - 72) 
 + 40) 
 
 ;'!£ !>■'»<'• 
 
 4 
 
 - 
 
 -190) 
 + 20 \ 
 
 -3S°°h.tod. 
 + 400 \ 
 
 „ 
 
 -► 
 
 — no 
 
 — 2200 d. to 1. 
 
 t> 
 
 — 
 
 -190 
 
 — 3S00 d. to 1. 
 
 9 
 
 - 
 
 -272 
 + 28 
 
 :ig!'-'»^- 
 
 9 
 
 - 
 
 -312 1 
 + 55 i 
 
 -f 1 100 \ 
 
 „ 
 
 -* 
 
 — 400 
 
 — Scoo d. to 1. 
 
 „ 
 
 — 
 
 -179 
 
 — 35S0 d. to 1. 
 
 18 
 
 - 
 
 -299) 
 + 21 \ 
 
 -598° h. tod. 
 -1- 420^ 
 
 
 Coil still farther away (g-g)- 
 
 " 
 
 
 -630 
 
 — 12600 d. to 1. 
 
 18 
 >> 
 
 - 
 
 - 89) 
 
 + lyi 
 -142 
 
 -•"4o[i.tod. 
 -t- 1020 \ 
 
 — 8520 d. to 1. 
 
 ^ In this, as in all experiments of this nature, if a rheochord be used in order to 
 obtain weak polarising currents, the breali must lie between the tissue and the 
 rheochord, not between the latter and the battery. (See Pfliiger's 'Archiv,' vol. vi. p. 318.) 
 
 ^ The fractions in the brackets indicate the relative delicacy in each particular 
 
PHENOMENA OF MUSCLE AND NERVE. 301 
 
 There is then abundance of proof of the fact overlooked by du 
 3ois-^e jvaond t/iat the ■\- phase of the after-current hi muscle prevails 
 only in the region of the anode. 
 
 In the case of nerve the experiments which have been pre- 
 viously cited showed that there was no difference of behaviour 
 whether the leading'-off occurred close to the anode or close to the 
 kathode. This should not surprise us since, in nerve, anelectro- 
 tonus extends with strong currents quite up to the kathode. In 
 spite of this it seemed desirable to employ upon nerves, the methods 
 used in the case of muscle in order to obtain more decided differ- 
 ences between the behaviour of nerve at the anode and kathode : in 
 point of fact such differences were obtained. 
 
 For this purpose a polarising and a leading-off electrode were 
 laid upon a destroyed spot in a nerve. The middle portion of a 
 nerve preparation, made as before described by laying two nerves 
 together, was destroyed by dipping it into hot water (the prepara- 
 tion being first folded in the middle so that the two halves lay side 
 by side). As however the vapour severely injures the remaining 
 portions of the preparation^ as described by me in an earlier work ; 
 this method was abandoned \ and in the later experiments the 
 piece of nerve-preparation was effectually squeezed between the 
 blades of a small vice. Still, the first method of treatment gave 
 results invariably like the second. 
 
 It appeared that in nerve also, with appropriate strength of 
 current, the + phase of the after-current regularly failed when the 
 physiological anode and one leading-off electrode lay on the artificial 
 cross-section. 
 
 Exaynple. Two sciatic nerves placed side by side in opposite 
 directions ; and squeezed in the middle for 9 mm. (the squeezed 
 part is indicated in the schema by the dark line). Polarising- 
 carrent led in at AB. 
 
 B 
 
 
 c'' 
 
 c 
 
 (/ 
 
 A 
 
 1 
 
 1 
 
 1 
 
 i 
 
 case (due to particular distance as well as length of coil), the highest attainable 
 delicacy being taken as unit. The distances were so arranged that this ratio should 
 be as nearly as possible a round number. With my galvanometer, the bobbin of 
 which is immovable, another extra bobbin was used. See Pfliiger's 'Archiv,' vol. 
 xxiv. p. 254. 
 
 ^ Loc. cit. vol. xxiv. p. 254. 
 
302 
 
 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 Led off at 
 
 cc'. 
 
 Led off at cc". 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current. 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current. 
 
 Current of rest ->• 437 s 
 
 scale = o-oi 49 D. 
 
 Current of rest ■*— 394 scale = 0-0099 ^^ 
 
 I 
 
 - 
 
 - 34 living 
 
 to dead 
 
 — 32 dead to 
 
 1 
 
 I 
 
 - 
 
 — 97 dead to 
 
 living 
 
 - 74 living 
 
 » 
 
 
 living 
 
 " 
 
 
 to dead 
 
 4 
 
 •H- 
 
 -118 1. tod. 
 
 4 
 
 +- 
 
 — 210 d. to 1. 
 
 „ 
 
 — >• 
 
 — 121 d. to 1. 
 
 ,, 
 
 — »■ 
 
 — 200 1. to d. 
 
 9 
 
 
 i'-^«h-tod. 
 
 9 
 
 ■^ 
 
 — 202 d. to 1. 
 
 
 + 58 ) 
 
 ,j 
 
 — >• 
 
 — 5!;o 1. to d. 
 
 
 ^ 
 
 — 206 d. to 1. 
 
 
 
 
 18 
 
 - 
 
 ;s-ji.tod. 
 
 18 
 
 Half the coil u 
 
 *— 
 
 -433 d. tol. 
 
 >> 
 
 
 -552 d. tol. 
 
 
 
 C-54i) 
 
 
 
 
 >> 
 
 ♦— 
 
 -395 d. tol. 
 (-494) 
 
 
 
 
 " 
 
 —*■ 
 
 (:':0 
 
 
 
 
 
 
 As in the last-mentioned experiments on muscle, the two pairs of 
 electrodes were now identified, and the one united pair was placed 
 upon a destroyed spot, the other upon a living* portion of a nerve- 
 preparation. 
 
 In this case also, with a few exceptions in which very strong 
 currents were used, the + after-effect appeared only with polarising- 
 currents directed from living to dead tissue, and on account of the 
 favourable leading-off it revealed itself after a weak current such as 
 that of one zinc carbon cell. 
 
 Example. Two sciatic nerves placed together as before and 
 crushed at one end (dark line in schema) united leading-in and 
 leading-off electrodes at CC. 
 
 C 
 
PHENOMENA OF MUSCLE AND NERVE. 
 
 303 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current. 
 
 Actual readings. 
 
 Calculated for 
 greater delicacy. 
 
 
 Current of rest — 334 scale = o-oi D. 
 
 I 
 
 ► 
 
 -165 1 
 + 23^ 
 — 202 
 
 ^^ I living to dead 
 + 23i * 
 
 — 202 dead to living 
 
 4 
 
 ^ 
 
 -co) 
 
 + 49 i 
 
 - CO 
 
 ~ °° I living to dead 
 
 + 49i 
 
 — 00 dead to living 
 
 
 
 Half the coil used (i). 
 
 " 
 
 ► 
 
 -106 ) 
 + 24 \ 
 — 140 
 
 ^^^ living to dead 
 + 30 
 
 — 175 dead to living 
 
 
 
 Coil at a distance (^V)- 
 
 9 
 
 >> 
 i8 
 
 5J 
 
 A 
 
 ►• 
 
 •* 
 
 - 53) 
 + 2$ 
 
 - 61 
 -III 
 
 - 142 
 
 -^°^°! living to dead 
 
 + 40 ) 
 
 — 1220 dead to living 
 
 — 2220 living to dead 
 
 — 2840 dead to living 
 
 T/iere can therefore le no douU that in nerve also the + intrajwlar 
 after-current is a_phenomenon connected with the anektroto?iic state. 
 
 The text shows that exceptions to this rule sometimes occur, as, 
 for instance, when strong currents have been employed, in which 
 case the + phase appears not only with currents directed from 
 living to dead tissue, but also with those directed from dead to 
 living; but this is not perplexing, for, as I have repeatedly stated 
 in former communications, nerve cannot be stretched to such a 
 deo-ree as to get rid of the wavy form of its fibres, hence, with 
 currents directed through its length, there must be multiple anodes 
 
 and kathodes. ^ , /. ^ , i - a 
 
 Biedermanni has recently drawn attention to the fact that, under 
 exactly similar conditions, currents directed from dead to living 
 tissue must in part enter the fibres in their length instead of their 
 cross-section owing to unavoidable bends, etc. Muscle can be much 
 more stretched without risk. In spite of this I hold the question 
 as not finally decided whether or no anelectrotonus fails m the case 
 of the strongest currents which are directed strictly from dead to 
 Uving tissue ; it involves principles of great importance and should 
 be approached by special experiment. 
 
 ' ' Sitzungsber. d. Wiener Acad.,' Part 3, vol. Ixxxv. p. 144. 
 
304 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 For the present sufficient material has been obtained to determine 
 the correctness of my explanation of du Bois-Reymond's observa- 
 tions. 
 
 The phenomena of the extrapolar-currents, to the investigation of 
 which we now turn, will render the decision of the point much 
 more certain. 
 
 4. Investigation of extrapolar after-currents in Nerves 
 and Muscles. 
 
 The extrapolar after-currents, following- the passage of a current 
 through a given portion of tissue, have been hitherto only inves- 
 tigated in the case of nerve. Fick was the first to observe that 
 on either side of the polarising-current there appeared an after- 
 current opposed to it in direction ^. Shortly afterwards I entered 
 on the same enquiry- and found that the two extrapolar regions 
 were completely different in their behaviour ; on the side of the 
 anode I found, like Fick, an after-current opposed in direction to 
 the pre\Tious current, on that of the kathode, however, an after- 
 current, similar in direction to the previous current ; the former 
 was much more marked than the latter. On repeating his experi- 
 ments Fick obtained results similar to mine ^. Du Bois-Reymond 
 mentions these phenomena, but does not state whether he has ever 
 experimented upon the subject or not. 
 
 Since the publication of the work I did at that time, I have often 
 bad the opportunity of observing again the extra j^olar after-currents 
 of nerve, especially in the years 1871, 1872, when I was investigat- 
 ing the extrapolar after-currents in the case of platinum wires 
 surrounded by fluid ^. The new work of du Bois-Beymond and the 
 explanation which, I believed, should be given to his supposed ' posi- 
 tive polarisation,' prompted me to a systematic re-investigation. As 
 a matter of fact, the old experiments could not be considered as 
 exhaustive, since in these the multij)lier was used, the sluggish 
 needle of which would not answer to a rather rapid change of 
 current-direction ; moreover, du Bois-Reymond's results encouraged 
 making experiments with high strengths of current (which were 
 
 ^ ' Centralblatt f. d. med. Wissensch.' 1S67, p. 436. 
 
 ^ ' Untersuchungen zur Physiologie der Muskeln und Nerven,' Part 3, p. 71. 
 Berlin, 1868. 
 
 * ' Untersuchungen aus dem physiol. Labor, d. ZiiricherHochschule,' Part i, p. 129. 
 Vienna, 1S69. 
 
 * Pfliiger's ' Archiv,' voL v. p. 267 ; vol. vi. p. 31 S. 
 
PHENOMENA OP MUSCLE AND NERVE. 305 
 
 formerly viewed with distrust) and with very short durations of 
 closure. Finally, the investigation had also to be extended to the 
 case of muscle. 
 
 The method of experiment was that described previously in this 
 communication, the leading"-ofF electrodes being- now extrapolar. 
 The insulation of the two circuits by means of a paraffin commuta- 
 tor proved quite effectual also for these experiments, althoug-h, as I 
 knew from previous work, the disturbing effect of bad insulation is 
 more marked in the case of extrapolar than in that of intiaj)olar 
 experiments. This latter circumstance is easily explained ; for if the 
 material of the commutator is ab all conductive, then, during the 
 closure of the battery ciicuit, derived currents will pass through the 
 galvanometer which, in the case of extrapolar leading-off, must, 
 however derived, be similarly directed through the instrument, 
 whereas in the case of intraj)olar leading-off these would be opposed 
 in direction, and consequently would in part compensate each other ; 
 in this last case it is as though the galvanometer was introduced 
 into the cross wire of a Wheatstone's bridge. 
 
 A large number of experiments, with currents of from i to 1 8 
 zinc carbon elements and of durations of closure from i to 60 
 seconds, establish the following propositions. 
 
 The deflection is in the case of nerve always in accordance with 
 my previous observations ; that is, on the side of the kathode it is 
 + , similar in direction to the previous polarising current ; on the 
 side of the anode it is — , opposed in direction to the previous 
 polarising current ; in other words, it is, in both situations, directed 
 away from the led-through portion of tissue. The anodal after- 
 current is without exception stronger than the kathodal as I had 
 previously found. But further, the light magnet of the galvano- 
 meter brought into view a new fact, viz. : the appearance, before the 
 — anodal after-current, of a short + effect. The anodal after- 
 ctirrent is thus diphasic, i. e. first + , then — ; the kathodal after-current 
 is entirely + . Moreover the — anodal after-current is of long dura- 
 tion, the magnet after reaching the maximum of its deflection 
 returns slowly and incompletely, often remaining stationary, 
 whereas the kathodal + deflection ceases immediately and com- 
 pletely. 
 
 In the case of muscle, (the sartorius stretched out on cork being 
 generally used in these experiments) the law of the extrapolar after- 
 currents, which are always very well marked, is in general the 
 same as in that of nerve. A fact, as we shall see, of great import- 
 
 X 
 
306 
 
 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 ance is the streng-th of the first + anodal after-current relatively to 
 the second — effect, the former with weak currents frequently ap- 
 pearing alone ; further the second phase is not, as in the case of 
 nerve, stronger than the kathodal after-current, but on the contrary 
 is always markedly weaker ; it is, however, always of longer dura- 
 tion. The law for muscle may therefore be expressed as follows : 
 on breaking a j^olarising current, there appears on each side an after- 
 current having the same direction as the previous polarising one [i. e. 
 -f ) ; this is followed on the side of the anode by an opposed — de- 
 -flection of long duration, whereas on the side of the kathode the -f- 
 after-current shows no such opposite phase, bttt attains a very marked 
 strength and then comjdetely disappears. In a few instances (of which 
 the second example is one) I saw with weak currents, a double 
 effect, first + then — on the side of the kathode. The follow- 
 ing are examples of the facts described. 
 
 Example of an experitnent on nerves. Two sciatic nerves placed 
 side by side in alternating directions (as in previous experiments). 
 Extrapolar leading off at CC Duration of closure short as possible. 
 
 B| 
 
 B' 
 
 |c 
 
 Ic' 
 
 Long polarised region AB. | 
 
 Short polarised region AB'. 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current 
 in cc'. 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current 
 in cc'. 
 
 I 
 3 
 9 
 i8 
 
 •< 
 
 < 
 
 < 
 
 ► 
 
 •4 
 
 ► 
 
 -25* i 
 
 + 21 C 
 
 -117 *a 
 + 52 c 
 + 20 
 -154* ^ 
 + 75° 
 
 + 54 c 
 
 - 70* 5 ^ 
 
 + 32 c 
 
 I 
 
 5> 
 
 3 
 )j 
 9 
 
 )) 
 18 
 
 >> 
 
 * 
 
 •< 
 
 ► 
 
 •« 
 
 + ^la 
 + 6 c 
 
 + 23 c 
 
 - 9i*r 
 
 4- 50 c 
 
 -108* p 
 + 39 c 
 
PHENOMENA OF MUSCLE AND NEEVE. 
 
 3or 
 
 Another example on nerve. Arrang-ement as in preceding experi- 
 ment, polarising current led through AB and AB' alternately. Short 
 duration of closure. 
 
 
 
 After-current in cc'. 
 
 Number 
 of cells. 
 
 Polarising 
 current led 
 
 With direction 
 
 of polarising 
 
 current. 
 
 With direction 
 
 of polarising 
 
 current. 
 
 
 
 < 
 
 *■ 
 
 I 
 
 AB 
 
 -\V 
 
 + 5c 
 
 jj 
 
 AB' 
 
 - 6 P 
 
 + 14 c 
 
 3 
 
 AB 
 
 - 29* P 
 
 + 13 c 
 
 >) 
 
 AB' 
 
 + ^ la 
 - 6i*P 
 
 + 27 c 
 
 9 
 
 AB 
 
 + 12 ) 
 
 - 65* r 
 
 + 33C 
 
 j> 
 
 AB' 
 
 -114* a 
 
 + 50 c 
 
 i8 
 
 AB 
 
 - 70*!- 
 
 + 48 c 
 
 J) 
 
 AB' 
 
 - 93* r 
 
 + 30 c 
 
 X 2 
 
308 
 
 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 JExamples of experiments on muscle. Sartorius, extrapolar leading'-ofF 
 contacts at cc'. Polarising current led in at AB ; short duration of 
 closure. 
 
 A I I B Ic i</ 
 
 Example i. 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current 
 in c/. 
 
 + 46 a 
 
 - 4c 
 + 20 a 
 
 - 4c 
 + 32a 
 
 o c 
 
 + 3 
 
 - 15* 
 + 156 c 
 
 + 15 
 
 - 32* 
 + 162 c 
 
 + 30 
 
 - 60* 
 
 + 153C 
 
 + 17 
 -136* 
 
 + 339C 
 
 + 17 
 -183* 
 
 + 328 c 
 
 Example 2. 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current 
 in cc'. 
 
 + 35 a 
 + 2| 
 
 - 44) 
 + 30 a 
 + I ) 
 
 - 5M 
 -t- 30 a 
 + I 
 
 - 28 
 
 + 55 a 
 
 + 170 
 
 -f- 41 a 
 
 + 48 c 
 
 -I- 42 
 
 - 16* 
 
 + 394° 
 -I- 64 
 
 - 47* 
 + 436C 
 
 + 35 
 
 - 15 
 + 485C 
 
 + 58 
 
 - 12 
 
 + 402 c 
 
 + 30 
 -255* 
 + 542C 
 
 + 30 
 -185* 
 
 + 468C 
 
 We postpone the explanation of the appearances observed to the 
 next section. From experiments mth intrapolar leading-off, it was 
 to be expected that the phenomena would also appear more forcibly 
 with extrapolar leading-off, if one leading-off electrode was united 
 
PHENOMENA OF MUSCLE AND NERVE. 
 
 309 
 
 with one of the polarising' electrodes, and the extrapolar led-off 
 part brought thus immediately up to the polarised region. One 
 leading-ofF electrode was therefore placed with its kaolin point on 
 the kaolin of one of the leading-in electrodes. 
 
 In this case the phenomena appeared still better and with more 
 regularity than they did with the ordinary method. 
 
 Exa7nple of an experiment on nerve. Two sciatics lying side by 
 side in alternating directions. Leading-in contacts at AB. Lead- 
 ing-ofl' contacts at cc'. Electrodes c and B united. Short duration 
 of closure. 
 
 A[ B!j l^ 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 cuiTent. 
 
 After-current 
 in cc'. 
 
 I 
 j> 
 4 
 
 5J 
 
 9 
 
 )> 
 i8 
 
 )) 
 
 -- > 
 
 + 8c 
 + 3 la 
 
 - 200* \ 
 + III C 
 
 -247*r 
 
 + i86c 
 + 9° la 
 + 2960 
 
310 
 
 THE SO-CALLED SECONDABY ELECTEOMOTIVE 
 
 Example of an experiment on muscle. Sartorius, led in and off as 
 in the preceding case of nerve. 
 
 Number 
 of cells. 
 
 Direction of 
 
 polarising 
 
 current. 
 
 After-current in cc'. 
 
 I 
 
 * 
 
 -I27M* 
 
 )i 
 
 ► 
 
 + 56 c 
 
 4 
 
 ~ 
 
 + 20 ) 
 — 00* j 
 
 » 
 
 *■ 
 
 + 123c 
 
 Bobbin removed to 
 further distance 
 from galvanometer (yV). 
 reduced 
 values. 
 
 + 2 ) + 36 
 
 j> 
 
 
 _ 38* P " -684* 
 
 5J 
 
 ► 
 
 + 5c = + 90 
 
 9 
 
 •« 
 
 - 53* a = -954* 
 
 j> 
 
 ► 
 
 + II c = + 198 
 
 i8 
 
 — 
 
 + 14 ) „ _ +252 
 - 43M -784* 
 
 )> 
 
 * 
 
 + 12 c = + 216 
 
 Finally, it may be stated that I have made a number of experi- 
 ments with Herr von Gendre's help, upon nerve and muscle led-off 
 intrapolarly and extrapolarly at the same time, in order to deter- 
 mine the time relations of the two phases. But as these experi- 
 ments offer no new point of view, I omit further reference to them. 
 
 5. Explanation of the intrapolar and extrapolar after-currents 
 of Nerves and Muscles, together with further experiments. 
 
 A. The after-currents loJiich arise from true polarisatmi. 
 
 Since its discover}^ by Peltier, it has been accepted by all obser- 
 vers that the opposed — after- current, which appears in the 
 intrapolar region of a length of nerve or muscle, through which a 
 current has been led and broken, is due to internal galvanic polari- 
 sation. This current shows, in fact, unmistakeably all the peculia- 
 rities of a polarisation current. Especially noticeable is its rapid 
 decline, as evidenced by the fact, that only slight traces of it can be 
 found when a short time is allowed to elapse between the breaking 
 of the battery and the closure of the galvanometer circuit. The 
 
PHENOMENA OP MUSCLE AND NERVE. 311 
 
 aperiodic magnet shows this characteristic most plainly by its mode 
 of movement, which does not at all follow the law^s which hold for 
 persisting" currents^, the first part of the deflection being- very rapid 
 as compared with the later part, and in the case of light magnets 
 often only a momentary jerk ; evidently the current has perceptibly 
 decreased before the deflection has had time to occur ; hence the 
 deflection is smaller with strongly damped than with slightly 
 damped magnets, apart from the delicacy of the galvanometer. 
 (Compare the above-mentioned experiments with two galvano- 
 meters.) Moreover, the relation of the opposed — after-current 
 with the strength and duration of the polarising current is quite 
 of the same kind as in ordinary polarisation ^. 
 
 As is well known, du Bois-Reymond was the first to indicate that 
 the seat of this ' internal ' polarisation is by no means entirely con- 
 fined to the electrodes of the polarising current, but is spread over 
 the whole led-through region^. Since many porous materials, 
 when moistened with electrolytic substances, show an internal 
 polarisation, which may be explained as due to the formation 
 of ions, du Bois-Reymond applied this explanation to the in- 
 ternal polarisation found in led-through muscle and nerve, these 
 tissues being both porous and electrolytic (du Bois-Reymond, 
 Sect. 3). 
 
 By experiments with my conducting axis schema, I have demon- 
 strated and explained the phenomenon of internal polarisation, and 
 I have further shown that known physical conditions are effectual for 
 its production in the case both of nerve and of muscle. This former 
 explanation of mine receives now additional support from the fact 
 that it can also be applied to the extrapolar after-currents of nerve 
 and muscle. For brevity I denote by conducting axis schema, a 
 wire enveloped by an electrolytic material, into the envelope of 
 which a current is led, it being indifferent whether the envelope is 
 the twisted thread soaked in fluid used by Matte ucci, or the fluid con- 
 tained in a small tube used by myself. A wire thus enveloped in 
 fluid shows, just as muscle and nerve show, opposed polarisation 
 after-currents throughout its intrapolar region ^. (An exception 
 presents itself when the connection is unpolarisable, such as an 
 amalgamated zinc wire in solution of a zinc salt.) 
 
 • See du Bois-Reymond on aperiodic magnets. 
 
 ^ See Tigerstedt's paper on internal polarisation, p. 80 of this volume. 
 ^ ' Untersuchungen iiber thierische Elektricitat,' i. p. 377 ; ii. 2, p. 378. 
 
 * See Pfliiger, ' Archiv,' vi. p. 319, 
 
312 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 Before passing- on to the extrapolar phenomena shown by the 
 conducting axis schema, it is first necessary to supplement in some 
 points the theory and experiments previously set forth. 
 
 The polarisation after-airrents in ilie conducting axis schema 
 ( Kern leiterm odell) . 
 
 (a). Experimental. 
 
 Several circumstances made it desirable that the experiments? 
 carried out in 1871-72, upon the after-currents of the conducting 
 axis schema, should be repeated and supplemented. First, that 
 these had been made only with moderate strengths and long closures 
 of current ; consequently, having regard to du Bois-Reymond's experi- 
 ments, the effects of strong currents and short closures had still to be 
 investigated. In the second place, in my former experiments I broke 
 the polarising current and then closed the galvanometer circuit by 
 a key ; it was therefore desirable to arrange for a more rapid succes- 
 sion than this gave, so that the conditions immediateh^ following 
 the break might be determined. Finally, it remained to be seen 
 whether the use of the galvanometer with light magnet, which I 
 now employ, would bring into \'iew new phenomena, such, for 
 instance, as change of direction of after-currents. 
 
 For most of my experiments I used tubes filled with solution 
 of sulphate of zinc, and provided wdth side openings ; the tubes 
 were 4f mm. in internal diameter, and from 40 to 196 cm. in length^. 
 Through the tubes ^vere drawn platinum wires either very fine or of 
 ordinary thickness. The four electrodes were amalgamated zinc 
 w^ires with bayonet-shaped bent ends, thrust into the side tubes ; the 
 bayonet-shaped bends gave such support that the wires only reached 
 as far as the upper surface of the main tube. The alternate closure 
 of battery and galvanometer circuits was effected by means of the 
 previously described paraffin commutator. The battery was 1-18 
 zinc carbon elements. The delicacy of the galvanometer w^as dimi- 
 nished to the requisite degree by distancing its coils. 
 
 I have previously stated that, with a combination polarisable in 
 two ways, the extrapolar after-current is opposed in direction to 
 
 * In the latter case they were made up of several shorter pieces of tube joined by 
 elastic tubing. Some of my tubes have many side openings close to one another, 
 others have a few some distance apart. Long tubes were generall)' so made up that 
 the side openings were near one another at the two ends, far apart in the 
 middle. 
 
PHENOMENA OF MUSCLE AND NEKVE. 313 
 
 the polarising" current, and that it is connected with the actual pro- 
 duction of two oppositely polarised axial regions ; that, on the 
 other hand, with one-sided polarisation the extrapolar after-current 
 is in the same dii'ection as the polarising* current. The first con- 
 dition is present when a platinum wire is immersed in sulphate of 
 zinc solution or weak sulphuric acid, and the polarising- current is 
 led in at two points in the fluid ; the second occurs when a 
 copper wire is immersed in weak sulphuric acid, and the current is 
 led in at two points in the fluid, also when a platinum wire is 
 similarly immersed^ and one leading-in electrode is connected directly 
 with the wire ^. 
 
 These two main facts were supported by the new experiments. 
 In addition, however, the following' facts were ascertained : — 
 
 (i) With platinum wire in zinc sulphate solution light magnets 
 show an extrapolar after-current which is exclusively — on both 
 sides, provided the polarising' current is of moderate strength ; when 
 strong-er currents are used, this — current is preceded, on the anodal 
 side, by a rapid + effect, in consequence of which the prevailing* 
 — after-current appears to be much weaker here than on the 
 kathodal side, where no such previous effect appears. When very 
 strong currents are used the + effect on the anodal side may quite 
 overpower the — one. 
 
 (2) With platinum wire in weak sulphuric acid the extrapolar 
 after-current generally remains — even with the strongest 
 currents. The arrangements for this experiment in which a fluid, 
 other than zinc sulphate, is employed, have been most advantage- 
 ously modified so as to do away with the tubes bound round with 
 membrane (see Pfliiger, ' Archiv,' vi. p. 319). The side openings of 
 the experiment tube having been filled to the brim with zinc 
 sulphate solution, were closed by elastic tubing caps into which 
 glass plugs were fitted ; into the four side openings used for leading 
 in and off", amalgamated zinc wires instead of glass plugs were 
 introduced and fixed. The main tube was now turned through 1 80° 
 so that the side tubes pointed downwards, and the zinc solution was 
 poured out of the tube, the side tubes alone remaining full ; the 
 fluid to be experimented with was now injected or aspirated into the 
 tube : it was thus brought above the zinc without disturbance of 
 any kind. 
 
 (3) With a very thin envelope of fluid surrounding the platinum 
 
 ' See the method of experiment as described in Pfliiger, ' Archiv,' vol. vi. p. 312, 319. 
 
314 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 wire the extrapolar currents are + on both sides of the polarising 
 current. For experiments of this kind I used Matteucci's arrang-e- 
 ment, i.e. platinum wires surrounded with twisted cotton soaked in 
 sulphate of zinc ; leading* in and off was effected by means of 
 amalgamated zinc wires bent into hooks and applied to the stretched 
 wire. I have also used two cables made up of many fine platinum 
 wires which behave exactly like the single covered wire. A very 
 long, fine platinum wire, thickly covered with fine silk, was cut 
 into a number of equal lengths, and these bound together so as to 
 make a cable, which was then itself covered with silk ; one such cable 
 contained 30, the other 100 wires (those of the first were 28, of 
 the second 57 cm. long). Both cables, after soaking well in sulphate 
 of ziuc, gave on each side of the polarising current + extrapolar 
 after-currents. The same results were obtained after soaking the 
 single wire, or the cable, in other fluids. Kaolin electrodes were 
 used for leading in and off". 
 
 (4) The ditference, stated above, in the behaviour of thin and 
 thick fluid envelopes, and the theoretical considerations connected 
 with this diflTerence, made it desirable to observe the changes in the 
 behaviour of the extrapolar after-currents with varying thick- 
 nesses of fluid envelope. 
 
 I have not as yet been able to accomplish this in the case of the 
 tubes and wires. The end seemed more attainable in those cases in 
 which the conducting combination had a wide-spread surface. To 
 obtain this, I covered the floor of a Bunsen's mercurial bath with 
 mercury, and poured upon this a thin film of sulphate of zinc. For 
 leading off" and in the zinc wires dipping into the fluid were not 
 used, as these were liable to disturb the mercury, but tube electrodes 
 were employed, the kaolin points of which, saturated with zinc 
 sulphate, rested upon the fluid. 
 
 With axially arranged leading in and off* contacts, the entire 
 extrapolar surface of the fluid showed an after-effect similarly 
 directed to the polarising current -f . As the film of fluid was 
 gradually increased in thickness, the after-effect in the neighbour- 
 hood of the electrodes became — or -|- — , at a distance from the 
 electrodes it remained 4- , by further increase in thickness, the after- 
 effect everywhere became + — , or — only. There is, therefore, a 
 certain thickness of fluid with which the extrapolar after-current 
 shows at a certain distance from the electrodes a ' turning-point,' or 
 ' line,' inside which it is — , outside + . The thickness in question 
 is dependent upon the strength of the polarising current ; reversal 
 
PHENOMENA OF MUSCLE AND NERVE. 315 
 
 of current gives predominantly double, + — , effects. The obser- 
 vations are very variable, owing- to inequalities in the surface of the 
 mercury, so that I have not followed out these experiments further. 
 They sufficiently establish the above-mentioned ' turning point.' 
 
 (b). Theoretical. 
 
 I have previously shown that a current led through a portion of 
 the schema spreads, on account of resistances round the core, very 
 widely into the extrapolar regions, so much so that + currents are 
 indicated in galvanometer circuits connected with the schema at 
 considerable distances from the led-th rough part ; these correspond 
 to the electrotonic currents of nerve and muscle. At the same 
 time polarisation occurs through the formation of ions round the 
 core, the distribution of which may be expressed by a double curve. 
 This curve has its positive maximum at the anode of the polarising 
 current, the negative maximum at the kathode ; it cuts the abscissa 
 at a point in the intrapolar region (the indifference point), and 
 approaches the abscissa in the extrapolar regions asymptotically. 
 
 This distribution of polarisation is the key by means of which 
 the physiological effects described as electrotonic appearances become 
 intelligible. 
 
 On breaking the polarising current this polarisation equalises 
 itself by inner currents, which are indeed present during the passage 
 of the led-through currents and can weaken or strengthen the latter 
 according to their direction. 
 
 If we take an extrapolar region by itself without reference to 
 the rest of the conducting schema, then this must, as is easily 
 seen, show a + after-current (i.e. in the same direction as the polaris- 
 ing current). On the side of the anode, for instance, the core is, in 
 the extrapolar region, polarised positively (the platinum wire, if in 
 zinc sulphate, being covered with zinc), in less and less degree 
 the fm-ther from the anode, consequently every point nearer the 
 anode is positive to one further off. A leading-off circuit connected 
 with this region will therefore lead off a current directed through 
 it, from the anodal to the further point of connection ; in the same 
 direction, that is to say, as the polarising current has in this circuit. 
 Similarly, on the side of the kathode, every point nearer the kathode 
 is negative to one further off (in the schema becoming covered with 
 SO4, i.e. sulphuric acid and oxygen), so that here again a circuit 
 must lead off a + current. 
 
316 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 This relation is present in all cases where there is no disturb- 
 ance from the formation of oppositely polarised reg-ions ; above all, 
 in cases in which the current is led through the schema with one 
 ]eading"-in electrode attached directly to the core. On each side 
 of the remaining- sing-le electrode the core is similarly polarised, 
 and this in reg-ularly decreasing- degree, so that on each side of the 
 electrode a current is found which, without exception, is directed 
 either from or to it, according as it is an anode or a kathode (taking- 
 the direction of the current in the leading'-off circuit). I have con- 
 vinced myself by special experiments that it is indifferent with which 
 end of the core the other pole of the polarising current is joined. 
 It is easily conceived that in this simplest ' unipolar ' case the dis- 
 tinction between intra- and extra-polar leading--ofF, and the descrip- 
 tion of after-currents as + and — are inadmissible, since both 
 expressions change when the polarising circuit is directly joined first 
 to one then to the other end of the core, without any variations 
 being- manifested in the after-currents. 
 
 A second case in which the extrapolar relations appear undisturbed 
 as above, is that in which the combination is only capable of polari- 
 sation on one side, and the extrapolar leading--ofF is on the side of 
 the polarisable electrode. 
 
 A third case will be mentioned later on. 
 
 These extrapolar conditions, causing- currents which, for short, I 
 may be allowed to describe as ' idiopolar,' will be disturbed by the 
 presence in the core of another oppositely polarised region. The 
 two regions seek to equalise their polarisations by currents which 
 run through the fluid from the anodal to the kathodal parts, and 
 which are thus when led-off" — in character (i.e. directed through 
 the leading-off circuit in opposite manner to the polarising-current) ; 
 these I propose to describe as ' bipolar ' currents. Eleven years ago 
 I proved experimentally that these opposed currents were connected 
 with the presence of two oppositely-polarised regions of the core. 
 
 In the intrapolar region, the bipolar current evidently must 
 always be directed similarly to the idiopolar currents which would 
 be shown on either side of the indifference-point in each half of the 
 region. In the intrapolar region, therefore, the after-currents under 
 all circumstances, as well in combinations, polarisable on one side only 
 as in these, must be — , and this is borne out by all experiments. 
 
 Much more complicated are the conditions of the extrapolar 
 regions with doubly polarisable combinations. Here the bipolar- 
 current is opposed to the idiopolar, and the ultimate condition must 
 
PHENOMENA, OP MUSCLE AND NERVE. 317 
 
 therefore, depend upon the extent to which the former can spread 
 into these regions. It is evident that this state of affairs must be at 
 a minimum for the distance in which the extrapolar polarisation is at 
 its maximum. The maximal polarised reg-ions in themselves will 
 have no idiopolar-currents, but only currents between one another, 
 which, if led off, would be — in character, just as two pieces of 
 metal in fluid give a considerable bipolar-current. Mere observa- 
 tion readily shows that when the fluid cylinder is large, the depo- 
 sition of zinc on the one side, and the development of gas on the 
 other, spreads, even with moderate strength of current, far out into 
 the extrapolar regions, whereas, on the other hand, in the intra- 
 polar region both are invisible quite close to the electrodes, provided 
 that this region be not very long. There can be no doubt 
 that where the ions are visible the maximum of polarisation has been 
 reached, and that this maximum may stretch still further out. But 
 on account of the good conduction of the core, the bipolar-current 
 must spread in the fluid beyond the region of maximal polarisa- 
 tion, the effect of which is that the surfaces of potential proceeding 
 from the maximal polarised portions, independent of fhe decreasing 
 tension due to the proper polarisation of the parts superficial to the 
 core, run almost concentrically along the core, and in consequence 
 the idiopolar-GUYYent of the extrapolar region must be almost or 
 quite concealed. As a matter of fact I have never succeeded, after 
 numberless experiments with tubes i^ metre long, and with the 
 finest wire core, in ascertaining a ' turning-point ' in the extrapolar 
 region, although such a point certainly must theoretically exist. The 
 bipolar-current, when it is present, spreads so far into the extra- 
 polar region that on reaching a position where the idiopolar 
 current alone might occur, the polarisation effects have dwindled 
 almost to nothing. (See Pfliiger, ' Archiv,' vol. vii. p. 317.) 
 
 It is quite otherwise in the case of very thin envelopes of fluid. 
 In the first place, the fall of polarisation effect is here much steeper, 
 since the intensity of the lines of current entering the core must 
 decrease much more rapidly in proportion to their distance from the 
 led-through region, this more rapid decrease being due to the much 
 more marked inequalities in their length. In the second place, 
 a greater resistance is offered to the passage of the bipolar currents, 
 and moreover, the conditions are not favourable for their extrapolar 
 spread. In this case, too, the distribution of tension upon the 
 surface of the thin fluid envelope must be a true expression of 
 the distribution upon the core itself. So then here the idiopolar 
 
318 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 must come into operation with but slig-ht disturbance, and the 
 ' turning'-point ' must nearly coincide with the electrode itself. It 
 is thus that wires with thin fluid envelopes give + extrapolar — 
 intrapolar after-currents. 
 
 It mig-ht thus be predicted that if the thickness of the fluid 
 envelope could be increased by successive steps, then, with a ' two- 
 sided ' polarisable combination, a ' turning'-point ' would be found 
 in the extrapolar region, which would rapidly move to a greater 
 distance from the electrode in proportion as the envelope increased 
 in size. This prediction is fufilled by the results of the above- 
 mentioned experiments with the mercury trough. 
 
 Considering the unavoidable conflict in all cases of ' two-sided ' 
 polarisation between the idio- and bi-polar currents, it is not sur- 
 prising that after-currents are found directed both ways, although 
 it may be difficult to explain the phenomena with exactitude. An 
 extrapolar after-current, first + and then — , such as occurs with 
 currents of medium strength at the anode of a platinum zinc 
 sulphate schema, would thus mean that in the first moment after 
 the polarising-current was broken the idiopolar current had the 
 mastery, but that in a short time it gave way before the bipolar 
 current ; whilst with stronger currents only the latter can be 
 observed ^ "With certain exceptions the decrease in the extrapolar 
 tensions obviously offers a ground for this expression of the state 
 of affairs ; but its actual truth might be shown by special experi- 
 ments, for which however there is no urgent present necessity, 
 since, as we have seen, this special condition does not occur in the 
 case of nerve and muscle. The fact that the biphasic extrapolar 
 after-current of the platinum zinc sulphate schema is more easily 
 produced on the side of the anode than on that of the kathode is 
 possibly connected with the covering of the wire on the kathodal 
 side by a badly-conducting gaseous layer. In favour of this view 
 is the fact mentioned previously that with a platinum in sulphuric 
 acid schema, in which there is development of gas at both poles, 
 the phenomenon appears but seldom and then only in a very rudi- 
 mentary manner. 
 
 Here, then, I conclude the investigation of the polarisation 
 after-currents of the schema. To give a complete theoretical re- 
 presentation of the tension and current distribution would be im- 
 
 * This may be expressed also as a moving of the 'turning-point' further out. 
 The connection of this phenomenon with one long observed by me in such schemata 
 will be brought forward in another treatise. 
 
PHENOMENA OF MUSCLE AND NERVE. 319 
 
 practicable, as everyone versed in the mathematical laws of the 
 subject will admit, since for its expression it would be first of all 
 necessary to state the distribution of polarisation upon the surface 
 of the core at the moment of break. This distribution, however, 
 changes with iJie very currents which tend to equalise matters, and 
 these currents themselves vary under the influence of the resistance 
 of the gaseous ions. The principal points are, I believe, sufficiently 
 explained by the foregoing experiments and theoretical considera- 
 tions, I therefore refrain from giving the approximate distribution 
 charts which I have prepared of the chief cases. 
 
 Let us now go back to muscle and nerve and see what polarisa- 
 tion after-currents are to be expected in these tissues. 
 
 The polarisable combination of core and envelope in muscle and 
 nerve is, as I have previously shown ^, a ' doubly polarisable ' one. 
 We infer this — 
 
 (i) From the approximately equal size of the electrotonic current 
 branches in the anodal and kathodal regions, 
 
 (2) From the existence of two separate oppositely polarised 
 regions, with intermediate indifference-point, as testified to by the 
 well-known excitation phenomena of an- and catelectrotonus. 
 
 The envelope may be regarded as very thin, the conductivity of 
 the core as in no way relatively great. Without doubt, therefore, 
 with these tissues the state of matters is this, that the polarisation 
 bipolar after-current does not spread into the extrapolar regions. 
 So that these only show the idiopolar after-current as in the case 
 of the platinum wire schema with thin fluid envelope. 
 
 The polarisation after-current in muscle and nerve must therefore 
 he — {opposed to polarising-current) in the intrapolar, + in the extra- 
 polar region. 
 
 B. The after-currents proceeding from the break-excitation. 
 
 A number of facts have established that it is the anelectrotonic 
 region which is excited both in muscle and nerve at the moment 
 of break, and that this region may remain in prolonged excita- 
 tion on the break of a very strong current or one of long duration. 
 Notably, Pfliiger has shown, by a beautiful experiment, the con- 
 nection of Ritter's breaking tetanus in nerve with the excitation 
 produced by a disappearing anelectrotonic state ; in muscle, indeed, 
 
 ' See Pfliiger, ' Archiv,' vols, v., vi., vii. 
 
320 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 the prolong-ed breaking excitation reveals itself at the anode by 
 the bulg-ing- of the tissue ^. 
 
 There is no doubt that this excitation is strongest at the anode 
 and decreases in intensity in i:)roportion as the distance from the 
 anode increases. According, therefore, to the law^ of currents of 
 action, nerve and muscle, after the break of a sufficiently strong 
 current, must be the seat of currents of action which are directed in 
 the tissue away from the anode on both sides. 
 
 For completeness it may be further stated that another current 
 of action may be i:)resent after the break of a current of veiy short 
 duration, namely, the action-current produced at the kathode by 
 the make of the polarising-current. This will undoubtedly be 
 present during the earl}^ part of the time of closure, and there is no 
 decided reason why it should not continue longer than the polaris- 
 ing current if the time of closure of the latter is very short ; it 
 certainly would not be of marked character, since, in the first place, 
 the polarisation opposed current is unfavourable to the persistence 
 of the kathodal excitation (the kathode now becoming the anode) ; 
 and, in the second place, I have never been able to see after short 
 closures any bulging of muscle at the kathode, which would remain 
 synchronous with a bulging at the anode. 
 
 This kathodal action-current would evidently be opposed to the 
 polarising current when intrapolar, + W'hen extrapolar; it has, 
 therefore, in all cases the same direction as the polarisation after- 
 current, so that it could be differentiated from the latter at most 
 by its longer duration. As however such long duration is not 
 observable we may omit this current, and therefore it will be disre- 
 garded in what follows. It will be referred to once again later on. 
 
 C. Combination of the polarisation and excitation after-currents. 
 Exjplanation of the secondary electromotive 'phenomena. 
 
 The anodal current of action is similarly directed to the polarising- 
 current {i.e. +) in the intrapolar, oppositely directed {i.e. — ) in the 
 extrapolar region, it compjletely fails in the kathodal extrapolar region. 
 Whenever it appears, it is, therefore, oppoosed to the pjolarisation after- 
 currents and comes into co7iflict with these. 
 
 The whole relation is seen at a glance by means of the 
 following schema, in which the arrow P indicates the polarising 
 current, A and C the anode and kathode. 
 
 ^ See ' Handbuch/ Hermann, vol. i. p. 93. 
 
PHENOMENA OF MUSCLE AND NERVE. 331 
 
 Polarisation after-currents 
 (of short duration) 
 
 p" P P P' 
 
 Action after-currents in nerve ... .«_ <_ — > ' 
 
 (of long duration) a a a 
 
 Action after-cuiTents in muscle ... ^~ ~^, 
 
 a a' 
 
 It lies in the nature of the excitation after-current that it should 
 he much more prolonged than the polarisation after-current, which 
 as well in the tissues as in the conducting- axis schema, in all cases 
 rapidly disappears. Nothing- is move natural therefore than that 
 the conflict of the two currents should express itself in the double 
 character of the after-currents actually observed, the first phase of 
 which always indicates the polarisation, the second the action- 
 current. Further, the more the action-current is developed in 
 comparison with the polarisation-current, the more will the first 
 phase of the doubly-directed current tend to disappear, as also 
 must be the case when the magnet has a slow swing. 
 
 We have therefore to expect that — 
 
 (i) The intrapolar region should show an after-current which is 
 first — (opposed to the polarising-current), and then + : 
 
 (2) The anodal extiapolar region an after-current first + and 
 
 then — : 
 
 (3) The kathodal extrapolar region an after-current which is 
 
 always + . 
 
 Now the relations established by experiment agree in all points 
 with these. That given above under i is the actual condition as 
 stated in du Bois-Reymond's treatise, those given under 2 and 3 
 were stated previously by myself and A. Tick, and are fully estab- 
 lished by the present investigations. 
 
 It is further readily intelligible that the intrapolar action- 
 current, du Bois-Reymond's second positive phase, should only 
 appear after currents which are strong and of short duration ; 
 strong, because only then is the breaking excitation sufficiently 
 prolonged to overlap the polarisation after-current, and of short 
 duration, because the latter current is not so markedly developed by 
 short as it is by long closure, and thus the complete concealment of 
 the action-current by the polarisation is not so easily produced.^ 
 
 Further proof of the correctness of my explanation is furnished 
 by the circumstance discovered by du Bois-Reymond himself; that 
 his positi\e phase, our current of action, is intimately connected 
 
 Y 
 
322 THE SO-CALLED SECONDAKT ELECTROMOTIVE 
 
 with the vitality of the tissue, whilst the first negative phase, our 
 polarisation-current, occurs in dead tissues so long* as their structure 
 is not destroyed. The unproductive speculations of du Bois-Rey- 
 mond connected with these relations are thus very simply settled. 
 
 Finally, the most con\ancino> proof lies in the fact, quite over- 
 looked by du Bois-Reymond, and now established by myself, that 
 the intrapolar + phase of muscle is connected in the most pro- 
 nounced manner with neig-hbourhood to the anode, whilst there 
 is no trace of the — phase there. It is ob\aous that the in- 
 difference-point, even with the strongest currents, does not move 
 so far towards the kathode in muscle as in nerve ; so also the 
 sjireading- of polarisation and the electrotonic currents do not 
 extend so far in muscle as in nerve. 
 
 It is therefore not to be wondered at, as has been already remarked, 
 that the + phase in nerve should appear equally well in all parts 
 of the intrapolar region, since with currents of the strength used 
 iu these experiments the indifference-point is brought, as Pfiiiger 
 has shown, quite close up to the kathode \ 
 
 This is the place to mention a set of experiments which I have 
 conducted on consideration of the fact that Bitter's tetanus is pro- 
 duced even by currents of moderate strength when these are kept 
 closed for some time, half an hour for instance. In such a way I 
 hoped to be able to bring out the + intrapolar phase with weak 
 currents (one zinc carbon cell). The nerve was left attached to the 
 leg so that the breaking tetanus might be present as a control. But 
 after half-an-hour's closure a Tery strong and abnormally long — 
 deflection was all that appeared during the violent breaking tetanus ; 
 that is to say, the long closure had developed polarisation of a much 
 more marked and more slowly vanishing character than would 
 otherwise have been the case, so that this fully compensated the 
 current of action. This experiment however shows the interesting 
 result that Hitter s tetanus is connected with a polarisation of the 
 nerve which remains for a long time after the hreah. 
 
 The above statements also disfavour the theory of the breaking 
 excitation recently set forth by Tigerstedt and Griitzner ^, which, 
 after reading Griitzner's detailed work^, I hold to be erroneous. 
 We see after the short closure of a strong current, the intrapolar 
 current of action immediately appearing ; that is to say, the 
 
 ^ See previous remarks upon electrotonic spread. 
 
 ' See my previous remarks in Pfiiiger, ' Arcliiv,' vol. xxxi. p. 99. 
 
 ' Pflfiger, ' Archiv,' vol. xxxii. p. 357. 
 
PHENOMENA OF MUSCLE AND NEEVE. 323 
 
 breaking excitation here lasts mueli longer than the polarisation 
 after-current. The breaking excitation cannot therefore be directly 
 dependent at any rate upon the establishment of this polarisation- 
 current. 
 
 To this it may be objected that the long prevalence of an action- 
 current does not prove that a polarisation-current is not present, it 
 may be there all the time but overpowered. On the kathodal 
 side, however, we see how quickly, in all cases, the polari- 
 sation-current vanishes. Further, so long as the action-current 
 lasts, even though a portion of the polarising-current be present, 
 so long is the anodal region practically one where current enters, 
 not leaves, as the above theory requires. The break-excitation is 
 therefore dependent upon changes which the current has left in the 
 tissue upon the disappearance of positive polarisation, whether this 
 disappearance occur by a polarisation opposed current or by any 
 other process of removal of electrolytic products. A number of 
 objections to Griitzner's last publication may be reserved for a 
 later opportunity ^. 
 
 The extrapolar phenomena of muscle and nerve can be as com- 
 pletely explained as the intrapolar. The facts discovered by me 
 some time ago with the multij^lier, namely -1- after-current in the 
 kathodal, stronger — after-current in the anodal extrapolar region 
 of nerve, are now recognised as polarisation and action-currents 
 respectively ^. By using a lighter and more aperiodic magnet we 
 
 * A few points only I must mention here. First, with regard to the experiments 
 made by me in 1875 upon the connection of the polarisation opposed current with 
 the breaking contraction, and now first communicated. These experiments, as I can 
 show most accurately from the records, were made with leading-in contacts at a 
 great distance from the cross-section. Happily the positions of the electrodes were 
 marked and fixed on the myograph papers which have the muscle tracings. In the 
 second place, the conclusion which I draw from these experiments is not of recent 
 date, but was arrived at eight years ago, the whole experiment being, indeed, conducted 
 with this in view. I am able, by documents, to prove this to any who may 
 be interested in the matter. Its present publication is due to the fact that an investi- 
 gation into the nature of nerve-excitation, reaching back to the year 1871, and for 
 which I have collected a mass of experimental material, has only recently appeared 
 to me sufiiciently advanced to permit of its publication ; some part has already 
 appeared in this 'Archiv.' In the third place, Griitzner has cited a discrepancy of 
 Tigerstedt's as in disagreement with my law of the polarisation increment. I have 
 already shown in another place the insufiiciency of the discrepancy — a fact which 
 Griitzner cannot have known (Hofmann and Schwalbe's ' Jahresbericht,' 1882, ii. p. 19.) 
 
 ^ Following out a suggestion made by Pfliiger, I assigned this latter explanation, 
 eleven years ago, to the phenomena of the catelectrotonic region, as I then erroneously 
 believed that the extrapolar polarisation currents in nerve must be — in character, 
 being led to this by my experiments with the schema having an envelope of thick 
 fluid (see Pfltiger, 'Archiv,' vol. vi. p. 357). 
 
 y 2 
 
324 THE SO-CALLED SECONDARY ELECTROMOTIVE 
 
 have leamt that the former rapidly disappears, the latter remains 
 for some time persistent ; and, further, that the latter is preceded 
 by a rapid + condition, which is the polarisation-current of the 
 anodal extrapolar region. Moreover, we have established the same 
 laws for the phenomena in muscle also. There is here only a 
 quantitative difference. In nerve the anodal — action-currrent is 
 so strong that, although it has to contend with a -j- polarisation- 
 current, it is still able to exceed in magnitude the undiminished 
 -f polarisation-current of the kathodal region. In muscle this is 
 not the case ; the action-current is indeed strong, but when weak- 
 ened by the polarisation-current, it sinks below the magnitude of 
 the full polarisation-current of the other region. In the schema 
 (given under C) this may be indicated by stating that in nerve 
 a' — p' > p'', or if p'= p", then a' > 2 p' ; whilst in muscle, a' — p' < p'', 
 ora'<2p'. The weak development of the extrapolar breaking- 
 excitation in muscle may be due either to a weak development 
 or to a rapid disappearance of the anelectrotonic state. It is note- 
 worthy that we find the intrapolar spread of anelectrotonus to be 
 less in muscle than in nerve. 
 
 In his second publication, on the extrapolar after-currents of 
 nerve, Tick suggests that the kathodal after-current, which he had 
 stated in his first communication to be — , like the anodal, might 
 possibly be — in the first moment after opening, and then -|- , thus 
 being double in character. This suggestion now appears to be a 
 faulty one. As a matter of fact it is the anodal and not the 
 kathodal-current which is double in character, being first + and 
 then — . We are able now to substitute for the view, which 
 led Fick to this suggestion, something more certain, viz. the 
 complete explanation of the causal relations. 
 
 Finally, we obtain some explanation of a long-known pheno- 
 menon. Pfliiger by excitatory'^, and du Bois-Reymond^ by gal- 
 vanometric methods, found that during the closure of a polarising- 
 current the extrapolar anelectrotonic state tends to increase, the 
 catelectrotonie state to decrease. This may be now fully explained 
 as follows. The extrapolar polarisation (as in the metal and fluid 
 envelope schema, vide Pfliiger, ' Archiv,' vol. vi. p. 320) must 
 strengthen the extrapolar currents on both sides the longer it 
 continues. On the side of the kathode, however, we have at the 
 commencement of closure an action-current of similar direction 
 
 * ' Untersuchungen iiber die Phyfiologie des Electrotonus,' p. 319. 
 ^ 'Archiv ' f. Anat. u. Phygiol. 1867, p. 441. 
 
PHENOMENA OF MUSCLE AND NERVE. 325 
 
 to the polarising- current (see previous remarks) which tends 
 rapidly to decrease. In consequence of this the kathodal current 
 must first become stronger and then tend to decrease. The 
 phenomena, during- closure of the polarising--current, are there- 
 fore no less in accordance with the principles laid down than are 
 those followino- its break. 
 
 6. Concluding Remarks. 
 
 The foreg-oing- investig-ation has accomplished the task of ex- 
 plaining the phenomena of intrapolar after-currents as given by 
 du Bois-Reymond ; it has also introduced new facts ; and further, it 
 has widened our knowledg-e of the extrapolar after-currents, has 
 extended these to muscle, and fully explained them. 
 
 There remain unnoticed, (with the omission of facts connected 
 with electrical organs, in regard to which I could make no 
 investig-ation, and of the after-currents which, at any rate, allow 
 of explanation by means of our principles), certain statements of 
 du Bois-Reymond as to the way in which ' positive polarisation ' is 
 favoured by the use of currents, the direction of which coincides 
 with the natural direction of the excitatory wave ; upon these 
 phenomena the author lays very great weight (for muscle, see du 
 Bois-Reymond, Sect. lo ; for nerve. Sect. 17). 
 
 In respect to nerve, du Bois-Reymond's own statements are 
 dubious, but in respect to muscle they are pretty decisive, although, 
 strangely enough, the most conclusive method of experiment has 
 not been employed, viz. that of leading through the entire muscle 
 and leading-ofF the after-currents from two intrapolar regions in 
 the two muscle-halves at the same time. But, before acquiescing 
 in the far-reaching conclusions which du Bois-Reymond draws from 
 these phenomena, it might be well to inquire what share the above 
 enunciated simple laws, which have entirely escaped du Bois-Rey- 
 mond's notice, may have in the production of the phenomena. I 
 prefer, however, to leave this inquiry to du Bois-Reymond himself, 
 merely hinting here at conditions which would suffice at once to 
 explain the whole of this wonderful thing. 
 
 Du Bois-Reymond had, at the extremities of both muscles, fibres 
 of which the ends were for the most part injured, as his intimations 
 with regard to the currents of rest show. This circumstance which 
 
326 THE SO-CALLED SECONDAKY ELECTROMOTIVE 
 
 has SO often proved a source of danger ^, must, in accordance with 
 the above-disclosed law, more or less impair the development of the 
 + after-current, in the case of abterminal (from dead to living-) 
 direction of current, so that it would appear to be favoured in the 
 case of atterminal direction of. current, i.e. a direction like that 
 of the excitation wave. 
 
 There is, therefore, no subject in the whole range of electro- 
 physiology in which the connection of the phenomena is more 
 transparent than that of the ' secondary electromotive phenomena,' 
 which now extend to the extrapolar region. Du Bois-Reymond's 
 ' curiosity' as to what 'auxiliary hypothesis' I shall advance to dis- 
 pose of his positive polarisation-currents, will now be fully satisfied. 
 The question, how far it is true that ' all that I have discovered 
 concerning electrotonus ' has ' been exploded,' that ' the elucidation 
 of electrotonus must be approached anew,' and, finally, that I 
 have been ' compelled ' to ' import some changes into my scheme,' 
 as to this everyone can now judge for himself. Happily none 
 of these suggestions are true. The confusion which the mole- 
 cular theory caused, in that it overlooked, whilst expatiating, what 
 was most obvious, is luckily done away with now in this department 
 of the subject, as it has long been in other departments, and every- 
 thing has so turned out as to furnish fresh support to my expla- 
 nation of the natu.re of the phenomena of animal electricity. When 
 du Bois-Reymond deplores that he did not, in 1867, disclose to me 
 the fact ' that there is positive polarisation in the intrapolar region,' 
 because then ' the course of science in this direction would have 
 been difierent, and perhaps more fruitful,' I can only agree with him, 
 for then, possibly, we should have come to an understanding years 
 ago as to this ' positive polarisation,' and du Bois-Re}- mond would 
 have allowed the molecular theory, of which this is the last prop, to 
 fall. 
 
 The molecular theory owes its production to the error of fact 
 that uninjured muscles show as strong a muscle-current at their 
 natural cross-section, as injured muscles do at their artificial cross- 
 sections. When this was found to be an error, the theory was not 
 given up, but the mischievous compromise of ' parelectronomy ' was 
 invented. Then, when I found the total absence of current in 
 uninjured tissues, and parelectronomy no longer sufficed, (the ' cold 
 hypothesis ' brought to its aid being false as to its facts.) still the 
 
 * See for instance Pfliiger, ' Archiv,' vol. xvi. p. 243. 
 
PHENOMENA OP MUSCLE AND NERVE. 327 
 
 theory was not abandoned, but every inconvenient case of absence 
 of current was, on various pretexts, explained away. The mole- 
 cular theory next introduced untold confusion into the subject of 
 action-currents, where it literally turned things upside down ; errors 
 of fact had to be first removed, and then new facts discovered be- 
 fore this confusion could be got rid of, and before this department 
 of the subject could be so cleared up that its law, like that of the 
 current of rest, could be expressed simply. 
 
 The molecular theory was especially unhappy in its application 
 to electrotonus. It was incapable of explaining either the galvanic 
 or the excitatory phenomena, although it pretended to explain the 
 former and held out the prospect of explaining the latter, together 
 with the function both of nerve and of muscle in its entirety, in the 
 future. Now it has come to this, that on the one hand the theory 
 claims to have discovered a fact basis for the supposed intrapolar 
 arrangement of molecules (of course, only by disregarding obvious 
 explanations), while on the other it abandons its explanation of 
 extrapolar currents, notwithstanding that this intrapolar rotation 
 was invented (it need scarcely be said without any basis of fact, 
 or discussible theory) for the very purpose of explaining the extra- 
 polar currents. 
 
 The theory was equally unfortunate in its application to the currents 
 of the skin and secreting organs ; here it has favoured the setting 
 up of an almost inverted law of secretion-currents, founded on a 
 false conception of action-currents in nerve and muscle, and so has 
 led to results which are wrong in fact. Finally, it has shown 
 itself an utterly incapable guide in the study of electrical fish, the 
 best that it could do being to bring into existence an easily demo- 
 lished fabric of rash speculations. I will not refer to the 
 currents of plants and parenchyma which, as they are in the highest 
 degree inconvenient to the molecular theory, are wilfully neglected, 
 or of which the facts are misrepresented. 
 
 I challenge the mention of a single statement contained in this 
 indictment against the molecular theory which I am not able to 
 substantiate and confirm by quotations. I maintain, however, that 
 this theory which was sufficiently dangerous to lead its author (the 
 creator of this rich field, the discoverer of its methods, the man who 
 has helped to teach his contemporaries in medicine to think in 
 accordance with physics) from error to error, that this theory must 
 now at length be given up. It must be given up completely ; all 
 toying with it in handbooks must cease, even at the cost of losing a 
 
328 THE SO-CALLED SECONDARY ELECTROMOTIVE, ETC. 
 
 couj^le of pages of elegant schematizing. Nothing is so elegant as 
 the simple truth. 
 
 But is there a single subject in physiology to which more simple 
 and beautiful explanations are applicable, even to the finest details, 
 than that of electro-physiology ? 
 
 Four short, and more or less interdependent, propositions suffice 
 to explain all the galvanic phenomena, with the exception of the 
 shock of electrical fish and the ' action- currents ' of the retina, 
 neither of which are contradictory to them, although they may not 
 as yet, admit of exj^lanation by them. These propositions are : — 
 
 I. Localised death in continuity of protoplasm, whether caused 
 by injury or by metamorphosis (mucous, horny), renders the dead 
 part negative electrically to the unaltered part. 
 
 1. Localised excitation in continuity of protoplasm renders the 
 excited part negative electrically to the unaltered part. 
 
 3. Localised warming in continuity of protoplasm renders the 
 warm part positive, localised cooling the cold part negative to the 
 unaltered part. 
 
 4. Protoplasm is strongly polarisable on its limiting surfaces (first 
 shown as regards the protoplasm enclosed in tubes of muscles and 
 nerves) ; the polarisation constant decreases on excitation (and on 
 dying). 
 
 Besides, having disposed of the secondary-electromotive pheno- 
 mena, the above investigation has furnished some not unimportant 
 contributions to the physiology of muscle and nerve. Pfluger's 
 statements as to the movement of the indifference-point have been 
 verified for strong currents in a new manner, as well as the results 
 of Biedermann, of Engelmann, and of Van Loon as to the effect of 
 demarcation-currents upon the efficiency of ' physiological electrodes.' 
 The electrotonus of muscle intra- and extrapolar has been estabbshed 
 anew, and its complete analogy with the phenomena in nerve has 
 been shown, together with the quantitative differences as regards 
 its spread from the electrodes, especially from the anode. The 
 time relations of the electrotonic spread in nerve have been as 
 satisfactorily explained as the electrotonic after-currents. The 
 intimate connection of breaking tetanus with long persistent polari- 
 sation has been proved, and a new position in the investigation of 
 the nature of the breaking tetanus has been gained. 
 
 In relation to many further questions new methods of experiment 
 have suggested themselves which will be better discussed when 
 these experiments are published. 
 
ON 
 
 THE PHENOMENA OE INHIBITION 
 
 PKODUCED BY THE 
 
 ELECTEICAL EXCITATION OF STEIATED MUSCLES, AND 
 ON POSITIVE lATHODIC POLARISATION. 
 
 By Dr. WILHELM BIEDERMANN, 
 
 Extraordinary Fiofessor of Physiology in the German University, Prague. 
 Sitzungsberichte der kais. Akad. der Wissensch., Vienna. Part III. p. 142 (1885). 
 
CONTENTS. 
 
 PAGE 
 
 1. Introduction 331 
 
 2. Changes of form which muscle poisoned by Veratrine undergoes when 
 
 excited by. a Battery CuiTent ......... 333 
 
 3. The Secondary Electromotive Phenomena of Veratrine muscle excited 
 
 electrically 343 
 
 4. Positive Kathodic Polarisation in normal striated muscles . . .352 
 
1. Introduction. 
 
 ^ In the investigation of the secondary electro-motive phenomena 
 in the adductor muscle of Anodonta ^ the most striking- fact seemed, 
 that here, quite contrary to the usual behaviour of the striated 
 trunk muscles of the frog", not only positive anodic but also positive 
 kathodic polarisation is observed, for on leading off from the anodic 
 as well as from the kathodic half of the muscle traversed by the 
 current longitudinally, occasionally an after-current of the same 
 direction and frequently of very considerable strength appears, 
 without simultaneous internal polarisation of the intrapolar tract 
 of any importance at all. Hence, the inference is, that the cause 
 of this phenomenon is to be sought solely in alterations of the 
 substance of the muscle produced by the polarising current at its 
 points of exit. 
 
 The anodic persistent excitation which follows on opening a 
 current of corresponding intensity and duration of closing, explains 
 positive anodic polarisation in a thoroughly satisfactory manner, as 
 has been convincingly shown in investigations already published ^, 
 whilst the negative kathodic after-current, which, under ordinary 
 circumstances, is prominent in striated muscle, might be proved to 
 be in the main a result of the alteration of the kathodic spots of 
 fibre by the preceding closing excitation. 
 
 It is however intrinsically more difficult to attain a definite 
 
 ^ [This paper is the eighteenth in the series of 'Contributions to the general 
 Physiology of Muscle and Nerve,' which have been published during the last six 
 years by Professors Hering and Biedermann of Prague. Ed.] 
 
 2 As I have since learnt from a copy of Professor Bernstein's Dissertation, ' De 
 animalium evertebratorum musculis nonnulla,' Berolini, 1862, which he kindly sent 
 to me, but which was unfortunately inaccessible to me at the time when I wrote my 
 work upon the adductor muscle of Anodonta, he was the first to draw attention to 
 the peculiar ' Tonus ' of this preparation, and he had investigated its electromotive 
 actions also before Pick. 
 
 3 See No. VII and No. IX of this work. Also Hering, ' tjber Veriinderungeu 
 der Muskeln,' &c. Sitzungsber, vol. Lxxsviii. p. 41 5- 
 
332 INHIBITION BY EXCITATION AND 
 
 conception in regard to the cause of the negative anodic and posi- 
 tive kathodic after-curreuts, especially as both phenomena, as >vill be 
 shown, appear distinctly in striated muscle, only under certain 
 conditions. It seems that here, certain subsequent phenomena 
 of electrical excitation of tonically contracted muscles must be 
 taken into consideration in the first place ; these do not constitute 
 excitation phenomena in the ordinary sense, i. e. phenomena of 
 contraction, but much rather their opposite, that is, as relaxation 
 of previously contracted parts, caused by an inhibition of ex- 
 citation. 
 
 In this connexion I will only call to mind the phenomena of 
 inhibition which result on electrical excitation of the muscle of the 
 heart in systolic contraction, and which are so easily demonstrable 
 and at the same time so remarkable. In studying these, one is 
 necessarily led to the assumption of two processes antagonistic to 
 the polar phenomena of excitation which I have denoted as ' inhibi- 
 tion of closing and opening,' or ' anodic and kathodic inhibition,' 
 because the former has its orig-in at the point of entrance of the 
 current, and the latter at its point of exit ^. 
 
 I further called attention to the fact, that if we assume a per- 
 sistent tonic condition of excitation, then the result of the electrical 
 excitation of such a muscle in respect to the secondary electromotive 
 phenomena, must necessarily manifest itself under the circumstances, 
 as a positive kathodic after-current, that is to say as a negative 
 anodic one. Thus the strong positive kathodic polarisation, which 
 under fitting circumstances is observed uniformly on electrical 
 excitation of the adductor muscle of Anodonta, in consideration of 
 its very marked ' tonus,' would be also easily explained by the 
 hypothesis of a kathodic opening inhibition. The legitimacy of this 
 conception, with which moreover other facts stated in the work men- 
 tioned are in complete accordance, was nevertheless necessarily left 
 doubtful, since I had not succeeded in demonstrating the existence 
 of polar phenomena of relaxation in so unequivocal a manner in 
 the object named, as I had done in the muscle of the heart. Still 
 this circumstance could hardly be maintained as a fundamental 
 objection, because the tonus of this preparation is as a rule so 
 strong, that a local and not very marked relaxation might easily 
 escape demonstration by means of graphical methods, and all the 
 more as the changes of form follow only languidly and slowly. 
 
 My effort has since been directed to gain certain and mfallible 
 
 ' ' tjber das Herz von Melix ^omatia,' Sitzungsber, vol. Ixxxix. p. 19. 
 
KATHODIC POLAEISATION OF MUSCLE. 333 
 
 Bupporfc for the theoretical conception of the secondary electromotive 
 phenomena of the smooth muscle of Anodonta, and especially of 
 positive kathodic polarisation in it, by the investigation of objects 
 better suited to the purpose. 
 
 As there is such a regretable want of suitable organs with 
 smooth muscles, experiments on the heart, especially of inverte- 
 brate animals, seemed at first to promise a good result ; they fail, 
 however, chiefly on account of the small size of the object. The 
 heart is moreover very little suited for experiments on polarisation 
 by reason of the complicated course of the fibre. Thus the only 
 thing left was to consider the possibility of artificially putting 
 striated trunk muscles into a continuous condition of excitation, 
 lasting for a sufficient time, and comparable to the ' tonus ' of 
 certain smooth muscles, and during this condition to investigate 
 the changes of form resulting on electrical excitation on the one 
 hand, and the secondary electromotive phenomena on the other. 
 I have endeavoured to attain this end by poisoning with veratrine, 
 and I will communicate the results of my experiments in this matter 
 on the sartorius muscle of the frog, in the following paragraphs. 
 
 I must not omit to mention here, that at the suggestion of 
 Professor Hering, I made experiments several years ago, upon the 
 effect of electrical stimulation on muscles in persistent contraction ; 
 although I had repeatedly succeeded at that time, by closing an 
 electrical current, in producing a perceptible relaxation in muscle 
 preparations which had been previously put into persistent con- 
 traction by the action of vapour of ammonia, yet the results seemed 
 too uncertain to permit of making further deductions from them. 
 
 2. Changes of form which muscle poisoned by veratrine 
 undergoes when excited by a battery current. 
 
 Since von Bezold^ first determined accurately the remarkable 
 effects of veratrine upon striated muscle, they have been repeatedly 
 the object of detailed investigations. Their main result is the 
 theory that the extremely striking after-effect produced by every 
 stimulus, however short its action, is founded solely on a changed 
 condition of the substance of the muscle itself, and as Fick^ thinks, 
 
 ^ 'TJnters. aus d. Wiirzburger Laboratorium,' 1867. 
 
 ^ ' Arbeiten aus d. physiolog. Laboratorium der Wiirzburger Hochschule,' ii. Lief- 
 erung, 1873, p. 142 f. 
 
334 INHIBITION BY EXCITATION AND 
 
 depends probably upon an ' intensification of the process of excita- 
 tion beyond the normal amount.' 
 
 It is unnecessary to give an exact description of the vari- 
 ations in the contraction curve caused by veratrine poison- 
 ing-, because a detailed examination of them is found in many 
 works, and I have nothing- essentially novel to add. I shall there- 
 fore limit myself here to a brief discussion of the method of 
 poisoning which showed itself best suited to my purpose in the 
 course of the investigation. As it was of the first importance to 
 me to obtain a strong- contraction of the whole transverse section of 
 the sartorius, and esj)ecially that this should last as long as possible, 
 the most appropriate way seemed to be to poison the whole animal ; 
 thus the only question was to find the most suitable dose on the 
 one hand, and on the other to determine the point of time which 
 should be considered the most favourable for the jireparation of the 
 muscle. 
 
 The extraordinan^ sensitiveness of the muscle of the frog for 
 minimal amounts of the poison introduced through the blood, 
 made me refrain for a long time from using stronger doses, because 
 I feared a too rapid decrease of excitability. However, in the end, 
 the use of larger amounts with a shorter period of action proved 
 to be best. As some of the experiments took place at the begin- 
 ning of the warm season, when the muscles of Kana esculenta are 
 little suited on the whole for electro-physiological experiments as 
 I had found from previous experience, I used later almost exclusively 
 the muscles of R. temporaria. I had no ground for regret on this 
 account, for they proved to be much more excitable, and are 
 therefore better for the purpose than those of R. esculenta even 
 at the most favom-able season of the year. 
 
 Although, as Kolliker has already shown, a previous administra- 
 tion of curare is not sensibly prejudicial to the action of veratrine, 
 I confined myself as a rule to the use of the latter poison alone, 
 because certain appearances of movement which present themselves 
 in the muscles of the poisoned animal, enable one to recognise the 
 proper stage of the action of the poison with the greatest certainty. 
 However, I assured myself eveiy time by special control experi- 
 ments, that the results of the experiments about to be described are 
 in no way prejudicially affected by the simultaneous use of curare. 
 
 I introduced as a rule 6-7 drops of a i "/o solution of acetate of 
 Veratrine into the dorsal lymph sac, and killed the frog 10 minutes 
 afterwards at latest. Generally ^-'j minutes suffice to produce 
 
KATHODIC POLAKTSATION OF MUSCLE. 335 
 
 those symptoms, which, as already mentioned, characterise the 
 proper stag-e of the action of the poison. Those which chiefly 
 claim attention are more or less marked tetanic stretching- spasms 
 of the lower extremities, which succeed each other at rather short 
 intervals, and are ushered in by violent restlessness of the animal 
 and by spasmodic opening* of the mouth frequently repeated. It must 
 be considered the rule, to kill the animal before reflex movements 
 are completely extinguished, and at a time when the heart still beats 
 strongly and the circulation is in action throughout the body. The 
 exposed muscles oug-ht to be transparent and to show a g-ood red 
 colour. Another certain criterion of the fit state is, that the muscles 
 of the belly pass into a state of long continuing tetanic contraction 
 on mechanical excitation, such as nipping" with the forceps. The 
 same thing* should also happen on dividing* the nerve in preparing* 
 the sartorius. In doing- this, it is often seen that after a short 
 pause a more slowly increasing* contraction succeeds the rapid 
 twitching* at the instant of the section of the nerve ; this per- 
 sists for some time and only gives way quite gradually to renewed 
 relaxation. If the frog has been killed at a time when voluntary 
 movements have already ceased and reflex excitability is almost 
 entirely exting-uished, then as a rule the muscles are scarcely in a 
 fit condition ; they still react indeed with a tetanic contraction on 
 excitation, but this has not the same long* duration, as in an earlier 
 stage of poisoning. 
 
 Certain effects of poison on striated muscles, e. g*. the results of 
 treating with solutions of potash or soda salts, can, as is well 
 known, be almost removed by washing* with an indifferent liquid. 
 Now it is remarkable that this is not the case for veratrine, or at 
 least not nearly to the same extent, whether it is introduced into 
 the course of the blood or is applied directly to the muscle. This 
 circumstance is of importance, because it is thus possible to repeat 
 the experiments of excitation several times in succession on the same 
 preparation, but for reasons to be discussed later, this can only be 
 done between long* intervals of rest, during which the preparation 
 remains plunged in a solution of Na CI, or must at least be kept 
 moist with it. 
 
 If we record the change of form of the sartorius poisoned in the 
 way described with veratrine, which may be effected by securing the 
 ends of bone to which it is attached, either in the usual manner or 
 by means of Hering's double myograph without the middle clamp, 
 one of the two movable non-polarisable electrodes being fixed, curves 
 
336 INHIBITION BY EXCITATION AND 
 
 are obtained which are essentially the same, whether the excitation is 
 effected by an indnction-shock acting- anywhere along- the muscle, or 
 bj' the shortest possible closure of a battery-current in either direction. 
 In both cases, the wave of contraction remains, so to speak, arrested 
 on its way through the muscle and produces tetanus which con- 
 tinues for a longer or shorter time at about the same strength. 
 When we consider the hitherto complete absence of proof of the 
 discontinuous nature of this contraction, it will be better to describe 
 it as a 'tonic ' shortening of the muscle in all its parts. 
 
 As von Bezold and Fick have already remarked, various forms of 
 the contraction of veratrine muscle can be distinguished, of which 
 I mention only one of those of most frequent occurrence, viz. that 
 in which the proper tonic persistent contraction is preceded by an 
 initial twitch of short duration. In this case, just as I have 
 described after section of the nerve, there occurs at the moment 
 of excitation a rapid contraction of the muscle to a maximum ; this 
 is immediately succeeded hj a considerable elongation, followed 
 closely by another slow contraction which gradually gives w'ay to 
 relaxation, 
 
 I have hardly ever failed to observe indications of this character- 
 istic mode of shortening, especially after the preparation has lain 
 a long time in dilute solution of common salt. As Fick has already 
 shown, it is out of the question to explain the initial twitch 
 mentioned, by an indirect excitation of the muscle through the 
 intermuscular nerves, and to refer the succeeding persistent con- 
 traction to direct excitation of the muscle exclusively ; for just the 
 same forms of curve are observed after previous curarising. The 
 phenomenon may possibly be accounted for by what Griitzner ^ has 
 recently asserted as to muscle being composed of two kinds of 
 fibre, morphologically and physiologically different, corresponding 
 to red and white muscles. We may perhaps consider as in support 
 of this vieW' , the fact that the same sort of contraction-cuixes with 
 two apices are not unfrequently observed under other circumstances 
 e. g. after treating- locally with Nag CO3, or even in quite normal 
 curarised muscles. Griitzner ^ goes so far as to believe, that such 
 curves represent the rule for the fresh sartorius of the frog. 
 
 The form of shortening just discussed does not seem the most 
 favourable for the experiments to be described. A more suitable 
 form is that in which the contraction, at first rapid, is gradually 
 
 * Recueil zoologique Suisse, vol. i. No. 4 (zur Anatomie u. Physiol, quergestr. 
 Muskeln). '^ Loc. cit. p. 675. 
 
KATHODIC POLARISATION OF MUSCLE. 337 
 
 retarded, and in which the gradually increasing relaxation is de- 
 ferred for several seconds. 
 
 If in such a ease, after the maximum of shortening- has been 
 reached, a battery current of medium strength (say two Daniells) 
 is closed for 1-2 sees., especially if it be ascending, there is observed 
 almost without exception an elong-ation of the muscle which is 
 sometimes rather considerable. This elong-ation coincides in time 
 with the moment of closure, and betrays itself by a sudden steep 
 fall of the curve. Whilst the closure lasts, the curve either con- 
 tinues horizontal or shows a slig-ht inclination towards the abscissa, 
 and falls ag-ain more steeply only after the current is opened. It is 
 the exception under these circumstances, to observe on opening the 
 exciting current, a feeble renewed rising of the curve as an expres- 
 sion of the opening excitation of the muscle. 
 
 If the current is closed at a time when the muscle is no longer 
 at its maximum shortening, then as a rule, a more or less evident 
 closing contraction is observed. If the same excitation is repeated, 
 by closing the current for a short time in unchanged direction 
 at different phases of the shortening and renewed lengthening of 
 the muscle, it is then seen that its proneness to excitation at closure 
 is generally diminished in proportion to the degree in which it 
 is shortened at the time of excitation. But even the completely 
 relaxed muscle immediately afterwards scarcely responds to the 
 same stimulus which shortly before produced a powerful contraction. 
 In the majority of cases, however, the increase of the results of 
 closing excitation keeps pace exactly with the gradually increasing 
 relaxation of the muscle, so that closing twitches produced during 
 relaxation at equal intervals, which almost always pass away 
 rapidly, all rise to the same height above the line of descent of 
 the curve which the muscle would have drawn had it been only once 
 excited. Tick also made analogous observations of a frog's muscle 
 poisoned with veratrine, by indirect excitation through the nerve, 
 and I shall return to these presently. 
 
 If we observe more closely a veratrine muscle when in persistent 
 contraction during electrical excitation, we easily see that the 
 above described relaxation revealing itself in the curve by an 
 almost rectangular descent when the current is closed, constitutes 
 essentialy a local phenomenon, and indeed one confined to the 
 immediate neighbourhood of the anode, with which moreover the 
 small amount of descent in comparison with the total height of 
 the curve is in agreement. 
 
 z 
 
338 
 
 INHIBITION BY EXCITATION AND 
 
 This fact may however be established with greater certainty and 
 in a more complete manner with the aid of the simple artifice 
 
 Fig. 8. 
 
 Fig. 8 represents the effect of exciting the 31. sartorius of Bana temporaria 
 after previously poisoning it with veratrine. A single Daniell was always used as 
 the source of the current ; a rheochord was not inserted in the circuit. The muscle 
 was stretched in Hering's double myograph, and it was fixed at the middle. The 
 tracing A is that of the anodic, and K that of the kathodic half of the muscle. Before 
 the special excitation (at C), a tonic contraction of the muscle, lasting a long time, 
 was produced by closing a battery current as short a time as possible, or by a single 
 induction shock. The moments of closing and opening the exciting current are marked 
 by the letters C and 0. Its direction was i . In order to economise space, the 
 tracings are shortened in such a manner, that the descending part corresponding to 
 the renewed lengthening of the muscle is only reproduced in part. The phenomenon 
 that the kathodic half of the muscle, on opening the current, is suddenly more 
 relaxed, whilst the anodic half is either shortened or shows no alteration in length, 
 is especially clear here. Both halves of the curve show at C and exactly 
 opposite changes of form. The divisions on the Time Line indicate seconds. 
 
 of fixing the muscle in the middle in such a way, that while the 
 propagation of the excitation is unhindered, the variations in form 
 of the one half are prevented from aflfecting the other. In doing 
 this, I employed the same method which has already been dis- 
 cussed in detail in an earlier communication^ and I need not 
 therefore give a minute description of it here. 
 
 ' Hering, ' tjber die Methoden zur Untersuchung der polaren Wirkungen des 
 elektrischen Stromes,' &c. Sitzungsber. vol. Ixxix. 
 
KATHODIC POLARISATION OF MUSCLE. 339 
 
 * 
 
 The phenomena which I hoped, vainly however, to show on 
 electrical excitation of tonically contracted muscle of Anodonta, 
 appear under these circumstances with convincing- clearness in a 
 frog's sartorius poisoned with veratrine. 
 
 If the muscle is fastened with proper precaution in a completely 
 relaxed condition in the double myograph, and the middle is fixed 
 with great care between oil clay, and if a single induction-shock 
 of corresponding strength is sent either through the whole length 
 of the muscle or through only a part of it, then, as a rule, both 
 halves are nearly equally shortened, and the attached levers (one 
 upwards and the other downwards) draw curves, which, except 
 as to differences in size, are in every respect similar to those of 
 freely twitching muscle described above. If, whilst the muscle 
 continues at its maximum contraction, a battery current of sufficient 
 strength is closed in the ascending direction, the anodic half is 
 seen to lengthen considerably at once, the curve corresponding' 
 to it rapidly sinking, whilst as a rule, the kathodic half of the 
 muscle shortens a little more, or undergoes no perceptible change 
 in length. If now after the closing has lasted a short time, the 
 current is again opened, in a successful experiment, exactly the 
 opposite changes of form present themselves. The anodic half 
 now shortens to an often not inconsiderable extent, this con- 
 traction being obviously the expression of the opening excitation, 
 whilst at the same time the half corresponding to the kathode 
 is distinctly more relaxed than would have been presumably 
 the case had it not been stimulated. If the excitations are 
 rapidly repeated with the same direction of current, the same 
 phenomena, though diminishing in amount, appear as at the 
 beginning of the experiment, provided that the muscle is still 
 shortened to a considerable extent. A fact mentioned above repeats 
 itself here, viz, that the readiness of the kathodic half of the muscle 
 to respond to the excitation of closing, does not always increase 
 in proportion to the progress of relaxation ; it often indeed seems 
 actually diminished, when the muscle has already regained its 
 original length. After a longer pause however, the original 
 excitability is recovered again without exception, and thus one is 
 in a position to repeat the experiment several times in succession 
 on the same preparation with like results. 
 
 It is hardly necessary to remark, that nothing essential is changed 
 in the preceding experiment, if from the beginning, one excites with 
 a descending instead of an ascending current, except that in the 
 
 z 2 
 
340 
 
 INHIBITION BY EXCITATION AND 
 
 former case, the effects of the excitation on the anodic half of the 
 muscle seem less marked than those on the side of the kathode ; this 
 may be attributed in the main to the greater density of the current at 
 
 Fig. 9. 
 
 Fig. 9. Tracings obtained in the same way as in Fig. 8 when the exciting- 
 current was directed \ . 
 
 the point of exit. The sudden relaxation of the kathodic half of 
 the muscle on opening- the current is then shown with extreme 
 distinctness and is recognised at once as a phenomenon equivalent 
 to the anodic inhibition of closing. 
 
 From this description of the variations in form which are observed 
 when a veratrine muscle in tonic contraction is excited electrically, 
 and from a consideration of the examjiles of the curves which are 
 given, we perceive at once that we are dealing here with essentially 
 local changes of the muscle, confined to the immediate neighbour- 
 
KATHODIC POLARISATION OF MUSCLE. 341 
 
 hood of the 'physiological kathode' and 'anode' respectively, 
 and that these experiments give no information as to the pos- 
 sible extension of the changes beyond the place of their direct 
 origin. 
 
 As far as my experiments hitherto go, it seems however that 
 a propagation of the excitatory inhibition does not take place to 
 any great extent in the circumstances under consideration, either 
 from the anode (on closing) or from the kathode (on opening) 
 the exciting current ; the possibility of such a propagation is 
 nevertheless not to be denied, for instances in favour of it actually 
 exist. (Compare my observations on the muscle of the snail's 
 heart.) 
 
 The relaxation of the kathodic end of the muscle, which occurs 
 simultaneously with the opening of the excitation current, explains 
 also the above-mentioned inhibitory after-effect of the electrical exci- 
 tation of the muscle when not fixed in the middle, in case the opening 
 excitation at the anode is inconsiderable. For it is obvious that 
 the nature of the variation in form of the whole muscle at the 
 opening of the current is conditioned solely by the mutual relations 
 of the two antagonistic variations at the poles. If the kathodic 
 opening inhibition preponderates, as is the rule when a muscle 
 is much shortened, then a rapid further elongation must result 
 on opening the current, and this constitutes apparently an after- 
 effect of the anodic closing inhibition, which is usually more 
 marked. 
 
 It follows hence directly, that a complete knowledge of the 
 way in which the above-described mechanical results of excitation 
 of persistently shortened muscle are in conformity with law, can 
 only be gained with the help of the double myograph, and by 
 using the artifice of fixing the muscle in the middle. 
 
 I do not, I think, go too far in regarding the variations of form 
 of a sartorius thrown artificially into a ' tonus-like ' condition 
 by veratrine, as a complete counterpart of phenomena which I 
 discussed as resulting from electrical excitation of the systolically 
 contracted cardiac muscle. Here as well as there, besides the 
 usual polar phenomena of excitation, which certainly appear less 
 clearly than during the condition of rest and under some circum- 
 stances are suppressed altogether, polar processes of inhibition may 
 be also directly demonstrated ; these manifest themselves by the 
 cessation or the diminution of an already existing condition of 
 excitation and a relaxation of the muscle conditioned thereby, 
 
343 INHIBITION BY EXCITATION AND 
 
 which is primarily local. The elong-ation which takes place in 
 a muscle which is in the state of persistent opening contraction, 
 when the current is closed in the same direction as before, has 
 been long- known to occur, and must certainly be considered as a 
 kindred phenomenon ; the only difference is^ that the question here 
 concerns the inhibition of a condition of excitation produced at 
 the physiological anode, by the after-effect of the cui'rent previously 
 flowing through the muscle. 
 
 Notwithstanding the great sensitiveness of the veratrine muscle 
 in repose, to the closing excitation, yet in general this excitation 
 develops effects so much the feebler as the muscle remains more 
 forcibly contracted at the time of excitation ; now and then, after 
 the return of complete expansion, it fails to produce any effects. 
 Considering these facts, there seems in this behaviour a further 
 analogy with systolically contracted cardiac muscle, and especially 
 with the tonically shortened muscle of Anodonta. In the case 
 of the former, INIarey^ showed that it is during systole quite unex- 
 cita.ble by feeble stimuli which are capable of exciting the diastolic 
 heart to contraction, and that this excitation occurs the more readily 
 the later the stage of relaxation in which it acts. This recalls 
 directly the observation of Fick^, already mentioned, that in indirect 
 rhythmical excitation of a veratrine muscle in persistent contraction, 
 all the twitches produced by it ' rise to very nearly the same height,' 
 so that all their base points lie in a curve, which, as it appears, 
 corresponds closely to the gradual re-expansion of the muscle after 
 the first twitch. 
 
 In regard to the adductor muscle of Anodonta which consists of 
 smooth fibre cells, I have myself found, that in a state of tonic 
 contraction, it does not react sensibly even on very strong closing 
 excitation, although on opening the current it sometimes contracts 
 vigorously. On the other hand, when it is as much as possible 
 relaxed, it behaves in regard to the current (leaving out of con- 
 sideration the difference of period of the contraction) exactly like 
 a normal striated muscle, for then the sensitiveness to the closing 
 excitation is far greater than to the opening one. 
 
 The facts abeady mentioned appear to speak very decidedly in 
 favour of the assumption, that in regard to the phenomena resulting 
 from electrical excitation, far-reaching analogies exist between 
 striated muscles of the trunk thrown into persistent contraction 
 
 ^ 'Physiologic exp^rimentale (Travaux du laboratoire de Marey,' 1876, p. 63 f.). 
 * Loc. cit. p. 146, 
 
KATHODIC POLARISATION OF MUSCLE. 343 
 
 by veratrine on the one hand, and the tonically shortened adductor 
 muscle of Anodonta and the systolically contracted heart on the 
 other. This assumption is further materially supported by the 
 ag-reement which, as will be shown, exists in regard to secondary 
 electromotive phenomena, between the muscle of Anodonta and 
 striated muscle poisoned with veratrine. 
 
 3. The secondary electromotive phenomena of veratrine 
 muscle excited electrically. 
 
 Du Bois-Reymond-^ has already occupied himself with the ex- 
 perimental solution of the question, how the polarisation phenomena 
 of striated muscle behave during* tetanic excitation, and arrived at 
 the result, that ' tetanised muscle undergoes less strong (positive) 
 polarisation than muscle in repose.' 
 
 I also now directed my attention to the investigation of the 
 secondary electromotive phenomena of a muscle tetanised through 
 the nerve, and I selected the sartorius as the object presumably 
 best fitted for the purpose. However I soon became con- 
 vinced that satisfactory results are hardly to be expected in this 
 way, because it proved that as a rule, the strength of tetanus is 
 maintained at the same point, too short a time to allow of ex- 
 perimenting with sufficient certainty. Moreover it seemed 
 desirable to replace the discontinuous tetanus by a more constant 
 condition of excitation comparable to tonus, and I therefore had 
 recourse again to the means of poisoning with veratrine already 
 tried. 
 
 The method of poisoning the entire muscle through the cir- 
 culation had proved best adapted for the investigation of the 
 variations of form on electrical excitation, but it is less fitted in a 
 case Like the present, when the results of the excitation have to 
 be examined by means of the galvanometer. The chief reason for 
 this is, that the force of the excitation, as well as its period of 
 passing away, are presumably not quite the same at all points, and 
 hence there will be differences of tension in the continuity of 
 the muscle ; these do not admit of being estimated, and make it 
 impossible to interpret the result of the observation with certainty 
 in each case. I have therefore since adopted exclusively local 
 poisoning of the muscle by direct application of a correspondingly 
 
 * See No. V. of this work. 
 
344 INHIBITION BY EXCITATION AND 
 
 dilate solution of acetate of veratrine, and by this method I have 
 attained quite unequivocal results. 
 
 Before I proceed to state these results, I will briefly mention the 
 experimental means which were available for my experiments. 
 I had at my disposal a new larg-e reflecting" galvanometer with bell 
 magnets by Edelmann of Munich ; this presents a very substantial 
 advantage for the intended experiments, inasmuch as its extremely 
 strong damping renders nearly perfect aperiodicity possible without 
 the use of Hauy's bar magnet. The magnet comes to rest in 
 little more than a second. 
 
 The bobbins of the Edelmann instrument have however, on 
 account of their large dimensions and the great number of turns, 
 a resistance much too considerable for the present experiments, and 
 I therefore used in their stead two bobbins with a less number of 
 turns, made after Professor Hering's design. Each of these consists 
 essentially of a copper hemispherical cup with thin walls, and when 
 the bobbins are as close as possible to each other, the cavities of 
 the cups almost enclose the siDherical damper ; the wires are wound 
 directly over these copper capsules, and thus the greatest part of 
 the space which would be injurious, and which is left empty even 
 when the Edelmann bobbins are in actual contact, is made of use ; and 
 at the same time the coils lie much closer to the magnet here than 
 there, so that notwithstanding their smaller number, the sensitive- 
 ness of the instrument is considerably increased. 
 
 . When I led off" from the artificial transverse section of a sartorius 
 and from a corresponding point at about the middle of the muscle, 
 with the bnish electrodes already mentioned, I obsen^ed on an 
 average a deflection of 40-50 scale divisions when the scale was 
 at a distance of about 2-5 metres. The sciatic nerve of a larger 
 frog gave deflections of 5-8 scale divisions, when led off" from 
 a tract about i centim. long under similar circumstances. 
 
 This amount of sensitiveness proved to be quite sufficient to 
 demonstrate with all the certainty desirable, electromotive effects 
 even much more insignificant than those concerned in the ex- 
 periments which are about to be described. At the same time, 
 the instrument offers the advantage which ought not to be under- 
 estimated, that the rapidity with which the magnet obeys the 
 directing forces, makes it possible to recognise differences of tension 
 which change their signs very quickly, 
 
 I may refer to previous communications for all the rest of the 
 arrangements for the experiments, because they agree precisely 
 
KATHODIC POLARISATION OF MUSCLE. 345 
 
 with the description given there ^. However, I was able here, as 
 well as in the polarisation experiments on muscle of Anodonta, 
 to dispense with the use of the pendulum rheotome, because the 
 closing- time of the excitino- current must not be allowed to sink 
 below a limit attainable with the aid of a double switch, if the 
 effects to be described are to come out with sufficient clearness. 
 A more exact determination of the transmission time in each case 
 is of no importance in the experiments now under consideration. 
 Battery currents were used exclusively for excitation, and as a rule 
 I employed two Daniells as the source of current. 
 
 I have already shown in an earlier communication^, that by the 
 direct action of properly dilute solutions of acetate of veratrine 
 in o-6°/o solution of common salt, it is easy to induce a locally 
 limited condition of the muscular substance, which is identical 
 with that observed in the muscles in their totality when the 
 whole animal is poisoned, and this behaviour serves to make the 
 law of polar excitation by the electrical current directly evident. 
 I also called attention later ^ to the sometimes extraordinarily strong 
 'action current' which may be observed after momentary excitation 
 at any point of a muscle which has been partially treated with 
 veratrine, when the leading off electrodes are applied to the 
 poisoned end of the muscle and to some point of the longitudinal 
 surface higher up. The difference of tension to which this is 
 due is, as I then found, very persistent, and under any circum- 
 stances lasts longer than the visible local persistent contraction of 
 the muscle. 
 
 For the present object, which is solely concerned with the 
 investigation of galvanic changes occurring under the influence 
 of an electrical current in a length of muscle alternately in repose 
 and excited, it was therefore an essential preliminary condition to 
 make the negativity of the latter last as long as possible as an 
 expression of excitation, in order to be able to carry out at least 
 one polarisation experiment during this time. For this purpose 
 a special method of local poisoning is required, and as such, I 
 recommend the following mode, which hardly ever fails if the 
 preparation is of sufficient sensitiveness. 
 
 As in the experiments described in paragraph 2, I used here 
 also almost always the sartorius of R. tem,poraria, and as a rule I 
 first poisoned the animals with a strong, quickly-acting dose of 
 
 1 Sitzungsber. vol. Ixxxviii. ^ Loc. cit. vol. Ixxx. 
 
 ' Sitzungsber. vol. Isxxi. p. 107 (Biedermann). 
 
346 INHIBITION BY EXCITATION AND 
 
 curare, without however discovering- any difference when the 
 curare was dispensed with. The muscle, with the two stumps 
 of hone left attached to it in the usual way at both ends, was then 
 stret<5hed immovably in a horizontal position before it was excited : 
 the lower thin end had been previously dipped up to about 5 mm. of 
 its length for 5 to 10 minutes in a o-6°/o solution of common salt, 
 in which o-oi-o-03yo of acetate of veratrine had been dissolved. 
 
 It is of special importance in these experiments to make it a 
 rule to choose only preparations, the muscles of which look in- 
 tensely red. as otherwise one cannot count on perceiving* the 
 phenomena resulting from the excitation with complete and con- 
 vincing clearness. 
 
 When the end of the muscle is dipped into the solution of 
 veratrine, contractions occur generally after a time ; these are best 
 prevented by a slight weighting of the muscle as it hangs vertically, 
 so as to prevent other points of the surface besides the tract which 
 is immersed from coming into contact with the solution. 
 
 Notwithstanding the fact that the veratrine solution is so very 
 dilute it may still act so as to excite the muscular substance chemi- 
 cally, bringing about a persistent contraction of the immersed tract 
 of muscle which often lasts after the time of immersion and occurs 
 even in the best preparations. In this case, by leading off from the 
 lower end and a point corresponding to about the middle of the 
 muscle, a strong ascending action current is at once observed and it 
 only gradually passes away. To exclude this, and also for reasons 
 which shall be given later, I found it advisable not to make the 
 polarisation experiment directly after removing the muscle from the 
 solution, but only after leaving it for some time (j hour and longer) 
 in a physiological solution of common salt. 
 
 As I have already remarked, the very characteristic action of 
 veratrine, in contrast with the changes effected in muscle substance 
 by treating locally with potash or soda salts, is not removed even 
 after long continued washing with indifferent liquids, but persists 
 in almost unchanged strength until the death of the muscle, 
 whether it is poisoned as a whole or only locally. If the muscle 
 has not been injured in the preparation, and if the negati^aty, 
 occurring as a result of whatever persistent contraction of the 
 veratrinised part there may be, has been got rid of by leaving 
 it for some time in a o-6% solution of common salt, then, pro\'ided 
 that the muscle has been very carefully stretched out, no appre- 
 ciable differences of tension can be observed between the poisoned 
 
KATHODIC POLARISATION OF MUSCLE. 347 
 
 end and the middle of a muscle, none at all events whicli the 
 galvanometer is capable of showing-. If the exciting- electrodes 
 are now applied to the bones at both ends, and the current of two 
 Daniells is closed for a very short time in a descending' direction, 
 so that it traverses the whole length of the muscle, then if the 
 galvanometer circuit is closed directly afterwards by turning the 
 switch, there results a very strong deflection in the direction of 
 a current opposed to the polarising • one. The intensity of this 
 current is always incomparably greater than that of a negative 
 kathodic polarisation current which appears in normal muscle 
 preparations under similar conditions but after a much longer 
 period of closure. It even exceeds that of the demarcation current. 
 
 It can hardly be doubted that this difference of tension is caused 
 by the strong and protracted excitation of the tract of muscle 
 which has been treated with veratrine, and the current produced 
 by it must accordingly be considered as an ' action current.' 
 
 The development of this cuiTent, as regards time, very frequently 
 results in such a manner (and these are exactly the most favourable 
 cases for the following experiments) that immediately after opening 
 the exciting circuit and closing that of the galvanometer, the 
 deflection attains a certain amount very quickly, remains constant 
 at this amount for a short time, and then increases gradually 
 during several seconds, until finally it passes away slowly. If 
 now, after compensating as quickly as possible, the excitation is 
 repeated with a descending current in the same way as before, 
 whilst the position of the magnet remains unchanged or is still 
 in the act of increasing, then, ^vithout exception, there follows a more 
 or less considerable hachwarcl swing, indicating a polarisation current 
 which has the same direction as that of the stimulating current, and 
 is therefore positive. The image of the scale then either retm-ns to 
 its former position more slowly, sometimes passing beyond it, or it 
 takes up a new position, which indicates a persistent diminution 
 of negativity of the fibres at the kathodic spots. 
 
 Whether the one or the other happens, depends essentially upon 
 the stage of the development of the action current when the ex- 
 citation takes place. If the polarising current is closed immediately 
 after it begins or only a little later, then after an excitation of 
 short duration, the diminution of the negative after-current caused 
 by it, is seen almost uniformly to be succeeded by a corresponding 
 increase ; and the same experiment can be repeated several times 
 in succession with the same result. If however the excitation 
 
348 INHIBITION BY EXCITATION AND 
 
 takes place at a time when the spontaneous decrease of the after- 
 current is imminent, or has already begun, then as a rule only 
 a more or less considerable and sudden acceleration of the return 
 is effected ; if the excitation takes place later still, neg-ative 
 polarisation is again produced, although it is much feebler than at 
 the commencement of the series of experiments. Sometimes in this 
 case also, the opposite positive action becomes still perceptible by 
 a transient delay of the negative deflection. 
 
 It is noteworthy that if directly after the passing away of the 
 action current, which always takes a rather long time, the same 
 descending current, which at first produced such powerful effects, 
 is closed, only comparatively trifling deflections in the dii'ection 
 of a negative polarisation current are obtained. If however, the 
 muscle has a longer period of repose, being left in a physiological 
 solution of common salt in the interval, then on the same excitation 
 exactly the same phenomena described above are repeated. This 
 is very much in favour of the assumption of a local fatigue, caused 
 by the strong and prolonged state of excitation of the poisoned 
 part of the muscle ; and another fact is in accordance with it, viz. 
 that the mechanical results of excitation under the same circum- 
 stances often seem to be seriously impaired. 
 
 The following tabulated results of experiments will serve to 
 explain the foregoing more clearly. 
 
 The horizontally stretched muscle is to be supposed throughout as 
 having its lower end turned towards the left ; two out of the three 
 galvanometer electrodes were in direct contact with the tendinous 
 ends, and the third was in contact with a corresponding point of 
 the middle of the muscle. The exciting electrodes were applied to 
 the two stumps of bone. The leading off" was as a rule only from 
 the left kathodic half. The direction of the polarising current is 
 indicated in the tables by arrows. 
 
 1. Eana tewporarla, curarised, the lower end of the sartorius 
 dipped in o-oi°/o solution of veratrine for 5 min., afterwards treated 
 with 0'6°/o Na CI. solution for 15 min. CI. T = closing time. 
 M. St. = momentary stimulation effected by a rapid closing and 
 opening of the exciting circuit by means of the double switch. 
 Before the first stimvilation, the muscle is found to be without 
 current on leading off' from the kathodic half The strong negative 
 action current after the first, fifth, and ninth excitations is rapidly 
 compensated every time. 
 
KATHODIC POLARISATION OF MUSCLE. 
 
 349 
 
 Number 
 of cells. 
 
 Cl.T. 
 
 Left half. 
 
 2 Dan. 
 
 2 Dan. 
 
 2 Dan. 
 
 M.St, 
 I sec. 
 
 M.St. 
 2 sec. 
 
 M.St. 
 I sec. 
 3 „ 
 5 » 
 
 fquick to —40, 
 \slow to — 74 sc. 
 + 18, quick to o 
 + 32, „ +4 
 + 9> j» o 
 
 — 32, slow to —56 
 + 32, quick to O 
 + 16, „ -8 
 + 24, „ +3 
 
 quick to —24, slow to —47 
 + 26, slow to — 2 
 + 8, slow to o 
 — 12, slow to o 
 
 Right 
 half. 
 
 Remarks. 
 
 After a pause of 
 10 min. 
 
 M. C. (muscle cur- 
 rent) = — 8 sc. 
 
 Pause of 10 min. 
 M.C. = — 7 sc. 
 
 2. Bana temporaria, curarised ; everything' as in the previous 
 experiments. 
 
 Number , 
 of cells. 
 
 Cl.T. 
 
 Left half. 
 
 Eight 
 half. 
 
 Remarks. 
 
 2 Dan. 
 
 M.St. .. 
 
 quick to —21, slow to ■ 
 
 -54 
 
 \ 
 
 
 5> 
 
 M.St. * 
 
 + 12, slow to 
 
 
 
 
 5> 
 
 I sec. •< 
 
 + 23, slow to —5 
 hesitates then —3 
 
 
 ) 
 
 
 >> 
 
 5 » " 
 
 — 14, slow to —6 
 
 
 / 
 
 Pause of 15 min. 
 MC.= — 12 sc. 
 
 2 Dan. 
 
 M.St. '> 
 
 quick to — 18, slow to 
 
 -35 
 
 \ 
 
 
 » 
 
 » 
 
 I sec. ■< 
 
 + 20, slow to — 2 
 — 8 + 15, slow to 
 
 
 
 
 )> 
 
 4 . — 
 
 — 18, slow to —12 
 
 
 } 
 
 Pause of 10 min. 
 
 ,, 
 
 M.St. .. 
 
 quick to — 25, slow to 
 
 -.^Q 
 
 \ 
 
 After crushing the 
 
 >> 
 
 2 sec. -< • 
 
 + 24, slow to — 5 
 
 
 \ 
 
 lower (left) end 
 of the muscle and 
 
 " 
 
 2 „ " 
 
 + 36, slow to + 3 
 
 
 \ 
 
 compensating the 
 
 j> 
 
 2 „ >. 
 
 + 28, slow to + 10 
 
 
 ) 
 
 current. 
 
 Thus, as is directly proved from the preceding- series of experi- 
 ments, strong positive kathodic after-currents may be observed 
 
350 INHIBITION BY EXCITATION AND 
 
 in striated muscle when excited by batteiy currents under suitable 
 conditions. It appears to be essential that the spots of fibre, which 
 represent the physiolog-ical kathode, shall be already in a uniform 
 state of strong excitation at the moment they are stimulated, and 
 this is obviously in complete accord with the theoretical views 
 ali'eady mentioned. 
 
 It might be advanced as an objection, that the question in these 
 experiments is not so much that of a kathodic opening inhibition 
 comparable to the anodic closing effect, but rather that of the 
 result of a local fatigue of the muscle, due to previous excitation 
 at the points of exit of the current, the fatigue having enfeebled 
 the already existing excitation transitorily or permanently. Against 
 such a view however the fact speaks decisively, that the magnitude 
 of the positive deflections throughout does not increase in proportion 
 to the duration of closure of the polarising current, as would 
 certainly be the case if the relative positivity of the kathodic points 
 of the fibre were conditioned only by a local fatigue. 
 
 Thus a glance at the tables before us shows that an increase 
 of the positive kathodic effects with increasing duration of closure, 
 takes place only within very narrow limits and soon changes to 
 the opposite, although the tonic excitation still exists in sufficient 
 amount, and therefore excitations of short duration are succeeded 
 afterwards as before, by positive deflections. It is often possible 
 to observe that when the duration of closure is gradually increased, 
 the deflections which were at first entirely positive, next become 
 diphasic (positive deflection with a negative preceding it) ; finally 
 the second phase gradually diminishes, until at last it gives way 
 to effects wholly negative in direction (comp. Table 2). 
 
 It is quite as much out of the question to explain the positive 
 kathodic after-current by an opening excitation propagated from 
 the anode up to the middle electrode of the galvanometer. Any 
 such assumption is seen from the first to be inadmissible, when we 
 consider the comparatively slight strength of the exciting current, 
 combined with the short time of closure, and the fact that the de- 
 scending direction of the current is unfavourable for producing strong 
 opening excitation of the sartorius. It is however easy to con\'ince 
 oneself in every case, that on leading off" in the continuity of the 
 muscle, at about the middle third, no considerable differences of 
 tension are to be found under the given conditions of experiment. 
 
 Thus, so far as I see, the hypothesis already mentioned alone 
 remains, viz. that ai the moment of ojpening the exclthuj current, an 
 
KATHODIC POLAEISATION OF MUSCLE. 351 
 
 inhibition of the state of excitation already existing- at the physio- 
 logical kathode develops itself, giving- rise to relative positivity of 
 the kathodic spots of fibre, and that in the case now before us, the 
 positive kathodic after-current is conditioned by this inhibition. 
 
 The experiments in regard to the variations of form of electrically 
 excited veratrine muscle, spoken of in the second section of this paper, 
 stand in perfect agreement vfith this interpretation of the observed 
 phenomena. It accords no less completely with facts which remain 
 to be described. 
 
 As already frequently mentioned, the resulting phenomena of the 
 changes of the stimulated muscle substance, produced under the 
 influence of the anode whilst the current remains closed, resemble 
 in every respect those observed under the same circumstances at 
 the kathode on opening the current. This holds good, not only 
 with regard to the variations in form of the muscle, which may be 
 characterised in both cases as a locally limited relaxation, but also in 
 regard to the accompanying electromotive phenomena which are 
 characterised by the relative positivity of the point of entrance or of 
 the point of exit of the current, and give rise in the one case to 
 a negative anodic, in the other to a positive kathodic after-current. 
 
 Since the method of the investigation of secondary electromotive 
 effects only permits of determining the results of electrical excitation 
 after opening the polarising current, it is clear, that as soon as the 
 conditions for producing distinct opening excitation are given 
 (especially when stronger currents and longer period of closure are 
 used) the positive anodic after-current caused thereby will come 
 into prominence, whilst the negative after-current can only some- 
 times make itself apparent as a preliminary jerk. Only in the 
 case in which the accomplishment of the opening excitation is 
 impeded in any way, or is altog-ether prevented, can one expect 
 to observe stronger effects in the direction of a negative anodic 
 polarisation current, as e. g. is actually the case with exhausted 
 preparations, or after the anodic ends of fibre have been killed. In 
 this connection, it is of interest to notice that under the same 
 circumstances in which on opening the current, strong kathodic 
 effects are observed in the sartorius veratrinised at one end, 
 negative anodic after-cm'rents are also found to exist if the 
 current enters at the poisoned end of the muscle, and this happens 
 under conditions in which in normal preparations, purely positive 
 anodic polarisation of considerable strength might have been ex- 
 pected to present itself. (Comp. Tables i and 2.) 
 
352 INHIBITION BY EXCITATION AND 
 
 It is further very remarkable, tliat the positive Jcathodic polarisa- 
 tion in muscles veratrinised at one end is not only not itnpaired by 
 killing that end of the muscle dxiring the existence of the state of ex- 
 citation there, but, as I had also previously observed in the muscle of 
 Anodonta, it very often increases considerably. 
 
 Even directly after an action current produced by a momentary 
 excitation has passed away, it is possible at once to renew its power 
 of g-iving positive kathodic after-currents, in a muscle prepared 
 in the manner indicated. This may be done by killing- the poisoned 
 fibre ends corresponding to the point of exit of the exciting* current, 
 by crushing them with the forceps and compensating the strong 
 demarcation current produced thereby. By this procedure it is 
 well known that the production of a closing excitation is materially 
 prevented, while, as it appears, the results of the inhibitory actions 
 come into play as strongly after as before. However, before I 
 consider in detail the meaning of this phenomenon, which is sur- 
 prising at first sight, I must discuss some other facts which relate 
 to the existence of joositive kathodic after-currents on exciting 
 electrically normal muscles that have not been poisoned. 
 
 4. Positive kathodic polarisation in normal striated muscles, 
 
 Hering^, who used exclusively muscle of H. esculenta, mentions 
 that sometimes in quite fresh muscles after the first excitation, 
 extremely feeble deflections of the magnet occur in the direction of 
 a positive kathodic polarisation, but he did not proceed to a more 
 detailed investigation of these trifling effects at that time. Since 
 then, on repeating similar experiments on preparations of the 
 sartorius of frogs {R. esculenta) which had been cooled in ice, 
 and the excitability of which had attained a very high degree, 
 I succeeded several times in getting stronger positive kathodic 
 effects, but they are still far less considerable than those which may 
 be observed under similar circumstances in preparations of R. tem- 
 poraria. 
 
 Thus, on using the Edelmann galvanometer I was surprised that 
 normal sartorius preparations of R. temporaria, on being excited 
 with currents of medium strength and short period of closure, and 
 on leading off" from the lower (kathodic) end, and from a correspond- 
 ing point of the middle of the muscle, gave at first very often 
 exclusively positive kathodic after-cui-rents of considerable strength. 
 
 * Sitzungsber. vol. Ixxxviii. p. 430. 
 
KATHODIC POLARISATION OF MUSCLE. 
 
 353 
 
 The deflections, which I observed in the most favourable case under 
 these circumstances, never amounted to more than a few (at most 
 five) scale divisions ; still there covild hardly be any mistake about it, 
 as all sources of error were excluded, and it only remained to make 
 the feeble effects more clearly perceptible, by using' an instrument 
 so delicate as to make a more exact investigation possible. 
 
 I will give first of all some values of positive kathodic polarisation 
 observed in a fresh, not poisoned sartorius of li. temporaria, when the 
 Edelmann g-alvanometer was used, in order to compare them with 
 the earlier experiments on vei'atrine muscle. 
 
 3. Arrangement of experiments as in i and 2. 
 
 Number 
 of cells. 
 
 Cl.T. 
 
 Left half. 
 
 Right 
 half. 
 
 Remarks. 
 
 2 Dan. 
 
 M.St. -. 
 
 2 sec. -• 
 
 M.St. .. 
 
 I sec. •« 
 
 3 sec. ■« 
 
 + 5 sec. slow to 
 + 2 „ ,, 
 -14 :. „ -3 
 
 + 11 5, J, 
 
 + 4 » J) 
 — 12 „ quick to + I 
 
 \ 
 
 1 
 
 After crushing the 
 lower (left) end 
 and compensating 
 the current. 
 
 In all the experiments which will be described in the following- 
 paragraphs, I used a mirror galvanometer constructed by Meyer of 
 Zurich after Hermann's specifications, and which left nothing to be 
 desired as regards sensitiveness. I refer for a description of it to 
 Hermann's paper in Pfliiger's Archiv ^ and will only remark that 
 the new instrument used by me has 1000 turns more than that of 
 Hermann. The current from the longitudinal to the transverse 
 section of the sciatic nerve of a medium-sized frog' with its lower 
 end lying on the brush electrodes, produced on an average, a 
 deflection of 120-150 scale divisions when the instrument was 
 made aperiodic by Hauy's bar magnet, the leading'-ofF electrodes 
 being about i cm. apart and the distance of the scale from the 
 mirror about 2-5 metres. The demarcation current of the sartorius, 
 when similarly led off, naturally drove the scale at once far out of 
 the field ; consequently whenever there were any considerable differ- 
 ences of tension at the beginning of the experiments of excitation, 
 it was necessary to adjust the compensator approximately before 
 closing the galvanometer circuit. 
 
 ' Pfliiger's ' Archiv,' vol. xxi. 
 A a 
 
354 
 
 INHIBITION BY EXCITATION AND 
 
 Under these circumstances it could be shown with complete 
 certainty that positive kafJiodic polarisation is a conseqiience of 
 electrical excitation with hattery currents even in striated muscles 
 wliicTi are entirely uninjured. Under favourable conditions this can be 
 always demonstrated. 
 
 In proof of this statement I give the three following- series of 
 experiments, which require no further explanation to be understood, 
 
 4. Sartorius of B. temporaria curarised. The arrangement of 
 experiment is as in i, 2, and 3. The sign > denotes that the 
 image of the scale moves back wdth decreasing speed. The muscle 
 current of the left half amounted at first to — 70 se. and was com- 
 pensated. 
 
 Number 
 of cells. 
 
 Cl.T. 
 
 Left half. 
 
 Right 
 half. 
 
 Remarks. 
 
 2 Dan. 
 
 M.St. .. 
 
 1 sec. -< 
 
 3 » 
 
 2 ,, * 
 
 + 108 sc. > 
 + 87 > 
 
 - 23 + i4> -42 
 
 - 12 + 55 '> -22 
 
 
 
 
 J ., 
 
 + 260 > + 50 
 
 \ 
 
 
 
 5 V ^ 
 
 + i40> + 12 
 
 
 After crushing the 
 lower end of the 
 
 
 10 ,, ■< 
 
 + 58> + 32 
 
 r 
 
 muscle. 
 
 
 10 „ 
 
 - 27 + 8 
 
 
 (MC. compensated.) 
 
 
 M.St. ► 
 
 + 226 > 
 
 ' 
 
 
 5. Sartorius of R. temporaria curarised. Arrangement of experi- 
 ment as before. MC before the commencement = — 76 sc. 
 
 Number 
 of cells. 
 
 Cl.T. 
 
 Left half. 
 
 Right 
 half. 
 
 Remarks. 
 
 2 Dan. 
 
 5> 
 
 M.St. * 
 
 1 sec. ■< 
 
 2 » " 
 
 M.St. * 
 
 I sec. * 
 
 + 53 quick -22 
 + 36 ,, -30 
 — 82 slow —18 
 
 + 13 quick —II 
 -3 + 8 quick —15 
 
 
 Pause of \ hour. 
 MC. = + 82 sc. 
 
 After crushing the 
 
 „ 
 
 I !. ■• 
 
 + 56> +12 
 
 \ 
 
 lower end of the 
 
 
 3 » ■" 
 
 — 42 slow —24 
 
 muscle. 
 
 
 
 
 
 (MC. compensated.) 
 
KATHODIC POLAKISATION OF MUSCLE. 
 
 355 
 
 6. Sartorius of R. temporaria not curarised ; the upper end of the 
 muscle tied off, everything* else as in the previous experiments. MC 
 at the commencement —52 sc. 
 
 Number 
 of cells. 
 
 Cl.T. 
 
 Left half. 
 
 Right 
 half. 
 
 Remarks. 
 
 2 Dan. 
 
 M.St. -> 
 
 M.St. .. 
 
 + 48 > + 2 
 
 + 53 quick + 9 
 
 
 Pause of 3 min., 
 MC. = 0. 
 
 
 
 » ■< 
 
 + 32 „ -12 
 
 
 
 
 
 I sec. * 
 
 — 70 slow — 24 
 
 
 
 
 
 M.St. 
 
 1 sec. * 
 
 2 ,, •« 
 
 I „ ■< 
 
 4 ,< * 
 
 + 52>-43 
 
 + 35 >o 
 
 — 5 slow — 16 
 
 + 49 >o 
 -46 
 
 i 
 
 After crushing the 
 lower end of the 
 muscle and com- 
 pensating the cur- 
 rent. 
 
 An inspection of the preceding tables shows that the positive 
 kathodic polarisation of uninjured muscles may sometimes attain to 
 very considerable values ; even if such powerful effects as e. g". in 
 Table 4 are reg-arded as exceptional, we shall nevertheless scarcely 
 ever fail to find after-currents of the kathodic half of the muscle, 
 on excitations with battery currents of medium strength and of 
 short duration, provided that the preparations are sufficiently sensi- 
 tive. These after-currents either appear at once after closing 
 the galvanometer-circuit, or are preceded by a more or less marked 
 negative preliminary jerk, which sometimes only shows itself as a 
 delay of the positive deflection. One may say in general, that the 
 kathodic polarisation phenomena of fresh muscles depend upon 
 the closing time of the exciting current in this sense, that as a 
 rule at first only unmixed positive after-currents appear, with 
 which afterwards negative ones are associated, increasing rapidly 
 with the period of closing, so that whereas at first they present 
 themselves as preliminary jerks, they subsequently entirely suppress 
 the positive effects. Hence these latter sometimes, so to speak, look 
 like a transitory ' positive variation ' of a negative polarisation- 
 current, after the expiration of which, the magnet swings again 
 almost uniformly with decreasing velocity beyond the zero point, so 
 that, finally, a more or less strong negative deflection remains 
 behind, and this disappears only quite gradually. (See Tables 
 4 and 5, and below 7 and 8.) Sometimes the positive effect seems 
 
 A a 2 
 
356 INHIBITION BY EXCITATION AND 
 
 to be indicated only by slight hesitation of the negative deflection. 
 (See below, Table 8.) 
 
 Though kathodic polarisation in fresh uninjured muscles varies 
 in some details as to its mode of manifesting itself, still under all 
 circumstances, it is alvrays easy to determine the fact that the 
 intensity of the led-off positive after-current is directly dependent 
 upon the time of closure of the polarising current, in such a way 
 that it always decreases rapidly as the latter increases. Beyond a 
 certain duration of closing time, a purely negative kathodic polar- 
 isation is observed without exception ; and in less excitable pre- 
 parations this is the case from the first. Positive kathodic polarisa- 
 tion also shows itself dependent upon the strength of the exciting 
 current as well as upon the closing time. Very feeble cm-rents, if 
 they have any eflTect at all, produce always only feeble negative 
 after-currents ; and the same thing is true, on the other hand, of 
 very strong ones, by which the muscle substance at the kathode 
 is evidently subject to profound changes, even when the duration of 
 closing is shortei*. 
 
 Other facts which will be communicated later, support the idea 
 that a certain relation exists between the positive kathodic polar- 
 isation current and the electromotive behaviour of the muscle in 
 repose, of such a nature, that the current is more easily and more 
 strongly developed if a normal current of repose is present from 
 the very first ; e. g. when, as in our case, the lower tendinous end is 
 negative to the middle of the muscle. Although, however, on the 
 ground of results which remain to be discussed, such a relation is 
 not to be denied, yet it can by no means be held as a universal law ; 
 and therefore I desire to lay special stress on the fact that I have 
 observed positive kathodic polarisation, not only in preparations 
 which from the first showed normal differences of tension in the 
 sense of an ascending demarcation current, but also in muscles 
 wholly without cmTcnt, as well as in those which were electro- 
 motive in the opposite direction. 
 
 If we look through a long series of polarisation experiments on 
 fresh sartorius prepai'ations of R. temporaria, as uninjured as possible, 
 we cannot fail to see, that even if positive kathodic after-cm-rents 
 are observed in the majority of cases at the first excitations, 
 3^et as a rule these are essentially of inferior strength to the 
 other secondary electromotive phenomena and particularly to 
 the i:)ositive anodic after-currents. Cases like that mentioned 
 in Table 3, of which I have observed several, are nevertheless 
 
KATHODIC POLARISATION OF MUSCLE. 
 
 357 
 
 exceptions, at least at the season (early spring-) when my experi- 
 ments were performed. Later in the year, when the animals have 
 been for a long-er period in a higher external temperature, and their 
 muscles no longer show that beautiful red tint which is so 
 characteristic of their good condition, even slight traces of positive 
 kathodic polarisation are only exceptionally observed. 
 
 Now it is very remarkable and of the greatest interest for the 
 interpretation of positive kathodic after-currents, that after Icilling 
 the end of the muscle corresponding to the '■ jjhijslological kathode,' it is 
 easy, even in less sensitive 2^reparatio7is, to observe tolerably strong 
 jpositive after-currents when atterminally directed battery-currents are 
 used for stimulation. 
 
 A glance at the above Tables 4, 5, and 6 shows, that after the 
 kathodic end of the muscle has been killed, positive polarisation 
 at the points of exit of the current is just as possible here as 
 in the veratrine muscle. In the majority of cases indeed a con- 
 siderable strengthening of the + deflections took place after the 
 injury. This justified the expectation, that in cases also where 
 positive kathodic effects are not demonstrable at all, they might be 
 produced, so to speak, artificially, by killing the kathodic end of the 
 muscle. The following series of experiments show that this is iu 
 fact the case. 
 
 7. Sartorius of H. temjwraria, curarised. 
 ment as before. MC = —%o se. 
 
 Arrangement of experi- 
 
 Number 
 of cells. 
 
 C1.T. 
 
 Left half. 
 
 Eight 
 half. 
 
 Remarks. 
 
 2 Dan. 
 J) 
 
 I sec. < 
 
 M.St. -H- 
 M.St. ■>- 
 
 I sec. ■< 
 
 3 » ^ 
 
 5 .. " 
 
 6 „ ^- 
 I ., ■* 
 
 — 6+14 quick to —25 
 
 — 34 slow to —18 
 + 83> + 2 
 
 + 112 > + 25 
 
 + 91 > + i8 
 
 + 24 slow to + 1 1 
 
 — 20 ,, 
 + 54 > + 6 
 
 
 Third and following 
 excitations after 
 crushing the lower 
 (left) end of the 
 muscle. 
 
 Current compen- 
 sated. 
 
358 
 
 INHIBITION BY EXCITATION AND 
 
 8. Sartorius of H. temporaria, curarised. Arrangement of experi- 
 ment as before. MC = -F 9 sc (reversed). 
 
 Number 
 of cells. 
 
 Cl.T. 
 
 Left half. 
 
 Eight 
 half. 
 
 Eemarks. 
 
 2 Dan. 
 
 M.St. .. 
 
 M.St. 
 
 I sec. -« 
 
 5 » " 
 
 I » 
 
 I „ " 
 
 quick to —3+12-15 
 f ,, —7 hesitating 
 [then quick to —22 
 
 -f 76 > + 8 
 — 63 (lasting) 
 + 66 > + 1 2 
 -(-97>o 
 
 - 
 
 The third and fol- 
 lowing excita- 
 tions after killing 
 the lower end by 
 concentrated so- 
 lution of NazCOa. 
 
 (MC. compensated.) 
 
 9. Sartorius of R. temporaria, curarised. Arrangement of experi- 
 ment as before. MC. before commencing = — 43 sc. 
 
 Number 
 of cells. 
 
 C1.T. 
 
 Left half. 
 
 Right 
 half. 
 
 Eemarks. 
 
 2 Dan. 
 
 5) 
 J> 
 
 M.St. * 
 
 M.St. < 
 
 I sec. •« 
 
 3 „ 
 
 M.St. " 
 
 M.St. .. 
 
 „ 
 
 -32>-5 
 + 46 > — 1 1 
 
 + 23> 
 
 -26 > +3 
 
 + 33>-4 
 + 36> 
 
 — 114 (lasting) 
 -f-24> -2 
 
 - 
 
 The second and following 
 excitations after killing 
 the lower end with con- 
 centrated solution of 
 NaCl. 
 
 M.C. compensated (460°). 
 
 After a pause of ^ hour. 
 
 M. C. compensated(2 8o°). 
 
 As the preceding series of experiments show, positive kathodic 
 polarisation in a muscle of which the kathodic end has been killed, 
 obeys exactly the same laws in regard to its dependence upon the 
 intensity of the exciting cun^ent and its duration of closure as 
 muscle veratrinised at one end, and, apart from differences of 
 degree the corresponding phenomena accord throughout. The 
 negative preliminary jerk, which often ushers in the positive 
 kathodic after-current in normal uninjured muscles, is wanting in 
 both cases throughout, a circumstance which, as it seems, makes the 
 near connexion which doubtless exists between the kathodic closing 
 excitation and the negative after-current, come out in clear relief. 
 
KATHODIC POLARISATION OF MUSCLE. 359 
 
 That is to say, if the negative preliminaiy jerk is to be reg-arded as 
 essentially an immediate after-effect of the closing excitation, then 
 like this, it must be altogether or partially suppressed by killing at 
 the kathodic end, as is in fact always the case. On the other hand, 
 it was shown earlier, that also during the existence of a persistent 
 excitation of the kathodic parts of the fibre of a muscle treated 
 locally with veratrine, a battery current of corresponding direction 
 produces only a feeble closing excitation or none at all. 
 
 If we take into account the very striking agreement which exists 
 in regard to the conditions of commencement and mode of mani- 
 festation of positive kathodic polarisation in veratrine muscle on the 
 one hand, and on the other, after the death of the kathodic ends of 
 fibre in normal striated muscle, it certainly seems a very natural 
 idea to refer the phenomenon to the same cause in both cases. 
 Hence, here as there, we should have to deal with inhibition of 
 an existing state of excitation, and with the consequent relative 
 positivity of the points of exit of the current as compared with 
 other points in the muscle, this inhibition being produced at the 
 physiological kathode on opening the exciting current. 
 
 However little reason there may be to doubt the correctness of 
 this conception as regards locally veratrinised muscle, the generalisa- 
 tion thus indicated requires a more complete justification. 
 
 It cannot be denied, that after killing at one end the terminations 
 of the fibre of a normal regularly constructed muscle, the nearest 
 excitable transverse sections of it are in a condition of more or less 
 strong persistent excitation, and this betrays itself moreover often 
 macroscopically by local contraction, but in all cases, it is easily 
 recognised by means of the microscope. 
 
 But under this hypothesis, the occurrence of positive kathodic 
 after-currents, when a muscle injured at one end is traversed 
 by an atterminal current, is no longer surprising ; for under the 
 hypothesis of an opening kathodic inhibition, this is an immediate 
 and necessary result which ought to follow the procedure to which 
 it has been subjected. Such a preparation does not behave differently 
 in any essential point immediately after a momentary excitation, 
 from a muscle which has been treated locally with veratrine ; and 
 everything which has been previously communicated as to the 
 behaviour of secondary electromotive phenomena in such a muscle 
 can be applied directly to the case now before us. 
 
 In connection with the preceding discussion it is worth noticing 
 that the way in which the end of the muscle is killed is not quite 
 
360 INHIBITION BY EXCITATION AND 
 
 a matter of indiflerence as reg-ards the positive katliodic polarisation 
 which results from the excitation. I observed the strono-est positive 
 kathodic after-currents uniformly, after the local action of con- 
 centrated solutions of soda salts (especially Na CI. and Nag CO3) 
 on the kathodic end of the muscle. Mechanical injury by crushinti- 
 has a decidedly feebler effect, and still more unsatisfactory is local 
 treatment with dilute solutions of potash salts, by which the ex- 
 citability is at once speedily and considerably impaired. It is well 
 known, that the soda salts named, strong-ly excite the muscle 
 substance at the commencement of their action, or at least increase 
 its excitability. Now when they are aj^plied locally, a slow distribu- 
 tion of the salt solutions takes place owing- to diffusion ; the first 
 effect must therefore be to bring" fresh sections of the muscle in 
 the vicinity of the demarcation surface successively into a condition 
 of excitation. It is evident that under these circumstances the 
 conditions for the development of positive kathodic after-currents 
 are most favourable. Thus it is precisely in the first period after 
 killing the kathodic fibre ends by concentrated solution of common 
 salt, when the demarcation current which it produces is still in the 
 act of slow increase, that we usually see the strongest positive effects. 
 
 The capability of such a preparation of yielding positive after- 
 currents which can be led off from the kathodic half of the muscle, 
 under the given conditions, is moreover a very persistent one, and 
 continues, though in diminished degree, even after prolonged 
 washing with dilute salt solution. The muscle then (after a quarter 
 to half an hour), on its electrical excitation, often shows a behaviour 
 quite similar to that after local poisoning with veratrine : thus 
 after a battery current, atterminally directed, has been closed for 
 the shortest possible time, on leading off from the kathodic half, 
 there is at first a very strong deflection in the direction of an 
 increase of the (compensated) demarcation current, that is a negative 
 polarisation, but a moment later, the same stimulus produces a 
 more or less strong positive polarisation. (See above. Table 9.) 
 
 In all cases, in which afteV the death of the muscle at one end, 
 positive kathodic polarisation prevails over negative kathodic with 
 atterminal direction of the current, it can always be shown that 
 with an abterminally directed exciting current, an exactly opposite 
 reciprocal relation exists between the positive and negative anodic 
 after-currents, the negative invariably predominating. 
 
 I hardly need to emphasise further, that in none of the cases 
 before us were any considerable differences of tension in the con- 
 
KATHODIC POLARISATION OF MUSCLE. 361 
 
 tinuity of the muscle within the interpolar tract observed as results 
 of electrical excitation, so that it may be reg-arded as a firmly 
 established fact, that the polarisation phenomena discussed are of 
 exclusively polar origin. This is noteworthy, because otherwise it 
 would be natural to consider the experiments on positive kathodic 
 polarisation of striated muscles as supporting- the views maintained 
 by du Bois-Reymond as to the existence of an internal positive 
 polarisability of muscle. 
 
 If, as I believe, we may maintain that we have in the main, 
 succeeded in referring to the same cause, the positive after-currents 
 which are observed on the electrical excitation of locally veratrinised 
 muscle on the one hand, and those in a muscle of which the 
 kathodic fibre-ends have been killed on the other, there still 
 remains the question to be answered, how positive kathodic polar- 
 isation of currentless muscles, which have been injured as little 
 as possible is to be understood. 
 
 This ought not to be compared with the corresponding phenomenon 
 in the uninjured muscle of Anodonta without further ground ; for 
 this latter is in a state of persistent excitation in all its parts, 
 whilst normal striated muscle may be assumed to be in repose. 
 If in the former case we are dealing only with the phenomena 
 resulting from inhibition of tonic excitation appearing at definite 
 spots, and a relative positivity of those spots caused thereby, then 
 in the latter case it is necessary to assume a local change in the 
 muscle substance in repose, which displays itself by its becoming 
 positive in regard to other unchanged parts of the fibre. As is 
 at once obvious, such a change at the kathode, under the existing 
 circumstances, can be regarded only as a phenomenon consequent 
 upon the preceding closing excitation, by which the same points 
 of the fibre become without doubt first of all strongly negative, 
 so that the idea immediately forces itself upon us, that we are 
 dealing here, so to speak, with a reaction of the living substance 
 against the preceding excitation. 
 
 It is not without interest for the study of the changes of muscle 
 substance at the kathode, of which we are speaking, that we have 
 it in our power in every muscle, by treating any portion of it with 
 certain chemical substances, not only, as I formerly showed, to make 
 it negative^, but also positive as compared with other parts ; and 
 
 ^ ' tJber die Abhangigkeit des Muskelstromes von localen chemischen Verander- 
 ungen der Muskelsubstanz,' Sitzungsber. vol. Ixxxi, p. 74. 
 
362 INHIBITION BY EXCITATION AND 
 
 in this way to produce artificially what may be called a reversed 
 muscle current. 
 
 In the work just quoted, I mentioned experiments which I 
 performed at that time with Nag CO3 and other agents calculated 
 to increase the excitability of muscle substance, in the hope that 
 I might succeed in making by their local action, certain points 
 positive in regard to others. The low degree of sensitiveness of 
 the instrviment, which alone was then at my disposal, an old 
 Meyerstein mirror galvanometer, did not make it possible to answer 
 the question decidedly. 
 
 I have recently resumed these experiments, and have convinced 
 myself that after local treatment with dilute solutions of Nag CO3, 
 Na CI, or veratrine, the differences of tension anticipated can be de- 
 monstrated very easily and with great certainty. 
 
 If the uninjured lower end of a sartorius is dipped for a short 
 time (5-10 min.) in a 0-5-170 solution of Nag CO 3 (a 2-4% NaCl 
 solution proves less efficient) it will be fovmd more or less positive 
 in regard to the middle of the muscle. Under these circumstances 
 employing muscles that were previously currentless, deflections of 
 + 200 scale divisions and upwards with the Hermann galvanometer 
 which I used, were by no means of rare occurrence. Feeble normal 
 currents are got rid of or even reversed. The (reversed) ' soda 
 currents ' may be again got rid of by a rather long washing with 
 a physiological solution of common salt, though less easily than 
 ' potash currents.' 
 
 Now, disregarding other differences, one very essential difference 
 exists between the result, just discussed, of the direct action of 
 certain chemical substances, and the positive kathodic polarisation 
 which arises from the electrical excitation of uninjured currentless 
 muscles. In the latter case, as has been already mentioned, it is 
 obvious that we are dealing with an indirect effect of the current, 
 with a ' Reaction ' of the muscle substance from the primaiy 
 process of excitation produced at the kathode. The chief fact in 
 support of this is, that after a longer duration of closing or with 
 less favourable preparations, a negative preliminary jerk often 
 precedes the positive deflection, as was showm above, and that in 
 less sensitive muscles (especially in those of R. esculent a) there seems 
 to be only a trace of positive kathodic polarisation, or it is altogether 
 absent. With reference to this, I must remark fui'ther, that the 
 positive reaction at the kathode does fiot, as it seems, stand in 
 direct dependence upon the condition of excitability of the kathodic 
 
KATHODIC POLARISATION OF MUSCLE. 363 
 
 spots at the time, but on other, for the present unknown, conditions. 
 Otherwise, after local treatment of the kathodie end of the muscle 
 with dilute solutions of Na2C03, by which excitability is enor- 
 mously increased, we should expect to find, with a corresponding- 
 direction of current, positive kathodie after-currents in those cases 
 also, in which previously no trace of them could be demonstrated. 
 This is however never the case, the result of such a procedure being- 
 always merely to favour in a remarkable manner the negative kathodie 
 and positive anodic polarisation (as expressions of the closing and 
 opening- excitations respectively), whilst positive kathodie and 
 negative anodic after-currents do not appear even with the shortest 
 possible duration of closure attainable by means of the switch. 
 It still remains to be investigated, how the secondary electromotive 
 phenomena would behave under the same circumstances, on exci- 
 tation with induced currents. It must be left for further investi- 
 gations to decide whether the positive reaction of muscle substance, 
 which has been assumed here, is exclusively determined by electrical 
 excitation and only occurs at the spot directly excited (the physio- 
 logical kathode), or whether it may also be observed when other 
 excitants are employed and at a distance from the excited spot 
 namely in the continuity of the muscle, as if it followed the propa- 
 gation of the wave of excitation. 
 
 The results at which I arrived, at the time of my investigations 
 on the consequences of electrical excitation of the muscle of the 
 heart (of the snail and of the frog) as well as of the adductor 
 muscle of Anodonta, gain in significance from the fruits of the 
 present work. For I am now in a position to generalise the 
 conclusions to which I was then led by the observation of the 
 effects of exciting the cardiac muscle when in different states of 
 contraction. 
 
 The assumption of two processes of inhibition, antagonistic to 
 the polar processes of excitation, which seemed to be incontestable 
 for the cardiac muscle in the state of systole, now proves to be 
 that which is capable of affording the simplest explanation of the 
 phenomena following on electrical excitation of striated muscle. 
 
 The same thins- is true for results of mechanical stimulation 
 as for electromotive after-effects. The two methods of investigation, 
 viz. the examination of the changes in form of the excited muscle 
 on the one hand, and the determination of the condition of polar- 
 isation at the close of excitation on the other, mutually supplement 
 each other in the desired way, so that a satisfactory insighf into 
 
364 INHIBITION BY EXCITATION, 
 
 the nature of the chang-es effected by the current is in fact to be 
 attained only by the combination of both methods of investig-ation. 
 In this connection it should be distinctly noticed, that a direct proof 
 of the presence of an antagonistic process following or preceding- 
 excitation, and indicated by corresponding changes in foim of the 
 muscle, is ob\'iously only possible in the course of a persistent 
 contraction of the muscle which has already come into existence ; 
 without this, it could only be inferred in a most indirect way, 
 by observations of the changes of excitability. On the other 
 hand, the investigation of secondary electromotive phenomena serves 
 to demonstrate with complete certainty, the existence of polar 
 antagonistic processes in muscle when in a state of repose. 
 
 Positive anodic polarisation coupled with negative kathodic on 
 the one hand, positive kathodic polarisation coupled with negative 
 anodic on the other, owe their origin to polar antagonistic 
 alterations of the muscular substance : one set of these alterations 
 produces negativity and the other set produces positivity of the 
 affected spots. The closing and opening- contraction rej)resent the 
 mechanical exeitatoiy effect of the former, and the opening and 
 closing relaxation the inhibitory effect of the latter (when a tonic 
 condition of contraction is present). Both contraction and relaxa- 
 tion are no doubt conditioned by chemical changes of the excitable 
 muscle substance, evoked by the influence of the current, though 
 nothing definite can be said at present as to the nature of these 
 changes. But while the changes which take place on closing the 
 current are directly produced by it, the effects of opening it are 
 essentially reaction phenomena, which have their origin in the 
 altered muscular substance ; a statement which applies not only to 
 the anodic opening excitation, but also to the kathodic opening 
 inhibition. 
 
Part III. 
 
 RESEARCHES 
 
 RELATING TO THE 
 
 ELECTEICAL ORGANS OF MALAPTEEUEUS 
 
 AND 
 
 TOEPEDO. 
 
XI. 
 
 OBSEEVATIONS AND EXPEEIMENTS 
 
 ON MALAPTEHURUS BEOUGHT TO BERLIN ALIVE. 
 By E. DU BOIS-REYMOND. 
 
 Gesammelte Abhandlungen zur allgemeinen Muskel- und Nerven-physik, 
 • vol. ii. p. 60 1. 
 
CONTENTS. 
 
 PAGE 
 
 1. Introduction ............ 369 
 
 2. Living Malapterurus at Berlin. How they may be kept .... 372 
 
 3. Natural History. Habits of the Malapterurus 374 
 
 4. Mode of experimenting with living Malapterurus 378 
 
 5. Subjective testing of the shock of the Malapterurus 387 
 
 6. Of the direction of the shock in the Malapterurus 388 
 
 7. Physical investigation of the shock of the Malapterurus .... 390 
 
 8. More detailed investigation of the distribution of tensions in the active 
 
 organ of Malapterurus 397 
 
 9. On the influence of the length of the linings of the leading-off covers upon 
 
 the strength of the branch current in the experimental circuit . . .402 
 
 10. More precise examination of the value of the Frog-interrupter in experi- 
 
 ments on the Malapterurus 403 
 
 11. Of the relative immunity of Malapterurus from electric shocks . . . 404 
 
 12. Experiments on living electrical nerve and organ after separation . . 410 
 
 13. Of the chemical reaction of the electrical organ of the Malapterurus . . 41 2 
 
XL 
 
 1. Introduction. 
 
 The unusual interest which electrical fishes ^ have always excited, 
 has been increased still further at the present day, by its becoming 
 known that the endowment which made these animals so long an 
 object of inquisitive wonder does not belong to them alone, but 
 appears to be a special application of a common property, such as 
 is frequent in the animal world. Since we have known that all 
 nerves and muscles of all animals are capable of electrical actions, 
 electrical fishes have indeed lost somewhat of their wonder, but this 
 loss is abundantly compensated by the hope which now attaches to 
 their investigation, that the solution of the great problems of the 
 general physics of nerve and muscle may be promoted by it. 
 
 Until lately we had reason to regret, that of the three electrical 
 
 ' My researches on electrical fishes are found in the following : I. Nachricht von 
 einem nach Berlin erlangten lebenden Zitterwelse. Monatsberichte, etc., 1857, 
 pp. 424-429 ; Moleschott's Untersuchungen zur Naturlehre des Menschens und der 
 Thiere, 1S58, vol. iv. p. 97. II. Ueber lebend nach Berlin gelangte Zitterwelse 
 aus Westafrika. (Lecture given in the Friedrichsitze of the Academy on Jan. 28, 
 1858 ; containing a popular account of what had been ascertained up to that date 
 about the Malapterurus). Monatsberichte, etc., 1858, pp. 84-1 11 : Moleschott's 
 Untersuchungen, etc., 185S, vol. v. p. 109. III. Bemerkungen iiber Reaction des 
 elektrischen Organes und der Muskeln, Archiv filr Anatomic, u. s. w. 1859, 
 pp. 846-848. IV. Ueber Jodkalium Elektrolyse und Polarisation durch den Schlag 
 des Zitterwelses. Monatsberichte, etc., 1861, pp. 1105-1128; Moleschott's Unter- 
 suchungen, etc., 1862, vol. viii. p. 549, V. Beschreibung einiger Vomchtungen 
 und Versuchsweisen zu elektro-physiologischen Zwecken. Abhandlungen der 
 Akademie, 1862, Berlin, 1863, 40. Physikalische Klasse. In §§ xviii and xix of 
 this paper, pp. 148-162, the Frog-alarum and Frog-interrupter, of which I made use 
 in the experiments on the Malapterurus, are described. VI. Ueber die raumliche 
 Ausbreitung des Schlages der Zitterfische. Monatsberichte, etc., 1864, pp. 31 7-354 > 
 Moleschott's Untersuchungen, etc., 1865, vol. ix. p. 437. VII. Experimentalkritik 
 der Entladungshypothese tiber die Wirkung von Nerv auf Muskel. Monatsberichte, 
 etc., 1874, pp. 519-560. No. V is printed in the first volume of the Collected Papers, 
 pp. 213-227. IV, VI, and VII follow here as Abh. XXIX, XXX, and XXXI. The 
 present paper, XXVIII, takes the place of the three first numbers for the reasons 
 stated in the preface to this volume. 
 
 Bb 
 
370 OBSERVATIONS AND EXPERIMENTS 
 
 fishes, one of them had remained as g-ood as unknown to us. 
 The electrical ray of European coasts ^ had been many times in- 
 vestigated since the revival of science in the seventeenth century, 
 in all its relations, and each time with more perfect resources. 
 The electrical eel had been investigated in its own home by 
 Alexander von Humboldt, and this fish had already been brought 
 often to Europe, and had been observed by the first experimenter 
 of all times, by Faraday himself. On the other hand, the elec- 
 trical sheath fish^ or Malapterurus electricns^, which inhabits the 
 rivers of Africa, which was known to the ancients, and which is 
 still no rare appearance in Bulak, the fish-market of Cairo, and 
 which thus, next to the electrical ray, seemed the most accessible to 
 science, was notwithstanding very little known. Until the year 
 1857 we possessed only isolated anatomical details which left even 
 the coarse anatomy of its organ unexplained. As to its electrical 
 action, there existed indeed already a valuable experiment on it, as 
 will be shown hereafter, but this was so concealed in literature, that 
 it might escape even the most earnest student of the subject. 
 (See note to p. 388.) 
 
 A fortunate chain of circumstances has in so far supplied this 
 want, that now the Malapterurus ranks with the Torpedo as one of 
 the two electrical fishes of which the anatomy is best known ; and 
 that not only the most pressing physiological questions relating to 
 it are answered, but light has been thrown on entirely new aspects 
 of the phenomenon. 
 
 Theodor Bilharz, professor of anatomy in the medical school in 
 Cairo, who has since unfortunately been snatched away from science 
 by death, published in 1857 an anatomical description of the 
 Malapterurus, worked out with the aid of modem appliances ^. In this 
 investigation he first obtained an insight into the structure of an 
 electrical organ. Setting aside the parts intended for support and 
 nourishment, such an organ, according to Bilharz, is to be re- 
 garded as a direct continuation of the nervous system. In an 
 electrical organ a great number of small plates are arranged above 
 
 ^ For the sake of brevity, I speak of the family of the Torpediniae, so rich in 
 genera and species, and which was increased by Linnaes' species Itaja Torpedo, by 
 the name of one of the three electrical fishes, i. e. Torpedo. 
 
 " Herr W. Peters has brought into use the name Maloptenirus, which is at least 
 somewhat more correct than the name framed by Lacepfede, Malapterums (instead 
 of Malakopterurus). [Considering that Malapterurus is familiar to zoological stu- 
 dents, it has been thought best to adopt that mode of spelling. Ed.] 
 
 ^ Das elektrische Organ des Zitterwelses anatomisch beschrieben, etc. Leipzig, 
 1857, Fol. 
 
ON LIVING MALAPTEKURUS. 371 
 
 and by the side of each other, the substance of which, according to 
 Bilharz, does not distinctly differ from the substance of gang-lion 
 cells. These are connected with the electrical nerves, and are the 
 seat of the development of electricity ; i. e. at the command of the 
 electrical nerves the surfaces of all the little plates which look in 
 one direction become positively electrical, and the opposite surfaces 
 become negatively so. On this account the little plates are called 
 electrical plates. The direction of the shock is always vertical to 
 the plane of the plates. In the Torpedo, as the plates lie hori- 
 zontally in the natm-al position of the fish, the direction of the 
 shock is vertical in the organ from the belly to the back. In 
 the Gymnotus, where the plates lie vertically, the direction of the 
 shock is horizontal in the organ from the tail to the head. In the 
 Malapterurus, according to Bilharz, the plates also lie in vertical 
 planes. Thus we might infer that in this fish, as in the Gymnotus, 
 the electricity will flow horizontally. But what will be the direc- 
 tion of the shock ? Will positive electricity flow from the tail 
 towards the head, in other words, will the front surface of the plates 
 be the positive, and the back be the negative, as in the Gymnotus, 
 or conversely ? 
 
 On this point also Bilharz' investigations already permitted 
 a conjecture. The connexion of the electrical plates with the 
 nervous system is as follows. The electrical nei*ves, by continuous 
 division, resolve themselves into numberless terminal branches, 
 which finally sink into one of the surfaces of each electrical plate 
 and are merged into its substance. As Signor Pacini has already 
 observed ^, this embedding of the ends of the nerves in the case of 
 the Torpedo and of the Gymnotus, is exclusively into that surface of 
 the electrical plates which becomes negative in the discharge. 
 Thus in the Torpedo this is the lower, and in the Gymnotus 
 the posterior surface of the plate. Now in the Malapterurus, the 
 nerves also seem to enter the posterior surface of the plates. 
 Hence Bilharz inferred that in this fish, as in the Gymnotus at the 
 moment of discharge, the posterior surface becomes negative, the 
 anterior positive, that is, that the discharge in the organ would 
 pass from back to front ^. He was obliged to let the matter rest 
 there, without being able to put this conclusion to the test of 
 experiment. According to his account and that of Herr Mar- 
 
 ^ Sulla Struttura dell' Organo elettrico del Gimnoto e di altri pesci elettrici. 
 Florence, 1S52, p. 25. 
 ^ Loc. cit. p. 44. 
 
 B b 2 
 
S72 ^ OBSEKVATIONS AND EXPERIMENTS 
 
 kusen^, the procuring- of living' Malapterurus in Cairo is attended 
 with extraordinary difficulties, which are caused by the restrictions 
 to which the sale of fish in Bulak is subject in consequence 
 of taxation. 
 
 Thus the investigation came to a standstill in Egypt at the 
 moment when it acquired the greatest interest, by the discovery 
 of a connexion between the arrangement of the nerves and the 
 direction of the discharge in the electrical organ, having been 
 brought within view. But unexpectedly, the way was made plain 
 for its prosecution from another quarter. 
 
 2. Living Malapterurus at Berlin. How they may be kept. 
 
 As early as 1855, Scotch missionaries sent some specimens of the 
 Malapterurus in spirits to Edinburgh from Creek Town, a settle- 
 ment situated about 90 kil. up the Old Calabar River. Mr. Andrew 
 Murray thought that he recognised in them a new kind, which he 
 named ^ Malapterurus Beumensis, after the Bay of Benin into which 
 the Old Calabar river empties itself. In the summer of 1857, Mrs. 
 Anderson, the wife of one of the missionaries, undertook to bring 
 three living Malapterurus from Creek Town to Edinburgh, and 
 carried out her project successfully in spite of shipwreck suffered on 
 the way. In Edinburgh the fish came into Prof. Goodsir's hands, 
 who was on the point of travelling to Berlin. He was so very 
 obliging as to bring one of them with him, and as he saw that 
 I was ready to devote myself to the subject, he had the other two 
 sent subsequently. 
 
 This took place in August 1857. On a former occasion I de- 
 scribed the arrangements which I made, in order to preserve the fish 
 in life and health. It would be useless to repeat this descrij)tion, as 
 the same difficulties will not occur again. Besides, at first, my 
 exertions were not very successful. Only one of the three fish lived 
 until Januaiy of the following year. I will, however, give the 
 method which I finally adopted for the preservation of malapterurus, 
 as they may possibly be useful. 
 
 In the summer of 1858, I received three from Goodsir and my 
 friend Dr. Bence Jones of London", but one of them died veiy 
 
 * Bulletin physico-math^matique de 1' Academic de St. Petersburg, t. xii. 1854, 
 p. 203. 
 
 * The Edinburgh New Philosophical Journal, New Series, 1855, vol. ii. pp. 49, 
 379 ; vol. iii. p. 188 ; Eeport of the British Association, etc., 1855. Transactions of 
 the Sections, p. 114, 
 
 ' They belonged indeed to the same consignment. Bence Jones sent two from 
 
ON LIVING MALAPTERURUS. 373 
 
 soon. In the summer of 1859, Bence Jones sent two fish through 
 the captain of a ship by way of Liverpool. The ship, which 
 brought them from Hull to Hamburg, had such a stormy j)assage, 
 that on arrival here, one fish was dead, and the other apparently 
 dying, with white corneas and bent barbels. I succeeded in restoring 
 it, by exposing it for weeks to a stream of fresh water, and it was 
 just this fish which lived longest of all, till the autumn of 1864, and 
 it grew considerably in confinement. 
 
 I kept the Malapterurus in a trough four feet (125 cm.) long, 
 one foot and a half (47 cm.) broad, and two feet (^6<^ cm.) deep, the 
 floor of which was formed by a slab of slate, and its walls were of 
 plate glass. The trough was filled with tap-water up to 5 cm. 
 from the edge, which was renewed at a sufficient rate. In order to 
 maintain the temperature at the proper height, the trough stood in 
 a zinc case five feet (157 cm.) long, two feet (6g cm.) broad, and 
 thirteen inches (34 cm.) deep, encased in wood with a layer of 
 saw-dust, and its lid was fitted closely to the glass plates by means 
 of a packing of tow. The water in the zinc case was warmed by a 
 small copper boiler arranged at the side, the source of heat being 
 an Argand gas burner, so that a thermometer floating in the trough 
 indicated I7'5 — 2o"C. Goodsir had told me that 70° Fah. = 2i-i°C 
 is the proper temperature for the fish. But I found accidentally, 
 that they bore a temperature of only 15°, and at one time, when I 
 fed the trough with spring-water, the Malapterm'us which was 
 then the sole inhabitant of it, always sought out the very spot at 
 which water at only about 11° was flowing in. As the animal was 
 in a state of inanition, perhaps its instinct impelled it to diminish 
 its respiration and thereby its need of nourishment ; for Liebig's 
 assertion : ' It is not a diflScult task to endure hunger for a long 
 time at the Equator, but cold and hunger wear away the body in a 
 short time ' ^ — does not hold good for animals of variable tem- 
 perature. 
 
 Malapterurus take refuge in the dark whenever they can. I 
 covered the trough therefore with a wooden black varnished lid, 
 which shut out the light almost entirely, but admitted air. Water 
 plants in the trough and earth at the bottom of it make cleanliness 
 more difficult, without proving to be of any use. The trough must 
 
 London by way of Hamburg, and Dr. Turner, at that time Goodsir's demonstrator 
 of Anatomy, had the goodness to bring the third with him by way of Leith and 
 Hamburg. 
 
 ^ Chemische Briefe, 3rd Edn. Heidelberg, 1S51, p. 401. 
 
374 OBSERVATIONS AND EXPERIMENTS 
 
 be divided by a fine trellis of varnished iron wire, into as many- 
 divisions as there are Malapteriu'us to be kept, for these animals 
 carry on violent battles, in which they chase each other, bite one 
 aiiother and will not let go their hold, and thus naturally wear 
 themselves out. They also attack other fish furiously and kill uhem 
 with electric shocks (see the following' section). Whether they dis- 
 charge electric shocks in their battles with each other, I do not 
 know. 
 
 It fortunately proved very easy to feed the fish. As, in the 
 specimen in S2:)irits of wine which reached Edinburgh, the in- 
 testine contained remains of fresh water Crustacea, an efibrt was 
 made both there and here to procure similar kinds [gammarus, 
 asellus, daphnia, etc.) for them. But as this was difficult, earth- 
 worms were tried in Edinburgh as food, and as this was successful, 
 worms were given to the fish here also, sO long as there were any to 
 be had, in the autumn. One of them took the worms out of the 
 pincers, swallowed them down without giving them a shock, with 
 a quick sucking motion, and rose to the surface as if asking 
 for more, so that one might consider it as, to a certain extent, 
 tame. When later in the year we fell short of worms, the idea 
 occurred to Graff, the keeper, to give the fish strips of beef like 
 worms, and they not only devoured them eagerly, but this food 
 also evidently agreed with them very well ; perhaps even better 
 than worms, which generally swarm with parasites, and which 
 therefore were discontinued also in summer for later consignments 
 of fish. 
 
 Exact information is wanting as to how the Malapterm*us ac- 
 complished the journey from Africa to Scotland, They were 
 brought here from Scotland or England, each one singly with some 
 water-plants in a gold-fish bowl wdth a cover of basket-work, which 
 fitted so exactly that the cover seemed as if plaited over the bowl. 
 It was hung up by the handle of the cover in the cabin or the railway 
 carriage. The water-plants answered the purpose of lessening the 
 shaking of the water and protecting the fish from knocks. 
 
 3. Natural History. Habits of the Malapterurus. 
 With the aid of the rich material of the Berlin collections, Herr 
 Peters, who had observed the living Malapterm-us in South-east 
 Africa, answered in the negative the question, whether that from 
 
 ^ Monatsberichte, etc., 1868, p. 121 ; Naturwissenschaftliche Eeise nach Mozam- 
 bique, etc., Zoologie IV, Flussfische. Berlin, 186S, 4°. p. 41. 
 
ON LIVING MALAPTERURUS. 375 
 
 West Africa is a new species. I refer to his^ paper for the zoo- 
 logical characteristics, which at first sight and with less complete 
 comparison, had led to this suggestion. 
 
 Our specimens were all small, without barbels, between 12-5 and 
 27-5 cm. long, whilst Bilharz measured one of 32''= 57-5 cm. 
 The quick growth of one which was kept alive longest, seems to 
 show that the creatures were still young. Most of them were 
 females. 
 
 It is impossible to form a good idea of Malapterurus from 
 specimens in spirits of wine and from drawings taken from them. 
 In drawings, the position in which the barbels are held is par- 
 ticularly incorrect, as they are represented as hanging down in a 
 flabby manner and curled, whilst in the healthy fish they are 
 extended in straight lines and far apai-t (see Figs. 10, 11). 
 
 Another characteristic in the appearance of the Malapterurus, 
 which is no longer recognisable in specimens in spirits of wine, 
 consists of beautiful, regular, broad transverse folds, which in our 
 fish were always 5-8 mm. in size (unquestionably more in larger 
 fish), and which show themselves on the concave side of the fish 
 when it bends laterally. They are made by the organ which 
 envelops the fish in the form of a thick walled tube, the outer 
 layers of which arrange themselves in folds over the contracted 
 muscles of the flanks, whereas in other fish, the thin and firmly 
 attached skin always follows exactly the contour of the body. 
 
 In life, the organ as well as the skin covering it, is transparent, 
 so that in the light one looks into its reddish depth. In death, 
 especially in specimens in spirits, it is dim and opaque. The 
 general colour of such specimens is gray; in life, oui- fish were 
 brown except the black spots with which they were, as it were, 
 sprinkled on the back ; the weaker ones were yellowish brown, the 
 stronger ones were rather deep red brown. The colour even of the 
 same fish varied. In the dark they became blackish, under the 
 effect of light, bright again. When fatigued by experiments they 
 looked pale, after resting for a few days i,they appeared again deeper 
 coloured. 
 
 The animals usually remained motionless in the darkest part of 
 the bottom of the tank with barbels far extended. At night 
 they seem to become more lively ; at least in Creek Town, fishes 
 which were kept in readiness to be shipped off", perished through 
 jumjjing out of their vessels during the night. When the fish had 
 fresh water given to them in the tubs in which I originally kept 
 
376 OBSEKVATIONS AND EXPERIMENTS 
 
 them\ they swam about briskly in the little whirlpool, and as the 
 ' frog'-alarum ' showed (see below, p. 384), discharged their 
 batteries repeatedly, whether to ward off fancied dang-er, or as an 
 expression of content, it would be difficult to say. One of the fish 
 had evidently taken a dislike to the electrodes of the frog'-alarum, 
 and often attacked it with bites, which it accompanied with several 
 discharges following rapidly on each other. The sight of red 
 colours, e. g. of a stick of sealing-wax, did not appear to excite the 
 Malapterurus in the same way as it does frogs and some other 
 animals. 
 
 Generally the Malapterurus answer every touch, even by a non- 
 conductor, with one or more discharges. Sometimes, however, they 
 escape from the hand by a violent movement, without giving a 
 shock. This shows that in the former case the discharges are 
 voluntary and not merely reflex. The frog-alarum also showed (see 
 below, p. 384) that the fish discharged many times without any 
 assignable cause. 
 
 Other fish, which are put into the water with the Malapterurus, 
 are immediately attacked by them with electrical broadsides. One 
 sees the fish immediately lose their balance and drift apparently 
 lifeless with the side towards the light. If they are withdrawn at 
 once, such fish come to themselves again. Given over to their fate, 
 they die. The three first Malapterurus which reached Edinburgh, 
 killed all the gold-fish in the tank of a hot-house. 
 
 One afternoon (in September, 1857) I placed in each of the tubs 
 just mentioned, with the Malapterurus already in it, a tench {tinea 
 chrysitis) about 15 cm. long, and a loach {cobitis fossilis) of the same 
 length. A violent tumult at once arose in the three tubs. Here 
 and there a tench leapt into the air, whilst the loaches, twisting 
 themselves in an eel-like fashion, coursed round the circumference 
 of the surface of water as if driven by deadly fear, and at last, one 
 after the other, threw themselves over the edge of the tub, 2-5 cm. 
 high, between the tub and the wire netting covering it. When 
 they were put in again, they escaped repeatedly, until I made the 
 edge twice as high, by letting off the water. Naturally, each 
 electrical fish of prey had equally distressed both tench and loach, 
 and the water had become turbid in each case, by the stirring 
 up of the earth with which at that time I still kept the bottom of 
 the tubs covered (see above p. ^']2))- I should have remained in the 
 
 * They were of the same kind as those described in the following paragraphs ae 
 experimental tubs. 
 
ON LIVING MALAPTERURUS. 377 
 
 dark as to the precise course of events, if I had not taken the pre- 
 caution to place the electrodes of the frog'-alarum in one of the 
 tubs. But this betrayed clearly enoug-h what took place. Its bell 
 went on ringing continuously, indicating now a strong, now a weak 
 branch current in the nerve, either because the fish discharged with 
 different strength, or because the electrodes took up a more favoui'able 
 position at the moment of the discharge. The hammer seemed 
 sometimes to stick to the bell, indicating that the electrical fish was 
 tetanising its victim. Now followed an interval of peace, until 
 presumably, the tench, aroused from their stupefaction, began again 
 to move, and the Malapterurus also rested, felt disposed for a fresh 
 attack. Ever and anon the tumult started again, now in one 
 vessel, now in the other ; the individual periods, however, became 
 always shorter, and were separated by longer intervals. I then 
 left the field of battle. When I came into the laboratory the next 
 morning, the loaches lay dead upon the ground. Thus during the 
 night, they had again escaped even over the edge of the tubs 5 cm. 
 high. In the tubs of the two larger fish, the tench were dead. 
 They must have been dead, a long time, because they were stiff and 
 their corneas had begun to cloud. The water was clear, thus peace 
 had long prevailed. The little adventure did not seem to have 
 affected the Malapterurus. The smallest had not been able to kill 
 its tench, but it died very shortly after, although I put it in 
 another vessel. A fourth couple of tench and loach, which I kept 
 as a check on the experiments in a fourth tub, under circumstances 
 quite the same in other respects, lived for weeks afterwards. 
 
 A frog, under the effect of the discharge of the Malapterurus, 
 stretches itself as in tetanus from strychnine. But on account of 
 the poisonous skin secretion of the batrachians ^, I did not make 
 any experiment with them, similar to the foregoing one. The fish 
 allowed earth-worms to cm"l themselves about their barbels without 
 discharging (comp. above, p. 374). But in general they seemed not 
 to give a shock when their barbels were touched, but to draw back ; 
 on the other hand, as already observed, when the skin was touched, 
 they discharged almost invariably. 
 
 When the fish became ill, they changed their habits very much. 
 They forgot their fear of light, and without allowing themselves to 
 be frightened away by the proximity of an observer, they kept 
 themselves at the surface by a quick and laboured action of their 
 
 ^ Untersuchungen, etc., vol. ii. Abth. i. p. 17. Anin. 2. Comp. Gratiolet et Cloez, 
 Comptes rendus, etc., 1851 ; t. xxxii. p. 592. 
 
378 OBSERVATIONS AND EXPERIMENTS 
 
 pectoral fins, and by supporting' the tail on near objects. At the 
 same time they breathed extremely fast, half water, half air, with 
 the upper jaw out of the w^ater. They expelled the water mostly 
 from the mouth, more rarely from the gills, so that at each breath 
 they threw up bubbles, and soon a wreath of foam of large bubbles 
 surrounded them. Many fish, among others the Gymnotus^, rise to 
 the surface to take in air, but this throwing up of bubbles was 
 recognised by our keeper Graff as a sign of distress and weakness 
 in fishes. Malapterurus in this condition no longer discharged on 
 being touched ; the barbels were no longer stretched out in straight 
 lines, the skin became wrinkled and ulcerated ; blood issued from 
 the fins and the gill-covers ; sometimes at last a violent motion 
 convulsed the jaws and gills. Fish so affected inevitably died. 
 Aeration of the water by means of bellows remained without 
 result. Such fish ought to be used systematically in good time for 
 experiments, instead of delaying, as I unfortunately often did, until 
 they were found dead, in which case it is not only impossible to 
 experiment, but the event may be observed so late, that the fish are 
 useless even for histological purposes. One can, however, be never 
 quite secure against this mishai?, for the fish may die without clear 
 premonitory symptoms, and as they generally lie quiet deep down, 
 their death may remain unnoticed so long, that nothing more can 
 be done with them. 
 
 4. Mode of experimenting with living Malapterurus. 
 
 After the above experience there is no doubt that of the three 
 electrical fishes, the Malapterm-us is the best fitted to be kept in 
 confinement. The Torpedo, as an inliabitant of the sea, cannot bear 
 comparison with it'; the Gymnotus, while it can be kept equally 
 
 ' Comp. Edw. Bancroft, An Essay on the Natural History of Guiana, London, 
 1879, p. 192. 
 
 Hugh Williamson, Philosophical Transactions, etc., 1775> P- 94 > Alex. Garden, 
 ibidem, p. 102 ; Humboldt, Recueil d' Observations de Zoologie et d'Anatomie 
 compar^e, etc., Paris, 181 1, 4". p. 61 ; Faraday, Experimental Researches in Elec- 
 tricity. Reprinted from the Philosophical Transactions, etc., vol. ii. London, 1844, 
 p. 3. Ser. XV. § 1753. (The paper, frequently quoted in the following, by Faraday, 
 upon the electric eel, is also found in Poggendortt's Annalen, etc., 1842, Erganzungs- 
 band, p. 391.) 
 
 ■^ UntU the formation of zoological stations, no one in my knowledge has succeeded 
 in keeping the Torpedo alive longer than fifteen days, and John Dary is the only one 
 who was so fortunate. (Researches physiological and anatomical, London, 1839, 
 vol. i. p. 48; comp. Faraday, Experimental Researches, etc., loc. cit. p. 2, § I752-) 
 John Todd, who made his observations at the Cape of Good Hope, and therefore 
 presumably on Astrape. Cupcnds 'MuW. Henl., (Philosophical Transactions, etc., 1S16, 
 
ON LIVING MALAPTEKURUS. 379 
 
 long", is inconvenient on account of its size and the force of its 
 discharges. 
 
 But of the three electrical fishes, it is again the Malapterurus, 
 which is the best suited for all experiments in which the electrical 
 nerves have to be dealt with ; as, for example, when these have to 
 be employed in connexion with the organ, in the same way as a 
 muscle nerve is used in connexion with the muscle. In the Malap- 
 terurus, each lateral half of the organ is provided with a single 
 nerve, which one cut exposes to a considerable length almost with- 
 out bleeding-. The electrical nerve is, as it were, already prepared by 
 nature ; indeed the same cut suffices for both nerves. In the Torpedo, 
 each organ contains four, comparatively much shorter nerves, which 
 demand far more skilful preparation and greater injury in order to 
 expose them, than the one nerve in the Malapterurus. Moreover, in 
 the Gymnotus, each half of the chief and accessory organ is provided 
 with over two hundred nerves, all of which it is impossible to pre- 
 pare and deal with experimentally at the same time in the living 
 animal. I do not speak of the unusual interest which the investi- 
 gation of the electrical nerve of the Malapterurus offers on account 
 of its structure, hitherto unique in the animal world, and of its 
 origin from a giant ganglion celP. Add to this, that the tissues of 
 the Malapterurus, inclusive of the electrical organ, appeared to me 
 to possess great tenacity of life, recalling that of the frog. But in 
 order to make the most of these advantages, one must have at com- 
 mand any required number of animals, and this would be possible 
 only in their own home. With two or three fishes, whose life must be 
 spared because much still remains to be ascertained about the living 
 animal, one cannot set on foot researches in which one needs a fresh 
 preparation for almost each experiment, as in the investigation of 
 nerve and muscle. Imagine what it would be in the investigation 
 of nerve and muscle, to be limited to the use of two or three un- 
 injured living frogs. Such a consideration may serve to undeceive 
 
 p. 120), and Saint-Linari (Bibliotlifeque universelle de Genfeve, Avril, 1S37, '^ol. 
 viii. p. 395) kept them alive only five days, Matteucci only three days (Essai sur les 
 Ph^nomfenes electriques des Animaux. Paris, 1840, p. 52.) 
 
 ^ One might suspect, that in the Gymnotus also, the electrical nerves are the 
 Deiters' processes of as many giant ganglion cells. M. Schultze, however, saw in the 
 electrical nerves, ordinary nerve tubes, and in the anterior half of the spinal-cord, 
 a remarkable number of large ganglion cells (Zur Kenntniss der elektrischen Organe 
 der Fische. Specially reprinted from the 4th vol. of Abhandlungen der Naturfor- 
 Bchenden Gesellschaft in Halle, 1858, pp. 30, 32, 33). This leads us rather to infer 
 the existence of a structure similar to that of the Lobus electricus in the brain of the 
 Torpedo (comp. M. Reichenheim in the Archiv fiir Anatomie, etc., 1873, p. 751). 
 
380 OBSERVATIONS AND EXPEEIMENT3 
 
 those who form an exag-g-erated idea of what can be determined 
 with a few live Malapterurus in the laboratory, and to whom my 
 results may seem too meag-re. 
 
 Add to this, that the methods of investigation had to be almost 
 entirely contrived. I will relate first, how I was provided in this 
 respect. 
 
 I. Observation, of the shock of the Malnpterurus with reflecting 
 mar/net, scale, and telescope. 
 
 By tbe term experimental circuit, will hencefortb be understood 
 the system of conduction by which the discharg-e of the fish is led 
 off, in order to examine any of its effects, whether the circuit be 
 interrupted or not. A Wiedemann's g-alvanometer was usually 
 introduced into it in order to measure the discharge. 
 
 In all earlier experiments with Malapterurus, ordinary multi- 
 pliers with astatic pairs of needles had been used. Apart from the 
 fact that they are ill-fitted for measurements, such an instrument is 
 here especially unserviceable. At times I found that the needles 
 are caught by the shock when they are at a great angle to the 
 direction of the coil, by which they are either rendered no longer 
 astatic, or may be even almost entirely demagnetised or reversed. 
 Observation of the deflections with mirror, scale, and telescope could 
 never be more advantageously made than here, and Herr Eckhard 
 and I introduced them simultaneously, he at Trieste in experiments 
 on the Torj)edo, and I here in those on the Malapterurus. When a 
 Wiedemann's galvanometer was used, and one of the thermo- 
 bobbins containing ^'^ turns of wire was allowed to act at a 
 distance of 25-75 mm. on a heavy reflecting magnet not rendered 
 astatic, I obtained measured deflections from the discharge of the 
 Malapterurus, without risk of a disturbing effect on its magnetism. 
 The aperiodic magnet would have been of great advantage in these 
 observations, but at that time I had not got so far. 
 
 2. The leading -off cover. 
 
 The shock of the fish in the experimental circuit is weakened by 
 derivation, in proportion as the mass of water in which the fish is 
 contained is greater. The action of the fish is also physiologically 
 violent as the quantity of water is less. Thus they were less able 
 to kill other fish in the trough than in the tubs. That they were 
 able to do so in the Edinburgh basins, which were certainly not 
 
ON LIVING MALAPTERURUS. 381 
 
 smaller than our troug-hs, must have depended on their greater 
 strength at that time (see above, p. ']'j6). 
 
 In their own home, where the supply of Malapterurus is un- 
 limited, an observer would not scruple to take the fish out of the 
 water, and thus get rid of the derivation due to immersion in water. 
 When the fish is in the air, this derivation is confined to its body. 
 But I could not venture to do this. I could only as far as possible 
 diminish the water-derivation. 
 
 The fish to be observed was caught in the landing-net and 
 removed to a shallow cylindrical tub, the experimental tub, filled 
 with water. This was made of coarse porcelain (Gesundheitso-e- 
 schirr) having a diameter of 295 mm. and a depth of 125 mm. 
 That the removal was not effected without violent shocks was 
 proved in a way which will be stated later. For reasons which 
 will also be stated later, an unsilvered glass plate having the same 
 diameter as the tub was placed at the bottom of it. The water 
 was drawn off" by a siphon, until the surface just touched the back 
 of the fish. The mass of water surroundmg the fish was thus 
 diminished as much as seemed compatible with its well-being 
 during the experiment, and the discharge could be led off" as 
 advantageously as possible without raising the fish into the air 1. 
 
 In order to lead off" the discharge, the means used by Faraday 2 
 for the Gymnotus first suggests itself; this is, to apply to the fish 
 two metal saddles which are insulated excepting at the surface of 
 contact. A telegraph wire, insulated by guttapercha, is soldered to 
 the middle of the arch of the saddle, and serves both as a handle 
 and for leading-off" the current. The saddles are made of different 
 shape and size, and of diffferent metals ^ as occasion requires. Gener- 
 ally they are bands of platinum 5 cm. long, i cm. wide, bent to 
 the contour of the fish at different parts of the body. But even 
 the inequalities of zinc mostly vanish as compared with the shock ; 
 if they do not, they are compensated. 
 
 If it is desired to lead as large a proportion as possible of the 
 shock into the experimental circuit, then the contrivance for lead- 
 ing-off* must be somewhat diff^l-ent. Faraday, who needed to be 
 even more careful with his Gymnotus than I with my Malapterurus, 
 
 ^ The late A. v. Bezold, who was helpful to me in many ways in these experiments 
 proposed to replace ordinary tap water by aerated distilled water, in order to in- 
 tensify the discharge. 
 
 2 Experimental Eesearches, etc., loc. cit. p. 5, § 1758. 
 
 2 See the following paper, §§ i, 3, 4. 
 
382 
 
 OBSERVATIONS AND EXPERIMENTS 
 
 endeavoured to diminish the derivation by the water, by insulating 
 the fish from the water at the moment of the shock. But he 
 attempted this with only the two parts of the fish within the 
 saddle, whereas I proposed the same thing- for the part also between 
 the saddles. In this respect the Malapterurus offered great ad- 
 vantages over the long Gymnotus with its lateral movement. 
 
 Fig. lo shows the contrivance with which I attained my object. 
 It has been not inaptly compared to the cover of a mummy 
 coffin. It consists of a guttapercha cover, which, when the fish 
 rests on the glass plate at the bottom of the tub, fits as closely 
 as possible to its body above and at the side, and touches the plate 
 
 Fig. lo. 
 
 with its lower free edge, a a' is the surface of the water in the 
 tub ; for the sake of clearness, it is shown higher than was usual 
 (see above). In order to prepare such a cover for any particular 
 fish, I cut out a model of the fish in lime wood. This served as a 
 last over which the guttapercha, softened in hot water, could be 
 moulded. When the guttapercha had hardened, it was cut off in a 
 line with the lower edge of the last, the cut edge was again softened 
 and flattened on a glass plate wettM with hot water. A handle of 
 guttapercha is welded on to the back of the cover. Inside this cover, 
 at the ends corresponding to the head and tail of the fish, are linings 
 of tinfoil stuck on with shellac ; these are indicated in the figure by 
 dotted lines. At the ends k and s of the cover, these linings are 
 connected with strips of tinfoil, which, protected by insulating 
 material, pass along the back of the cover to the handle, where 
 
ON LIVING MALAPTEEUKUS. 
 
 383 
 
 they are soldered to the small copper plates in which the wires end. 
 These wires, after an insulated course, come to view {k\ s') at the 
 upper end of the handle. The cover is open before and behind in 
 order to leave the barbels and the tail free. 
 
 In using this contrivance, one waits for the moment when the 
 fish is lying nearly across the experimental tub. The cover is 
 then held horizontally over it, until it settles in the most 
 favourable position, and is then quickly placed upon it. In 
 order that the air between the tinfoil lining and the fish may 
 escape, two holes are bored in the cover at its inner edge with a 
 hot punch. Almost at the instant that the cover touches the fish, 
 it gives a shock ; but as it is insulated in every direction excepting 
 where the head and tail are exposed, a considerable strengthening 
 of the discharge is produced, as the following experiment teaches. 
 
 I prepared a cover, which diffex*ed from another intended for the 
 same fish in this respect, that the guttapercha wall between the 
 two lined ends was removed and was replaced by three glass rods. 
 The results in divisions of the scale were as follows : — 
 
 
 I 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 Mean. 
 
 With the partial cover 
 
 With the complete cover. . . 
 
 73 
 
 191 
 
 82 
 
 141 
 
 ?^ 
 
 121 
 
 75-3 
 151.0 
 
 The insulation of the fish, by means of the cover, doubles therefore 
 the strength of the shock in the experimental circuit. The covers 
 would undoubtedly have worked still better, if they had also in- 
 cluded the tail. That the covers often enclosed the fish before it 
 discharged, is shown by the observation that when they were applied, 
 the frog-alarum not unfrequently remained inactive from the in- 
 sufiiciency of the branch current in the excitation tube. If the fish 
 discharges, as it generally does, several times after the excitation, the 
 leading-off cover has also the advantage, that the following shocks 
 are sure to be led through the experimental circuit, whereas the fish 
 cannot be held firmly with the loose saddles, without hurting it. 
 
 Between every two experiments, the cover is put in another tub 
 of water, in order that inequalities in the tinfoil linings may be 
 removed. They are not put in the same tub, because the fish is 
 thereby disturbed, and, probably taking the cover for another fish, 
 discharges frequent electric shocks at it (see above, p. ^76). 
 
384 OBSERVATIONS AND EXPERIMENTS 
 
 3. Frog-alarum and frog-interrupter. 
 
 One of the greatest improvements which I made in the method 
 of these experiments, consisted in perfecting- a contrivance which 
 Galvani had already used in 1797 in his experiments on the 
 Torpedo^, namely, in employing a nerve muscle preparation of the 
 frog, in order to make the twitches produced by the shock serve a 
 double purpose. 
 
 Fii-st, the preparation indicates that the fish has discharged, as 
 to which no certainty can otherwise be had, as the shock pro- 
 duces no visible effect in the fish itself, and the absence of any 
 action in the experimental circuit may be dependent on defect in 
 the circuit. Galvani and Matteucci were content with la\dng 
 the frog preparations on the Torpedo itself and observing the 
 twitches. I diverted a part of the discharge passing through the 
 mass of water in the experimental tub, led it through the nerves 
 which were in contact with the ring electrodes of the ' moist exci- 
 tation tube,' and let the muscle indicate its contraction by ringing 
 a bell. This was the origin of the ' frog-alarum",' which is 
 described in detail, and of which a diagram is given in a former 
 paper, and I will therefore not repeat it here. Within the limits 
 of accuracy there stated, the frog-alarum affords a means of watch- 
 ing the electrical activity of a fish in the experimental tub for 
 many hours together (comp. above, p. 377). If the electrodes of 
 the excitation tube were connected by wires with the iron hoop of 
 the landing-net and with a small iron disk at the bottom of the 
 net, the frog-alarum indicated the discharges made by the fish 
 during its removal from the trough to the experimental tub, but 
 for the most part, these did not begin to act until the fish emerged 
 from the large mass of water in the trough ^. 
 
 * Memorie sulla Elettricita animale . . . al celebre Abate Lazzaro Spallanzani, 
 Bologna, 1797, 4^ p. 74 ; Opere edite ed inedite ec. Bologna, 1 841, 4°. pp. 411, 41 2. 
 Matteucci made use of the same contrivance without naming Galvani (Essai sur 
 les phenomfenes electro-physiologiques des Animaux, Paris, p. 144 ; Lezioni di 
 Elettro-Fisiologia ec. Torino, 1856, p. 2 ; Corso di Elettro-Fisiologia ec. Torino, 1861, 
 
 p. 115)- 
 
 ^ See 'Description of some apparatus and modes of experiment,' etc., Gesammelte 
 Abhl. vol. 1. p. 213. [Fig. II with its description are interpolated from a previously 
 published paper, not included in this volume. Ed.] 
 
 ^ Comp. above, p. 380. — Gay-Lussac had already observed the intensifying of the 
 discharges in Torpedos, when they were lifted out of the water (Humboldt, Obser- 
 vations, etc., p. 79). The fact recalls the disappearance of the current of a gastro- 
 cnemiua immersed in a dilute salt solution. 
 
ON LIVING MALAPTERUEUS. 
 
 385 
 
 The second purpose for which the frog-preparation is adapted in 
 experiments on electrical 
 fishes is the following. The 
 frog-alanim teaches that 
 the excited Malapterurus, 
 if it is at all vigorous, 
 seldom discharges once 
 only. Generally two or 
 three strokes of the bell 
 occur, sometimes follow- 
 ing each other closely, 
 sometimes separated by a 
 longer interval. Without 
 having recourse to other 
 means, it is therefore im- 
 possible to ascertain the 
 efiect which any given 
 condition exercises on the 
 strength of the branch 
 current led into the ex- 
 perimental circuit. For it 
 always remains doubtful, 
 whether the differences 
 observable result from 
 that condition, or from the 
 varying number and suc- 
 cession of the discharges. 
 For reasons easily under- 
 stood, nothing could be 
 effected here with the 
 usual mechanical means. 
 But it is obvious that the 
 frog-preparation might be 
 relied on for the timely 
 opening of the experi- 
 mental circuit after exci- 
 
 Fig. II. E and E,, zinc electrodes in the trough 
 containing the fish. F, H, G„ the bell, hammer, 
 and muscle arranged to form an alarum, a, p, q, 
 lever of the frog-interrupter. ff,„ its muscle, p, the 
 4- i.' 1, i.1. £ J. 1- 1 platinum point which rests upon the supporting 
 tatlOn by the brst shock, plate. q, amalgamated copper wire which dips intS 
 With a given nerve- mercm-y. K, K„ keys. B, galvanometer. The 
 , ,, , 1 ty action of the interrupter is such, that only one 
 
 lengtJl and a muscle Ot shock can pass through the galvanometer, for by 
 given size and effi- *^® contraction of the muscle G,, the galvanometer 
 . ,, , . circuit is broken &t q and not again closed. The 
 
 ciency, 
 
 , ^ . circuit is broken at q and not again closed, 
 
 the muscle raises alarum of course goes on ringing. 
 
 c c 
 
386 OBSERVATIONS AND EXPERIMENTS 
 
 a given overweig-ht in an interval which, after an inappre- 
 ciably short excitation of the nerve, is always the same, and this 
 action could be employed to open the experimental circuit. The only 
 question was, how the time relations of the twitch corresponded to 
 those of the shock, about which nothing* was then knowTi^. The 
 first experiments taug-ht me that the duration of the shock is of the 
 same order with the duration of the twitch, aud this led to the 
 construction of the ' frog--interrupter,' which has also been already 
 described in detail and accurately drawn, and its properties tested". 
 I need not therefore dwell upon it here. As was there stated, it is 
 a more convenient form of the same contrivance used by Helmholtz 
 in measuring- the velocity of excitation in the nerve after Pouillet's 
 method, and in accordance with this more general character it has 
 acquired importance^ in many relations. In the present experi- 
 ments it serves the following purposes : i. it allows only the first of 
 the several shocks which the fish successively makes to pass into the 
 experimental circuit ; 2. it admits a proportion of the first shock, 
 which is greater or less according to the overweight used, and 
 which, provided that the shocks have similar course but different 
 strengths, is also proportional to the strength ; 3. it may be used 
 to break a derivation circuit by which the shock had previously 
 been kept out of the experimental circuit. We shall investigate 
 later (in § 10) with what degree of certainty the frog-interrupter 
 executes the second task. As, in general, it affords entrance for 
 a part only of the shock into the experimental circuit, it is evident 
 that when it is used, greater sensitiveness of the galvanometer is 
 requisite. 
 
 Each fish was always worked with, only after intervals of several 
 days. I usually put the saddles or covers on it every ten minutes, 
 and I went on for an hour and a half or two hom-s^. Including the 
 
 ^ Herr Eckhard showed at the same time in Trieste with Torpedos, that the shock 
 must last longer than an opening induction shock. (Beitrjige zur Anatomic und 
 Physiologic, vol. i. Giessen, 1858. 4°. p. 168, 169 tf). Since then, Herr Marey, in 
 Naples, has investigated the duration of the shock in Torpedos, with the Pendulum- 
 myograph, and found it to be 1/14", which confirms my result (Comples rendus, etc., 
 1871, vol. Ixxiii. p. 918). I do not need to remark, that myograph experiments on 
 living uninjured fish would meet with almost insuperable difiiculties, so that I had 
 to forego them. 
 
 * Gesammelte Abhandl., vol. i. p. 215 fF. PI. III. Fig. 12. 12a. Comp. the following 
 paper, § 3, Fig. 44. 
 
 ^ Loc. cit. vol. i. p. 227, note, where several applications of the frog-interrupter 
 are mentioned ; Holmgren's investigation of the course of the gastrocnemius variation 
 carried out with its help, is not alluded to. 
 
 * Gesammelte Abhandl., vol. i. p. 221. 
 
ON LIVING MALAPTERURUS. 387 
 
 removal of tlie fish from the troug-h to the experimental tub and 
 back again, it was excited eleven to fourteen times ; however, it 
 gave at least twice or thrice as many shocks. In the course 
 of such a series of experiments, the fish became evidently fatigued. 
 It became pale (see above, p. 375), and at last responded with one 
 discharge only when the cover was put on. In certain respects this 
 was advantageous ; it could not, however, be depended on, as the 
 fish not unfrequently grew impatient towards the end of the series, 
 and unexpectedly gave several exceedingly strong shocks following 
 each other rapidly. 
 
 One difiiculty of experiments with Malapterurus in the form 
 which I gave to them, lies in the necessity for three observers — 
 one to manipulate the fish, one to sit at the telescope, and one to 
 watch the frog-interrupter. I take this opportunity to express my 
 tardy, but not less hearty thanks to Messrs. Pfliiger, G. Quincke, 
 and Rosenthal, who gave me their assistance at that time. 
 
 5. Subjective testing of the shock of the Malapterurus. 
 
 In comparison with its size, the shock of Malapterurus is sur- 
 prisingly violent. If the head and tail of a powerful fish are 
 touched with the fore-fingers in the water, the shock does not 
 extend beyond the knuckles. If it is seized with hands thoroughly 
 wetted, a severe shock is received up to the elbow. If it is touched 
 with one hand, a pricking sensation is experienced in the skin, 
 a burning one in wounds, and a painful shock is felt in all the 
 joints of the submerged parts. The best way to take the shock is 
 to hold with wei hands, ordinary metal handles which are connected 
 by long wires with the linings of a leading-off cover, and to let an 
 assistant put this on the fish. As one is accustomed to test electric 
 shocks in this way, and is not disturbed by anxiety to get proper 
 hold of the fish without hurting it, or by the feeling of repulsion at 
 laying hold of it, one can better judge of the sensation caused by 
 the shock. The shock does not seem so sharp as that of a Leyden 
 jar, but has a somewhat swelling character. Several maxima may 
 be frequently distinguished in it. By sending opening shocks of 
 an induction apparatus, with two Groves in the primary circuit, 
 through the water of an experimental tub, by means of copper 
 plates plunged into it, having first approximated the secondary 
 coil to the primary as closely as possible, the shock which I ex- 
 perienced when my hands were immersed in the water, was certainly 
 
 c c 2 
 
388 OBSEKVATIONS AND EXPERIMENTS 
 
 not stronger than that of a vio^oroiis fish^. An opening- shock 
 with the coil quite pushed in and one Grove in the circuit, taken 
 directly thi-ough the handles, has about the same strength as such 
 a shock. 
 
 6. Of the direction of the shock in the Malapterurus. 
 
 The simplest observation teaches that, in accordance with Bil- 
 harz' prediction, the motion of electricity in the organ of the 
 Malapterurus occurs in the direction of its length. But the further 
 sm-mise, inferred by Bilharz from Pacini's rule, that the current 
 in the organ of this fish would flow from the tail to the head, 
 as in that of the Gymnotus (see above, p. 371), has not been 
 confirmed. 
 
 The very first experiment which I made on the 13th August, 
 1857, with the fish brought with him by Prof. Goodsir, and which 
 I communicated ^ to the Academy on the same day, was the reverse 
 of what Bilharz had inferred from the microscopical appearances, 
 with what seemed to be such complete justification. It has since 
 then been established in numerous experiments, that the shock in 
 the organ of the Malapterurus is invariably directed from the head 
 towards the tail. Just as a column of the organ of Torpedo would 
 have to bend forwards with its upper end in order to become a 
 column of the organ of Gymnotus, so it would have to bend back- 
 wards with the same end in order to become one of the organ of 
 Malapterunis. 
 
 The hope of finding a constant relation between the nerve-ending 
 and the distribution of tension in the electrical plates seemed thus 
 shut out. But a remarkable way out of this difiiculty presented 
 itself. Ecker had observed shortly before in the pseudo-electrical 
 organs of certain species of Mormyrus (J/, dorsalis and anguilloides), 
 that the nerve-tubes do not pass directly into the sm'faces of the 
 pseudo-electrical plates tm-ned towards them, but go into the plates 
 as though through sharply punched holes ; they then swell knob-like, 
 
 ^ Comp. a coiTesponding experiment on the Gymnotus, Ges. Abh. xxx. § 4. 
 
 ^ Monatsberichte, etc., 1S57, p. 424. As early as 1855, *1^® Florentine surgeon, 
 Ranzi, who died shortly afterwards, and who had resided in Egypt for the sake of 
 his health, had determined the direction of the discharge of the Malapterurus by 
 irreproachable experiments, using silver spoons as electrodes. I was unacquainted 
 with these experiments, because they were only printed in the two first volumes 
 of the Nuovo Cimento, which were not to be had either in Berlin or Gottingen until 
 much later, so that I was obliged to obtain a copy of the essay from Italy (Comp. 
 Archiv fttr Anatomic, etc., 1859, p. 209). 
 
ON LIVING MALAPTERURUS. 389 
 
 and finally send ^ numerous ramifications backwaixis into the surfaces 
 originally turned away from their direction of distribution. Max 
 Schultze (then in Halle) reeog-nised in Bilharz' drawings of the 
 electrical plates of the Malapterurus, traces of a similar arrangement, 
 and the thought struck him, that this might be the explanation of 
 the discrepancy between Bilharz' result and that observed by me. 
 After I had sent to Schultze pieces of the organ of the Malapterurus 
 — some as fresh as possible, some preserved in suitable liquids — he 
 succeeded, as he believed, in confirming with certainty his previous 
 conjecture. 
 
 According to him, the nerve-tubes enter the posterior positive 
 surface of the electrical plates indeed, but do not sink into them. 
 They go through a hole in the middle of the plate to its anterior 
 negative side, and there spread out backwards in the shape of 
 a bell. The edge of the bell, as corresponding with the true nerve- 
 ending, blends with the negative surface of the plate at the 
 circumference of the hole^. Thus Pacini's rule might still hold 
 good, according to which, that surface of an electrical plate into 
 which the nerves bury themselves is always negative at the 
 moment of the shock ; but in the case of Malapterurus and of 
 those species of Mormyi-us, in virtue of the peculiar arrangement 
 described, the surface which, according to this rule, ought to be 
 negative, becomes positive, and conversely. 
 
 Schultze's successors reject this explanation. The character of 
 the membrane scarcely admits of its being considered as an expan- 
 sion of the nerve, itself flat, which spreads into it, and it must 
 be admitted that it is as often seen to continue into the plate in 
 a transverse plane, as to originate in the form of a bell ^ with its 
 edge to the anterior circumference of the hole. 
 
 Another reason, hitherto disregarded, in opposition to Pacini's rule, 
 has its weight. Robin succeeded apparently, where Joh. Miiller and 
 Matteucci failed, in eliciting a shock from the pseudo-electrical organ 
 of the common skate *. In this organ the nerves terminate at the 
 
 ^ Untersuchungen zur Ichthyologie, etc. Freiburg, vol. i. 1857, 4°. p. 29. 
 
 ^ Zur Kenntniss der elektrischen Organe, etc., loc. cit. pp. 14, 15. 
 
 ^ Comp. R. Hartmann in the Archiv flir Anatomie, etc., 1S61, p. 661. F. Boll in 
 Schultze's Archiv fiir mikrosoopische Anatomie, 1873, vol. x. p. 242. Babuchin in 
 Centralblatt fiir die medicinischen Wissenschaften, 1875, pp. 131, 132. 
 
 ^ See my Berichte in den Fortschritten der Physik im Jahre, 1846, etc. 
 II. Jahrgang. Berlin, 184S, p. 469; in the year 1847, III. Jahrgang. 1850, p. 440; 
 comp. Untersuchungen, etc., vol. ii. part i. p. 207 ; Peters never obtained a shock 
 from the Mormyri. The natives were also unaware of any striking property of 
 these fish. But he only carne across emaciated fish, the electrical organs of which 
 
390 OBSERVATIONS AND EXPERIMENTS 
 
 anterior side of the plates. The direction of the shock oug-ht 
 therefore to be, as Max Schultze already supposed, the same as that 
 in Malapterurus ^. According- to Robin, it would be the opposite ^. 
 He did not himself perceive the bearing- of his statement ; this is 
 not to be wondered at, as he did not even know that the direction 
 of the shock is different in the Gymnotus and Malapterurus ^ 
 
 7. Physical Investigation of the shock of the Malapterurus. 
 
 The discovery of the Leyden jar, towards the middle of last century, 
 quickly put an end to the theories of the older physicists as to the 
 shock of electrical fish, and its explanation by stupefying emanations 
 or mechanical shock. After Adanson on the Senegal, in 1751, had 
 observed the similarity of the discharge of the Leyden jar and of 
 the sensation produced by the Malapterurus ^, the endeavour arose to 
 prove the identity of the cause of both phenomena in other ways. 
 The attempt was made to reproduce by means of the shock 
 of electrical fish other electrical actions, and particularly to show 
 that the conductors and non-conductors of both are the same. In 
 spite of the efibrts of Walsh ^, Cavendish^, Configliachi '^ and others, 
 there still remained sufficient uncertainty in regard to this point 
 to induce Humphry Davy ^ to express shortly before his death, the 
 opinion that the electricity of electrical fishes might be different 
 from ordinary electricity. Faraday, indeed, only four years later, 
 
 had wasted away to the tendinous sheath. Unfortunately, he could not ascertain if 
 this arose from individual differences, or if it was a question of periodical conditions 
 dependent on the season of the year and spawning time (Naturwissenschaftliche 
 Eeise nach Mossambique, etc., loc. cit. p. 33). For further testimony as to the 
 inactivity of the organs of Mormyrus, see Hartmann, in the Archiv fiir 
 Anatomie, etc., 1861, p. 653. Herr Babuchin, however, records a shock of Mor- 
 myrus (Centralbl. etc., loc. cit. p. 163). Goodsir communicated to me his surmise 
 (if he has printed it, I know not) that the pseudoelectric organs of the common 
 skate are electrically active only during sexual excitement. 
 
 1 Archiv fiir Anatomie, etc., 1858, p. 213. 
 
 ^ Comptes rendus, etc., 1865, vol. Ixi. p. 239; Journal de I'Anatomie et de la 
 Physiologie, etc. By M. Ch. Robin 2™*' Annee, 1865, pp. 507, 577. 
 
 ^ 'Ainsi dans Vappareil ilectrique des rales, comme sur le Gymnote et le Malap- 
 Urure, le courant est constamment dirigi de Vextremitd cephalique vers Vextrimite 
 caudale^ (i.e. in the wire of the multiplier). 'La direction du courant viontre que 
 le pole positif est tovjours virrs sa partie antirieare, et le pole negutif vers sa partie 
 posterieure.^ Journal, etc., I. c. p. 588). 
 
 * Histoire Naturelle du S(5n^gal, Adanson. Paris^ I7£7» P- 135- 
 ^ Philosophical Transactions, etc., 17731 P- 461. 
 
 * Ibidem, 1776, p. 196. 
 
 ' L'identitii del Fluido elettrico col cosi detto Fluido galvanico vittoriosamente 
 dimostrata, etc. Pavia, 1 814, 4°. p. 141, § iii. 
 
 * Philosophical Transactions, etc., 1829, p. 17. 
 
ON LIVING MALAPTEKURUS. 391 
 
 laid down as the first principle of the science of electricity, the 
 identity of all electricity, whatever be its origin ; hut as reg-ards 
 the five electricities — Voltaic, friction, magneto-, thermo-, and 
 animal electricity — in enumerating the eight effects which ought 
 to be obtained in order to prove their identity,- Faraday was com- 
 pelled in the case of animal electricity to leave four blanks, viz., the 
 spark, thermic effect, attraction and repulsion, and conduction 
 through hot air ^. Since then, the three first of these effects have 
 been obtained in the Torpedo, and in part by Faraday himself, in 
 the Gymnotus, and the whole position of the matter is such that no 
 one would seriously doubt that the electricity of the Malapterurus 
 is also ordinary electricity. I thought, nevertheless, that I ought 
 not to let the opportunity slip, to complete Faraday's table as far as 
 possible for this fish also. A part of the experiments to be described 
 were made with simple metal saddles, and not with the leading-ofF 
 cover. As they would have been completely successful with the 
 covers, I have not needlessly hampered the description of them by 
 an account of the mode of leading-off used each time. 
 
 I. Electroli/sis. 
 
 The physiological effect and the deflection of the magnetic needle 
 have been already spoken of. Concerning electrolysis, I first 
 observed, with aid of the frog - interrupter, polarisation of the 
 electrodes by the shock of Malapterurus. This observation is so far 
 of importance, as it enabled me to dispense with the decomposition 
 of water, which latter has been unsuccessful in my hands, if indeed 
 it has succeeded in any other ^. 
 
 In the electrolysis of iodide of potassium, the remarkable cir- 
 cumstance presented itself, which must have been overlooked by 
 observers of the Torpedo and Gymnotus, that a patch appeared under 
 each point. The two effects, polarisation and secondary iodine 
 patch, which are so closely connected together, were made by me 
 the subject of a special investigation which will be found in the 
 following paper. It was here particularly that the frog-interrupter 
 rendered me help which could scarcely otherwise have been obtained. 
 
 In order to observe the decomposition of water, I enclosed two 
 
 ' Experimental Eesearches, etc., vol. i. London, 1839, P- 99 ^l- Series III, 1883, 
 §35isq. 
 
 ^ John Davy states that he decomposed v?ater with gold and platinum wires 
 (Researches, physiological and anatomical, etc., vol. i. p. 15), Saint-Linari with iron 
 wires (Bibliothfeque universelle de Genbve, 1837, t. viii. p. 395). 
 
393 OBSERVATIONS AND EXPERIMENTS 
 
 very fine platinum wires in a guttapercha column, out of which 
 copper wires connected with them projected. I cut throug-h the 
 column vertically to the wires, which now appeared as two platinum 
 points at a distance of 3-5 mm. on the cut surface, moistened the 
 surface with the liquids used by John Davy — sulphuric acid 
 (SO4 H2 : HgO : : I : i) and saturated solution of common salt — and 
 observed the platinum points with a suitable magnifier. A single 
 discharge of the induction apparatus developed bubbles at the nega- 
 tive point ; but all efforts to get the same result with the shock of 
 the fish failed. I succeeded just as little in observing a deposit of 
 copper from a sulphate of copper solution on the negative electrode. 
 I might have tried to replace the positive point by a plate, but I 
 did not do so. 
 
 2. Melafion of the shock of Malapterurus to resistances 
 opposed to it. Its striking distance. 
 
 The attempt failed completely to conduct the discharge through 
 two copper plates 75 mm. long, 40 mm. broad, which stood opposite 
 each other at a distance of 1-3 mm. in the flame of a Berzelius' 
 lamp ; in the experiment, Hankel's flame-currents showed them- 
 selves in the galvanometer ^. 
 
 I never succeeded in seeing the discharge pass over the smallest 
 gap between stationary metallic conductors with the formation of a 
 spark. I was indebted to the kindness of my friends, Messrs. 
 Siemens and Halske, for a spark-micrometer, in which I could 
 make two platinum points approach each other up to o-oioo mm. ; 
 I also made slits with a knife in strips of tinfoil stuck on microscope 
 slides, which were not wider than 0-0033 — 0*0050 mm. I never 
 succeeded, whilst observing with the microscope in the dark, in 
 getting a spark from the Malapterurus. 
 
 On the other hand, I found to my surprise, that the secondary 
 cmTent of the induction apparatus passed over this slit with forma- 
 tion of sparks, even when the secondary coil was at 90 mm. distance 
 from the primary coil. But my surprise increased still more, when 
 it was found that these discharges caused no sensation either 
 between handle-electrodes, or even on the tongue, although the 
 sparks continued to appear. Even gastrocnemius muscles in the 
 
 * Poggendorff's Annalen, etc., 1850, vol. Ixxxi. p. 213: — Elektrischen Unter- 
 Buchungen, Ueber das Verhalten der Weingeistflamme in elektrischer Beziehung. 
 (Aua den Abth. d. math.-phys. Classe d. Kgl, Sachsischen Gesellachaft d. Wiss.). 
 Leipzig, 1859. 
 
ON LIVING MALAPTERURUS. 393 
 
 circuit remained unexcited. If their nerve was in the circuit, 
 twitches presented themselves, as also when the nerve of a nerve- 
 muscle preparation was placed on a gastrocnemius included in the 
 circuit, thoug-h the muscle did not itself twitch. But the latter 
 twitches were of unipolar origin, as it was easy to prove ^. On 
 frequent repetition of the experiment, and especially if the shocks 
 were somewhat strengthened, the slit easily fused together. 
 
 In these experiments, the discrepancy which had always been 
 observed between the strength of the shock in Malapterurus as 
 otherwise tested, and its striking distance, manifests itself afresh, 
 and if I mistake not, more clearly than hitherto. It was ever a 
 matter of surprise that such a powerful shock as that of Gymnotus 
 should show itself so sensitive to hindrances in its path, that it will 
 not pass through flame ^, nor even through loosely hanging chains 
 of one of the baser metals, and that it is scarcely perceived with dry 
 copper handles 3. On account of this behaviour, the identity of the 
 force emanating from electrical fishes with ordinary electricity, was 
 doubted for a long time, and Cavendish sought in vain to remove 
 this diffculty in those theoretical investigations which were far in 
 advance of his time, in which he compared the shock of the 
 electrical fish to that of a battery of great capacity but small 
 tension, which also does not pass readily through a chain *. 
 
 I now give the direct proof of the imcomparable superiority, as 
 regards striking distance, of the shocks produced by our apparatus 
 to those of the electrical fish. I am also in a position to give the 
 solution of this enigma, which has escaped all earlier investigators 
 of such fish. 
 
 Let two equal currents /„ = /,, flow into two conductors A and B 
 of the resistance A. Let the ends of the conductor £ be connected 
 by a side conductor of the resistance Aj, whilst the conductor A 
 forms a part of an unbranched cii'cuit. If the resistance A of the 
 two conductors is increased by the same magnitude 8, it can be 
 shown that I^ loses thereby more than /„ in strength. Let JE be 
 the electromotive force in the unbranched circuit, of which A forms 
 a part, let ;« be a constant > i , and w the actual resistance in both 
 circuits, then we have 
 
 * Untersuchungen, etc., vol. i. p. 4230". 
 
 ^ Humboldt in Observations de Zoologie et d'Anatomie compare'e, etc. Paris, 
 181 1, 4°. p. 84. ' 
 
 ^ Faraday, loc. cit. p. 5, § 1760. 
 
 * Loc. cit. pp. 217-222, 224, 225. Comp. Untersuchungen, etc., vol. i. p. 563. 
 
394 OBSEKVATIONS ANT) EXPERIMENTS 
 
 E nE.X, 
 
 If we add 6 to A on both sides, we g-et 
 
 E ^ nE. A, 
 
 w + K + b'^ {w + k^){\ + b) + w Aj 
 
 If we put for n its value from the above equation, the left side of 
 the inequality is found to be the g-veater. 
 
 The proof may also be got at by differentiating* with respect to 
 A the expressions 
 
 w + \ ?^A^ + wA + AAj ' 
 
 and thus obtaining- the quotients 
 
 f^4 ^. fll^ 
 -J— and. —J- • 
 
 Without regard to the sign, the latter is the greater ; and by both 
 methods one finds that the difference in favour of the conductor A 
 is so much the greater as the actual resistance w is greater. 
 
 This is an illustration of what occurs in comparing the sources of 
 electricity of our laboratories with the shocks of electrical fish. In 
 those we avoid short circuiting as carefully as possible ; in the 
 electrical fish, even when removed from the water, there is 
 always derivation through the body (see above, p. 380). As 
 I emphatically remarked in my first work, its shock is always 
 obtained by derivation^. The actual resistance of the organ is very 
 considerable. Consequently an induction shock with a com- 
 paratively inconsiderable diminution of strength, can bear much 
 greater resistance than that of an electrical fish of equal or even of 
 much greater strength. Thence may be derived the important rule 
 in experimenting with electrical fish, not to be deceived by the 
 great actual resistance, but always to diminish the resistance of the 
 experimental-circuit as far as possible. This is also the reason of 
 the usefulness of an apparatus, which, like the leading-ofF covers, 
 diminishes derivation in the experimental circuit as far as possible. 
 
 3. Passage of Spai'ks. 
 
 In order to observe sparks, the experimental circuit must be 
 opened suddenly at the instant of the shock and as nearly as possible 
 
 1 Vorliiufiger Abriss, etc. PoggendorfF's Annalen, etc., 1843, vol. Iviii. p. 30. § 76. 
 
ON LIVING MALAPTERUEUS. 395 
 
 when it is at its acme. Faraday attained this originally in the 
 simplest manner, by fastening- a file to each of the saddles on the 
 Gymnotus, and rubbing one file along the other. He afterwards 
 made use of a toothed- wheel, the circumference of which was rubbed 
 by a spring ^. I imitated this mode of experiment, at first by making 
 use of a Savart's wheel of 400 teeth, turned by hand 3-4 times in 
 a second. Then, in order to be relieved of the trouble of turning, I 
 took a small toothed-wheel of 50 teeth, which was turned 9-10 
 times in a second by clock-work, constructed by Herr Griiel, for 
 experiments with colour discs ^. The spark due to the shock of the 
 Malapterurus was perceived tolerably often in both ways, better 
 however, as far as could be judged, with the Savart's wheel than 
 with the smaller wheel. Perhaps this resulted from the 3-4 times 
 greater number of interruptions with the Savart's wheel; more 
 probably, according to the principle stated above, because the 
 slighter contact of the spring with the teeth and the inferior con- 
 duction in the axle-bearings in the case of the small toothed-wheel, 
 weakened the shock more than with the Savart's wheel. 
 
 As the frog-interrupter with a sufficient overweight breaks the 
 shock at about the middle of its course, it ought to furnish a new, 
 and, as it seems, the surest mode of observing separation sparks. I 
 have not yet pursued this thought sufficiently. In the few experi- 
 ments carried out with this object, the place between the point and 
 the plate on which it rests is, in the present form of the interrupter, 
 so unfavourable for observation that no conclusion can be drawn 
 from their failure. 
 
 4. Induciion. 
 
 If a coil of 720 turns of copper wire i-i mm. thick, the interior 
 of which was 25 mm. wide, and was filled with soft iron wires, was 
 put into the circuit, the extra current naturally made the sparks 
 much stronger. 
 
 Induction in a secondary coil by the shock of the fish had never, 
 so far as I know, been before noticed, and it gives the oppor- 
 tunity for an instructive observation. When the shock was sent 
 through the primary coil of a Ruhmkorff^s inductorium, the 
 divisions of which were connected so as to have half the length and 
 double the thickness, an observer, placed between copper handles in 
 the cii-cuit of the secondary coil, received a shock. If the platinum 
 
 ^ Loc. cit. p. 7. § 1767. 
 
 * Poggendorff's Annalen, etc., 1848, vol. Ixxv. p. 525. 
 
396 OBSERVATIONS AND EXPERIMENTS 
 
 points of the spark-micrometer were put in the latter circuit, two 
 sparks, one larger, immediately followed by a smaller, were seen to 
 pass regularly between the points. As these experiments were still 
 made without the frog-interrupter, it is possible that the two sparks 
 may have proceeded from two different shocks, but the regularity of 
 their appearance leads me to believe that they belonged to the begin- 
 ning and ending current, which corresponded with the rise and fall of 
 one and the same shock. But it is interesting and important, that 
 in accordance with the above principle, we can in this manner 
 obtain from the fish a current which passes over a gap between 
 fixed metals. That this principle actually prevails here, seems 
 to follow from the circumstance, that the sparks failed to pass when 
 the lengths of wire of the primary coil were connected, so as to have 
 double length and half thickness. 
 
 5. Magnetisation. 
 
 The effect of the introduction of the wires in the coil used 
 for obtaining the passage of the spark is to be attributed to the 
 electro-magnetisation of soft iron. 
 
 Six completely unmagnetised sewing needles 37 mm. long, 0-7 
 mm. thick, separated from each other by wax, were put in the 
 hollow axis, 8 mm. wide, of a coil which consisted of 735 turns of a 
 copper wire, 0-4 mm. thick. Three shocks of ISfalapterurus sufficed 
 to magnetise the needles decidedly strongly, though not to 
 saturation. Ten single opening shocks of the ordinary induction 
 apparatus (one Grove, with the coil quite pushed up) did not make 
 the needles perceptibly magnetic ; this was only accomplished when 
 the spark-current had passed for some time through the coil. This 
 current was, however, somewhat more effectual than the shock of 
 the fish. 
 
 6. Electrical Attraction. 
 
 The surest indication of electricity has always been its longest 
 known effect — electroscopic attraction and repulsion. I pass over my 
 unavailing experiments, which were however not long continued, to 
 charge a Leyden jar or a condenser by the fish. Since then, this 
 task has been very perfectly accomplished by Armand Moreau 
 with the Torpedo^. But I frequently observed the electrical 
 attraction of two gold leaves by the electricities of the shock 
 
 ^ Comptes rendus, etc., 1862, vol. liv. p. 963 : Annales des Sciences naturelles, 4^™® 
 Serie, Zoologie, 1862, vol. xviii. p. 11. 
 
ON LIVING MALAPTERURUS. 397 
 
 of Malapterurus in the beautiful manner indicated by Mr. Gassiot^. 
 The above-mentioned spark-micrometer was so arrang-ed that, in 
 place of the two platinum points which approached each other, two 
 copper wires bent like a swan's neck were used, from which gold 
 leaves hung down opposite each other. A glass case shielded it 
 from any draught. At a distance of 3 mm. from each other, the 
 movement of the leaves at the moment of the shock was doubtful. 
 At a distance of 3 mm. they obviously attracted each other. At a 
 less distance they flew together, a magnificent green flash appeared 
 and left the leaves fused together. It was then often possible to 
 tear them asunder uninjured by means of the micrometer screw, and 
 to repeat the experiment. 
 
 Unipolar action, to be discussed in Sect. 3 of the following paper, 
 is concerned here. 
 
 8. More detailed Investigation of the distribution of tensions 
 in the active organ of Malapterurus. The posterior half of 
 the organ acts more feebly than the anterior. The Equator 
 of the organ. 
 
 It is the custom to speak of the polar surfaces of the electrical 
 organs. This is however a mistake, unless the organ is cut out of 
 the animal, raised into the air, and regardedas endowed with equal 
 electromotive forces at all points. As long as the organs are 
 under water, or are adjacent laterally to animal parts which have 
 similar relations as conductors, there can be no question of polar 
 surfaces in the strict sense of the word. If we repj-esent such an 
 organ diagrammatically as a cylinder, the current curves now cut 
 the coat of the cylinder, instead of issuing only from its basal 
 surfaces. We may, nevertheless, speak of the polar surfaces of the 
 organ with the same accuracy as we do of those of a magnet, and with 
 this proviso we shall make use of this mode of expression. Minuter 
 details as to the special distribution of the discharge of the electric 
 fish will be found in my Collected Papers, No. XXX. 
 
 1 On the attraction force manifested by the electricity of the Gymnotus electricus. 
 In the Transactions and the Proceedings of the London Electrical Society, from 1837 
 to 1840. London, 1841, 4°. p. 197. Matteucci had described a similar experiment 
 on the Torpedo somewhat earlier. A wire supports the one gold leaf above the edge 
 of a metal disc, on which the fish lies. The other is suspended by a similar wire at 
 the edge of a metal cover put on the back of the fish. This cover is moved in order 
 to excite the fish, then the gold leaves may be seen to attract each other (Biblio- 
 thfeque Universelle, etc. Nouvelle Serie, 1837, t. xi. p. 392 ; Essai sur des Ph^no- 
 mfenes ^lectriques des Animaux, 1840, p. 51, Fig. V; Traits, etc., p. 154, pi. III. 
 Fig. 27). It is difficult to comprehend how one can judge of the attraction of the 
 gold leaves, when one moves them one's self and moreover in the open air. 
 
398 OBSERVATIONS AND EXPERIMENTS 
 
 With the help of a pair of platinum electrodes, to which a tele- 
 graph wire insulated with guttapercha served as a handle (see 
 above, pp. 380, 381), I readily observed that any point of the 
 organ nearer the tail behaves positively to any point nearer the head, 
 no matter where the point may lie on the circumference of a given 
 transverse section of the fish, whether on the back, flank, or belly. 
 The fish allowed a strip of platinum to be pushed quietly from the 
 flank under the belly without giving a shock, and only discharged 
 when the other electrode was put on it. 
 
 It follows from these experiments, that the polai" sm-faees of the 
 organ lie at the head and the tail, and on account of the tubular 
 shape of the organ, the poles themselves must be sought some- 
 where in the body of the flsb within the annular anterior or 
 posterior limit. It follows further, that the shocks must increase 
 with the distance of the contacts in the fish, or as we call it in 
 muscle or nerve, with the span of the leading-ofl" arch. The result 
 confirmed this surmise. On a large Malapterm'us I placed two zinc 
 saddles, first at a distance of i cm., then of 2 cm. from each other 
 at theii' inner edges, then so that the contacts were at the ends of 
 the organ, finally so that they touched the mouth and tail. I ob- 
 tained deflections of 20, 50, 116, 140 se. respectively. In the last 
 experiment the fish gave two shocks, and, as I still worked without 
 the frog-interrupter, it remains doubtful whether the greater strength 
 of the effect was due only to the double shock, or also to the greater 
 diSerence of tension between the points from which it was led ofll 
 
 These facts had been already observed by Ranzi in Malapterm-us 
 (see above, p. 388, note 2), and perfectly similar ones by Faraday 
 in the Gymnotus, but with reversed signs. But whereas De la 
 Rive remarks expressly that the anterior and the posterior halves 
 of the organ of Gymnotus discharge with precisely equal strength^, 
 it was found in Malapterurus that the anterior half of the organ 
 exceeds the posterior half in acti\nty. and indeed in the ratio of 
 II : 6. This ratio is very regular under ordinaiy circumstances, as 
 will be seen from the numbers to be given immediately. 
 
 When I first met with this difference, I imagined that it must 
 be due to a difierence of structure, perhaps to a difference in the 
 number of plates in a unit of length. But I convinced myself, in 
 common with Max Schultze, that nothing of the sort exists^. Then 
 
 ' Arcliives de I'Electricit^ 1845, vol. v. p. 505; Traits de I'^jfelectricit^ th^orique et 
 appliqu^e, etc., vol. iii. Paris, 1858, p. 76. 
 
 '^ Comp. M. Schultze in the Abhandlungen der naturforschenden Gesellschaft in 
 Halle, etc., loc. cit. pp. 16, 17. 
 
ON LIVING MALAPTEKURUS. 
 
 399 
 
 I considered whether the less strength of the posterior half might 
 not depend on the distribution of a single nerve-tube in the whole 
 organ, in consequence of which, the wave of innervation would be 
 weakened before reaching the posterior part of the organ. 
 
 However, before such a theory of the phenomenon can be pro- 
 pounded, another question must be solved. Does the posterior half 
 of the organ act with less electromotive force than the anterior, or 
 does the feebleness of its action result simply from its smaller 
 transverse section, and consequently greater resistance ? Not only is 
 the organ somewhat thinner at the tail, but the diameter of the tube 
 formed by it decreases. The organ indeed decreases somewhat in 
 transverse section towards the head also, but much more gradually. 
 
 It seemed possible to decide by experiment the question, whether 
 this is the reason of the behaviour observed. I prepared a leading- 
 off cover, w^hich had a third lining in the middle of its length, and 
 determined the relation of the shock proceeding from the two 
 halves, both with small and great resistance in the experimental 
 circuit. In the latter ease, the circuit contained a tube 142 mm. 
 long, i'6 mm. wide, filled with distilled water, to which some 
 spring water was added. The bobbins of the galvanometer (12,000 
 turns) were obliged to be close up in order to obtain perceptible 
 deflections. The frog-interrupter worked with an overweight of 
 100 gr. in this and in all the following experiments, where it is not 
 otherwise stated. With only an instantaneous excitation, this 
 would correspond to about 0-0168'''^, according to Helmholtz' 
 determinations ; but I have already before explained the reasons 
 why these determinations are not directly applicable ^. The num- 
 ber of shocks indicated by the frog-alarum every time the cover 
 was put on, although always recorded in my note-book, is omitted 
 in the following tables, because when the frog-interrupter is used, 
 it makes no difference. 
 
 Resist- 
 ance. 
 
 Half of 
 the Organ. 
 
 I 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 Mean. 
 
 A :P. 
 
 Small j 
 Great | 
 
 Anterior (A) 
 Posterior (P) 
 
 Anterior (A) 
 Posterior (P) 
 
 25 
 5 
 
 12 
 
 5 
 
 22 
 8 
 
 14 
 
 7 
 
 19 
 6 
 
 10 
 
 8 
 
 7 
 
 10 
 
 9 
 
 22 
 12 
 
 7-5 
 7.0 
 
 1 =100:54.55, 
 J or=ii : 6. 
 
 [ = 100:93.33. 
 
 ^ Archiv fiir Anatomie, etc., 1850, p. 325. 
 ^ Ges. Abhandl., vol. i. pp. 226, 227. 
 
400 OBSERVATIONS AND EXPERIMENTS 
 
 Thus, with increasing- resistance of the experimental cii'cuit, the 
 ratio between the effects of both halves actually approaches to 
 unity. There would be no reason for this if the different strengths 
 of the shocks of the two halves were occasioned by difference of 
 electro-motive force, and we may therefore conclude that the less 
 activity of the posterior half of the organ is due to greater 
 resistance. 
 
 I succeeded in demonstrating the superiority of the anterior over 
 the posterior half of the organ in another very instructive manner. 
 I fixed two zinc saddles on to one of the fish at such a distance 
 from each other, that they corresponded to the poles of the organ, 
 and I made a third saddle movable in such a way that it could 
 divide the distance between the two first (125 mm.) in any desired 
 proportion. The middle saddle was connected by the frog-inter- 
 rupter with the one terminal A, and the two other saddles were 
 connected with the other terminal B of the galvanometer, the 
 thermo-bobbins of which (213 turns) were close up. With this 
 arrangement, each of the two sections of the organ sent a current 
 in an opposite direction through the galvanometer, I now tried to 
 give such a position to the middle saddle that both currents should 
 be equal, or that the mirror should remain at rest. The results of 
 the experiments are shown in the following table, in which a de- 
 flection in the direction of the anterior half is denoted positive and 
 of the posterior negative : — 
 
 Eatio of length of the anterior ^ ~ . , aalvanometer 
 
 section to the posterior. •" ^ 
 
 62.5 
 
 50 
 45 
 40 
 
 62-5 Positive deflection, not measurable. 
 
 75 + 42 scale divisions. 
 
 80 +22;— 3; +23. 
 
 85 Strong negative deflection. 
 
 The ratio 45 : 80 seemed nearest to that sought. In order to 
 learn to what extent the equilibrium found was comp)lete, the 
 effect of the anterior section, 45 mm. long, and of the posterior, 
 80 mm., was measured separately, with only ^"^ turns. The former 
 was found = + 48, or multiplied into 4.= + 192 ; the latter = — 
 34, or multiplied into 4 + — 136 sc. Equilibrium was certainly at- 
 tained as nearly as possible, when the diflBculty of the experiment 
 is considered. 
 
 This experiment acquaints us with the equator of the organ, or 
 
ON LIVING MALAPTERURUS. 
 
 401 
 
 its neutral transverse section in presence of the leading-off cover, 
 but independently of the branch -current throug-h the galvano- 
 meter. The resistance of the conducting wires leading from the 
 poles to the terminal B of the galvanometer was the same, and 
 thus the mean tension between the tensions at the two poles, i.e. 
 tension zero, prevailed at the point B; consequently, when no 
 current was present in the galvanometer, the tension at the 
 terminal A was also nil, and, according to Bosscha's law, as there 
 was no current in the galvanometer-circuit, its removal would make 
 no difference (see Collected Papers, vol. ii. pp. 83, 84, note). 
 
 I went a step fm-ther. I put a pair of saddles, having a constant 
 space between them (3 cm. between the edges of the bands opposite 
 each other) on the fish, at different points of its length. I got, as 
 the means of several experiments, the following deflections :— 
 
 From the anterior third 26 scale divisions. 
 
 „ middle „ 29 „ „ 
 
 „ posterior „ 13 
 
 Thus the curve of the tensions, the axis of the fish being taken as 
 the axis of abscissae, rises almost in a straight line from the front 
 backwards as far as the end of the middle third, and then becomes 
 convex towards the axis. It follows also from this, that the equator 
 IS shifted towards the anterior end. Whether the somewhat smaller 
 amount of the deflection from the anterior third, in comparison with 
 the amount of the deflection from the middle third, was due to 
 the diminution of the transverse section of the organ towards 
 the anterior end (p. 399), or whether it was accidental, I cannot 
 say. 
 
 I prepared a cover like the former one, except that all the gutta- 
 percha between the three linings was removed except three narrow 
 strips, one on the back and one on each side. In this case, the 
 ratio A : P ought to be greater than with the complete cover. 
 I obtained the following result : — 
 
 Halfof 
 the Organ. 
 
 I 
 
 2 
 
 3 
 
 12 
 
 4 
 31 
 
 5 
 14 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 Mean. 
 
 A -.P. 
 
 Anterior 
 Posterior 
 
 36 
 
 14 
 
 35 
 
 21 
 
 33 
 
 12 
 
 19 
 
 30.8 
 16.8 
 
 = 100:54-55, 
 or 1 1 : 6 
 
 The experiment, as is seen, was a failure, inasmuch as the ratio 
 11:6 recurred with remarkable exactness. Evidently some chance 
 
 Dd 
 
402 OBSERVATIONS AND EXPERIMENTS 
 
 came into play, for such uniformity is not to be thought of here. 
 Unfortunately I never repeated the experiment, and therefore a 
 certain obscurity is left in this respect also. 
 
 9. On the influence of the length of the linings of the leading- 
 oflf covers upon the strength of the branch current in the 
 experimental circuit. 
 
 A series of experiments have their place here, the object of 
 which was to determine the most favourable conditions for the 
 construction of the leading-off cover. The question is the leng-th ^ 
 to be given to the linings. In Fig. lo, these have a medium 
 length which was given to them by guess. But the length of the 
 linings affects the strength of the branch current in the experi- 
 mental circuit in a two-fold way. First, its strength increases 
 with their surface, inasmuch as this increase of surface diminishes 
 the resistance of the circuit. Secondly, if we suppose the linings 
 not in electrical connexion to be applied to the electromotive 
 surfaces of the organ, it is clear that the tensions of the latter will 
 in part equalise themselves through the linings. Thus the strength 
 of the cm-rent in the experimental circuit will be directly pro- 
 portional to the mean tension of the surface touched by them, and 
 inversely projDortional to the resistance of the experimental circuit. 
 Accordingly, it seems as though, when the resistancie of the 
 experimental circuit is greater, shorter linings would be more 
 advantageous, and longer when the resistance is less : and in 
 order to obtain the strongest possible effect from the fish under 
 different circumstances, I imdertook to decide this question by 
 experiment. 
 
 Of two covers fitting the same fish, I provided the one with 
 short, the other with long linings. In the latter case, a gap of 
 only 8 mm. was left between them ; the short linings were about 
 1 cm. wide. With these linings it was easy to prove ^ the correct- 
 ness of the above conclusion. The great resistance here was a 
 tube 280 mm. long, 1*4 mm. wide, filled with solution of sulphate 
 of zinc, and it was again necessary to push the bobbins of the 
 galvanometer close up (see above, p. 400). 
 
 ' Comp. Ge8. Abhandl., vol. i. p. 215. 
 
 ^ These experiments also were not performed until after the lecture in the 
 Academy, where I could only express the surmise that tbis would prove to be the 
 case. (ioc. cit. p. 104.). 
 
ON LIVING MALAPTERURUS. 
 Small resistance. 
 
 403 
 
 Lining. 
 
 I 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 II 
 
 Mean. 
 
 Long ... 
 Short... 
 
 20 
 
 74 
 
 57 
 
 50 
 
 64 
 
 28 
 
 41 
 
 37 
 
 40 
 
 30 
 
 40 
 
 51-5 
 34-4 
 
 
 
 
 
 Great resistance. 
 
 
 
 
 
 
 Long ... 
 Short... 
 
 73 
 
 54 
 
 74 
 
 75 
 
 100 
 
 
 
 
 
 
 
 64-5 
 82.3 
 
 If shocks were taken from the fish alternately with covers with 
 long and short Kning-s, the greater strength of the latter was very 
 clear. 
 
 10. More precise examination of the value of the Frog- 
 interrupter in experiments on the Malapterurus. 
 
 I do not return here to a detailed consideration of the questions 
 as to the time relations of the shock, hut refer to what has already 
 been said on this matter in discussing the frog-interrupter, p. 385. 
 
 In the ' Description of certain Apparatus and Methods of Experi- 
 ment,' it was shown that the frog-interrupter serves the purpose of 
 opening the circuit at any required short period after the excitation 
 with sufficient accuracy. Consequently the great irregularities 
 which were always observable in the effects of the shock of the fish 
 were the more obvious. They finally led me to decide on testing 
 the action of the inteiTupter in experiments on the fishes them- 
 selves also, although it seemed inconceivable that it should not do 
 here what was expected from it. 
 
 At first, I simply made series of experiments with increasing 
 overweight, in order to see how the deflections caused by the fish 
 shock would increase correspondingly to the overweight. Thus I 
 obtained (with ^'^ turns at 75 mm. distance) the series immediately 
 following, in which W.I. signifies ' without interrupter.' 
 
 Overweight. 
 
 2gr. 
 
 W.I. 
 
 2 
 
 50 
 
 100 
 
 150 
 
 200 W.I. 
 
 250 
 
 300 
 
 350 
 
 450 
 
 500 
 
 I 
 
 8-2 
 
 125 
 
 10.7 
 
 8.4 
 8 
 
 16.3 
 
 8-8 
 
 17.8 
 
 lO-O 
 
 25.0 
 9.0 
 
 80 
 
 14 
 
 14-6 
 
 24.8 
 
 22-6 
 
 22 
 
 II 
 
 The fall of the deflections at the close of the second series may 
 be explained thus : at the time of the interruption, a great part of 
 
 D d 2 
 
404 OBSERVATIONS AND EXPERIMENTS 
 
 the shock was over, consequently the fatigue of the fish was more 
 influential than the prolongation of the time of closure by in- 
 creasing" the overweight. 
 
 But I now began other series of experiments, in which I pro- 
 posed to prove directly, the regularity introduced by the frog- 
 interrupter into the results of the fish experiments. I wished to 
 show that the mean error of a series of deflections caused by the 
 fish was diminished by the frog-interrupter. With this object, 
 I made the wires proceeding from the cover, divide into two con- 
 ductors, the one of which led to the galvanometer, guarded as usual 
 by the frog-interrupter, and the other led to an unguarded galvano- 
 meter. The guarded galvanometer, until its circuit was opened, 
 weakened the current in the unguarded one. However this could 
 have no other eflfect on the result, than to make the deflections in 
 the unguarded galvanometer still more irregular. I made only 
 one such experiment, but it was unsuccessful, inasmuch ds the 
 guarded galvanometer, for some incomprehensible reason, gave less 
 regular deflections than the unguarded one. Soon afterwards the 
 last of my fishes died, and I remained in the dark as to the reason 
 of failure. 
 
 11. Of the relative immunity of Malapterurus from electric 
 
 shocks. 
 
 As long ago as 1842, in my ' Vorliiufiger Abriss,' I discussed the 
 question ^, how it happens that an electrical fish strikes other fish 
 to death, but neither itself nor, according to v. Humboldt's ^ and 
 Colladon's^ experience its fellows ; why the Torpedo, which is vivi- 
 parous, does not kill the young in its uterus ? I was led to this 
 question by the consideration, that according to indubitable 
 physical laws, the shock of necessity passes through the body of 
 the fish itself, in fact that this body is more favourably situated in 
 regard to the organ for the reception of the shock than the body 
 of another animal approached by it. 
 
 This statement might perhaps be doubted in regard to the 
 Gymnotus, on account of its head being so far in front of the 
 anterior polar surface of the organ. It is certainly correct in the 
 case of the Torpedo, and it is hardly invalidated by the remark, 
 
 * Poggendorff's Annalen, etc., 1843, vol. Iviii. pp. 29, 30, § 75. 
 " Recueil d'Observations de Zoologie, etc., p. 80. 
 
 ^ Comptes rendus, etc., t. iii. p. 490; L'Institut, etc., t. iv. No. i8x, p. 350; 
 PoggendorfPs Annalen, etc., 1836, vol. xxxix. p. 413. 
 
ON LIVING MALAPTERUKUS. 405 
 
 that the accumulated discharges of both organs strike the central 
 nervous system vertically to its axis. But the truth of the state- 
 ment is not so obvious in any of the three electrical fishes as in 
 the Malapterurus. 
 
 Nevertheless, I did not omit to put it to the test of experiment, 
 and to prove that the discharg-e actually goes through the body of 
 the fish. For this object, I put insulated copper wires into the 
 body through the mouth and vent of a Malapterurus doomed to 
 death, while an assistant held it with an indiarubber glove. When 
 it was placed in a small parallelopiped glass trough full of water, 
 with the wires in its body, and its skin was excited with a glass 
 rod, it discharged, and the current always showed itself in the 
 circuit of the wires in the right direction and strength. 
 
 The only conceivable reply to the question, why the electrical fish 
 does not kill itself, was evidently that it is very little or not at all 
 sensitive to electric shocks. I had made in the year 1852 experi- 
 ments with Faraday on the Gymnotus of the Polytechnic Institution, 
 in order to discover if the fish is as insensitive to currents of 
 external origin as to its own. We did not succeed in exciting the 
 fish perceptibly ; however it was found that generally even in a 
 man, om* apparatus was too weak to produce any appreciable 
 effects ^ in the large mass of water which contained the animal. 
 
 I was now naturally very eager to obtain from the Malapte- 
 rurus the reply to this question, which had been in my mind for 
 fifteen years. With this object, I let down into the water of the 
 experimental tub, a pair of zinc electrodes as the terminals of the 
 secondary coil of the induction apparatus, and in addition, for a 
 reason to be explained immediately, a pair of platinum electrodes 
 in connexion with half of the length of the nerve multiplier. 
 Two Groves were placed in the primary circuit of the induction 
 apparatus. As I had expected, the multiplier was not affected by 
 the alternating currents of the inductorium, as the current from it 
 was too weak to produce a deflection in successively opposite 
 directions. 
 
 When I put into the tub, river-fish of this country, tench, perch 
 {Perca Jluviatilis), chub (Carassins Gihelio), burbot {^Lota Jtmnafilis), 
 pike [Esox Lucius), Silurus [Silurus Glmiis), and exposed them to 
 the current, they tilted over on their side when the coil was 
 pushed up half-way, did not move their gills any more, and drifted 
 about unconsciously. If this condition was allowed to continue, 
 
 * Comp. above, p. 380, and Ges. Abh. vol. ii. xxx. § 4. 
 
406 OBSEBVATIONS AND EXPERIMENTS 
 
 the fish died ; some sorts of fish were more easily killed than 
 others. Frog-s were strongly tetanised, but recovered easily^. 
 
 Now, if I put a Malapterui'us in the tub instead of or along- with 
 the fish, it behaved in a very different manner. By the shocks 
 which sufficed for the fatal stupefaction of the other fishes, it 
 appeared to be unaffected, just like a frog in a Guericke's vacuum 
 as compared with a g-uinea-pig or a linnet. If a fish or a frog 
 came too near it, it also discharged its battery, as was indicated by 
 the multiplier needle which was thrown against the stop and the 
 astaticism of which was destroyed. The only perceptible effect of 
 the shocks upon the Malapterurus was, that on opening the key, its 
 barbels were laid back, and it is worthy of remark, that the barbels 
 of the Silurus also move in a similar way when it is tetanised. 
 However, when the utmost force was imparted to the shocks, by 
 pushing the coil fully up, it was easy to see that the fish noticed 
 and avoided them. If it came in the neighbourhood of the elec- 
 trodes, where the current density was greatest, it withdrew hastily, 
 gave a shock or two, and sought with correct instinct, that position 
 in which its axis of length cut perpendicularly the least dense 
 current curves, as if it knew the laws of the distribution of current 
 in non-prismatic conductors. But it moved gills and fins with 
 usual rhythm in the midst of the strongest currents, and swam out 
 of their range without a single convulsive or involuntary move- 
 ment. 
 
 Tin electrodes in connexion with the induction apparatus, and 
 platinum electrodes in connexion with the half length of the nerve 
 multiplier, were put on a dying Malapterurus, which lay in a small 
 parallelepiped glass trough, which it almost filled (see p. 405). 
 Even with two Groves in the primary circuit of the induction 
 apparatus and with the coil quite pushed in, the fish seemed not to 
 perceive the shocks at all, breathed on quietly and did not dis- 
 charge. A deflection in successively opposite directions now 
 presented itself. 
 
 The result was substantially the same with the current of a 
 Grove's battery of thirty cells, in place of the alternating current 
 of the inductorium. The common river fish, whilst the key was 
 shut, lay on their side and their fins twitched. When it was 
 opened, a general convulsive movement ensued, but shortly after- 
 wards, if the current had not lasted too long, they swam about 
 
 ' Comp. similar experiment by R. Bottger in PoggendorflTs Annalen, etc., 1840, 
 vol. 1. p. 39. 
 
ON LIVING MALAPTERURUS. 407 
 
 ag-ain. Tetanising- evidently affected them more than the current 
 of comparatively constant density. The Malapterurus withstood 
 the effect of the constant current also, and this appears to be of 
 special interest, as it did not need to be accommodated to it. 
 However, here also, it sought that position which is theoretically 
 the most protected. 
 
 It is now clear why the Malapterurus does not suffer from its 
 own current, and the same reason will naturally hold good for other 
 electrical fish. They all possess, at all events relative immunity^ as 
 regards electric currents, for the exposed muscles and muscle nerves 
 of electrical fishes, and even the electrical nerves themselves are 
 excitable by the current, and consequently their immunity must 
 have a limit of curr.ent density inferior to that attained in experi- 
 ments. It would certainly be possible to kill a Malapterurus with a 
 Leyden battery of a certain size, if it were charged to sufficiently 
 high tension. Wherever this limit may lie, the difference between 
 electrical fish and other animals is great enough to be regarded as 
 one of the most enigmatical facts of physiology. According to 
 ancient accounts, crabs are said to possess immunity as regards the 
 shock of the Gymnotus^. I therefore subjected the cray-fish of 
 this country to the same experiments as the fish and the frog. 
 It is true that they bear a great deal, perhaps because their chitine 
 tunics protect them. "When they are tetanised, they twitch 
 violently, their body is contorted, their antennae are laid back, but 
 there is no such immunity as that of Malapterurus. I need not 
 refer to the well-known, and in part very naive tales concerning 
 the immunity of certain men from electric shocks^. 
 
 ' In M. Colladon's paper on the Torpedo, we read (p. 491) : 'M.J. Davy a 
 constats . . . que le courant d'une pile ne parait pas faire souffrir ceux de ces 
 poissons qui sont interpost^s dans le circuit.' This is a misunderstanding of Col- 
 ladon's. John Davy's statement about the absence of influence of the current on 
 the Torpedo refers solely to the electrical organs, and only assists him to conclude that 
 the organs are not muscles. (Philosophical Transactions, etc., 1832, p. 269; Ee- 
 Bearches, physiological and anatomical, London, 1839, vol. i. pp. 32-34.) 
 
 * Priestley, The History and present state of Electricity, London, 1769, p. 403. 
 G. H. Loskiel, Geschichte der Mission der evangelischen Bruder unter den Indianern 
 in Nordamerika, Barby, 1789, p. 124 ; also in Voigt's Magazin fiir das Neuste aus 
 der Physik, u. s. w., Gotha 1789. Bd, VI. St. ii. S. 171, Loskiel's statement 
 appears the more questionable, as there are no Gymnotus in North America. Alex. 
 V. Humboldt produced convulsive movements in crabs on the Lido by the simple 
 battery (Versuche viber die gereizte Muskel- und Nerven-faser, Berlin und Posen, 
 1797, vol. i. p. 285.) 
 
 ^ Comp. Humboldt, Versuche iiber die gereizte Muskel- und Nerven-faser, u. 8. w., 
 vol. i. p. 160 ; Recueil d'Observations de Zoologie, etc., p. 69. 
 
408 OBSERVATIONS AND EXPERIMENTS 
 
 There are many instances on record of the immunity of certain 
 animals from certain injurious agencies which are destnictive of 
 most others. I have collected a large number of such accounts, but it 
 would lead us too far from our object to mention them in detail, the 
 more so, that they are often very indefinite and not well authenti- 
 cated. Teleologically, the comparison with the immunity of vipers 
 from their own poison, proved by Fontana, is the nearest. Although 
 Montana had shown that non-poisonous snakes, lizards, toads, eels, 
 snails and leeches are also proof against the viper's poison like the 
 viper itself, yet the immunity of the viper suggested the notion 
 and afforded support to it, that animals are fundamentally proof 
 against a poison produced by themselves. But this is not always 
 the case. Fontana had already remarked, that animals with corro- 
 sive and acid secretions, like the hymenoptera are not proof 
 against their own poison^. 
 
 And this brings us to the point with which we are here con- 
 cerned. All those immunities relate to substances acting in a 
 specific manner, and which can therefore be resisted by a specific 
 organisation. That strychnine in such small amounts kills most 
 animals, is at bottom more remarkable than that the fowl^, the 
 rhinoceros bird and the sloth ^ can resist it more or less. But 
 immunity from a natural power like electricity, gives at first sight 
 almost the impression as if an animal were proof against a corrosive 
 or a metallic poison, against mechanical force or destructive heat. 
 
 This is however a deceptive appearance. The electric cuiTent 
 stands to nerve and muscle in as peculiar a relation, as according to 
 modern ideas, certain organic molecules do to certain limited 
 provinces of the nervous system, and it is ultimately not more 
 wonderful, that in this special relation variations occur in the case 
 of certain animals, than that other animals are relatively proof 
 against the strychnine molecule which is so formidable to all 
 others. Along with the eflTects which the current produces in 
 animals, in virtue of its special relation to nerve and muscle, it is 
 capable of other, more general effects, which, if increased beyond a 
 certain degree, become likewise destructive to the animal in its 
 thermic, electrolytic, cataphoric and anaphoric actions. That 
 electrical fishes are not exempt from these actions is clear. This 
 consideration does not in the least detract from the importance of 
 
 ^ Abhandlung liber das Viperngift, u. s. w., Berlin, 1787, 4°. p. 15 fF. p. 155. 
 ' Lehube in the Archiv fiir Anatomic, u. s. w., 1867, P- ^^9- 
 ' Ibid. 1868, p. 756. 
 
ON LIVING MALAPTEKUEUS. 409 
 
 the problem, of the immunity of electrical fish from the electric 
 current. On the contrary, it affords us an additional motive in our 
 efforts to solve it, in as far as it opens the prospect, that disclosures 
 of importance in relation to the general physics of muscle and 
 nerve will result from it. 
 
 Pfliig-er, who was a witness of my experiments (see above, p. 387), 
 and who at that time was working* at electrotonus, conjectured 
 that the Malapterurus may have power to put its nerves into a 
 state of anelectrotonus. But except under the influence of ex- 
 traneous currents, no trace of electrotonic conditions has ever been 
 observed in the nervous system. Pfliig-er's theory seems to be con- 
 tradicted also by the following" experiment. The electrical nerve of 
 the dying- Malapterurus in the g-lass trough (see above, p. 406) had 
 been ligatured, and therefore could no longer be anelectrotonised by 
 means of its giant ganglion cells. Notwithstanding, it was not 
 excited by very strong alternating currents. 
 
 There might be supposed to be some connexion between the 
 immunity of electrical fishes from the current, and the much dis- 
 cussed inexcitability of the spinal cord. This would be quite 
 arbitrary, for the same electrical nerve, which in the previous 
 experiment showed itself electrically unexcitable, gave a violent 
 shock at the moment of ligaturing the nerve, and in another 
 experiment, section acted in the same way. The spinal cord, on the 
 contrary, responds as little to mechanical as to electrical irritations. 
 
 Since Briieke showed that muscles with the nerve removed, are 
 little sensitive to currents of short duration^, the fact that the 
 Malapterurus is also proof against the constant current of a battery 
 assumed much greater importance (see p. 407). Its immunity is 
 therefore not, like that of the muscles, connected with the du;ration 
 of the current. 
 
 No other opinion seems to me to be left open, than that the 
 threshold of excitation lies higher for the nerves of the electrical fish 
 than for the nerves of other animals. I should have continued the 
 investigation further in this direction myself, if opportunity had 
 been afforded to me. I rejoice all the more that Boll has recently 
 followed this track in the case of the Torpedo, apparently with good 
 success^. 
 
 ^ Sitznngsberichte der Wiener Akademie, u. s. w., 1867, vol. Ivi. Part ii. p. 594 ; 
 1868, vol. Ivii. Part ii. p. 125; Vorlesungen iiber Physiologie, 2 Ed. 1875, vol. i. 
 
 p. 485- 
 
 ^ Archiv ffir Anatomie, u. s. w., 1873, p. 76. According to Steiner, the threshold 
 of electrical excitation in the Torpedo would lie below the thickness which a shock of 
 
410 OBSERVATIONS AND EXPERIMENTS 
 
 12. Experiments on living electrical nerve and organ after 
 
 separation. 
 
 The following experiments were made on a dying Malapterurus 
 which had been used for several previous investigations. 
 
 When the electrical nerve was laid on the platinum plate of the 
 leading-in apparatus^, and was tetanised, tetanus of a nerve-muscle 
 preparation followed, wherever on the organ of the same side, the 
 nerve was placed. If the stimulation of the electrical nerve is con- 
 tinued, the organ does not produce a continuous current, as one 
 might suppose, but a series of rapidly succeeding shocks, just as the 
 contraction and variation of current in a muscle, which are produced 
 by continuous excitation of its nerve are only apparently constant. 
 It is a proof of the small amount of discernment with which 
 Matteucci conducted his numberless experiments, that he left such 
 fundamental questions for me to answer^. 
 
 The needle of the multiplier, connected with the fish by 
 platinum electrodes, was deflected at the commencement of the 
 tetanus, but took up no fixed position. However, it might be 
 possible that with unpolarisable electrodes and with an aperiodic 
 light magnetj the result would be different. The positive polari- 
 sation of the organ by its own current^ might be matter for 
 consideration. After a few minutes rest, the tetanus was again 
 observed. I find no mention in my diary as to the distance of the 
 secondary coil, at which it occurred. 
 
 When a cun-ent of two Groves was sent through the nerve 
 nothing resulted. With five Groves, a shock at closing the cun-ent 
 took place, which showed itself by twitching of the nerve-muscle 
 preparation, and in the galvanometer, by a deflection _ in the right 
 direction. The great strength of current required for the stimu- 
 lation of the nerves was very surprising ; however, the nerve was 
 no longer fresh, and it was not possible to make comparative 
 observations. It therefore remained uncertain, whether it might 
 
 the Torpedo B, or even of the Torpedo A itself can reach in the Torpedo A, ibid. 
 1873, p. 684. 
 
 ^ Untersuchungen, etc., vol. i. p. 450, Tab. ii. Fig. 20, 21. 
 
 ^ Four years later than I, Moreau also stated that tetanising the electrical nerve 
 (in the Torpedo) produces an uninterrupted series of shocks, without saying however, 
 how this was observed (Annales des Sciences naturelles, etc., 4™®. S(3rie, Zoologie, 
 1862, t. xviii. p. 10). This statement was incorrectly regarded as new by the 
 reporters of his work, MM. Claude Bernard and Becquerel senior (Comptes 
 rendus, etc., 1862, vol. liv. p. 966). Comp. also Marey in the Comptes rendus, etc., 
 1871, t. Ixiii. p. 918. See moreover Boll, Archiv flir Anatomie, etc., 1873, p. 79- 
 
 ^ See Gesammelte Abhandlungen, vol. ii. xxxi. § 6. 
 
ON LIVING MALAPTEKURUS. 411 
 
 be concluded from this experiment that the threshold of excitation 
 for electrical nerves is abnormally hig-h. There was no twitching of 
 the muscles of the trunk, even when the coil was pushed quite up 
 and with two Groves in the primary circuit of the induction 
 apparatus ; on the other hand, the muscles of the head and intestine 
 contracted. But, as regards the former, it must be added that 
 they did not respond to the irritation of a satvirated solution of 
 common salt either. 
 
 After circulation had ceased for 2| hours, the organ, when 
 irritated through its nerve, acted upon the multiplier ; but when 
 irritated directly, either mechanically or with caustic potash, it 
 acted neither upon the multiplier nor upon the nerve-muscle pre- 
 paration. 
 
 The electrical nerve, with its longitudinal and cross section 
 properly laid upon the pads of the nerve-multiplier, failed to 
 exhibit any nerve-current, or on being tetanised, any negative 
 variation. This appears natural when we consider that the nerve 
 consists of a single nerve-tube, which is no doubt unusually thick, 
 but is also surrounded by a disproportionately strong ^ sheath, the 
 effect of which is to efface the electromotive action of the nerve- 
 tube by derivation. It is however conceivable, that a more 
 efficient electrical nerve of the Malapterurus might show traces on 
 the galvanometer of the nerve-current and of its negative 
 variation. 
 
 Electrotonic currents were clearly perceived with two Groves ; 
 after cutting through the nerve between the stimulated and led- 
 off tract and joining the cut ends, feeble actions in the wrong 
 direction, depending upon escape of current, afforded evidence of 
 the genuineness of the previously observed deflections. 
 
 Here, the experiments on the electromotive activity of the 
 organ at rest, and on its secondary electromotive actions come to a 
 close. A more detailed report of both is given in § 6 of the 
 ' Experimental criticism of the discharge hypothesis.' 
 
 * Bilharz, loc. cit. p. 21, states the thickness of the sheath containing the vessel as 
 nine times that of the empty sheath. The diameter of the fibres themselves is to 
 that of the empty sheath in the proportion of 2-6 : 24-6 according to his drawings, 
 loc. cit. Tab. iii. Fig. 9. Hence the ratio of the cross section of the fibres to that of 
 the whole nerve is shown to be i : 8950. Max Schiiltze gives the ratio of the fibres 
 to the empty sheath still more unfavourably in his illustration in Fig. 7, Tab. i. of 
 his paper (3-25 : 33-10.) 
 
412 OBSERVATIONS AND EXPERIMENTS 
 
 13. Of the chemical reaction of the electrical organ of the 
 Malapterurus\ 
 
 Max Scliultze had stated, when making- observations at Trieste, 
 that the electrical organ of living Torpedos has a distinctly acid 
 reaction -. I examined the electrical organ of the Malapterurus 
 repeatedly as to its reaction. In the fresh state, I found it neutral 
 according to the manner of the reaction of muscle, i.e. amphoteric ^ 
 and only later, from about the third day forwards, distinctly acid. 
 The red spots did not disappear on drying. On the fom-th day 
 (in one case) the organ was decidedly in a state of decomposition, 
 and its reaction was alkaline. 
 
 With a fresh organ, on the first day, I examined the effect of 
 higher temperature on the reaction. After keeping it for five 
 minutes in water at 40-5o°C., which by the way leaves the con- 
 sistency of the organ unaltered, the neutral reaction changed to an 
 acid one. In order, however, that this should appear clearly, the 
 organ must be in contact with the litmus paper longer than frog 
 muscles treated in the same way (see Collected Papers, p. 17). 
 A more important difference between organ and muscle is, that 
 pieces of the organ thrown into boiling water become acid, in- 
 stead of becoming more alkaline like muscles (p. 18). Thus, in 
 this respect, the electrical organ agrees with the nervous central 
 organs, which, according to Funke"*, also become acid at boiling 
 temperature. But whereas Funke considers that brain and spinal 
 cord become more acid at boiling temperature than at 40-50°, the 
 acid reaction of the boiled organ seemed to me less marked than 
 that of the organ heated only 40 or 50° ; this might be considered 
 to indicate a transition to the behaviour of muscles. 
 
 ^ With use of my ' Bemerkungen iiber die Reaction der elektrischen Organe und 
 der Muskeln,' in the Archiv fiir Anatomic, etc., 1S59, P- ^4^- (Comp. above, 
 p. 369.) 
 
 ^ Zur Kenntniss der elektrischen Organe der Fische. Zweite Abtheilung, Torpedo, 
 Halle, 1859, 4°- P- 27. 
 
 ' GesammelteAbhandlungen, note i on p. 10. Comp.Kiihne in Virchow's Archiv fur 
 pathologische Anatomie, etc., vol. xiv. 185S, p. 328, 334, 345, 347. [Heidenhain has 
 since distinguished this kind of reaction from neutral reaction, as amphichromatic 
 (Mechanische Leistung, etc., p. 153.) Recently, it has been called the amphoteric 
 reaction. Comp. Heintz, Ueber die Ursache der Coagulation des Milchcasein durch 
 Lab und iiber die sogenannten amphotere Reaction. Journal fiir praktische Chemie, 
 1872, vol. vii. p. 374. It was my friend Heintz, who, in the beginning of my ex- 
 periments on the reaction of muscles, called my attention to the surmise, that in 
 amphoteric reaction, the blue colour would not become red and the red blue, but that 
 both colours would alike become violet ; experiment confirmed this conjecture.] 
 
 * Archiv fiir Anatomie, &c., 1849, P- ^41- Comp. Gscheidlen, Ueber die chemische 
 Reaction der nervbsen Centralorgane. Pfliiger's Archiv, &c., 1874, vol. viii. p. 177. 
 
ON LIVING MALAPTERURUS. 413 
 
 It was certainly in the hig-hest degree improbable, that the fresh 
 organ of the Malapterurus should have an amphichromatic reaction, 
 and that of the Torpedo, an acid one. In order to explain Max 
 Schultze's statement, I therefore ventured the suggestion, that the 
 electrical organ might, like muscle, become more acid by con- 
 tinuous exertion. It is easy to conceive that Torpedos, from the 
 moment of their capture until they reach the observer's hands, are 
 exposed to a multitude of annoyances, to which they respond by 
 shocks, so that they themselves bring their organ into the con- 
 dition of the muscles of guinea-pigs and frogs which I had 
 tetanised to exhaustion (see Collected Papers, vol. ii. pp. 26 if, and 
 
 This suggestion has only been verified in part. Moreau found 
 that the organ of the Torpedo had a neutral reaction ^ ; Boll an 
 alkaline one ; thus in any case. Max Schultze is mistaken in regard 
 to the reaction of the fresh organ in a state of rest. After the 
 lapse of 6-10 hours. Boll saw that the organ of the Torpedo had 
 become acid. According to him, however, it does not become acid 
 like muscles, through exertion 2, and thus it remains uncertain how 
 Max Schultze was misled. 
 
 * Annales des Sciences naturelles, etc., loc. cit. p. 6. 
 
 * Archiv fiir Anatomie, etc., 1273, pp. 99, 100. 
 
XII. 
 
 LIVING TOEPEDOS IN BEELIN. 
 
 By E. Du BOIS-REYMOND. 
 
 Sitzungsberichte der Koniglich Preussischen Akademie der Wissenschaften. 
 Berlin. March 13, 1S83. 
 
CONTENTS. 
 
 1. Introductory 
 
 2. General Observations on the experiments on Torpedos 
 
 3. Method of leading-ofF Malapterurus-shocks into an experimental Circuit 
 
 4. Method of leading-off the Torpedo-shock into an experimental Circuit 
 
 5. Sluggishness of the Torpedo in Confinement 
 
 6. Efiect of Torpedo Shock on Man 
 
 7. On the distribution of the Current in the Torpedo 
 
 8. On Iodide of Potassium Electrolysis by the Torpedo 
 
 9. On the Organ-current in the Torpedo 
 
 10. On the secondary electromotive Actions of the Organ of the Torpedo 
 
 11. On the relative Strengths of the Homodromous and of the Heterodro 
 
 mous Current in the electrical Organ ..... 
 
 12. Dependence of the secondary electromotive Effects and of the Con 
 
 ductivity of the Organ on its Vitality ..... 
 
 13. Discussion of the foregoing Results ...... 
 
 14. On the electromotive Actions of the electrical Nerves of the Torpedo 
 
 Appendix ........... 
 
 PAGE 
 
 417 
 41S 
 420' 
 423 
 425 
 426 
 427 
 435 
 438 
 445 
 
 451 
 
 457 
 458 
 462 
 469 
 
XII. 
 
 1. Introductory. 
 
 Until recently it had not been possible, even under favourable 
 conditions, to keep Torpedos alive in captivity more than a few days. 
 On this circumstance was founded the opinion expressed by Faraday 
 five-and-forty years ago, that in certain respects the Gymnotus was 
 better adapted than the Torpedo for researches on electrical fishes^ ; 
 in writing on this subject later I placed the Malapterurus above 
 the Gymnotus as an object of research, observing that ' the Torpedo 
 being a sea fish could not be taken into account ^.' The state of 
 matters has since then entirely altered. 
 
 Not only is the art of keeping marine animals in the Aquaria which 
 have come into existence during the last twenty years so much ad- 
 vanced, that there is now no difficulty in preserving Torpedos in a 
 living state as long as is necessary ; but also the methods of railway 
 transport are now so perfect that it is possible to convey the in- 
 habitants of the ocean to great distances from the coast without risk. 
 At the very time I wrote the words above quoted, Ranvier had at his 
 disposal living Torpedos from Concarneau on the Atlantic Ocean ^. 
 It is, however, remarkable that in the exhibition in the Champs 
 Elysees, as well as those in Munich and Vienna, these ' oldest of all 
 electrical machines,' as they were called by George Wilson*, were 
 wanting ; their only representatives being the submarine instru- 
 ments of warfare which have been called after them. 
 
 Torpedos were first received at the Berlin Aquarium in the 
 summer of 1881 from Trieste. The director, Dr. Otto Hermes, 
 who makes the resources of the institution available for scientific 
 purposes with the utmost liberality and courtesy, at that time per- 
 mitted me to make some experiments. These were, however, 
 limited to such general observations as could be made without inter- 
 fering with the purpose for which the animals were kept. I have 
 repeatedly on former occasions explained why the possession of 
 
 ^ Notice of the character and direction of the electric force of the Gymnotus (1838). 
 
 * E. du Bois-Reymond, Gesammelte Abhandlungen, vol. ii. p. 6ii. 
 
 * Legons sur I'histologie du Systfeme Nerveux, Paris, 1878, vol. ii. p. 88. 
 
 * The Edinburgh New Philosophical Journal, Oct. 1857, p. 267. 
 
 E 6 
 
418 LITIXG TORPEDOS IN' BERLIN. 
 
 living- electrical fish is under these circumstances not so useful as 
 mig-ht be expected^. Dr. Hermes, however, was so good as to arrange 
 recently for supplying the Physiological Institute with Torpedos 
 through the Aquarium at Trieste, so that I might be able to make 
 such experiments on them as seemed desirable. It is scarcely neces- 
 sary to observe what an admirable prospect is thereby opened for the 
 advancement of the knowledge of electrical fishes, and consequently 
 of the physics of nerA'e and muscle. A single Torpedo at one's 
 disposal in a German physiological laboratory is possibly capable of 
 yielding more for the progress of om* science than, to adopt Shylock's 
 mode of expression, the Adriatic full of Torpedos could do at a place 
 where you must first set up a galvanometer, and where, on account 
 of some piece of apparatus left at home or broken in the journey, 
 the most admirable plan of experiment becomes futile. 
 
 In this way I have already worked out two Torpedos during last 
 summer, and three during the winter, and by means of them have 
 answered several questions which had been long present to my mind, 
 and I have also begun the observation of the secondary electromotive 
 actions of the electrical organ. Although these last researches are 
 not completed, I publish the results already obtained as, even in 
 their present condition they constitute an important supplement to 
 my pre\'ious communication on secondary electromotive phenomena-. 
 Moreover, the ver}' small measure of completeness which I was 
 able to give to the researches on the same subject in muscles and 
 nerves did not appear to be attainable as regards the Torpedo for 
 reasons which I shall give later (p. 446). 
 
 The animals employed belonged to the species Torpedo marmorata : 
 they varied in length from 25-36 cm. The latter is the measure- 
 ment of a medium sized European Torpedo. 
 
 2. General observations on experiments on Torpedos. 
 
 The transport of living Torpedos from Trieste to Berlin is not pos- 
 silile in all weathers. Frost is no less injurious than summer heat. 
 Consequently the season for successful conveyance of the animals 
 is limited to the months of April, !May, September, and October. 
 
 The Physiological Institute in Berlin possesses its own fuUy- 
 equipped Aquarium. However, I thought it better to take advant- 
 age of Dr. Hermes' kind offer to keep the Torpedos in the tanks of 
 the Berlin Aquarium, which is only a few minutes distant. Here, 
 
 * Monatsberichte der Akademie, 1858, p. 94. 
 ' .See No. VI. of this work. 
 
LIVING TOEPEDOS IN BERLIN. 419 
 
 if they arrive in g-ood condition and are kept quiet, they continue 
 healthy for six weeks without material loss of vig-our. Subsequently 
 some animals died, which from want of leisure I had not been able 
 to use for experiment, but others survived several months, although 
 the temperature of the sea water was only 12° C^. According- to 
 the statement of the keeper, the Torpedos as long as they are in 
 good health burrow in the gravel at the bottom of the tank and are 
 then difficult to distinguish from the ground^. They do not trouble 
 themselves about other fish in the tank. They were not seen to eat 
 any of the fragments of fish which were thrown into it as food. In 
 the free state Torpedos swallow, as Professor Fritsch ascertained by 
 examining the contents of the stomach, \evj large fish which they 
 previously paralyse by electrical shocks ^. It is obviously, however, 
 not easy to provide living small sea fish for them in captivity. 
 
 On the days of experiment the fish was transferred by means of a 
 landing-net into a tub of such size that two men could carry it 
 easily. And by the same means they were introduced into the ex- 
 perimental trough described below. The catching of the animal was 
 accompanied by violent struggles, in which, no doubt, the fish gave 
 repeated shocks, as was ascertained to be the case in the Gymnotus by 
 connecting it with a rheoscopic limb of the frog*. I tried to obtain 
 the same evidence here, but the rheoscopic limb remained motion- 
 less, no doubt in consequence of short circuiting by the sea water. 
 
 After the fish had been introduced into the experimental trough 
 it remained quite quiet, the only motion visible being that of the 
 spiracles. If it was put back into the trough after a long series of 
 experiments, it swam wildly round and round, sometimes endeavour- 
 ing by peculiar undulating movements of its body to get over the 
 wall of the trough. 
 
 ^ Compare my observations as to the temperature at which pcekilothermal animala 
 should be kept. Gesammelte Abhandlungen, vol. ii. pp. 605, 606, and in the Unter- 
 suchungen, p. 77. 
 
 ^ It was stated by Reaumur and Walsh that Torpedos at low tide burrow in the 
 sand by working their pectoral fins, and that fishermen who step upon them with 
 naked feet fall down (Untersuchungen, p. 132). Here a Torpedo incident told me 
 by Ehrenberg may be mentioned, which gives an idea of the power of the Torpedos 
 in the Red Sea (probably T. pantliera). Ehrenberg had waded with his Arab 
 servant on a coral reef deep into the water. Suddenly the man cried that a shark 
 had bitten his foot off, staggered, and was on the point of falling down. "With the 
 quick perception of the naturalist, Ehrenberg reassured him with the words : ' Do 
 not be afraid, there is no blood, you still have your foot, you have only trodden on 
 a " Raad.'" 
 
 ^ Sitzungsberichte, 1883, vol. i. p. 205. 
 
 * Gesammelte Abhandlungen, pp. 615-617. 
 
 E e 3 
 
420 LIVING TORPEDOS IN BERLIN. 
 
 3. Method of leading-off Malapterurus shocks into an 
 experimental circuit. 
 
 By the term experimental circuit I desig-nate as before a circuit 
 however arranged, whether continuous or otherwise, into which the 
 shock is led for the purpose of testing* any of its actions^. One of 
 the first problems in the experiment is obviously to arrange this 
 conduction as favourably as possible. When the animal is taken 
 out of the water this is not difficult ; but even when you have a 
 living- animal at your disposal, it is undesirable to risk its life for 
 one observation ; moreover, fish that have been taken out of water 
 are usually so violent that it is disadvantageous to experiment 
 with them. It is therefore necessary to obtain a solution of the 
 problem which is applicable to animals which are still in the water. 
 
 Faraday in his researches on Gymnotus used for this purpose 
 special arrangements. He applied to the fish two saddles of thin 
 copper, covered externally with indiarubber. The further these 
 saddles were apart from one another, the stronger was the derived 
 current. If very strong actions were required, he placed glass plates 
 on the bottom of the trough to which the indiarubber border of the 
 saddle was so attached, that the part of the fish grasped by it was 
 almost as completely insulated as if in the air^. Faraday does not 
 appear to have thought of insulating the fish between the saddles, 
 indeed, it would be hardly possible to apply a cover so as to enclose 
 the whole of so long and wriggling a fish. In consequence of the 
 great intensity of the Gymnotus shock Faraday's contrivance 
 answered for almost all purposes, it was, however, difficult to obtain 
 discharge shocks^'. I was not able to recommend any better 
 arrangement to Dr. Sachs for leading ofi" the shocks. The saddle 
 employed by him differed from that of Faraday in this respect, that 
 instead of being made of copper coated externally with indiarubber, 
 it was made of guttapercha lined with tin, after the model of the 
 leading-off plates which I used for the Malapterurus. On the left 
 side of the figure the lacquered strip of tin is shown which serves 
 to connect the lining with the leading-off wire. 
 
 In working with the small and relatively weak Malapterurus, 
 I found that the application of the saddles to the ends of the organ 
 was insufficient, and that it was necessary also to insulate the fish 
 between the saddles from the water. This was easily effected by 
 
 * Gesanimelte Abhandlungen, vol. ii. p. 612. ^ Loc. cit. p. 5. nos. 1758-1760. 
 
 ^ Loc. cit. p. 7. no. 1767, note 2. C'omp. Untersuchungen, p. 158. 
 
LIVING TORPEDOS IN BERLIN. 
 
 421 
 
 means of leading'-off plates made of guttapercha shaped like the covers 
 of mummy coffins, and lined with tin at both ends ; these are shown in 
 rig". 12 ; one was made for each fish according to a pattern cut to 
 proper size. In using the cover a plate of glass was laid at the bottom 
 of the experimental trough, this was filled to such a height that the 
 surface of the water a, a' (for the sake of distinctness put too high 
 in the figure) coincided with the back of the fish which rested on 
 the glass plate. I observed that the insulation of the fish between 
 the saddles more than doubled the shock in the experimental 
 circuit. The insulation was so complete that when the electrodes of 
 the frog-alarum were plunged into the trough at two diametrically 
 
 Fig. 12. 
 
 opposite points of its circumference no indication was obtained. It 
 would have been still more perfect if I had closed the posterior 
 opening of the cover wdth a cap arching over the caudal fin. 
 
 In contriving this cover the question arose what length should 
 be given to the lining in order to obtain the strongest current 
 in the experimental circuit. I found that it was advantageous 
 to have a longer lining when the resistance of circuit was great, 
 shorter when it was small ^. I had expected this, and it will be 
 seen from what follows that it must be so. 
 
 The connections h^ Tc and s, s of the two linings may be re- 
 garded as terminals of the experimental circuit. According to 
 Helmholtz's 'principle of the electromotive surface^,' the strength of 
 the current in the experimental circuit is equal to the difference of 
 potential of the points of the lining which are in contact with its 
 
 * Gesammelte Abhandlungen, vol. ii. pp. 636, 637. 
 
 * Poggendorff's Annalen der Physik und Chemje, 1S53, vol. Ixxxix. p. 211. 
 
422 LIYING TORPEDOS IN BERLIN. 
 
 ends, di\nded by the resistance of the circuit plus that of the organ, 
 of the fish, and of the surrounding* water, etc. between those points. 
 The linings serve not merely to lead off from the ends of the elec- 
 trical organs, but so far as they extend over their lateral surfaces, also 
 constitute good conductors between these surfaces and the ends. 
 As by this conduction part of the shock is short circuited, the 
 difference of potential at the ends is the less the smaller the 
 distance between the two linings. In other words the difference is 
 the less the longer the linings. Further it is not quite inde- 
 pendent of the position of the point at which the lining is con- 
 nected with the circuit ; for as the linings themselves conduct 
 from the ends of the organs to the opposite borders, the potential 
 varies in them in the same direction. Consequently the arrange- 
 ment adopted on constructive grounds, according to which the ends 
 of the circuit are respectively united to the anterior border of anterior 
 lining, and posterior border of posterior lining is theoretically the 
 most correct, although in practice the difference can scarcely tell. 
 
 The resistance of the organ, fish, etc. between the leading-off 
 points of the two linings, is according to Helmholtz's law the 
 same which the conducting mass would oppose to a current pro- 
 duced by an electromotive force acting in the experimental circuit. 
 But in relation to such a current the linings constitute in accordance 
 with known principles^ isoelectric surfaces. Thus the resistance in 
 question becomes larger the shorter the linings, smaller the longer 
 the linings : inasmuch as their extent determines the area of the 
 surface of application of the moist to the metallic conductor. 
 
 Let Pi and i]. be the differences of potential of the ends of the 
 experimental circuit with a long and short lining respectively, and 
 the corresponding resistance between the Knings IF and IFj. ; and 
 further, let the great resistance in the experimental circuit be B, 
 the small ;•, then, according to observation, we have the two 
 following inequalities : 
 
 P, Pi Pi P, 
 
 W^ + R ^ Wi + R' Wi + r ^ r, + /' 
 
 this involves the inequalities P^ >Pi, R > r, and as we have seen 
 the latter to be proved, the observed relations are explained. The 
 current flowing through the experimental circuit combines in every 
 point of the conducting mass between the linings, with the current 
 which would exist at the same point if there were no experimental 
 
 ^ G. S. Ohm, die galvanische Kette Mathematiscli bearbeitet, Berlin, 1S27, p. 128. 
 G. Kirchoff in Poggendorff s Annalen der Physik und Chemie, 1 843, p. 500. 
 
LIVING TORPEDOS IN BERLIN. 423 
 
 circuit, according" to the parallelopiped of forces ^. For a complete 
 theory it would be necessary to take into account induction and 
 the manifold polarisations which exist, the main fact, however, 
 would remain unaltered by these considerations. 
 
 4. Method of leading-off the Torpedo-shock into an 
 experimental Circuit. 
 
 Difficulties are met with in the endeavour to carry out the same 
 principles in leading'-olF the Torpedo-shock in the most satisfactory 
 manner. The dorsal and ventral surface of each organ must have 
 a metallic cover applied to it, and these covers must be led oif from 
 their median edges which, for reasons hereafter to be mentioned, are 
 the most positive points on the back and the most negative points 
 on the belly (see p. 430). 
 
 If the fish is in air nothing stands in the way of such a leading 
 off; and Boll who simply fixed his Torpedo, used guttapercha 
 plates lined with tinfoil after the pattern of my leading-off cover 
 for Malapterurus, one of these plates being in as close contact as 
 possible with the dorsal and the other with the ventral surface of 
 the electrical organ ^. But if the fish remains in the water it is no 
 easy problem to apply the plates to the ventral surface of the 
 organ ; and it is still more difficult to insulate the rest of the fish at 
 the same time, which is, how^ever, especially desirable on account of 
 the conductivity of sea water ^. All this may be effected with such 
 a fish provided that it is laid down in the experimental tub upon 
 a guttapercha plate of greater extent than its body, the plate being 
 provided with two metal shields corresponding to the organs, the 
 median edges of these shields being in connection with one end of 
 the experimental circuit. A guttapercha plate must at the same 
 time be placed upon the back and moulded to its convex form. 
 This plate must also bear two suitable shields applied and led 
 off in the same manner. The root of the tail must be left free, 
 but the two plates surrounding the body must be everywhere in 
 connection with each other. 
 
 Unfortunately this plan theoretically so faultless is by no means 
 convenient in practice. The fish refuses to lie on the lower plate 
 
 ^ See Helmholtz, loc. cit. p. 219. In the Gesammelte Abhandlungen, I gave an 
 explanation of the different action of the long and short lining which I had less 
 thoroughly thought out ; in this the insufficiently determinate conception of a ' mean 
 tension of the lining ' was introduced. In place of this explanation the one given in 
 the text must be taken. 
 
 ^ Eeichert und du Bois-Reymond's Archiv fiir Anatomie und Physiologic, 1S73, P- 1^' 
 * Compare Arthur Christiani in the Untersuchungen, pp. 411 et seq. 
 
424 
 
 LIVING TORPEDOS IN BERLIN. 
 
 as it ought to do, while the arrangement of the upper plate, on 
 account of its extent, its double curvature and slight arching is 
 very difficult. Besides which it bends in such a way that it is not 
 easy to connect its warped undulating edge with that of the lower 
 plate. Finally as it is not a question of experimenting as long as 
 possible on a living subject (as was the case with my Malapte- 
 rurus), an apparatus adapted only to a particular animal is not of 
 great service. I therefore contented myself with the less rational 
 but much simpler arrangement diagrammatically represented in 
 Fig. 13, which was in point of fact sufficiently satisfactory. A glass 
 
 vessel 30 cm. wide and 10 cm. deep was used instead of the experi- 
 mental tub. On the bottom of it rested a circular zinc plate of 
 about the same width as the body, forming a ventral shield v'v, a 
 portion of which vv' was bent and hung hook-like over the side of 
 the vessel. One end of the experimental circuit was brought into 
 contact wath this hook. The arrangement shown assumes that a 
 shock vnW be imparted to the human body ; accordingly there are 
 two handles in the experimental circuit, to one of which Hv, the 
 wire v'H \s conducted. A circular piece of flannel^' was laid on 
 
LIVING TORPEDOS IN BERLIN. 425 
 
 the ventral shield, soaked with sea water, to prevent the edg-e of 
 the dorsal shield (fd from ever touching" the ventral, the effect of 
 which would be to render the result in the experimental circuit as 
 useless as in Humboldt's and Gay Lussac's key experiment \ The 
 fish rests on the flannel, and is represented in the fig-ure as if in 
 cross section throug-h the organ. The dorsal shield is an arched 
 zinc plate with the edge turned up, the upper surface of which is 
 lackered, having a wooden knob in the middle through which the 
 leading-off wire (VHcl is conducted insulated to the second handle. 
 The diameter of the dorsal shield corresponding to the median plane 
 of the fish is indicated by a red mark. I had two shields for 
 Torpedos of different sizes, the one 22 and the other 18 cm. in 
 diameter. The sea water in the vessel was in sufficient quantity 
 for its surface just to touch the back of the fish. 
 
 5. Sluggishness of the Torpedo in confinement. 
 In the case of the Malapterurus the putting on of the shield 
 was sufficient to induce them to give shocks, and if not fatigued 
 they would give two or three in succession ^ ; if the cover was 
 pressed down upon them they would give more frequent shocks 
 with convulsive movements until exhausted. The Torpedos on the 
 contrary often failed to react when the dorsal shield was put upon 
 them, and indeed there was no certain method of exciting them to 
 give shocks when in this condition. They would allow themselves 
 to be squeezed, rubbed, pricked, pinched either on the dorsal or 
 ventral surface so far as these were accessible, without being excited 
 to action. No shock was induced even when I applied the fully 
 pushed up secondary induction coil connected with a Duchenne 
 brush, to a portion of skin which projected out of the water, 
 although at the same time the gastrocnemius of the frog-alarum 
 was tetanized. As these Torpedos had not been nearly so long in 
 captivity or kept without food as had been the case with the 
 Malapterurus, it must be assumed that there is a difference in the 
 nature of the two kinds of electrical fish. It would appear, how- 
 ever, from the statements of Walsh, Matteucci and Colladon as to 
 the behaviour of their freshly-caught Torpedos ^, that this difference 
 
 ^ Gilbert's Annalen der Physik, 1806, vol. xxii. p. 8. Untersuchungen iiber 
 thierische Elektricitiit, vol. ii. part i. pp. 15, 105. Gesammelte Abhandlungen, 
 vol. ii. p. 695. 
 
 * Gesammelte Abhandlungen, vol. ii. p. 617. 
 
 ^ Untersuchungen, p. 255. Matteucci, Traitd des Phenombnes ^lectro-physiolo- 
 giques des animaux, Paris, 1844, p. 145. 
 
426 LIVING TORPEDOS IN BERLIN. 
 
 consists only in a more rapid loss of power by the Torpedos ; and in 
 fact Matteucci and Boll report that ' the energy of the electrical 
 org-ans is rapidly diminished in captivity^.' 
 
 The uncertainty whether in individual experiments the fish had 
 been excited or not, made it most desirable to be assured on this 
 point by means of the frog-alarum. Fig. 13 shows how I usually 
 connected it with the fish. The diagram is easily understood — the 
 Gastrocnemius G, the bell T, the hammer //, the weight E. The 
 electrode v'' (not represented in the figure) of the exciting tubes 
 was connected by the wire v''v' with the ventral shield. The 
 other electrode d'" was connected by the wires d'", d" with a 
 circular zinc plate 3 cm. in diameter which extended horizontally 
 from the edge of the basin under the sea water ; it served for the 
 frog-alarum the purpose of the dorsal shield. In experiments in 
 which no ventral shield was used the ring electrode v" was con- 
 nected with a similar zinc plate v' (shown in the figure by a dotted 
 line) on the floor of the basin. The frog-alarum did not seem to 
 be so reliable as in the experiments on Malapterurus, a result 
 which was partly due to the greater conductivity of sea water 
 than of fresh. INIoreover it frequently happened that Torpedos 
 received in spring were more excitable, and, like the Melapterurus ^, 
 according to the e\'idence of the frog-alarum, from unknown causes 
 gave shocks of themselves. 
 
 6. Effect of Torpedo-shock on Man. 
 
 If the handles are grasped with wet hands, and the fish is 
 eSectually excited, a shock is felt which only affects the wrist, 
 for reasons already set forth ^j but which gives a better idea of the 
 phenomenon than any method of seizing the fish directly which can 
 be devised. By this means I communicated the shock to a number 
 of students at my lectures who. joined their wet hands together. 
 The weakness of the Torpedo-shock as compared with that of 
 Malapterm-us (still \dvidly present to my mind) was veiy striking. 
 As, however, in my method of receiving the first shocks, the 
 strength of the Torpedo-shocks is necessarily much diminished from 
 the absence of insulation and the good conductivity of sea water, 
 I cannot regard this observation as a confii'mation of Babuchin's 
 statement that the strength of the shock of Malaptenarus exceeds 
 
 ^ Archiv fiir Anatomie, 1873, p. 78. Professor Fritsch complains of the sluggish- 
 ness of Torpedos in winter, even when freshly caught in the Mediterranean. Sitz- 
 ungsberichte, 1882, vol. i. p. 500. 
 
 ^ Gesammelte Abhandlungen, vol. ii. p. 408. ' Ibid. vol. ii. p. 6iy. 
 
LIVING TOKPEDOS IN BERLIN. 427 
 
 that of the Torpedo ^. The influence of the good conductivity of 
 the medium on the strength of the shock was clearly shown in 
 Dr. Hermes' experiment with a Malapterui'us, which every time 
 that it was brought (for a curative purpose) into a physiological 
 salt solution, just as in my observations on the gastrocnemius of 
 the frog ", apparently lost its electrical properties 'K 
 
 The late A. von Bezold, who helped me in my experiments on 
 Malapterurus, suggested that the shocks might be strengthened by 
 the substitution in the experimental vessel of aerated distilled 
 water for river water *. As many fish ascend rivers from the sea 
 (e. g. the Narcine Brasiliensis brought to Humboldt in Cumana as 
 Tremblador or Gymnotus, was fished from the Manzanares^), it 
 might presumably be possible without great risk to transfer a 
 Torpedo for a time into fresh water. I would gladly have 
 attempted to make observations on the strengthening of the shock 
 which might be looked for under these circumstances, but on 
 account of the sluggishness of the Torpedo it would have been 
 fruitless to make comparative series of observations on the strength 
 of the shock under varying conditions. 
 
 7. On the distribution of the current in the Torpedo. 
 
 The accompanying woodcut is a faithful reproduction of a figure 
 by Cavendish, who, more than a century ago, in his endeavours to 
 imitate by ordinary electricity the ^ 
 
 \ 
 
 Torpedo-shock, first conceived the ^ . ^ 
 
 idea of current curves, and was so / ^■^~~-^^\ \ i / ^---^ \ 
 much m advance of his time in / / /.■^.^C^\ \ I ^\ / /"_ "^ \ \ 
 
 his understanding of this process I \ \ i ( cdSuFj]]]!!^ ' ^ • i 
 in the animal, that Faraday was \ \ ^^^C's^yW \ / 1 ^^^"''^^ / / 
 the first after him to attain to v^^-JlJ^^/ /\ \ "^ --_--■ '^ y 
 
 the same standpoint. Cavendish " / \^"^~- 
 
 sought to accomplish his piu'pose " " p- j ~~ 
 
 by means of a leather model of 
 
 the fish soaked with sea water ; he covered the spots corresponding 
 
 to the polar surfaces of the organ with tinfoil, and connected them 
 
 by insulated wires with a Leyden battery''. By this means the 
 
 polar surfaces became isoelectrical. Cavendish passed over in 
 
 ' Untersuchungen, p. 411. ^ Gesammelte Abhandlungen, vol. ii. p. 379. 
 
 ^ Untersuchungen, pp. 409, 411. * Gesammelte Abhandlungen, p. 613. 
 
 * Untersuchungen, pp. 76, 282, 41 1. 
 
 * The Electrical Researches of the Hon. Henry Cavendish, edited by J. Clerk 
 Maxwell, Cambridge, 1879, p. 194. 
 
428 LIVING TORPEDOS IN BERLIN. 
 
 silence the course of the current curves in the animal itself, nor 
 could he form any distinct conception of them, for it was not until 
 seventy years later that I showed that the assu.mption of the exist- 
 ence of insulating sheaths in electrical fishes was both superfluous 
 and anatomically inadmissible ^. We are now able to correct and 
 complete Cavendish's diagram in both directions. 
 
 In 1831 Daniel Colladon of Geneva, by his investigations at La 
 Rochelle, arrived at the three following propositions as to the distri- 
 bution of electricity on the surface of the Torpedo when giving a shock. 
 
 1. All points on the back are positive to any point on the belly. 
 The strength of the current diminishes in proportion to the distance of 
 these points from the organ, and almost entirely disappears at the tail. 
 
 2. Any two unsymmetrical points of the back, or any two of the 
 belly, almost always yield a current in the galvanometer. The one 
 nearest to the organ is positive in the back, negative in the belly. 
 
 3. When contact is made with two symmetrical points either 
 of the back or belly, there is no deflection in the galvanometer. 
 
 As Colladon was the first to determine the electrical condition of 
 the dorsal and ventral surfaces of the Torpedo, I have given the 
 name of Colladon's currents to the currents between points on 
 either of these sm-faces. 
 
 Colladon's observations w^ere published in 1836. A year later 
 Matteucci, who had previously denied the existence of such cm'- 
 rents, asserted (without mentioning Colladon) that the points of 
 the organ corresponding to the entrance of the nerves were positive 
 on the back, negative on the belly ; a statement which is connected 
 with the erroneous ojiinion he then held as to the origin of the elec- 
 tricity in the brain of the Torpedo ^. 
 
 As I had already foreshadowed to myself the electrical plates 
 since made known by Bilharz, I preferred another explanation". 
 The greater the length of a Torpedo column, provided that the 
 number of plates in a unit of length is the same, so much the 
 greater must be its electromotive force. The columns diminish in 
 height from the inner to the outer edge of the organ by about 0'6 
 
 ^ Gesammelte Abhandlimgen, vol. ii. p. 678. 
 
 ^ The literature, both old and new, of the subject is to be found in the Gesammelte 
 Abhandlungen, vol. ii. pp. 684 fF. I may remark that the concluding sentence of 
 par. vi. p. 690, is not quite exact as to Matteucci's frequently changing expressions, 
 but the correction of it would require a lengthy setting forth and is no longer of 
 interest to any one. 
 
 ^ Preliminary sketch of an investigation of the so-called frog-current and of 
 electromotive fishes. PoggendorfF's Annalen, 1843, vol. Iviii. p. 25, par. 64, 65. 
 Gesammelte Abhandlungen, vol. ii. pp. 668, 670. Untersuchungen, pp. 284, 285. 
 
LIVING TORPEDOS IN BERLIN. 429 
 
 (see Fig". 15) ; by so much, therefore, may the electromotive force of 
 the median columns be g"reater than that of the outer ones. This 
 explains why when a Torpedo gives shocks in the air, a current is 
 observed in the back from the median to the outside points, and on 
 the belly in the reverse direction. In Torpedos under water this 
 arrangement seems to be adapted to its purpose, inasmuch as the 
 curves of currents proceeding" from the hig-her columns have 
 almost the same strength as the shorter ones which proceed from 
 the lower columns. 
 
 As Matteucci gradually altered his opinions and allowed that the 
 current flowed on the back from the thicker to the thinner parts, 
 and also indicated the existence of a stronger current between back 
 and belly in the thicker than in the thinner, so he later (without 
 mentioning me) adopted the explanation given by me in the ' Pre- 
 liminary Sketch.' I did not, however, adhere to this explanation, 
 for I found later that it was not well founded. By an experiment in 
 which I suddenly plunged into water zinc platinum elements grouped 
 like electrical plates, I imitated, suflficiently for the purpose though 
 very imperfectly, a shock given under water, and thus tested my 
 conclusions regarding its distribution. These observations have 
 already been given in detail, and I will here confine myself to 
 recalling the results. 
 
 An arrangement of electrical plates which corresponds to a slice 
 of the organ of the Torpedo bounded by two cross cuts, in which 
 therefore the number of plates in the columns diminished from the 
 one side to the other, yielded quite regular currents which were 
 directed from the higher to the lower on the positive aspect of the 
 organ, and from the lower to the higher on the negative ^. It was 
 found however experimentally, as on theoretical grounds might 
 have been anticipated, that Colladon's currents occurred even when 
 there was no difference in the height of the columns. If an 
 electrical organ having columns of the same height were placed in 
 an unlimited quantity of water, the middle districts of the polar 
 surfaces would be relatively the most positive and the most nega- 
 tive. Let us suppose all the columns of the organs of the Torpedo 
 to be of equal height, and the organs to be brought together 
 in the middle plane and there united ; then the middle of the 
 median line would be most positive on the dorsal aspect, most 
 negative on the ventral. Let the organs be again separated, then 
 
 * Comp. XJntersuchungen iiber thierische Elektricitat, vol. i. pp. 644 fF. Gesammelte 
 Abhandlungen, voL ii. p. 688. 
 
430 LIVING TORPEDOS IN BERLIN. 
 
 the most positive and the most negative spots in each organ after 
 the separation would assume a position intermediate between the 
 median edge and the middle, which last would only be reached 
 if the organs were placed at such an indefinite distance from each 
 other that each could be regarded as alone existent. 
 
 Thus it is clear that even apart from the diminution in the 
 height of the columns towards the sides, currents would exist on 
 the back and belly of the Torpedo in the direction observed. The 
 thinning of the organ towards the sides would nevertheless have 
 the effect first of strengthening these currents, and secondly of 
 shifting the points of greater and lesser positivity and negativity 
 to the median borders of the organ. It would also appear that 
 differences of tension due to inclination-currents similar to those 
 observed by me in oblique sections of muscle, would act in the 
 same direction^. Another conclusion may be drawn from all this 
 which is especially important to me at the point at which I have 
 arrived. The median borders of the organ being the most positive 
 dorsally and the most negative ventrally, they must necessarily 
 be relatively more positive and more negative than the middle line. 
 So that there must be currents on the back from the borders towards 
 this line, on the belly from the latter to the former. Colladon's 
 second proposition and Matteucci's statement that the points of 
 nerve entrance on the back and belly are relatively the most 
 positive and the most negative (see p. 428), might appear to 
 imply that the authors had actually observed the state of things 
 indicated above. But it is unlikely that the means at their 
 disposal were sufficient, or that they would not expressly have 
 mentioned an observation so remarkable. However this may be, 
 the result will show that it was worth while to put our conclusions 
 on this matter to the test of experiment ; and for many years 
 I waited a suitable opportunity of doing so 2. 
 
 A Torpedo 29 cm, long was placed on its belly in a vessel filled 
 with just enough sea water to cover the back of the animal. The 
 following arrangement was used to lead off the shocks into the 
 experimental circuit. Two amalgamated zinc plates, each 10 cms. 
 long and 2-5 broad, enclosed in corks, could be placed in every 
 position and at any desired distance in the manner described by 
 Norremberg'^. Oblong pads^ of equal breadth with the plates were 
 connected with them ; they were soaked with sulphate of zinc 
 
 * Gesammelte Abhandlungen, vol. ii. pp. 93-127. 
 
 - Loc. cit. p. 690. ^ Loc. cit. p. 648. * Loc. cif. vol. i. p. 7. 
 
LIVING TORPEDOS IN BERLIN. ' 431 
 
 solution. To these others were attached similarly soaked with sea 
 water. These last were pointed, and the points projected beyond 
 the zinc pads so as to touch the suiface of the fish at the spot at 
 which it was to be led off. The zinc plates were connected by 
 wires with the galvanometer, the bobbins of which possessed 5000 
 turns, at a distance of 30 mm. from the very lig-ht reflecting magnet 
 which was quite aperiodic ^. If the fish gave no shock the mirror 
 was deflected only a very few degrees of the scale. These ine- 
 qualities were balanced by the round compensator. The frog- 
 alarum was brought into connection with two horizontal zinc 
 plates, one of which rested on the bottom of the glass vessel, while 
 the other just dipped into the water (see above, p. 426). I ob- 
 served the galvanometer. Prof. Fritsch who is accustomed to 
 handle Torpedos manipulated the fish, and Prof. Christiani took 
 charge of the rest of the arrangements. 
 
 When one point was placed upon the median and the other upon 
 the outer edge of one organ, in a line which bisected it about 
 equally, and the fish was excited, a deflection was observed almost 
 simultaneously with the sound of the frog-alarum, and this deflec- 
 tion (through the galvanometer) was always from the median to the 
 outer edge. The extent of the deflection was very unequal, e. g. on 
 one day it was 135 right and 100 left, while on another it was 100 
 right and 500 + a; sc. left ; that is to say, in the latter case the scale 
 disappeared from the field of vision. These differences may have in 
 part arisen from differences of resistance in the circuit caused by 
 the greater or less pressure of the pointed pads ; for the most part, 
 however, they depended on the irregular efforts of the fish. When 
 deflections were observed beyond the limits of the scale it is 
 possible that several shocks occurred in such rapid succession that 
 they were confused in the frog-alarum. 
 
 If one point was placed upon the median edge while the other 
 was placed upon the median line, deflections occun'ed constantly in 
 the direction previously observed, namely from the edge of the organ 
 to the median line. Thus in four cases the amount of deflection was 
 
 Right. Left. 
 
 1 30 
 
 2 15 
 
 3 20 
 
 4 19 sc. 
 
 shewing as much regularity as could have been anticipated. The 
 
 * I used Min-or I of my first Treatise • Ueber aperiodische Bewegung gedampfter 
 Magnete.' Loc. cit. vol. i. pp. 308, 309. 
 
432 LIVING TORPEDOS IN BERLIN. 
 
 fish was then laid on its back. The frog-alarum showed that the 
 violent efforts of the fish to turn over ag-ain were accompanied by 
 numerous shocks. In this case the points were on the one side 
 upon the median line, on the o.ther upon the median border of the 
 org-an corresponding* to the gill openings, and the results of the 
 shock were 160 right sc. and 70 left sc. in the expected direction, 
 that is, from the median line to the border of the organ. 
 
 A shock between the back and belly always shot the scale far 
 out of sight, and in the first experiments, when so many points of 
 great importance pressed for observation, no comparison was made 
 of the strength of Colladon's currents with that of the shock be- 
 tween the polar surfaces. 
 
 On the other hand, in order to be perfectly sure of the existence 
 of the new currents, I did not fail frequently to lead off the shock 
 from symmetrical points. I had, however, a different result from 
 that put forward in Colladon's third projiosition (see p. 428), 
 namely, that in continuation of the last experiments there occurred 
 between two points on the ventral surface outwards from the nasal 
 openings a deflection of 160 sc. from right to left; on another day 
 between corresponding points on the dorsal surface a deflection 
 of 42 sc. from left to right, and between two symmetrical points 
 of the body some centims. behind the organs, 29 sc. in the same 
 direction. When I allowed the left point to stay where it was, 
 and placed the other on the posterior edge of the left pectoral fin, 
 a deflection of 230 sc. was seen in the expected direction, that is to 
 say, from the point nearest the organ towards the more distant one. 
 
 The results observed between symmetrical points show nothing 
 more than that the fish does not give a shock of equal strength 
 with both organs. In both experiments on the ventral surface, as 
 well as one on a later day on the dorsal surface, it usually gave a 
 stronger shock with the left organ. The reason why similar results 
 were not seen by Colladon may perhaps have been that his very 
 strong fishes innervated both organs equally. Perhaps also these 
 results were beyond the reach of the means of observation at his 
 disposal, for in this case the new cun-ents observed by us between 
 the middle line and the median border of the organs must neces- 
 sarily have escaped his notice. 
 
 The importance of the last named currents consists in what 
 follows. De Sanctis objected to my doctrine of the relative im- 
 munity of electrical fish from their own shocks, that he experienced 
 no shock when his hand was inside the body of a Torpedo which 
 
LIVING TORPEDOS IN BERLIN. 
 
 433 
 
 was out of the water. He concluded from this circumstance that 
 only an inconsiderable portion of the shock goes through the body 
 of the fish, without taking into account that this might be the 
 case with a fish exposed to the air but need not be so with one 
 under water. In the air, the current in order to go through the 
 body must take its way along the thin, perhaps half dry, skin, and 
 it is not therefore surprising that a hand inside the body should 
 feel very little of it^. Meanwhile even under these circumstances 
 Boll succeeded with the Torpedo, following my procedure with 
 Malapterurus, in detecting the shock in the interior of the body 
 of the fish by means of wires insulated to the points^. 
 
 The proof that the Torpedo-shock really penetrates the body of 
 the fish and attains a greater density than elsewhere in the brain 
 and spinal cord, and in the great nerve trunks, is not thereby 
 weakened. The currents, 
 which in the back flow 
 from the median borders 
 of the organ to the middle 
 line, and in the belly from 
 the middle line to those 
 borders, necessarily take 
 their course through the 
 brain and spinal cord, and 
 as this is the shortest path 
 between the most active 
 portions of both organs, 
 there exist no stronger cur- 
 rents in the Torpedo. As 
 in Malapterurus so here my assertion is confirmed, that the body 
 of the electric fish itself is more favourably placed for receiving 
 the shocks of its own organs than that of another creature in its 
 neighbourhood ^. 
 
 A glance at Fig. 15 makes this clearer, and also shows how in 
 consideration of the results we had obtained we needed to alter 
 Cavendish's diagram. The cross section represented, including 
 the columns which are only diagram matically indicated, is drawn 
 strictly from nature. The first fact to notice is that the current 
 
 Fig- 15- 
 
 ' Compare Untersuchungen, p. 128. 
 ^ Loc. cit. pp. 260, 261. 
 
 ^ Monatsbericlite der Akademie, 1858, p. 107. 
 ii. p. 638. 
 
 Ff 
 
 Gesammelte Abhandluugen, vol. 
 
434 LIVING TOIirEDOS IN BERLIN. 
 
 curves do not merely stream out from the so-called polar surfaces, 
 but also cut the lateral surfaces of the organ ^. They then run 
 inward and outwai'd thi-oug-h the body of the fish and occupy the 
 surrounding" space. The curves do not, however, as represented by 
 Cavendish, start from the polar surfaces of the columns as pro- 
 longations of their axes. But the curves in this direction, according 
 to the principle of the superposition of euiTents, blend with the 
 curves which tend from the thicker to the thinner portions of the 
 organ and through the latter to the ventral surface. Hence it 
 arises that the resulting curves are inclined towards the sides as 
 shown in the figure. A circumstance occurs here with which I 
 had previously to deal in my 'Experimental criticism of the dis- 
 charge hypothesis.' A nerve end-plate, regarded as an electrical 
 plate, which therefore has a stronger electromotive force the thicker 
 it is, and which becomes thinner from the middle towards the 
 edge, resembles the Torpedo organ, which in like manner decreases 
 towards the sides. Here, therefore, if there are really nerve end- 
 j^lates in this sense, the curves of the cuiTent must stream obliquely 
 out of the one surface and into the other ^. If you imagine a 
 sagittal section through a Torpedo organ, there being from front to 
 back no diminution in the height of the columns comparable with 
 that in the transverse direction, the direction of the curves in the 
 sagittal plane would be primarily that of the columns. The whole 
 system of curves could only be made clear by means of a model. 
 
 The inclination outwards of the median columns shown in the 
 figure (their electromotive action being parallel to their axes) must 
 influence the direction of the current curves on the back in the 
 same way as the greater height of the median columns, whereas on 
 the belly, as indicated in the figure, the inclination of the cm'ves is 
 lessened by the same circumstance. The oblique position of the 
 columns is more strikingly seen in the di-awing of the cross section 
 of a young Torpedo given by Fritsch in the Memoir alluded to on 
 p. 419. By this means the region over the middle of the fish is 
 apparently deprived of denser current curves and, to use a military 
 expression, becomes comparable to a dead angle (zu einem todten 
 Winkel). The teleological signification of this arrangement is 
 obscure, for when the fish, in its favourite position, lies half buried 
 at the bottom of the sea, the dorsal surface seems to require the 
 greatest protection, while the ventral surface is not in need of any. 
 
 ' Gesammelte Abhandlungen, vol. ii. pp. 629, 682. Untersuchungen, p. 148. 
 ' Ibid, vol. ii. p. 708, Fig. 47. 
 
LIVING TORPEDOS IN BERLIN. 435 
 
 8. On Iodide of Potassium Electrolysis by the Torpedo-shock. 
 
 When, following- Faraday's procedure with the Gymnotus, I 
 made use of iodide of potassium electrolysis in order thereby to 
 determine the direction of the shock in Malapteriu'uSj I came upon 
 the phenomenon of the secondary iodine spot (as already many 
 times related) ; that is to say, I saw a spot arise under the nega- 
 tive as well as under the positive platinum point. I at once per- 
 ceived that I had before me an occurrence with which I had already 
 become acquainted when I employed the same means for ascer- 
 taining the direction of the current in complicated induction-cir- 
 cuits^. The platinum points are polarised by the decomposition of 
 the iodide of potassium, and if the circuit is not instantly opened 
 again, whether in experiments with the induction coil or with 
 electrical fish, a secondary or polarisation- current follows on the 
 top of the primary-current in the opposite direction, and gives 
 rise to a spot on the formerly negative but now positive point. 
 I was induced by the experience of the Malapterurus experiments 
 to make the appearance of the secondary iodine spot the object of a 
 complete investigation, of the results of which I will recall only 
 one or two. In the first place, if the quantity of electricity in the 
 primary-current which is restored within the unit of time exceeds a 
 certain limit no secondary spot is seen. But there are circumstances 
 in which the secondary spot surpasses the primary, so that the 
 iodide of potassium electrolysis becomes a completely deceptive 
 indication of the original direction of the current. This is the 
 case if, besides the platinum points dipped in iodide of potassium 
 solution, another pair of platinum electrodes in say a weak solution 
 of sulphuric acid, are also in the circuit ^. 
 
 The possibility of the double iodine spot being caused by a 
 double shock of the fish was worth considering^. I satisfied myself 
 by means of the frog-interrupter that this was not the case. The 
 secondary spot was absent if the circuit was opened either imme- 
 diately after the shock or towards the end of it. Besides this, in 
 Malapterurus it was sometimes possible to make the secondary spot 
 the stronger of the two by leading-off' from the fish with platinum 
 electrodes. Thus the dependence of the secondary spot on polari- 
 sation was established in this case, and in this respect there remained 
 no obscurity as regards Malapterurus. 
 
 It is, however, incomprehensible that none of the observers who 
 
 * UntersuchungeDj voL ii. Part i. 1849, P- 4°0- 
 •^ Gesammelte Abhandlungen, voL ii. pp. 648, 666. 
 
 l' f 2 
 
436 LIVING TORPEDOS IN BERLIN. 
 
 before me undertook so many times iodide of potassium electrolysis, 
 both with Gymnotus and Torpedo, shoidd have mentioned the 
 secondary spot — not even Faraday, who made use of tliis means for 
 the purpose of studyin^^ in the fish the distriljution of tensions. 
 Perhaps in Gymnotus the shock is too strong to allow the formation 
 of the secondary spot. Even Sachs, on whom I pressed the clearin;^ 
 up of this point, failed to obtain the secondary spot in Gymnotus^. 
 
 I was extremely curious to see what the result would be of my 
 experiments on Torpedo. Nothing- being* needed for the experiment 
 except the iodide of potassium electrolyser and the frog-alarum, I was 
 able, through the kindness of Dr. Hermes, to undertake it as early 
 as the summer of 1881 in the Berlin Aquarium (see p. 418). Since 
 then I have often repeated it, and without exception have always 
 seen the secondary spot, very black and at the moment of its appear- 
 ance sharply defined, and I have also shown it to several other 
 observers. In the first instance the electrodes applied to the fish 
 were merely a couple of zinc plates, but afterwards the dorsal and 
 ventral shields above described were used. It need not be said that 
 care was taken that they should be homogeneous. The spots only 
 appeared at the moment when the frog-alarum struck. Both spots 
 seemed to me to be fainter than in the case of Malapterurus. 
 I endeavoui-ed by the use of platinum electrodes to make the 
 secondary spot the larger of the two. In this I was not successful, 
 although the spot was brought more distinctly into view. It was, 
 however, easy to hinder the formation of the secondary spot by 
 means of the frog-interrupter. An overweighting of 30 grammes 
 delayed the opening of the circuit sufliciently to permit the forma- 
 tion of a primary spot, which was not appreciably smaller, without 
 any secondary one. If the frog'-interrupter was partly short 
 circuited the secondary spot reappeared with the next shock. 
 
 Thus as regards the matter in hand all is clear with regard to the 
 Torpedo, and the only obscurity which remains is why the secondary 
 spot escaped the observation of all earlier observers, and especially 
 of Matteucci, who led off with platinum electrodes as well as with 
 platinum points, by which means the secondary spot is certainly 
 strengthened if not always made the stronger of the two. I cannot 
 imagine that the Torpedo-shocks by which John Davy and Mat- 
 teucci decomposed iodide of potassium, exceeded in strength the 
 powerful shocks of my Malapterurus to such an extent that the 
 secondary spot did not appear in consequence of the larg-er quantity 
 * Untersuchungen, p. 163. 
 
l.IVING TORPIiDOS IN BERLIN, 
 
 437 
 
 of electricity which was lost by derivation (see p. 435). The only 
 other possibility already considered by me in the conclusion of my 
 paper ' On Iodide of Potassium Electrolysis,' namely, that in 
 those experiments in which the Torpedos were observed out of the 
 water, the resistance of the circuit in which the polarisation 
 discharg-ed itself was too great, seems equally improbable consider- 
 ing- the sensitiveness of the iodide of potassium. The fish was kept 
 wet with well conducting sea water, the path of the current through 
 it was short and of large section, and its tissues jjrobably conducted 
 better than those of Malapterurus ; so that the resistance could 
 scarcely be greater than in my experiments on the last mentioned 
 fish ^ Under these circum- 
 stances I cannot avoid the 
 suspicion that the secondary 
 spot must have been seen, but 
 was passed over as an in- 
 comprehensible disturbance. 
 Some indication of this ap- 
 pears in Matteucci's paper ^. 
 In order to put others in 
 a position to repeat my ob- 
 servations in a similar man- 
 ner, I represent in Fig. 16 the 
 iodide of potassium electro- 
 lyser long used by me, and 
 which in mj^ opinion is an indispensable apparatus for the electrical 
 laboratory, but is seldom met with, although very easy to construct. 
 
 ' According to Leon Fredericq the blood of Octopus vulgaris and Astacus marinua 
 contains nearly four times as much salt as that of mammals ; and according to Boll 
 the tissues of Torpedos have a ' physiological ' concentration of 2-5 per cent, solution 
 of phloride of sodium (Untersuchungen, p. 133). Perhaps on this account the 
 tissues of marine animals conduct better. I have not yet found time to decide this 
 point as regards Torpedos. Any other marine fish out of the Aquarium would be 
 equally useful for tlje purpose. In accordance with Boll's statement the clay used 
 in the Torpedo experiments should have been kneaded with a 2-5 per cent, solution, 
 instead of the usual 0-75. I observed nothing, however, which could be regarded as 
 an unfavourable result of this neglect. A superiority of sea fish over fresh water fish 
 with respect to the ash of their muscles cannot with certainty be deduced from the 
 118 analyses of the flesh of fishes made by Atwater of Middietown, Conn., U. S. A., 
 and published in the Berichte der deutschen chemischen Gesellschaft (vol. xvi. 
 DOS. 12, 23, .July 1883). Tlie summary of Koenig, however, seems to favour the notion 
 (Die menschlichen Nahrungs- und Genuss- mittel, Berlin, 1883, 2nd ed. pp. 179, 180). 
 According to VVeyl the Torpedo organ yields 1-55 per cent, of ash, a quantity rather 
 greater than that of the muscles of river fish in these determinations (Monats- 
 berichte der Akademie, 1881, p. 382). 
 
 * Gesammelte Abhandlungen, vol. ii. p. 650. 
 
 i-iir. lO. 
 
438 LIVING TORPEDOS IN BERLIN. 
 
 The description will be found in the Introduction to the Treatise 
 ' Ueber lodkalium Elektrolyse.' It need only be mentioned here 
 that the sheet of plate-glass measures 150 mm. long and 75 mm. 
 broad, that the upright glass tube is 150 mm. in height, the hori- 
 zontal axis bearing the corks measured from the axis of the upright 
 tube is 125 mm. in length. As I have elsewhere remarked, the 
 secondary spot is best observed when it is seen by inclining a glass 
 plate placed at a suitable angle to the horizon, and looking at it 
 from behind. In order that the iodide of potassium paper may 
 allow the spot to shine through it, a single layer of the finest 
 blotting-jxaper must be used. 
 
 9. Of the Organ-current in the Torpedo. 
 
 By organ-current I mean, in contradistinction to the shock, a 
 current persistently generated through the organ and as a rule in 
 the direction of the shock. Its electromotive force will be referred 
 to as that of the organ-current. This action of the organ which 
 reminds one of muscle tone, was first observed in the Torpedo 
 by Zantedeschi, confirmed afterwards by Matteucci and denied by 
 Eckhard. I overlooked it in Malapterurus. On the other hand 
 Robin observed it even in the imperfect electrical organ of the 
 Skate, and Sachs found it constantly in the Gymnotus^. 
 
 I made a point of determining, if possible, the question as to the 
 existence of an organ-current in the Torpedo. The Torpedo used 
 in my first experiments of this kind was 36 cm. in length, and 
 had been living in the Berlin Aquarium for about five weeks from 
 the end of May. That it might be as little as possible exhausted 
 by shocks before the experiment it was killed in the following 
 manner. A sharpened borer prepared from a steel-tube 13 mm. 
 in diameter, was placed by Prof. Fritsch on that part of the bpny 
 skull of the fish (which was lying tranquilly in the tub) which he 
 knew corresponded with the electrical convolution of the brain. 
 He punched out this convolution by a single stroke of the hammer, 
 which drove the borer through the thickness of the fish into the 
 
 ^ Gesammelte Abhandlungen, vol. ii. pp. 672, 718, 722, 723. Untersuchungen, 
 p. 169. As I recently discovered, Galvani in his fifth letter to Spallanzi ascribes to 
 the Torpedo a persistent electrical action, which manifested itself in the twitchings 
 of frog preparations, the feet of which he had applied to the fish. In one instance 
 the frog preparation was suspended by a silk thread. Memorie sulla elletricitk 
 animale . . . al celebre Abbate Lazzaro Spallanzani, Bologna, 1797, iv. p. 75. It 
 must be remarked, however, that Galvani's statements are in some respects incompre- 
 hensible, and in others admit of a different explanation. 
 
LIVING TOKPEDOS IN BERLIN. 439 
 
 bottom of the tub. As a proof of the success of the operation the 
 convokition adhered to the borer. The fish twitched a g-ood deal 
 but gave no shocks. It was taken out of the water and suspended 
 with its body in a vertical plane by means of an anatomical hook 
 thrust through the edg-e of the pectoral fin. Thus both the 
 ventral and dorsal surfaces could be conveniently reached by the 
 pads described on p. 430 placed horizontally. They were constantly 
 so placed that the axes of the two coincided. The galvanometer had 
 the same sensitiveness as in the experiments on Colladon's currents. 
 
 A current showed itself with the g-reatest regularity in the 
 direction of the shock. It was strong-est when the hig-hest columns 
 on the median edge of the organ were between the pads, and 
 became weaker in proportion as the pads approached the thinner 
 outer edge of the organ ; e. g. if in the first case the result was 
 18 sc. it sank in the middle of the organ to 9 and at the edge 
 to 3 sc. It was even traceable in the same direction in the median 
 plane between the two organs, and at the edge of the body where 
 no organ exists. This is quite in order, for the persistent current 
 equalizes itself just as the momentary shock does in the skin and 
 body of the fish, by branch currents which follow every channel 
 that offers itself. In other words, if we put out of account the 
 smallness of the differences of potential, the electromotive surface of 
 the non shock giving fish differs probably from that of the same 
 fish when giving a shock only in the induction which accompanies 
 the shock. The results obsei-ved had nothing to do with the 
 wound caused by punching out the electrical convolutions. 
 
 There was still another way of investigating the organ-current. 
 In anticipation of experiments on Torpedos I had occupied myself 
 for a long period in considering how a bundle of columns of regular 
 limits could be so obtained as to serve for experiments such as 
 those on muscle or on the organ of Malapterurus. This last is by 
 nature so limited by the outer skin and inner tendinous membrane 
 that it is easy with scissors to cut regular strips of a given length 
 and breadth, whereas in dead Torpedos, which I had once received 
 from Trieste, the prisms of the organs broke up into hour-glass 
 form^. I had thought of all sorts of contrivances for encountering 
 this difficulty, such as a revolving sharp borer like a trephine, with 
 which cylindrical bits of the org-an might be cut out, troughs 
 of glass or ebonite, with removable sides, in which the bits could be 
 confined. When at length, last summer, I found myself in presence 
 ^ Gesammelte AbhaiKllungen, vol. ii. p. 721. 
 
440 LIVING TORPEDOS IN BERLIN. 
 
 of the reality, I discovered that all these preparations were super- 
 fluous. With a long, broad, straight knife, such as a ham-knife or 
 an English bread-knife, the use of which on the Torpedo I learned 
 from Prof. Fritsch, a strip ^-6 mm. thick can be cut from the 
 organ. This includes only a few columns, and if the cut-surface 
 is applied to strong cardboard it adheres firmly to it. From such a 
 strip, four-sided prismatic bits of the organ can be cut with long- 
 bladed paper scissors which are not too sharp. These four-sided 
 prisms, limited by a bit of skin on the back and belly 5 or 6 
 mm. square, contain a considerable number of columns. Such a 
 portion is placed upon the well-known three-cornered glass plate of 
 the ' Carrier^,' the bottom layers of skin are connected with the 
 clay shields of the conducting pads to which they adhere ; the pads 
 are gently separated so as to draw out the columns to their full 
 length. Such a preparation presents a very clean and neat object 
 of investigation, the length of which naturally depends on the size 
 of the fish and the part of the organ from which it was taken. 
 The longest columns which I manipulated were about 29 mm. in 
 length, and were therefore shorter than the group of muscles from 
 the thigh of the frog. The bundles of columns appeared somewhat 
 thicker. 
 
 My first idea was that the cut surfaces of the preparation would 
 pass through the columns parallel to their axes, not between them, 
 so that the lateral surfaces would be beset by shreds of electrical 
 plates which would either stick to them or to each other, or roll 
 themselves up. But the most careful examination with a lens in 
 fluid failed to discover any such shreds. On the contrary the 
 columns separate themselves from one another under the blades of 
 the scissors with the fibrous sheath undestroyed. It of course 
 sometimes happens that only the upper, under, or middle part of a 
 particular column is attached to the bundle, but these irregularities 
 can easily be got rid of. The shreds of plates if they existed would 
 have no noteworthy electromotive action, or if they had, would 
 retain it only for a very short time, so that they would only come 
 into account as aflfording channels of derivation. 
 
 Such bundles of columns will be used for our experiments on 
 polarisation, which I need hardly say are our most important 
 problems. These experiments take the following form ; as the 
 
 ' Untersuchungen, vol. i. 1S48, pp. 495, 496. [Professor du Bois' ' allgemeine 
 Trager ' is an instrument by means of whicli a preparation may be supported in the 
 moist chamber in any desired position. Tr.] 
 
LIVING TORPEDOS IN BERLIN. 441 
 
 bundles of columns do not twitch like muscle, it is not necessary to 
 immobilize them. It is sufficient to make the conducting- troug-hs, 
 between the clay shields of which the bundle of columns rests on 
 the triang-ular glass plate, the terminals of the polarising- circuit, 
 in order to lead the current in the manner best adapted to the 
 purpose through the ventral and dorsal surfaces to the columns. A 
 couple of non-polarisable conducting tubes serve for the purpose of 
 leading-off the polarisation-current, the clay points of these are in 
 contact with the preparation between the clay shields by which the 
 polarising-current is led in. 
 
 The organ-current then showed itself quite regularly in the 
 galvanometer-circuit just as Matteucci states that he observed it in 
 cubical bits of the organ only 2 mm. wide ^. This method of 
 observation has the advantage that there can be no suspicion of 
 any electromotive difference between the pigmented skin of the 
 back and the unpigmented skin of the belly. In preparations of 
 tolerably fresh organs the organ-current was always in the direction 
 of the shock. It was easy to demonstrate that a column-like 
 arrangement of the electromotive forces gave rise to it, for it 
 appeared in the same direction no matter where the clay points 
 were applied, and its strength increased with their distance from 
 each other. For example, between the middle and a point of the 
 lateral surface as near as possible to the dorsal end, it amounted to 
 + 13-5 sc. (The plus sign means that the effect took place in the 
 direction of the shock.) Between the middle and a point nearest 
 the ventral end, the amount was -H 10 sc. ; but between the two 
 outermost points + 23 sc. That is, owing to the considerable 
 resistance of the conducting tubes, only a little less than the sum 
 of the effects of both halves. In another case the corresponding 
 numbers were + 5, + 4, + 11 sc. If the connecting line of the 
 clay points cut the axis of the columns vertically no regular effect 
 ensued worthy of being taken into account. 
 
 In these experiments I frequently measured the force of the 
 organ-current and found it to be between 0-005 ^^nd 0-013 I^aoult, 
 considerably less therefore than the force of the nerve-current in 
 frogs. The current force in bits about 4 cm. long of the organ of 
 Gymnotus, were estimated by Sachs to be from 0-015 ^o 0-030 
 Uaniell, the mean therefore being 0-0325. The length of mv 
 bundles of columns varied from 29 to 12 mm. according as they 
 
 ' Archives des Sciences physiques et naturelles. Nouvelle Pe'riode, vol. xv. 1862, 
 pp. 41, 42. 
 
442 LIYING TORPEDOS IN BERLIN. 
 
 were taken from the middle or the outer edge of the organ in 
 larger or smaller fish, their average length therefore amounted to 
 
 2 cm., and. their average lorce was =0-009 Raoult, 
 
 or about 0-0085 DanielP. Double this length the force would 
 
 have been 0-0170, and thus only a little less than the average 
 
 force of much larger bits of the organ of Gymnotus. 
 
 This correspondence may indeed be followed still further. If 
 
 one reckons an average of ten plates to a millim. in the Gymnotus 
 
 organ, the medium organ-current force of the individual Gymnotus 
 
 1 j_ -n I- 0-0000375 + 0-00007 50 ^ ^ 
 
 plates will be '^^-^ '-^ = 0-00005625. On account 
 
 2' 
 
 of a coiTcction which is probably necessary to be made in all Sachs' 
 
 figures this value is raised to 0-00006. If in the Torpedo the 
 
 medium number of plates to the mm. is reckoned to be 30 this 
 
 gives 615 as the number of plates to the columns of medium 
 
 2Q 4-12 
 height, that is — = 20-5 mm., but only 500 of these would lie 
 
 between the conducting clay points. If we divide the average 
 organ-current force observed by us, viz. 0-0085 Daniell, by this 
 number, we have 0-0000117 Daniell, as the mean electromotive 
 force of a single Torpedo plate. This value is ^-^^ times less than 
 that found for a single plate of Gymnotus. According to my ideas 
 it ought to be about 8-5 times less on account of the thickness of 
 the Gymnotus plates being about 8-5 times greater ^. 
 
 A closer coincidence between two numbers obtained under such 
 circumstances is not to be expected. It is sufficiently surprising 
 to find that the numbers are in similar order. Possibly it may be 
 only a result of coincidence. Yet it may be regarded as a rare 
 coincidence that from the same assumptions as those on which the 
 above calculation is founded, I should arrive at the remarkable 
 conclusion according to which the electromotive force of the whole 
 Gymnotus bears to that of the whole Torpedo (in adaptation to the 
 conditions of each), almost the same relation as, according to 
 Christiani's determination, the resistance of fresh water to sea 
 water ^. 
 
 * Compare E. Kittler, ' Die elektromotorische Kraft des Daniell' schen Elements.' 
 Sitz. der math. phys. Classe der k. b. Akad. der Wissensch. zu Mfinchen, vol. xii. 
 1882, pp. 501, 502. Also Wiedemann's Annalen der Physik und Chemie. N. F., 
 1882, vol. xvii. p. 893. 
 
 ^ Untersuchungen, pp. 174, 175, 278-280, 286, 
 
 ' Ibid. pp. 414, 415. 
 
LIVING TORPEDOS IN BERLIN. 443 
 
 For the present we may conclude that the electromotive elements 
 are nearly of equal force in the two fish, and that the difference of 
 potential of the electrical plates increases with their thickness. 
 As regards Matteucci's statement that the force of the org-an- 
 cui'rent of the whole Torpedo lies between that of a single gas- 
 trocnemius of the frog and that of two such, it is sufficient to say 
 that the electromotive force of the uninjured gastrocnemius is, 
 through parelectronomy either nil or even negative, so that it 
 cannot serve as a unit of measurement. 
 
 The force of the organ-current diminishes as the organ dies ; 
 it possesses, however, great tenacity of life ^. In fish kept in the 
 cold, an organ-current force of 0-003 ^^ 0*003 Raoult is met with 
 after 24 or even 48 hours. The persistence of the organ-current 
 is a sign of the maintenance of functional activity. Later than 
 this the preparations are not only inactive but occasionally the 
 action is reversed, an observation already made by Zantedeschi as 
 regards the whole fish ^. Sachs made the same observation on 
 Gymnotus though under somewhat suspicious circumstances ^. 
 One might seek to explain this reversed action after the cessation 
 of the organ-current, by slight irregularities of a different kind, 
 which according to their direction sometimes assume the sem- 
 blance of remnants of normal organ-current, sometimes of a 
 reversed organ-current. But this reversal of the current seems to 
 occur too regularly to be so explained, and it recalls too distinctly 
 the similar phenomenon which takes place, also with great regu- 
 larity, in certain more delicate muscles as they gradually die *. 
 
 The same diminution and irregularity in the force of the organ- 
 current which is observed in the dying organ of Torpedo, appears to 
 occur in the living fish under unfavourable circumstances as regards 
 nomishment. For when I repeated the experiments in winter on 
 a Torpedo 29 cm. in length which had been about seven weeks 
 in the Aquarium and was killed as before by punching the 
 electrical convolutions, I did not obtain the same effects, but only 
 deflections indefinite in direction and extent, which might have 
 been dependent on slight differences in the skin or on other 
 disturbances. Even the bundles of columns from this fish were 
 either inactive or showed reversed action, though one of them 
 
 ' Berichte, 1882, voL i. p. 500. Compare Untersuchungen, p. 188. 
 
 ^ Coinptes rendus, 1842, vol. xiv. p. 489. 
 
 ^ Untersuchungen, p. 258. 
 
 * Ibid. vol. ii. pt. i. pp. 154, 3S3, 553. 
 
444 LIVING TORPEDOS IN BERLIN. 
 
 manifested a g-ood strong organ-current. In another Torpedo 
 26 cm. in length which had been even longer in captivity, and in 
 which I did not experiment on the organ-current of the whole fish, 
 I found the bundles of columns to act very regularly, and in some 
 instances with considerable force. 
 
 The diminution of force of the organ-current in badly nourished 
 animals would explain the fact that I found none in strips of 
 the Malapterurus organ even with the nerve-multiplier. As 
 regards what Eckhard considered his negative results, it will 
 be seen on closer examination that they are not altogether in- 
 consistent with ours. Of his experiments on the whole fish, of 
 which to prevent voluntary discharges and movements the brain 
 and spinal cord were destroyed, Eckhard remarks, ' The needle 
 was seldom completely at rest but the deflections were always very 
 small, and frequently at all events corresponded in direction to 
 such a current as would be induced in the organ by the excitation 
 of its nerves. There was, however, no permanent deflection.' Of 
 experiments on bits of an organ led off from the ventral and 
 dorsal surfaces he says, ' The deflections had often the same 
 direction as above noted ; they were not, however, to be com- 
 pared with those resulting from much smaller portions of muscle 
 of the same animal.' The organs and parts of organs gave 
 shocks on excitation of their electrical nerves ^ What this 
 amounts to is simply that Eckhard expected stronger currents than 
 those he actually observed, a circumstance which was due to the 
 defective sensibility of his galvanometric apparatus. The absence 
 of permanent deflections is explicable by the circumstance that 
 Eckhard still used platinum for leading-ofF. 
 
 In the bundles of columns taken from the median part of the 
 organ, it sometimes happens that nerve branches protrude from the 
 middle of their length, as out of the tubes of a muscle. If a bit of 
 nerve is cut off from the preparation a shock follows which may 
 even drive the scale out of sight. Matteucci described a similar 
 efl'ect, having observed twitchings in a nerve muscle preparation in 
 contact with bits of organ only as big as pin heads, when these were 
 mechanically excited^. I tried in vain, however, to excite the 
 bundles of columns with ammonia, although Sachs had succeeded in 
 doing so with portions of the organ of Gymnotus^. The reason is 
 clear, viz, that in bits of the Torpedo organ it would only be 
 
 ' Beitrage zur Anat. und Physiol., vol. i. pp. 159-162. 
 
 ' Untersuchungen, pp. 175, 176. ^ Ibid. pp. 177, 178. 
 
LIVING TORPEDOS IN BERLIN. 445 
 
 possible to excite effectually, in the manner above related, a single 
 column cut longitudinally. 
 
 In such preparations the organ-current often gradually di- 
 minishes. Matteucci in Torpedos, and Robin in the Skate observed 
 it to rise somewhat after each shock, and something similar was 
 also seen by Sachs in his observations on Gymnotus^. In the few 
 experiments in which I succeeded in exciting bundles of columns by 
 mechanically exciting the nerves, the rapid subsidence of deflection 
 due to the shock blended itself in such a way with the effect of the 
 organ-cm-rent, that the shock deflection presented itself as a mere 
 transitory increase of the latter. But Sachs has further shown in 
 Gymnotus that tetanus of the organ weakens the current ^. I had 
 no opportunity of making similar observations on the Torpedo. 
 
 10. On the secondary Electromotive Actions of the Organ 
 of the Torpedo. 
 
 I started upon this virgin soil with the greater zest because, as 
 the reader will remember, there seemed to be a contradiction 
 between the results of my experiments on this subject on Malapte- 
 rurus and those of Sachs on the Gymnotus, I longed to decide 
 which of these two antagonistic observations would accord with 
 those about to be made on the Torpedo. 
 
 The apparatus and methods of experiment which I used were the 
 same that served for my recent researches on secondary electromotive 
 phenomena in muscle and nerve. The manner in which the 
 preparations of the organ were exposed to the polarising-currents, 
 and in which the secondary-currents were led off has been already 
 described. At the first glance it appears to be an advantage that 
 the organ preparations do not twitch like muscle. On the other 
 hand, as in nerve experiments, one is deprived of the evidence of 
 the continuance of functional activity which the twitch gives in 
 muscle. Although some of the difficulties which I anticipated did 
 not appear, yet the preparations made in the manner described could 
 not be compared in regularity with such naturally formed objects 
 of experiment as we possess in nerves and in certain muscles. 
 Whereas it is not difficult to provide for comparison a bit of sciatic 
 nerve or a muscle of equal size and freshness, and uninjured, it is 
 impossible to be assured of an equal number of uninjured columns 
 in our preparation. In the same fish the length varies between the 
 
 * UntersuchungeTi, pp. 170, 173. . ^ Ibid. pp. 174, 187, 220. 
 
446 LIVING TORPEDOS IN BEKLIN. 
 
 median and outer edges of the organ ; the functional activity of 
 preparations cut from the same organ at short intervals of time is 
 for unknown reasons very different ; and in addition to this there 
 is the variable amount of short circuiting caused by the remnants 
 of cut columns which remain attached to the uninjured ones (see p. 
 440). These are the reasons (indicated on p. 41 tS) why even that 
 small degree of exactitude with which I was obliged to content 
 myself in muscles and nerves, cannot be attained in polarisation 
 experiments on the Torpedo columns themselves. A table of the 
 secondary electromotive effects of the electrical organ similar to the 
 ones w^hich I drew up for muscle and nerve, wath double entry, in 
 which one head should comprise increasing current densities wath 
 currents in both directions, another head increasing closing times, 
 could only be thoroughly carried out by someone who, like Boll at 
 Yiareggio, has at his disposal for weeks together at least five freshly 
 caught fish daily, and who has no other object in \dew^. 
 
 In w^hat follows the secondary electromotive effects will be 
 especially dealt with, which the current induces either in the 
 direction of the Torpedo-shock or in the opposite direction. The 
 former, according to the nomenclature employed by me in my book 
 on Gymnotus, I call absolutely positive, the latter absolutely nega- 
 tive^. Polarisation in the direction from belly to back is therefore 
 absolutely positive ; and according as it follows an absolutely posi- 
 tive or negative polarising-current it may be considered relatively 
 positive or negative. For the sake of greater facility in distinguish- 
 ing polarising-currents with respect to their direction, I will call 
 the current in the direction of the shock the Homodromous, and 
 that in the opposite direction the Heterodromous ^. The homo- 
 dromous current is indicated by an ascending arrow ( | ), the hete- 
 rodromous by a descending one ( j )• Plus and minus signs serve 
 to distinguish between relatively positive and negative polarisation. 
 By ' both currents,' I understand simply the homodromous and 
 heterodromous currents. Finally for brevity, it is desirable to call 
 the polarisation whether through the homodromous or the hete- 
 rodromous current, whatever its absolute direction may be, as 
 relatively homodromous and heterodromous polarisation. 
 
 ^ Archiv fiir Anat. und Physiol., 1873, p. 77. * Loc. cit. pp. 149, 213. 
 
 ^ A colleague learned in philology whom I consulted with regard to these expres- 
 sions, advised the use of ' isodromous ' and ' anisodromous.' I considered, however, 
 the forms introduced into the text more practical on account of the similarity of sound 
 of the others to ' isotropous ' and ' anisotropous ' which are used in the description of 
 the organ. 
 
LIVING TORPEDOS IN BERLIN. 447 
 
 The first result of these experiments is that internal polarisation 
 of the organ follows the passing* of a current through it in the 
 direction of the columns, which like the polarisation of muscles, 
 nerves and the organ of Malapterurus, is under different circum- 
 stances sometimes relatively positive, sometimes relatively negative, 
 the conditions required for the appearance of both polarisations 
 being generally the same in the latter as in the former. With a 
 long closing time, polarisation is in all circumstances relatively 
 negative ; with a short closing time, particularly of stronger 
 currents, relatively positive polarisation arises under certain condi- 
 tions. Under certain other conditions which cannot be intentionally 
 brought about, action in two directions results — first negative then 
 positive, thus proving that here also negative polarisation is the 
 more transitory. As the functional activity diminishes positive 
 polarisation disappears, though it is a considerable time before it 
 is quite absent and negative polarisation alone remains. All these 
 effects are in general observed in equal strength to whatever part ot 
 the columns the clay points are applied, provided that they are 
 kept at a constant distance. With a sufficient resistance of the 
 galvanometer - circuit, which is secured by the leading-off clay 
 points, these effects are the greater the greater the distance — in 
 other words, we have to do with electromotive forces arranged in 
 pile. Finally, with an oblique direction of the polarising-current 
 along with an oblique position of the leading-off clay points, only 
 weak results in indefinite direction are met with. The proofs of 
 this are put together in the Appendix. 
 
 The relations of polarisation to the direction of the polarising- 
 current are so marked in all these experiments, that one cannot go 
 into detail as to the phenomenon without first entering on this 
 point. My experiments had generally the same form as those on 
 the upper and lower half of the group of muscles, or on the anterior 
 and posterior roots of the spinal nerves^, that is to say the homo- 
 dromous and heterodromous currents are sent alternately through 
 the columns at definite intervals of time. The stronger the current 
 and the longer the closing time, the greater the interval must be 
 in order to give time to the preparation to return to a sufficiently 
 natural condition for comparative observations 2. 
 
 ' Sitzungsberichte, 1883, vol. i. pp. 363-383. 
 
 ^ Long continued series of experiments which have to be broken at definite 
 intervals are the more fatiguing from the necessity of constantly looking at the 
 clock to see whether the time for a fresh observation has arrived. After having 
 
448 LIVING TORPEDOS IN BERLIN. 
 
 The tables in the Appendix do not require any further explana- 
 tion than has already been given in the Memoir ' On secondary 
 electromotive phenomena.' The numbers in the horizontal series 
 indicated by S, are the deflections throug-h the secondary current, 
 those indicated by P, are the primary deflections. A glance at 
 these series, e.g. at series lo, will show that in this and the follow- 
 ing as far as 13, the homodromous current ( t ) induces both 
 absolutely and relatively positive polarisation, the heterodromous 
 current ( I ) absolutely positive and relatively negative. From 
 here onwards polarisation by both currents becomes relatively 
 negative, the heterodi'omous remains absolutely positive, the homo- 
 di'omous becomes even absolutely negative, but is at first much 
 weaker than the absolutely positive but relatively negative polarisa- 
 tion due to the heterodromous current. As the preparation becomes 
 exhausted this difference disappears more and more. 
 
 All the series of experiments with short currents sent in alternate 
 directions take as a rule this course. Whether the observation 
 begins with the homodromous or the heterodromous current, 
 polarisation by both currents becomes sooner or later relatively 
 negative — the sooner (that is after the smaller number of alternations) 
 the longer the closing time, but if the closing time has a certain 
 dm'ation even the first homodromous current produces polarisation 
 which is both absolutely and relatively negative. The same thing 
 occiu's when the functional activity of the preparation is too small. 
 Here from the very beginning only relatively negative polarisation 
 is obtained by both currents, but the homodromous polarisation is 
 appreciably the weaker of the two. Thus the condition arises at 
 once which with a better state of the preparation only gradually 
 appears in the course of a long series of observations, viz. that ' the 
 negative polarisation-current is always stronger in the direction of 
 the shock.' 
 
 We have thus insensibly arrived at an important result. For 
 these are the words in which Sachs summed up his researches on 
 homodromous and heterodromous polarisation in the Gymnotus^, 
 and of which I had hitherto supposed and repeatedly said that they 
 
 occupied myself for several years with such experiments, I had at length an experi- 
 mental Alarum clock made for me by Baltzar and Schmidt of Leipsic, which could 
 be set to various periods so as to indicate to the observer by the stroke of a bell 
 at intervals of i, i|, i|, 2, 2^, 37, 5 or 10 minutes, that the moment for a fresh 
 observation was come, he having been warned by another signal. 
 * Untersuchungen, pp. 217, 218. 
 
LIVING TORPEDOS IN BEKLIN. 449 
 
 contradicted my results with Malapterurus^. Now that I con- 
 stantly observed the same thing- in the Torpedo that Sachs did in 
 the Gymnotus, and at the same time saw how the phenomenon 
 came into existence, scales fell from my eyes. Not only is there no 
 contradiction between our results and those formulated by Sachs in 
 the above words, but both can be deduced from one very simple 
 assumption ; and as my results with Malapterurus are also in accord- 
 ance with the same assumption, it may be asserted that in all three 
 electrical fish the secondary electromotive phenomena are essentially 
 the same, and in a certain sense are understood. 
 
 We are further led to this assumption by the observation that in 
 all my experiments on Malapterurus and Torpedo the heterodromous 
 current never induced relatively positive polarisation. Even polari- 
 sation in two directions, first relatively negative and then positive, 
 is only seen in the homodromous currents. Let it be supposed 
 that both currents polarise relatively negatively in equal degree, 
 but that the strength of the homodi'omous current is so much 
 greater than that of the heterodromous that the relatively positive 
 heterodromous polarisation is constantly masked by the relatively 
 negative, and all becomes clear as may be seen in Fig. T7. 
 
 The assumptions under which this figure is drawn are similar to 
 those which illustrate Sachs' results in Table II of the book on 
 Gymnotus. The diagram represents shortly what happens in a 
 series of experiments in which, as in the series given in the Appen- 
 dix, the two currents are sent alternately through a portion of the 
 organ. The abscissal axis of course marks the time. The ordi- 
 nates in each section correspond to the moment of closure of the 
 galvanometer after the opening of the battery cii'cuit. Above the 
 abscissal axis they express positive polarisation and below it 
 negative. The ascending arrow means that, the current being 
 homodromous, the course of the curve of polarisation above the 
 abscissal axis is absolutely and relatively positive, and below the 
 abscissa absolutely and relatively negative. The descending arrow 
 implies that, the current being heterodromous, the course of the 
 curve of polarisation if above the line is absolutely positive, rela- 
 tively negative, and if below it, absolutely negative, relatively 
 positive. The resulting polarisation current which acts upon the 
 galvanometer is represented in each section by the shaded space 
 which lies between the curve resulting from the algebraic sum- 
 mation of both polarisations and the axis. 
 
 • Sitzungsberichte, vol. i. p. 395. 
 G o- 
 
450 
 
 LIVING TORPEDOS IN BERLIN. 
 
 In accordance with the reality it is seen that at first both 
 currents give absohitely positive polarisation (section i of the 
 fig-ure), the homodromous being- the strong-er. Then comes a period 
 (section 2) in which, althoug-h both polarisations are still absolutely 
 positive, the heterodromovis is the strong-er. In section 3 the 
 homodromous polarisation is seen to be absolutely neg-ative and 
 distinctly smaller than the heterodromous. In section 4 the two 
 j)olarisations approach equality. 
 
 I need scarcely point out that it was the phenomenon of section 
 3 which Sachs had seen, without having been aware of the two 
 preceding, and which he described in words which from their 
 apparent contradiction to my results caused me so much difficulty, 
 viz. that ' relatively negative polarisation is always stronger in 
 the direction of the shock.' There still remains however some 
 
 i I 
 
 ^ 
 
 M 
 
 3- 
 
 ik. 
 
 
 ^^ 
 
 ^ 
 
 Y 
 
 ■ " 
 
 ' 
 
 
 Fig. 17. 
 
 obscurity. Besides the polarisation experiments thus explained, 
 Sachs has given others which are shown graphically in Plate II of 
 the book on Gymnotus. In these experiments also polarisation by 
 either current was without exception first relatively negative, but 
 so strongly that the mirror shot out to the equator, in returning 
 from which, according to Sachs' description, it swung several times 
 on either side of the zero point. If the closing time exceeded a 
 certain duration it swung twice over the zero point into the 
 
LIVING TORPEDOS IN BERLIN. 451 
 
 relatively positive quadrant. The first of these two swing's may- 
 be explained by the laws of the aperiodic movement of damped 
 mag-nets. As the mag-net fell from a height which exceeded the 
 aperiodic district it was possible for it to pass the zero point, but 
 only once^. The second movement in the same direction which 
 Sachs also reg-arded as a swing-, could only have depended on the 
 change of sign of the resulting polarisation, provided that the 
 mirror was really aperiodic. In the Gj^mnotus book I succeeded in 
 drawing curves of the two polarisations, by combining which the 
 resulting curves observed by Sachs could be deduced without much 
 difficulty ; but this was done under the assumption that both the 
 heterodromous and homodromous currents induce relatively positive 
 polarisation. This assumption is inconsistent with the conception 
 which has now served to account for a much more numerous and 
 better series of observations, including those of Sachs' own differently 
 arranged experiments. The only traces of heterodromous relatively 
 positive polarisation in the three electrical organs would be that 
 second overpassing of the abseissal axis in some of Sachs' experi- 
 ments. Under these circumstances I am led to believe that the 
 latter were in fact what Sachs himself considered them to be, 
 namely, oscillations. His galvanometer did not stand quite steadily^, 
 and perhaps his magnet was not well centred. In such a case a very 
 strong current might throw the magnet into pendular oscillation 
 to such an extent that it might overpass the zero point twice. 
 
 However this may be, the assumption which lay at the basis of 
 my construction in the Gymnotus book, of the curves which express 
 Sachs' empirical results, cannot now be maintained, and it may be 
 considered proved that in all three electrical organs the homodro- 
 mous current induces relatively positive polarisation, if not ex- 
 clusively at any rate far more strongly. Before we attempt to 
 draw a conclusion it is desirable to take knowledge of other facts. 
 
 11. On the relative Strengths of the Homodromous and of the 
 Heterodromous Current in the Electrical Organ. 
 
 In my experiments on polarisation in the organ of Malapterurus 
 I had already been much struck with the different strengths of the 
 homodromous and heterodromous currents. ' In fresh strips in 
 which positive polarisation appeared in full force in the direction of 
 the shock, the descending current (homodromous in the Malapte- 
 
 ' Gesammelte Abhandlungen, vol. i. pp. 284, 324, 355. 
 
 * Aus den Llanos. Schilderung einer naturwissenscliaftlichen Keise nach Venezuela. 
 Leipzig, 1879, p. 198. Untersuchungen, p. 137. 
 
 G g 3 
 
45.2 LIVING TORPEDOS IN BERLIN. 
 
 inirus) was always considerably strong-er than the ascending' (hetero- 
 dromous) in the relation of lOO to 112, ii6, or even 125. In heated 
 or dying- strips the difference disappeared. This result could not 
 apparently be attributed to anything but the existence during- the 
 continuance of the primary current of a positive polarisation of 
 g-reat electromotive force, comparable to that of several Groves' 
 elements, which in the homodromous current added itself to the 
 force of the battery. Strips of the org-an vertical to the direction 
 of the shock (therefore cut obliquely from the fish) yielded, under 
 the influence for a short time of a battery of 30 cells, a weaker 
 positive polarisation, which however was equally strong- in both 
 directions ; and the streng-th of the polarising- current g-ave a 
 deflection in both directions to the same number of the scale^.' 
 
 So far had I arrived in 1857. How full of sig-nificance did it 
 now appear to me when in almost every experiment on the Tor- 
 pedo the superiority of the homodromous curi-ent manifested itself 
 even in a hig-her degree than in Malapterurus. Series 13 and 14 
 are striking instances in point, for here the homodromous cui'rent 
 of 30 Groves often appears more than twice as strong, as the hetero- 
 dromous. To explain this by a positive polarisation belonging 
 only to the homodromous current, while an equal strength is 
 attributed to the negative polarisation following each current, we 
 should have to ascribe to the homodromous positive polarisation an 
 electromotive force of more than 15 Groves. In series 20, with 50 
 Groves in the circuit, the relation between the heterodromous and 
 homodromous current is at first nearly 3 to 5 ; the positive polari- 
 sation must have attained a strength of 20 Groves. In series 10, 
 with 20 Groves, the same relation is as one to three ; the secondary 
 electromotive force in this instance is indeed only equal to about 
 13 Groves, but it must have amounted to double the primary. 
 
 When the current is directed transversely in the preparation 
 there is no difference between the two currents, and the same 
 remark holds good when the preparation is exhausted or has been 
 killed by boiling temperature (see Sect. 12, and series 17 and 18 in 
 the Appendix). In exposed preparations the difference can be ob- 
 served to diminish as the preparation gradually dies. 
 
 We unfortunately as yet know nothing certain about the electro- 
 motive force of electrical fishes, except that it must be very 
 considerable, least so, however, in the Torpedo ■'^. At all events it is 
 
 * Gesammelte Abhandlungen, vol. ii. p. 720. UntersuchuDgen, pp. 206, 218. 
 
 * Untersuchungeii, p. 276. Aum. 2. 411. 
 
LIVING TOKPEDOS IN BERLIN. 453 
 
 at the first g-lance incredible that so large an electromotive force 
 should prevail in our preparations of org-ans, and another cause must 
 be soug-ht for the different streng-th of the two currents. 
 
 A difference of current strength must depend upon one of two 
 causes — either upon unequal electromotive strength or upon unequal 
 resistance. It is conceivable that in the electrical organ an irre- 
 ciprocal resistance may exist ^, i.e. that it may conduct better in 
 the direction of the shock than in the other. Electrolytes conduct 
 only in so far as they are decomposed ; according to a well- 
 established law, however, electrolysis is preceded by a similar 
 arrangement of electropositive and negative elements, which 
 arrangement we have taken as the basis of our explanation of the 
 shock and of positive polarisation. If this arrangement occurred 
 more easily in the absolutely positive direction or in this direction 
 exclusively, it would be comprehensible why the organ should 
 conduct the homodromous better than the heterodromous current. 
 The dependence of the hypothetical irreciprocal resistance on the 
 vital condition might be explained on the ground that the in- 
 equality of the resistance in the two directions was an expression of 
 electrical properties of the organ molecules which depended on their 
 life. Thus we see ourselves in face of a problem which often recurs 
 in galvanism and generally remains unsolved, namely to dis- 
 criminate whether an observed difference of current strength arises 
 from a difference of electromotive force or of resistance. 
 
 In the present case one method of answering this question would 
 consist in introducing into the primary circuit so considerable a 
 resistance in addition to that of the preparation of the organ, that 
 the latter should disappear. If then the difference between the 
 two currents disappeared while that of the two polarisations 
 remained, this difference would be proved to be due to difference of 
 resistance. I planned an experiment of this kind, in which, besides 
 the preparation, I introduced as resistance a tube bent twice at 
 right angles, 7 1 centims. long and two millims. internal diameter, 
 containing physiological (normal) salt solution. The ends of the 
 tube were closed with clay and dipped into two vessels of zinc 
 solution. The difference of current strength in the two directions 
 disappeared, but unfortunately owing to the diminished functional 
 activity of the preparation, that of the polarisation current was also 
 absent. In this case nothing more could be done as to the 
 
 1 Comp. Arthur Christiani, Beitrage znr Elektricitatslelire. Ueber irreciproke 
 Leitung elektrischer Strome, Berlin, 1876. 
 
454 LIVING TORPEDOS IN BERLIN. 
 
 determination of this question, and I have not yet been able to 
 repeat the experiment, 
 
 I took an oj^portunity of approaching" the desired end in another 
 way. Instead of observing" the polarisation after opening the 
 primary current, I tried to form some notion of it during- its 
 persistence, by short circuiting" the preparation throug"h the leading- 
 off clay points and the galvanometer circuit. The battery con- 
 sisted of 20 Groves; the galvanometer coil {S) of 5000 turns had to 
 be placed at a great distance from the mirror in order to keep the 
 deflections within due limits. The closing time amounted at first 
 to o^'-oyG^, afterwards to i"-024. The result was astonishing ; for 
 while the homodromous current through the galvanometer (P) 
 appeared almost twice as strong as the heterodi-omous, the branch 
 of the homodromous cm-rent led off through the galvanometer (S) 
 appeared very much weaker than that from the heterodromous 
 (Series 25). When I replaced the preparation by a model of 
 physiological clay (a much worse conductor, by the way) the 
 diSerence had disappeared — a proof that irreciprocal resistance did 
 not exist anywhere in the circuit. 
 
 It thus seems settled that the difference depends upon irreci- 
 procal resistance. For as it is not conceivable that with an equal 
 resistance in the preparation the led-ofi" branch of the stronger current 
 should turn out to be weaker, one is led to suppose that this depends 
 exclusively on the better conductivity of the preparation for the 
 homodi-omous current. More precise considerations however, aided 
 by a schematic calculation, teach that this view is not sufficient, 
 and that at all events under the assumption taken as the basis of 
 the calculation, the result must be different. 
 
 For the sake of simplicity we assume that both currents produce" 
 relatively negative polarisation of equal strength, but that rela- 
 tively positive polarisation is produced only by the homodromous. 
 We assume the electromotive force of both polarisations to be pro- 
 portional to the density of the battery current in the preparation, 
 taking the transverse section and the length as each severally =1. 
 W^e will first enquire into what happens when the specific resistance 
 of the preparation is unequal, as we thus avoid the complication 
 which would arise from the assumption that the difference of 
 resistance in the two directions may be a function of the current 
 density. This assumption, though well worthy of consideration, is 
 very difficult to express mathematically. Also we look upon the 
 preparation as a linear conductor, or as if both the dorsal and 
 
LIVING TORPEDOS IN BERLIN. 
 
 455 
 
 ventral surfaces rested upon metallic non-polarisable electrodes, 
 with which the ends both of the battery circuit and of the gal- 
 vanometer circuit were in connection. See Fig*. i8, A and £, which 
 represents the state of thiug-s for the homodromous and hetero- 
 dromous currents respectively. The battery circuit with the battery 
 are indicated by 2, the galvanometer by (P), the galvanometer 
 through which the branch is led off by (S). The branches of current 
 which proceed from the battery are indicated by unbroken arrows, 
 those of the absolutely positive polarisation by broken arrows, aud 
 those of the relatively negative polarisation by dotted arrows. It 
 is unnecessary to remark that in this rough sketch we take no 
 account either of the duration of polarisation, or of induction. 
 
 Fig. 1 8. 
 
 Let E be the electromotive force of the battery : I, the resistance 
 of the battery circuit containing the battery and the galvanometer 
 (P) ; A, the resistance of the galvanometer circuit containing the 
 galvanometer (<S) ; w, w\ the resistance of the preparation when the 
 current is homodromous or heterodromous respectively ; K, the sum 
 of the products ivl + 2vk-\-kl with the homodromous current, 
 according to Kirchoff ; K', the corresponding sum with the hetero- 
 dromous current. Let P be the constant by which the cuiTent 
 density prevailing in the preparation must be multiplied in order 
 to give the electromotive force of the homodromous absolutely and 
 relatively positive polarisation ; IT, the corresponding constant for 
 the relatively negative polarisation ; /, the strength of the homo- 
 dromous current in the battery circuit ; 1', that of the hetero- 
 dromous ; ^, and i', the corresponding strengths of current in the 
 circuit derived through the galvanometer. 
 
 In the battery circuit there prevails a current strength of 
 E (w + \)/K which is due to the electromotive force of the battery. 
 To this is added whatever amount of current exists in the battery 
 circuit in consequence of the polarisations. The battery circuit has 
 
456 LIVING TORPEDOS IN BERLIN. 
 
 in the j^reparation the streng-th o? E\/K. Therefore the polarisation 
 induced (as the transverse section = 7) is [P — Y\)EX/ K, and the 
 branch current which originates in consequence in the column 
 current {{P—U)Ek/K]x[\/K\ As in this circuit P and E 
 are in the same direction, we have 
 
 I^§-,[{w + K)K+{P-U)K^-\. 
 
 As in the galvanometer circuit P and E counteract one another, 
 we have tt 
 
 i = ^,[wK-{P-u)xq. 
 
 1' and I are obtained respectively from / and i, by making 
 P = o and by transforming tvK into ivK'. Accordingly 
 
 r= -?- [(«;' + A) A'' -OA^], i'=^[tv'K'^UXq. 
 A 2 Ag 
 
 By observation it is known that i <i, I> /', thus we have the 
 
 inequality -. . ^, ., 
 
 ^ -^ I—i>r—i,ov 
 
 A-,[A+(P-n)(A + /)]>A-,[A'-n(A + /)] H. 
 
 The first question now is whether, according to our surmise, this 
 inequality is fulfilled by w' >w when P = o, which would imply 
 irreciprocal resistance in the preparation without absolutely positive 
 polarisation. Certainly this is the case. If you make P = o, and 
 for the sake of shortness A + ^ = a, A + ^=/3, the inequality may be 
 written thus : 
 
 (w' — w) [((3—2aU){fi + a(iv^ + w)) + aw'w'] >.o. 
 For this condition is satisfied by w' > w, so long as the quantity 
 within the square brackets does not become negative. This can 
 only happen if the constant n of negative polarisation exceeds 
 a certain value, which is for the moment of no interest to us. 
 
 Now, on the other hand, it may be asked whether, according to 
 the other part of our surmise and the conclusion we arrived at 
 jprima facie, the inequality may not also be fulfilled when w = to\ 
 when P is not = o as above, but of finite positive magnitude. 
 Inasmuch as_ it is obvious that the inequality (*) holds good so 
 long as P > o, that is to say, provided that homodi-omous absolutely 
 relative polarisation exists, the phenomena may present themselves 
 just as we observed them even without irreciprocal resistance. 
 Thus we have arrived at no conclusion ; the predominance of the 
 homodromous current can be explained as well by positive polarisa- 
 tion as by irreciprocal resistance. 
 
LIVING TORPEDOS IN BERLIN. 457 
 
 Naturally the conception remains possible that both explanations 
 have a foundation in fact ; it certainly is so if irreciprocal resistance 
 is in question ; for there can be no doubt of the existence of positive 
 polarisation, whereas irreciprocal resistance is a new property attri- 
 buted to the org-an, the existence of which along- with positive 
 polarisation still requires proof. Until this proof is g-iven, it must 
 be assumed that in an org-an preparation of scarcely the size of the 
 well-known group of muscles of the frog-, there is always an electro- 
 motive force of 20 Groves. It will be easier to believe this when 
 one remembers that in the organ of the living fish when giving 
 a shock, a much greater electromotive force is undoubtedly in 
 operation. But when the force of this organ is brought into 
 conflict with a known force like that of Groves' battery, the 
 astonishing value of this fact is impressed more forcibly upon us. 
 It has indeed only one counterpart, the surprising nature of which 
 we are apt to overlook, namely, the mechanical activity of which 
 the few grammes of water and organic substance which here act 
 electrically are capable when they take the form not of an electro- 
 motive organ, but of muscle^. 
 
 What gives in addition special interest to the experiments just 
 described, in which the preparation is short circuited during the 
 continuance of the current by the galvanometer circuit, is the 
 circumstance that Sachs made a similar experiment on Gymnotus 
 with, so far as can be judged, the same result. He kept the current 
 of his seventeen Groves closed through a bit of organ, and led off 
 a branch with the clay points through the galvanometer, the sen- 
 sibility of which was suitably diminished. ' When the cm-rent was 
 led through the organ in the direction of the shock, the deflection 
 amounted to 80 sc, in the other direction to 95 sc. The deflection 
 remained constant at this point 2.' Thus Sachs only observed 
 i and i\ as the means for observing I and /' were wanting. As 
 therefore he found, as I did in the torpedo, i<i\ it is very probable 
 that he would also have found I> V. 
 
 12. Dependence of the Secondary Electromotive Effects and 
 of the Conductivity of the Organ on its Vitality. 
 
 It has already been pointed out several times that the polarisa- 
 bility of the organ is intimately bound up with its vitality. I have 
 not yet made methodical experiments as to this dependence, and they 
 would be difficult to carry out for the reasons given on p. 445. The 
 
 ^ Comp. Moiiatsberichte der Akademie, 1858, p. 99. ^ Untersuchungen, p. 218. 
 
458 LIVING TORPEDOS IN BERLIN. 
 
 dead organ only shows weak relatively neg-ative polaris^ation which 
 is equally strong" in the two dii-eetions. The same fact is observed 
 in a preparation which has been boiled. The organ was no longer 
 fresh ; it had lain on ice for forty-eight hoiu-s, the air being of 
 a moderate summer temperature. The organ showed an alkaline 
 or neutral reaction to litmus paper, but after a sudden scalding and 
 boiling it became acid and turned the paper decidedly red. In this 
 resjject therefore it manifested the same difference from muscle as 
 the organ of ^lalapterurus^. The columns of the boiled organ 
 separated from one another at the slightest touch. In this con- 
 dition the preparation had lost every trace of organ current. 
 Before the boiling, the cm-rent of 30 Groves being closed for ^\ 
 the relatively negative polarisation was so strong that the scale 
 disappeared from the field ; after the boiling only a trace in the 
 two directions resulted. Thus it looked very much as if the 
 resistance of the preparation had diminished, just as I found to be 
 the case in muscle and in the tissue of plants-, and as I had also 
 stated with regard to the organ of INIalapterurus ^. The current 
 strengths before and after boiling were in the proportion of about 
 100 to 133; but as in boiling some of the columns had been 
 separated the proportion was undoubtedly smaller. Whether the 
 dead organ, like dead muscle, conducts better than the fresh, 
 I could not determine, but it is very probable, on account of the 
 acid reaction which appears as the organ dies. 
 
 13. Discussion of the foregoing Results. 
 
 However incomplete the experiments above given as to the 
 secondary electromotive phenomena of the electrical organ may 
 
 * Archiv ftir Anatomie, etc., 18,^9, p. 846. Gesanunelte Abhandlungen, vol. ii. 
 pp. 17 ff., 646, 647. Untersuchungen, pp. 70-72. 
 
 * Comp. J. Eanke, Tetanus. Eine physiologische Studie, 1865, p. 16. Gesammelte 
 Abhandlungen, vol. ii. pp. 95, 118. 
 
 ' Sitzungsberichte, 18S3, vol. i. pp. 392, 398. By an eiTor it is there stated that 
 a strip of Malapterurus organ has its resistance diminished by boiling, from 100 to 42. 
 It ought to be, as above in the text, that the strength of the current is raised by 
 boiling in the proportion of 100 to 238. That the proportion of the current strengths 
 before and after boiling was less in the case of the Malapterurus than in that of the 
 Torpedo arose from the circumstance (in addition to that given in the text) that the 
 conducting vessels with club-shaped pads and films of albumen, then used for leading 
 off, had a smaller resistance than the conducting tubes with clay points which were 
 employed later. But for this very reason, this proportion, in the experiments on 
 Malapterurus, approached more nearly to the actual (recipi'ocal) proportion of the 
 resistances of the preparation. 
 
LIVING TORPEDOS IN BERLIN. 459 
 
 be, they serve as the foundation of a considerable advance in the 
 knowledge of this organ. In place of the annoying- discrepancy 
 between Sachs' results with Gymnotus and mine with Malapterurus, 
 we see that in all the three electrical fish the secondary electro- 
 motive phenomena of the organ are the same. It has been possible 
 to follow these phenomena more in detail in the Torpedo than in 
 Malapterurus, and still more completely in Gymnotus. The mere 
 fact that the Torpedo organ, instead of being scarcely available for 
 these experiments, turns out to be well adapted to them, is of great 
 value, and in connection with the prospect of having at our disposal 
 in this laboratory living Torpedos, gives reason to hope that what 
 was only roughly sketched out in haste may be more completely 
 investigated. It might perhaps be advisable to postpone till then 
 the discussion of what has been already gained. It cannot do 
 harm, however, to set forth the actual condition of things, and to 
 indicate the chief points of view which have next to be taken. 
 The relatively negative polarisation, equally strong in the two 
 directions, which simply increases up to a certain limit with the 
 density and duration of the cuiTent, we will for the present con- 
 sider as ordinary internal polarisation, although as in muscle and 
 nerve its dependence upon vitality and its destruction by a boiling 
 temperature raises some doubt. 
 
 Our newer experience leads us to regard it as much more difficult 
 to form an idea of the nature of the absolutely positive polarisation 
 which is even provisionally satisfactory. There can be no doubt that, 
 in opposition to relatively negative polarisation, it appears as if it 
 were the more physiological process of the two. Its dependence on 
 vitality is greater, it is the first to disappear when the organ dies. 
 There are, however, two different ways of considering it. The first 
 consists in looking upon it along with the organ current as the after- 
 effect of a shock liberated by electrical excitation ; the second in 
 considering it as the consequence of a pile-like arrangement of the 
 electromotive molecules determined immediately by the homo- 
 dromous current. 
 
 At first sight there appear to be strong grounds for accepting the 
 former explanation. As we explain the shock itself by a pile-like 
 arrangement of electromotive molecules, it may be asked how this 
 arrangement may be distinguished from that indirectly generated 
 through the homodromous current (corresponding to absolutely 
 positive polarisation) and why the latter should not always assume 
 the character of a shock. The second interpretation, according to 
 
460 LIVING TORPEDOS IN BERLIN. 
 
 which absokitely positive polarisation may be indirectly affected by 
 the homodromous current but not by the heterodromous, attributes 
 a new and obscure property to the org-an. Indeed the first inter- 
 pretation has something" of a similar character in so far as it is 
 assumed that only the homodromous current liberates the shock. 
 But we can account for the same series of phenomena which Fig-. 1 7 
 was intended to illustrate, if we admit that a shock may be g-ene- 
 rated by the heterodromous current along with relatively negative 
 and absolutely positive polarisation. It must indeed remain much 
 weaker than that through the homodromous current, but for this 
 the law of contraction offers an analogy. 
 
 So much doubt, however, is attached to the first explanation that 
 I cannot unreservedly agree to it. It is to be remarked besides 
 that it rests upon a supposition which we made on account of its 
 simplicity, but which is still unproved, namely, that the hetero- 
 dromous current is incapable of g-enerating- relatively positive 
 polarisation. All that is certain is that such polarisation has not 
 yet been seen, but it may have been constantly masked by rela- 
 tively negative polarisation, because weaker than the homodromous 
 absolutely and relatively positive. This is indeed the conception 
 which lies at the base of our construction in Fig. 17. 
 
 Thus it is by no means so certain that there may not be two 
 conditions, both connected with the pile-like arrangement of the 
 molecules, which are the same in their external effect, but different 
 as regards what happens in the interior of the electrical plate, one 
 corresponding to the shock, the other to absolutely positive 
 homodromous polarisation. The organ-current no less than the 
 after-effect of the shock, according to the first assumption, demon- 
 strates that every absolutely positive effect is not a shock. 
 
 It sometimes comes about, however, in series of experiments with 
 fresh preparations that at first an action follows so exceptionally 
 strong that the scale shoots out of sight, and in this the after- 
 effect of a shock, if not the last portion of one, may undoubtedly be 
 recognised. This phenomenon presents quite a different aspect from 
 the ordinary absolutely positive polarisation which is experienced 
 when observations are frequently repeated on the same preparation, 
 inasmuch as it does not show a persistence proportional to its 
 original strength. A similar succession of phenomena are observed 
 in the after-effects of shocks liberated by mediate excitation of the 
 preparation such as cutting off protruding ends of nerve trunks. 
 The way in which the absolutely positive deflections gradually and 
 
LIVING TORPEDOS IN BERLIN. 461 
 
 slowly decline in strength with a regularity which perfectly cor- 
 responds to the nature of the shock and the functional activity 
 of the preparation, and finally fade away, affords no countenance 
 to the impression that we have here to do with a process of dis- 
 charge allied to muscle twitch. 
 
 This is the place to mention a point not yet expressly dealt 
 with, and which, although of great importance, has not been by 
 any means sufficiently explained in strict relation to facts, namely, 
 the dependence of absolutely positive polarisation on current den- 
 sity. Although I had long believed that the threshold of current 
 density for the generation of this polarisation must be high, I only 
 discovered too late with the preparations from the last fish sacri- 
 ficed, that under certain circumstances even the currents from one 
 Grove generated absolutely positive polarisation, with the pecu- 
 liarity, however, that this effect speedily came to an end, a fact 
 which had probably deceived me before. If now the number of 
 elements was suitably increased, the ordinary effects reappeared 
 (Series i). The variable phenomena which pressed upon me were 
 so numerous that I was unable to come to a certain conclusion. A 
 very curious movement of the mirror recurred from time to time with- 
 out my being able to penetrate into its meaning, still less to produce 
 it at will, namely, that the image of the scale instead of retreating 
 from the maximum deflection as usual was thrown back with a jerk. 
 
 A method of experiment suggested itself to me too late, by the 
 aid of which, when I have again Torpedos at my disposal, a 
 successful discrimination between the two conceptions of absolutely 
 positive polarisation which have been contrasted with each other, 
 ought to be possible. The experiment consists in tetanising the 
 organ preparation by alternate cui'rents instead of subjecting it to a 
 single current impulse. I began at once to set this going by 
 closing the secondary circuit ^ of the inductorium with the 
 ' Wippe ' instead of the battery circuit. If the absolutely positive 
 polarisation is nothing more than the after-effect of the shock, it 
 ought to appear with the greatest possible strength after tetanising 
 for a short time. This was, however, not the case. Even when 
 the induction was extremely strong, with the ordinary arrange- 
 ment of the inductorium only weak polarisation followed, the direc- 
 tion of which was such as would have resulted from exclusively 
 opening shocks ; that is to say, the polarisation was absolutely and 
 relatively positive when the opening shocks were homodi'omous, 
 
 ^ See p. 489, note. 
 
462 LIVING TORPEDOS IN BERLIN. 
 
 and relatively neg-ative but absolutely positive wben they were 
 heterodromous. After long'-continued tetanus of the gymnotus 
 organ Sachs found that the tetanus current was weakened rather 
 than streng-thened, as is the case after a sing-le shock (see above, 
 p. 445). But this does not explain the result of my experiment. 
 Unfortunately up to the present I have only been able to make 
 it a few times, with an organ in impaired state of functional 
 activity, so that I cannot altogether trust its result. 
 
 There is still one method of experiment at hand which under 
 certain conditions mig-ht lead to the attainment of the object, that 
 is to observe whether with a longer closing* time the homodromous 
 current maintains its ascendancy. If this is the case, positive 
 polarisation cannot be identified wuth the shock, for the latter, 
 when a constant cuiTent g-oes for a long time through the organ, 
 can only add itself to the battery current at the moment of closing. 
 I have in fact seen this ascendancy with a closing time of i'\ ^'\ 
 or even 20''' (Series 8, 15, 16, 25); but experiments of this kind 
 will not be conclusive until the hypothesis of an irreciprocal re- 
 sistance is completely set aside. 
 
 It is obvious that nothing remains but to set oneself patiently 
 to further investigations, the road to which is, however, clearly 
 indicated. The next consignment of Torpedos will bring us a step 
 nearer to the determination of the present question. 
 
 14. On the electromotive Actions of the Electrical Nerves 
 of the Torpedo. 
 
 The only researches as to the electromotive actions of electrical 
 nerves which had hitherto appeared were those which I had made 
 on the Malapterurus. The nerve showed no current between the 
 longitudinal and transverse section when at rest, and no negative 
 variation in tetanus ; but with two Groves it gave weak but 
 undoubted indications of electrotonus currents^. Its functional 
 activity was already considerably diminished ; it would not, how- 
 ever, be surprising if even the perfectly fresh nerve gave neither 
 rest current nor negative deflection, for the transverse section of 
 the single fibre which runs in the axis of the nerve, bears to the 
 total transverse section a relation varying from i : 90 to i : 104^. 
 In consequence of this extraordinary peculiarity in the construction 
 of the electrical nerve of Malapterurus, the result is just as if no 
 
 * Gefammelte Abliandlungen, ii. p. 645. 
 
 * Ibid., ii. p. 645. N.B. A misprint here puts 8950 for 89. 50. 
 
LIVING TORPEDOS IN BERLIN. 463 
 
 electromotive observation had been made with respect to electrical 
 nerves ; and unfortunately Sachs let the opportunity pass by of 
 filling' up the g-ap by his observations on Gymnotus^. 
 
 Even the Italian electrophysiologists did not make use of the 
 advantages offered them by Nature, and so it was, strange to say, 
 reserved for me move than forty years after the discovery of the 
 nerve-current, in the Physiological Institute of this northern 
 metropolis, to bring Torpedo nerve for the first time into the 
 galvanometer circuit. This occurred on the first occasion that I 
 had the opportunity of killing a Torpedo, viz. on the 13th of June 
 of last year. As it was uncertain how long the nerves would 
 remain alive in summer temperature, I begged Prof. Christian! to 
 undertake this investigation while I occupied myself with experiments 
 on polarisation in the organ. Professor Fritsch was so kind as to 
 prepare for experiment the nerves with which he was so familiar. 
 
 The eight electrical nerves of a large torpedo, four on each side, 
 form, as I had anticipated, an excellent object of investigation^. 
 They are easily prepared without branches, as long as from three to 
 4 cm., and at an average thickness of 2*5 mm. Even if they do 
 not as electrical nerves differ from other nerves, they will afford 
 the means of obtaining solutions hitherto unattainable of problems 
 in the general physics of nerves. 
 
 I. Ciirrent of rest of the Electrical Nerves of tJie Torpedo. 
 
 Shortly before undertaking the measurement of electromotive 
 force in the electrical nerves, a similar measurement with the same 
 apparatus was made of the N. ischiadicus of the frog, and yielded 
 the ordinary values of 0-014 to o-022 Rauolt. The portions of 
 electrical nerves were of the same length as those of the ischiadicus, 
 viz. 12 to 15 mm. From the greater thickness of the former it 
 might be expected that with an equal specific electromotive force 
 there would be a greater galvanometric effect. This was not the 
 case. Christiani obtained the following numbers from the electri- 
 cal nerves : — 
 
 L IL IIL lib. 
 
 0-0054 o-ooS6 0-0054 0-0089 
 
 The number under II b relates to another animal used on the 27th 
 
 of June. The greatest force obtained by Prof. Christiani is more 
 
 than twice as small as that of the nerves of the frog, more than 
 
 ^ Untersuchungen, p- 171. ' Wiener Sitzungaberlchte, 1883, i. p. 387. 
 
464 
 
 LIVING TORPEDOS IN BERLIN. 
 
 three times as small as that of the nerves of birds, or of mammalia 
 with the exception of those of the horse, and over five times 
 as small as that of the nerves of the lobster, as determined by 
 Fredericq^ 
 
 This striking result demanded fm'ther experimental proof, which 
 in the course of the winter I carried out on two Torpedos. In con- 
 sequence, perhaps, of the cold and hung-er from which the creatures 
 had suffered for more than two months, the average value which I 
 obtained was still less than that of Christiani. The mean of his 
 four measurements is 0-007075, of my sixteen only 0-005935. In 
 only one case I obtained a higher value, namely, in a very fresh 
 fish, in which the force reached 0-01123, ^^ amount observed even 
 in weaker nerves of the frog and of the toad, notwithstanding that 
 they are much thinner^. In so far as my measurements simply 
 confirm those of Prof. Christiani, I would not dwell further upon 
 them, had I not come in the com-se of them upon a point relating 
 to their behaviour, which seems sufficiently important to represent 
 in the following table : — 
 
 ELECTROMOTIVE FORCE OF THE ELECTRICAL NERVES OF THE 
 TORPEDO IN RAOULT. 
 
 
 
 Between 
 
 Equator 
 
 Between 
 
 the Two 
 
 
 
 
 and 
 
 Transverse Sections 
 
 
 
 
 Peripheral 
 
 Central 
 
 {P-C). 
 
 
 
 
 Transverse 
 
 Transverse 
 
 Observed 
 
 Calculated 
 
 D-A. 
 
 
 
 Section 
 
 Section 
 
 {D). 
 
 (^)- 
 
 
 
 
 ( + -P, !)• 
 
 ( + C, i). 
 
 
 
 
 
 ' (IV. 
 
 0.00529 
 
 0-00284 
 
 
 
 
 Oh 5 •- 
 
 I ) m- 
 
 0-00658 
 
 0-00400 
 
 — 
 
 — 
 
 — 
 
 
 0-00724 
 
 0-00458 
 
 + 0-00160 
 
 + 0.00265 
 
 — 0-00105 
 
 ^ S M 1 
 
 0-00366 
 
 0-00185 
 
 + 0-00194 
 
 + 0-00181 
 
 + 0-00013 
 
 
 .Hn. 
 
 0-00699 
 
 0-00489 
 
 + 0-001 19 
 
 + 0-00210 
 
 — 0-00091 
 
 e;i 
 
 0-00680 
 
 0-00583 
 
 + 0-00105 
 
 + 0-00097 
 
 + 0-00008 
 
 ro 
 
 ■73 -00 
 
 I. 
 
 0-01123 
 
 0.00970 
 
 + 0.00150 
 
 + 0-00153 
 
 — 0-00003 
 
 ° s 
 
 II. 
 
 0-00757 
 
 000577 
 
 + 0-00187 
 
 + 0-00180 
 
 + 0-00007 
 
 ^ 1 f' 
 
 Tri- ^ 
 
 geminus \ 
 
 ^ branch. J 
 
 
 
 
 
 
 b "^"^ 
 
 000361 
 
 0.00301 
 
 + 0-00195 
 
 + o-ooo6o 
 
 + 0-00135 
 
 "S 
 
 
 
 
 
 
 Ph M 
 
 
 
 
 
 
 
 ^ Archiv fiir Physiologie, 1880, pp. 68, 71. 
 Comp. Wedenskii, Notiz zur Nervenphysiologie der Krote. Archiv fiir Physiol. 
 1883, p. 310. 
 
LIVING TORPEDOS IN BERLIN. 465 
 
 The Roman figures are the numbers in order of the electrical 
 nerves, L and E stand for left and right, the rest of the table 
 speaks for itself. It is seen that in these experiments the peri- 
 pheral transverse section shows greater negativity as compared to 
 the equator. Having been struck by this in the two first experi- 
 ments, I thereupon led off the bit of nerve with the clay shields 
 from the two transverse sections. The last column shows how 
 exactly the force so obtained corresponds with the difference of 
 forces between the equator and the two transverse sections. 
 
 In order to test the correctness of the rule thus discovered, I 
 begged Prof. Fritsch to mark one end of the bit of nerve with a 
 thread of coloured silk, and to hand me the portion without telling 
 me which end was central and which peripheral. Without excep- 
 tion I indicated this correctly on the first glance, guided by the 
 ascending direction of the cuiTent in the nerve from section to 
 section. Even between symmetrical points of the longitudinal 
 surface in the neighbourhood of transverse sections the ascending 
 current could be observed. It would be important to investigate it 
 in uninjured nerves. 
 
 The first occasion on which this fact was seen was when the 
 Torpedo had been killed, as described on p. 438, by punching out 
 the electrical lobes ; and as I did not arrive at the investigation of 
 the nerves till some time later, there was a possibility that the 
 smaller negativity of the central transverse section as compared 
 with the peripheral might be due to the progressive exhaustion 
 of the nerve. In the fourth Torpedo, which yielded the figures 
 given in the three last rows of the Table, I obviated this sus- 
 picion by not removing the brain. I made instead a sagittal 
 incision with the knife mentioned on p. 440 at some distance from 
 the gills, and immediately after, a second cut through the gills 
 themselves close to the skull. Between these two almost simul- 
 taneous incisions lay the portion of nerve, both transverse sections 
 of which were equally fresh. In this case there could be no ques- 
 tion of the difi'erence being due to exhaustion, and yet the electro- 
 motive diflference was as marked as it was regular. Even in a nerve 
 branch detached from the organ, and so distant from the centre 
 that the local exhaustion could not as yet have reached it, I found 
 the ascending cm-rent between the two sections. 
 
 Some years ago (1867) I came on traces of a similar law in the 
 sciatic nerve of the frog, but I arrived at no conclusion ; the 
 greater number of my experiments indicated that here the central 
 
 Hh 
 
466 LIVING TORPEDOS IN BERLIN. 
 
 transverse section was the more negative, the nerve being- per- 
 meated from section to section by a descending- current. It is 
 obvious that this investigation must be resumed as regards various 
 nerves, both centripetal, centrifugal, and mixed. It will then soon 
 be manifest with what we have to do — whether with a peculiarity 
 of electrical nerves, whether with a fundamental law connected with 
 their centrifugal function, or finally, with a less essential difference 
 common to all nerves, dependent only on relations of nutrition. 
 The result seen in the sciatic of the frog-, a nerve of mixed function, 
 seems to show that function is not without influence. The last 
 records in the table, of the trigeminus branch, are only in apparent 
 contradiction with this idea, in so far as this branch being a 
 secretory nerve may be also regarded as centrifugally active. On 
 the other hand this experiment seems to show that the greater 
 negativity of the lower transverse section is not peculiar as such to 
 electrical nerves, as indeed was improbable. 
 
 It remains to be asked what connection there is between this and 
 the small electromotive force of the electrical nerves of the Torpedo. 
 The experiment on the trigeminus branch seems to show that this 
 peculiarity in them is not due to their electrical function. Yet it is 
 possible that it is connected with the electrical immunity of the 
 Torpedo and with the high excitatory limen of its muscular nerves, 
 as proved by BolP. Further investigations on this point are 
 necessary. A few experiments made at the same time on the 
 optic nerves of the carp and pike yielded no result worth recording. 
 
 2. Negative variation of the current of the Electrical Nerves of 
 the Torpedo when in a state of activity. 
 
 The nerve marked II in Christiani's experiments above mentioned 
 was in contact with the electrodes by a longitudinal and cross 
 section, and kept the thread at 60 of the scale. By means of the 
 round compensator the thread was brought back to zero. When 
 the nerve was tetanised by means of the inductorium, Christiani 
 obtained a negative variation to the amount of 5 sc, that is, j^ of 
 the original cun-ent strength. In another case, in which the nerve 
 had already been used for electrotonus experiments, the compensated 
 deflection amounted to I20 and the variation to 90 sc, fully f of 
 the current strength. When the nerve was divided between the 
 
 ^ Untersuchungen, p. 265. 
 
LIVING TORPEDOS IN BERLIN. 467 
 
 two pairs of electrodes and joined together again, no effect was 
 produced by tetanising it. 
 
 I made a similar experiment on the nerve marked R I. in the 
 Table, and obtained with a deflection of 53 sc. (Helmholtz' arrange- 
 ment) negative results as follows : — 
 
 With a distance of coil of 100 mm. the deflection was o, 
 
 n » 25 „ >, 5 SC. I 
 
 that is to say, a variation of xV* When a woollen thread steeped 
 in sea-water was substituted for the nerve no effect was seen. So 
 far everything was regular, but it is strange that both Christiani 
 and I observed, in addition to the regular negative variation, a quick, 
 positive, initial jerk, and I twice saw a similar terminal jerk, so to 
 speak. These positive jerks of the mirror could not be attributed 
 to errors in the arrangement (such as escape of current, action 
 at a distance or the like), because they were observed first by 
 Christiani and then by me, after half a year's interval of time and 
 with altogether different apparatus, and because no effect occurred 
 when the nerve was divided, or when 9, moist woollen thread was 
 substituted for it. Whether this indicates some peculiarity of 
 electrical nerves must be decided by further investigations. However 
 this may be, with regard to this unexpected collateral effect and 
 with regard to the search for the supposed fundamental errors of 
 experiment, an observation was neglected (a failure which will now 
 be repaired), viz. that of the propagation of the variation in two 
 opposite directions in a nerve acting purely centrifugally. Corre- 
 sponding observations were afterwards made by me on the spinal 
 nerve roots of the frog ^. 
 
 3. Mectrotonus currents in the Electrical Nerves of the Torpedo. 
 
 In the investigation of electrotonus currents in electrical nerves, 
 Christiani also met with in-egularities, which for the present can 
 be ranged under no certain law, and which require further researches 
 for their explanation. In the fourth right nerve of the fish killed 
 on the 1 3th of December, the total length of the three tracts being 
 10 mm., I observed with from one to five Groves in the electrotonis- 
 ing current, that the anelectrotonic and kathelectrotonic augmen- 
 tations appeared regularly, the first rising up to a certain point and 
 then gradually diminishing, the latter sinking immediately. When 
 the nerve was divided and brought together again between the 
 
 * LTntersuchungen iiber thierische Elektricitat, ii. pt. i. p. 589. 
 H h 2 
 
468 LIVING TORPEDOS IN BERLIN. 
 
 excited and led-off tracts, weak and sometimes reversed effects 
 remained, attributable to current escapes which, owing to the 
 thickness of the nerve, could not altogether be avoided. The 
 absolute amount of the augmentations observed was so considerable 
 that 10,000 turns and a distance of 50 mm. from the mirror gave 
 suitable deflections : their relative amount could not be determined 
 for want of a point of comparison. Observations of the spread of 
 electrotonus in directions alternately centiipetal and centrifugal 
 were unfortunately neglected. Such observations were made in 
 Christiani's more numerous experiments, but the irregularities 
 above mentioned appeared in both directions, and once more nothing 
 remains but to await patiently the results of new experiments. 
 
 The experiments as to the polarisation of the electrical nerves have 
 not been carried far enough to allow of anything certain being 
 communicated. I have not yet found time to measm-e the electro- 
 motive force of the muscles. It is to be hoped that a fresh consign- 
 ment of Torpedos in the spring will afford opportunity to cany out 
 more thoroughly and conveniently the investigations here mapped 
 out than was possible in the first instance, as well as to undertake 
 experiments of time measurement. 
 
APPENDIX. 
 
 In the following Tables, which contain only a part of the results of 
 my experiments, the Eoman figures indicate the number of Groves, 
 C T stands for the closing time, G for the force of the organ current, 
 L for the length of the columns of the preparation of the organ, P 
 (period) the time between each two immediately following experiments. 
 The mark || means that the experiment was continued on the same 
 preparation under the new conditions given. The deflections (with the 
 exception of those in series 25 where this would have had no meaning) 
 are reduced to the deflection which would have been caused by an 
 equal current strength in galvanometer {S) with a bobbin of 5000 turns 
 at a distance of 30 mm, from the magnet ; all the deflections in galvano- 
 meter (P) are reduced to the deflection which would have been produced 
 with 50 turns at a distance of 20 mm. This explains the large numbers 
 which appear when the currents are of great strength and persistence, 
 and which go far beyond the limits of the scale ^. 500 + as stands as 
 before for a deflection beyond the limits of the scale, which consequently 
 could not be reduced ; a stroke in place of an entry shows that from 
 some cause or another the observation failed ; C, L, and P are some- 
 times not indicated. 
 
 The figures i . . , 2 . . , &c. Torp. correspond to the order of the days 
 on which the animals Avere killed — June 13 and 27, Dec. 13 and 28 
 of last year, and Feb. 2 of this year. In the same order the lengths of 
 the fishes were respectively 29, 36, 29, 25, 26cm. The word 'fresh' 
 indicates that the experiment was made on the same day on which the 
 animal was killed ; * 2 . , day,' on the day after. 
 
 1. 
 
 5. Torp. 2. Day. OC still positive. 
 
 CT o".o764. 
 
 1.-8. I -234 t +80; +10 I -4 t +3 I o to; o; II X t +22 I ~^5t +25 1 -50 1 -12 1 -34 
 P. i 4 I 9-75 i 4-5 I 5 i 4-5 I 8; 8.5 || ^- | 82 i 71 | 78 j 75 j 79 | 65^ 
 
 This series shows that the homodromous current with only one Grove 
 may generate absolutely and relatively positive polarisation, but only for 
 a short time, after which a stronger current acts in the usual manner. 
 * Gesammelte Abhandlungen, p. 356. 
 
470 LIVING TORPEDOS IN BERLIN. 
 
 2. 
 
 As in Series 1. 
 CTo".o'j6^. 
 
 11.— S. t +357 I -"Of +^5 1 -5t +74 1 -o-5 t +62 I o 11 I -85 
 
 i'- I 7-5 i 7 1 8.5 i 8.5I 8 i 7.5 I 7-5 4 8 II i 
 
 This series appears to show that, with short closing time of a weak 
 current, relatively negative polarisation disappears so soon that almost 
 nothing is left but homodromous absolutely and relatively positive 
 polarisation, 
 
 3. 
 
 3. Torp. 2. Day. 0(7 00129. 2-'P. 
 
 CT o".oo63. 
 
 V. — 8. t +500 I —350 \ + 190 j —118 I + 160 1—179 
 
 P. ' +x I 2-25 I 4.2 I 3-6 4-2 2-8 
 
 I 5-5 i I ; I i 
 
 This series shows that under certain circumstances, with even five 
 Groves, positive and negative polarisation are beautifully produced. 
 
 4. 
 
 2. Torp. Fresh. L 23 mm. OC 0-0145. 2-'P. 
 
 CT o".o764. 
 
 Y.—S. t +35 I -126 t - I -122 t 5 - 3 I -70 t +28 
 
 .-+^5 1 
 P- I 29-5 i 23I 28 1 24.5 I 28 i 24.5 I 28 
 
 Here the less favourable result in spite of more favourable conditions 
 can scarcely arise from the closing time having been more than ten times 
 as long, but must have been caused by the unontrollable irregularities 
 of the experiment. The effect in both directions of the third homo- 
 dromous current impulse is to be noted. 
 
 5. 
 
 4. Torp. Fresh. 
 
 CT io".58o. 
 V. — S. I — 200 I —190, &c. 
 P. I 500 + X i 485 to 471 
 
 Only negative polarisation results from a longer closing time. The 
 
 polarising current sinks to a small amount during this period, in spite of 
 
 non-polarisable electrodes. 
 
 6. 
 
 3, Torp. 2. Day. ^z-'P- 
 
 CT o".ooi2. Cri".024. 
 
 X.—S. I -32t +19 I -54! +40 I -4t +39 I -3t +35 I -2|| I -ii3t "79 |--94.&c- 
 P. i 8-2 I 8-2 I 6-6 I 8-2 i 6-6 | 6-6 i 6-6 | 11.5! 6.6|| i 301 ] 318! 318 
 
 In this series, in which the closing time was the shortest possible, it 
 
 is seen that, as in series 2, negative polarisation subsides, whereas it 
 
 alone keeps the field when the closing time exceeds a certain limit. It 
 
 is striking, and reminds one of a twitch experiment to observe the 
 
 increase of the effects during the first four alternations of the current. 
 
APPENDIX. 
 
 471 
 
 2. Torp. Fresh. L 25 mm. OC 0-0079. 
 
 CTo"-o>]6^. 
 
 -42 t +44 I -30 
 
 With a closing time ten times longer than in the preceding series but 
 still short, negative polarisation already shows itself more strongly. 
 
 X.-^. t +38 I -140 1 ^55 I -58! +45 I 
 
 P. I 68.91 42-7 I 68.9! 58.8 I 62.4; 67-3 I 62.4; 59 
 
 X.— 5. 
 P. 
 
 8. 
 
 3. Torp. Fresh. OC 0-0133. Z-'P- 
 
 CT2i"--jS. 
 I —260 I —398 t —270 
 
 -424 
 564 to 515 
 
 1 564 to 515 1 589 to 564 I 548 to 484 I 573 to 548 
 This series shows pure positive polarisation with a longer closing 
 time, together with an appreciable sinking of the polarising current. 
 
 XX 
 
 3. Torp. 2. Day. 000-0044. i^—'F. 
 
 CT o". 002,2. 
 
 t +31 I -55t +57 I -5-5; 6| +58 I -3-5t44 I 
 . I 12 i 6 j 14 i 6-5; 6-5 I 12 i 6-5 I 12 i 
 
 The same result as with ten Groves and the same closing time 
 
 -3t +50 I -3-5»&c. 
 6 I 8.5 i 6.5 
 
 XX. 
 
 10. 
 
 2. Torp. Fresh. i 29mm. 2— 'P. 
 
 CT o"-o764. 
 
 -S. t -f28o I -45 t -f 50 1 -45 t +44 I -34 I +34 I -57 t +30 I -< 
 P. I 274 I 91 I 266 i 92 I 237 i 104 I 237 i 108 I 232 I II 
 
 10 
 
 -52 
 
 ^. t +26 I -42I +23 1 -37t -9 I -37 t -10 I -341 -M I 
 P. I 232 j 104 I 224 i 125 I 208 i 125 I 195 I 125 I 195 i 
 
 j — 20 
 P. j 191 i 129 j 191 i 141 1 191 i 141 i 183 i 145 j 170 \ 141' 
 
 20 
 
 -33 
 125 
 
 30 
 
 S. t -16 I -3ot -18-5 I -28 \ -20 I -31 I -21 I -30t -23 I -29 
 
 This series, in which thirty alternations of current at intervals of 
 two mm. each, lasted for an hour, expresses very fully the law graphically 
 represented in Fig. 17. After the 15th alternation the phenomenon of 
 polarisation showed iteelf in the manner described by Sachs in the 
 Gymnotus (see Sect. 10). 
 
 11. 
 
 XX.- 
 
 • t +520 1 
 
 2. Torp. 2. Day. 
 
 
 OC 0-0045. 
 
 2-'P 
 
 
 
 
 CT o".o764. 
 
 CT 2i"-78. 
 
 
 -1401 +65 1 -95 t +31 
 72 1 154 i 77 1 147 
 
 
 -1509 1 
 
 929 
 to 
 
 \ -355 
 
 ii6i 
 
 to 
 
 -1161 
 
 1161 
 
 to 
 
 -606 
 
 1393 
 to 
 
 -9541 
 
 1316 
 
 to 
 
 \ -632 
 
 1548 
 
 to 
 
 
 
 . 864 
 
 1071 , 
 
 • 1079 
 
 1213 
 
 1218 
 
 1342 
 
472 LIVING TORPEDOS IN BERLIN. 
 
 The exijeriments with a short closing time were merely a repetition 
 of the foregoing series, and serve to show that the preparation was still 
 in a condition of activity. The deflections of the galvanometer (P) with 
 the longer closing time correspond well with those obtained in series 8, 
 in which only lo Groves wei-e used and the closing time was the same. 
 The increase of the deflections as the experiment proceeded depended 
 on the heating of the preparation. The deflections due to negative 
 polarisation are remarkably large in the last half of this series. 
 
 12. 
 
 I. Torp. 2. Day. 06' 0.0059. 2h—'P. 
 
 CT o"-o62g. 
 
 XXX.— 8. I -58 t +99 I -50 t +50 I -40 t +28 1 -38 t +12 I -34 
 F. I 239 I 260 I 256 I 267 i 260 I 260 I 260 1 256 I 253 
 
 Essentially the same results are here repeated. 
 
 13. 
 
 2. Torp. Fresh. i 2omni. OC 0-0090. 2— 'P. 
 
 Cro"-o764. 
 
 XXX.— -S. 1 -128 t +80 I -i.Ht +130 I -73 T +49 
 
 -S. I -128 t +80 I -i.Ht +130 I -73 T +49 
 P. I 134 I 284 i 146 I 262 4. 151 I 268 
 
 M. 
 2. Torp. 2. Day. iiymm. OC 0-0053. 2— 'P. 
 
 XXX.— S. I -5oo-a;| +1-0 | -325! +5^ | -"51 -^o I -150 t -37 
 
 P. i 370 I 790 i 376 i 766 ; 370 I 684 i 380 I 654 
 
 15. 
 Torp. Fresh. L 14111111. 
 
 CT i".024. CT =^".202. 
 
 -500— a; 
 
 5475 
 
 XXX. — S. I +150* I —500— a; f —145 I —500— a; f — igoll | —500 — a; I 
 P. I 1712 \ 1012 I 1326 I 1012 I 1064 II I — I 
 
 The remarkable phenomenon which presented itself here, viz. the very 
 slow development of the absolutely and relatively positive deflection (*), 
 was obviously caused by the considerable diminution of negative polar- 
 isation, which however did not in this case occur with sufficient rapidity 
 to give occasion for a deflection in successively opposite directions. 
 
 16. 
 
 I. Torp. Fresh. Zi2mm. 3^ — 'P. 
 
 CT 5".202. 
 
 XXX.— ;8'. t -2288 I -3840! -2247 I -2105 
 P. I 2886 I 3105 I 3461 I 2434 
 
APPENDIX. 
 
 478 
 
 17. 
 
 1, Torp. 3- Day. i 15mm. OC? 2-'P. 
 
 CT o".o629. 
 
 XXX.— S'. t -87 I -84 t -90 I -89 
 
 P- I 513 i 513 I 544 i 560 
 
 The exhausted organ no longer gives positive polarisation. 
 
 18. 
 
 I. Torp. Fresh, but scalded. 2^—' P. 
 
 CT z". 202. 
 
 XXK.—S. t -30 I -21 t -21 I -13 
 ■P- I 877319830 I 11204! — 
 
 Negative polarisation is practically abolished by boiling temperature ; 
 on the other hand the resistance of the preparation is lessened, but not 
 in the proportion which might be expected from a comparison of series 
 16 (Sect. 12). 
 
 19. 
 
 I. Torp. Fresh, Z, 27mm. OC 0-0083. 2^ — 'F. 
 
 CT o".o629. 
 
 -L.—S. t +163 -1841 +45 I _i63t r-2o;f-4i I -1351 r-49 I -135 
 
 I I I |\ + l6\+Il| |\+2| 
 
 ■P- I 543 i 390 I 465 i 407 I 457; 457 i 394 I 39° i 407 
 
 This series is interesting from the appearance of effects in both 
 
 directions. 
 
 20. 
 
 I. Torp. Fresh. i 27mm. OC 0-0105. 2^-~'P. 
 
 CT o".o62g. 
 
 L.— ^. t +756 I -470 t +387 I -384? +153 I -322 t +88 I -337 t +20 
 -P- I 648 i 332 I 527 i 382 I 631 I 390 j 519 I 448 I 423 
 
 The points of interest here are, first the powerful positive initial effect, 
 and then the relation between homodromous and heterodromous current 
 strengths in the primary circuit. 
 
 21. 
 
 With L and 2i''.8 closing time the negative polarisation shot the scale 
 out of the field in both directions. It was not worth while to record 
 larger numbers than those in series 1 6. 
 
 22. 
 
 5. Torp. 2. Day. OC 0-0220. 
 
 Polarisation by a transversely directed polarising current. 
 Cro".o764. Cri".o24. 
 
 XX.— s. 
 p. 
 
 -85 
 
 + 25 
 665 
 
 -60 
 
 ■4 
 
 677 
 
 -35 
 
 -54 
 5465 
 
 -38 
 5539 
 
 42 
 5560 
 
 P. 652 665 669 665 677 — 5465 5539 5560 
 The great strength of the polarising current arises from the shortness 
 
474 
 
 LIVING TORPEDOS IN BERLIN. 
 
 and large sectional area of the path which it has to follow in the pre- 
 paration when it is placed crosswise. In a transverse direction (except 
 in the incomprehensible effect observed in exp. 4) there is no positive 
 polarisation and only very weak negative polarisation. 
 
 23. 
 
 5. Torp. 2. Day. OC 0-0030. 
 
 Polarisation when the current is led in transversely. 
 
 CTo".o'j64. 
 
 r. h. r. i. 
 
 XX. — ;S'. I +190 I —140 11 Led in | 32 I I7f o I o 
 
 P. I 134 4. 75 II transversely | 116 j 79 | 116 I 113 
 
 r and b signify red and black, the colours of the two halves of the 
 scale, and consequently the direction in which the mirror was deflected. 
 It is seen that under these circumstances there is scarcely any effect. 
 
 24. 
 
 5. Torp. Fresh. 
 
 Proof of the pile-like arrangement of the electromotive force in the polarised 
 
 preparation of the organ. 
 
 B- 
 
 -R 
 
 Sj Ji 60 o 7-0 Vi r-i 
 
 B R represents schematically the preparation as a linear conductor, 
 through which the shock of the organ current and the homodromous 
 current pass from B to R. If positive and negative homodromous and 
 negative heterodromous polarisation depend upon the pile-like arrange- 
 ment of electromotive forces, then in every case polarisation must appear 
 in the same direction between any two given points of the line BR, 
 which would be of the same strength if the distance of the leading-off 
 points is equal; and of greater strength if the distance is greater and 
 the external resistance is sufficient. B R-=L amounted to 22 vam.fh^r^ 
 18 mm., h^r^ 1-5 mm., oh^:=or^ 9 mm., h-^r^ about the same. In order 
 not to confuse the results, the positive organ-current force, although 
 observed constantly in every part, has been left out of account. 
 
 A. Homodromous currents, absolutely and relatively positive polarisation. 
 CT o".o764. CTo".o764. 
 
 XX. 
 
 t 6 »-2 
 -S. I +205 
 P. I 171 
 
 hu 
 
 6j r^ 
 
 hu 
 
 + 20 
 
 + 146 
 
 + 6 
 
 163 
 
 159 
 
 159 
 
 New I Ji r, 
 prepa- + 500 + x 
 213 
 
 ration. 
 
 60 ''o 
 
 &. 62 
 
 + 148 
 
 + 170 
 
 iSo 
 
 167 
 
 + 31; +12 
 159; 163 
 
 6, 6j 
 
 ? 
 
 147 
 
 B. Homodromous current, absolutely and relatively negative polarisation. 
 CT i".o24. 
 
 6, rj 
 
 -145 
 
 707 
 
 
 h n 
 
 i, n 
 
 6j rj 
 
 5j 
 
 r, 
 
 ioU 
 
 XX.-5. 
 
 -64 
 
 -125 
 
 — 160 
 
 -117 
 
 — 121 
 
 -45 
 
 ■P. \ 
 
 i 703 
 
 703 
 
 715 
 
 707 
 
 711 
 
 715 
 
 5j r, 
 + 20 
 134 
 
APPENDIX. 
 
 475 
 
 C. Heterodromous current, relatively negative, and absolutely positive polarisation. 
 Cro".o764. CT i".o24. 
 
 
 \u 
 
 &2 »'2 
 
 &2 
 
 or^ 
 
 \n 
 
 ^Ja^s 
 
 New 60 r„ 
 
 ^2 »-2 
 
 ^j 
 
 XK.—8. 
 
 — lO 
 
 -105 
 
 -30 
 
 -45 
 
 — 10 
 
 -47 
 
 prepa- — 78 
 
 -215 
 
 -70 
 
 P. 
 
 ■ 71 
 
 75 
 
 75 
 
 79 
 
 79 
 
 75 
 
 ration, | 761 
 
 744 
 
 757 
 
 62 »-2 
 
 -250 
 
 736 
 
 25. 
 
 5. Torp. Fresh. 
 
 The galvanometer circuit (S") forms a collateral closure to the preparation, as in 
 Fig., 8, A and B, and p. 455. 
 Cro".o764. CT\". 02^. 
 
 XK.—S. 
 P. 
 
 ?. I 230 t 150 I 216 t 158 I 193 t 162 I 190 II I 255 I 245 I 255 I 255 
 '. I 141 I 269 I 145 I 239 I 146 I 228 l 146 III 1162 I 1254 i 1179 1 1267 
 
 Distance of bobbin from mirror of galvanometer, 
 21 cm. 36 cm. 
 
 Number of turns 5000. 
 
XIII. 
 
 LIVING TOEPEDOS IN BERLIN. 
 
 By E. du BOIS-EEYMOND. 
 (Second Paper.) 
 
 Sitzungsberichte der Koniglich preussischen Akademie der Wissenschaften. 
 Berlin, July i6, 1885. 
 
CONTENTS. 
 
 PAGE 
 
 1. Introduction 479 
 
 2. On the electromotive behaviour of the skin of electrical Fishes . .481 
 
 3. On polarisation of the organ by alternating Currents .... 489 
 
 4. An attempt is made to compare the Negative Polarisation due to the 
 
 closing shocks of an inductorium with that due to the opening 
 shocks separately . . . ... . . . . .492 
 
 5. Continued discussion of the results of Polarisation of the organ by 
 
 alternating Currents ......... 497 
 
 6. Unsuccessful experiments to produce relatively positive Polarisation by 
 
 Heterodromous Currents ........ 498 
 
 7. The apparent Irreciprocity of- the Conduction of the electrical organ 
 
 increases with the Current Density ...... 503 
 
 8. The apparent Irreciprocity of Conduction in the electrical organ has its 
 
 seat in every transverse slice of the preparation, and increases with 
 
 the length of the piece of columns which is traversed by Currents . 505 
 
 9. The question, whether the appearance of Irreciprocal Conduction in the 
 
 organ depends upon polarisation or Conduction cannot be answered 
 
 by the introduction of additional resistance . . . . .510 
 
 10. Measurements of the resistance of the electrical organ of the Torpedo . 514 
 
 11. Dependence of the Irreciprocal Resistance and of the Resistance of the 
 
 organ in general upon its condition as to vitality . . . .520 
 
 12. The conduction of the electrical organ, when investigated by means of 
 
 constant Currents .......... 521 
 
 13. On the Law and Nature of Irreciprocal Conduction in the electrical 
 
 organ ............ 524 
 
 14. Teleology of Irreciprocal Conduction in the electrical organ . . . 528 
 
 15. Positive Polarisation in its dependence upon the density of the Polar- 
 
 ising Current. One case of Relatively Positive Polarisation by the 
 Heterodromous Current 535 
 
 16. On Conduction in both directions in the electrical Nerves of the Torpedo . 537 
 
 17. On the Secondary Electromotive Phenomena in the electrical Nerves of 
 
 the Torpedo ... 539 
 
XTIP. 
 
 1. Introduction. 
 
 During the past summer (1884) and this winter and spring- 
 (1884-5), throug-h the kind co-operation of the Director of the 
 Berlin Aquarium, Dr. Otto Hermes, I have again had at my 
 disposal several living- torpedos (21 marmoratd) from Trieste. 
 Those in the summer were four females, I, II, III, IV, and 
 their respective lengths were 29-5, o^^, 3I•5^ ?)^'^ cm. Torpedo I 
 arrived on June 9th, in a very sickly condition. As I worked with 
 it during the day its breathing was still regular, but it closed its 
 spiracles only partially, and it was evidently in a sort of opisthotonus, 
 so that its back was very concave in a sagittal direction. The 
 torpedos II, III, IV arrived on June 23rd, and were killed on 
 July 9th, July 1 6th, and August 13th respectively, after remaining 
 for about 2, 3, 7 weeks in the tanks of the Berlin Aquarium. 
 The winter supply which was brought on October 22nd consisted 
 of three females, V, VI, VII, and their lengths were 28-5, 26-5, 
 28cm. respectively; they were killed on the 21st and 30th De- 
 cember, 1884, and on the 6th May, 1885, after a period of about 
 8, 9, 28 weeks. An accident befell a first consignment in spring 
 this year, but on May 8th a female (VIII), 34 cm. long, arrived 
 in good condition, and was experimented with on June 3rd. 
 
 In the summer of 1883 I had also allowed three and five weeks 
 to elapse after the arrival of fish before killing them, and in the 
 winter of 1883-84 a period of ten and fifteen weeks. The reason 
 of these delays is, that in order to make full use of a torpedo it is 
 necessary to devote two whole consecutive days to the work 
 without interruption, and my occupations only permitted this at 
 wide intervals of time. I mention this because the functional 
 activity of the fish naturally suffers very much from a prolonged 
 
 1 See No. XII of this book. 
 
480 LIVING TORrEDOS IN BERLIN. 
 
 captivity, especially during- the summer. In the summer of 1883, 
 when they reached here earlier, they had kept in comparatively 
 good health. The fish of the spring consignment in 1884 re- 
 mained indeed quiet, huried in the gravel — a position which they 
 abandoned as soon as they became seriously ill (p. 419) ; they 
 looked quite well, breathed regularly, and struggled vigorously 
 when lifted up by the tail. But they gave shocks only after 
 repeated violent stimulation, and their electrical force was so much 
 lowered that on July 29th I did not succeed in making the shock 
 of the fourth torpedo perceptible to my audience in the manner 
 described in my former paper. The condition of the organ 
 corresponded obviously to its impaired functional activity, just as is 
 the case with frog preparations which do not react well ^. The organ 
 was soft, deliquescent, drops of liquid exuded from the cut surfaces, 
 and as it had lost its plumpness the organ preparations could not 
 be made so well as before. Its organ current appeared also less 
 strong and showed peculiar irregularities (see below, § 2). The 
 cause of this was undoubtedly that the fish fasted at a com- 
 pai'atively high temperature, but up to this time no mode of feeding 
 them had been discovered (p. 41 9). The keepers fancied that the 
 seventh toi-pedo, which had lived in the Aquarium the whole 
 previous winter, more than half the year, had taken nourishment. 
 They inferred this only from its apparently vigorous condition, but 
 I found the stomach and intestine quite empty. 
 
 Fish kept in winter at only 10-5-1 2°C. certainly sufier less in 
 confinement than summer fish. However, judging from their 
 electrical functional activity, they also are far from the normal 
 condition^. Even when put into warm sea- water but twenty-four 
 hours before the experiments, they gave feeble shocks only, and only 
 on strong excitation. 
 
 The field of our experiments on electrical fishes is much within 
 the limit beyond which the greatest amount of functional activity 
 becomes indispensable, and in spite of the bad condition of several 
 preparations, I succeeded, as I believe, in attaining some not unim- 
 portant results. I purpose to communicate these now, without 
 awaiting further consignments, as their safe arrival depends more 
 or less upon chance. On the other hand, the deficiency of 
 functional activity undoubtedly had for result, that certain other 
 questions could not be completely settled. I will however speak 
 
 ' Untersuchungen fiber thierische Elektricitiit, vol. ii. pt. i. p. l68. 
 ' Comp. i. p. 121 ; Untersuchungen, etc., p. 266. 
 
LIVING TORPEDOS IN BERLIN. 481 
 
 about these questions also, according to the position of the subject 
 at the present moment, because the scantiness of material and the 
 uncertainty of its condition compel me to deviate in this case from 
 the rule which I always strove to observe in treating- of muscles 
 and nerves, viz. to publish investigations, only when they had been 
 completed according to the best of my knowledge and power. 
 
 The preparations for the experiments were the same as in those of 
 the first communication. The animals were killed in all cases in 
 the way described there (p. 438), by punching out the electrical 
 lobe of the brain with a single blow of a hammer. This operation 
 presents no difficulty to any one who studies Savi's drawing^ closely, 
 even though he be not so conversant as Prof. Pritsch with the 
 anatomy of the torpedo. It is advisable to make a slit in the 
 sagittal plane in the skin first of all, as it easily slips laterally from 
 under the borer and takes the rod with it, so that the sagittal plane 
 does not divide the hole in half. The fish never gives a shock after 
 removal of the lobes, even if this has not been accomplished fault- 
 lessly, but it can still wriggle for a long time as well as perform 
 reflex movements. Any part of the organs not used on the day 
 when the fish was killed was kept on ice in hot weather. 
 
 Perhaps it is not useless to remark that the polarisation reverser 
 with its time disks ^ regulating the duration of a polarising current, 
 which I have used for many years as well as in the present experi- 
 ments, is to be found described and figured ^ in the concluding part of 
 my 'Untersuchungen Uber thierische Elektricitat ' recently published. 
 My investigations on polarisation at the junctions of dissimilar 
 electrolytes and in the interior of moist, porous semi-conductors 
 are also communicated there in full ; these form one of the most 
 important foundations for all researches on electromotive tissues, but 
 until now only extracts had appeared in the first volume of the 
 ' Gesammelte Abhandlungen.' 
 
 2. On the electromotive behaviour of the skin of Electrical 
 
 Fishes. 
 
 I have sought in vain, in organ preparations of Malapterurus 
 
 ^ Matteucci, Traits des Ph^nomfenes ^lectro-pliysiologiques des A.nimaux. Paris, 
 1844, Planche I, 
 
 2 Gesammelte Abhandlungen, etc., vol. i. pp. 3, 13, 34 ; vol. ii. p. 718. 
 
 2 Log. eit. vol. ii. Part ii. Berlin, G. Reimer, 1884, pp. 389-395, Plate vi. Figs. 151, 
 152, A, B. The means by which I can insure that a polarising or tetanising current 
 shall last for small fractions of a second only, are not yet given here ; this will be 
 done at the first opportunity. 
 
 I i 
 
482 LIVING TORPEDOS IN BERLIN. 
 
 electricus for an electromotive action during* rest, that is for an 
 oro'an-current. This failure must seem strang-e now that such 
 actions have been demonstrated^ in the Gymnotus and Torpedo, 
 and as it is alleg-ed even in the imperfect electrical organ of the 
 common skate. It can be explained, neither by the lowered 
 functional activity of the preparations, for they were in much 
 better condition than many preparations of the Torpedo which now 
 gave me the organ-current, nor by the imperfection of the experi- 
 mental method, for I made use of the nerve multiplier ; and the 
 resistance of the old conducting vessels and of the salt pads covered 
 with thin films of white of egg was certainly less than that of the 
 present unpolarisable apparatus. The polarisation of the platinum 
 in the old conducting vessels is all that remains to explain that a 
 constant weak action might escape me ; it did not, however, prevent 
 my perceiving a slight positivity of the external surface of the skin 
 in comparison with all other natural as well as artificial boundaries 
 of the organ preparation, an action with which, as will now be 
 shown, the organ-current if present would have been of the same 
 order. 
 
 In the Gymnotus, Sachs found points of the skin feebly positive 
 (0-0050 Daniell = 0-0053 Raoult ; see p. 440) in comparison with 
 points of natural longitudinal section not covered with skin and 
 situated in the same transverse plane, as also in comparison with 
 such points of artificial longitudinal and transverse section. In com- 
 parison with the surface of that part of the organ turned towards the 
 head, the skin is feebly positive, electrically the same, or very feebly 
 negative ; and in comparison with the surface towards the tail, it is 
 more strongly positive, as this gives the algebraical sum of the 
 organ current of the length situated between the leading-off points, 
 and of the skin current. In the first adjustment, for example, the 
 force amounted to a mean value of 0-0040, in the second to 0-0195 ; 
 whence by calculation the force of the skin current is found to be 
 0-0155/2 = 0-00775, and this approximates sufiieiently to the direct 
 measurement given above (0-005) ^. 
 
 The question natm-ally forced itself upon me, how the skin 
 behaves electromotively in regard to the interior of the organ in 
 the Torpedo. The case is not so simple here as in Malapterurus 
 which is not complicated by any perceptible organ-current, nor as 
 in Gymnotus, where the organ is bounded laterally by skin, 
 
 ' Untersuchungen, etc., p. 169 f. 
 * Untersuchungen, etc., p. 172. 
 
LIVING TORPEDOS IN BERLIN. 
 
 483 
 
 and where consequently the organ-current can be excluded by 
 leading- off from skin and organ from points in a line which is 
 isoelectrical as regards the shock and the organ current. In the 
 torpedo the polar surfaces of the organ preparations are covered 
 with sldn, and if on the one hand, the leading-off is from one of 
 these surfaces, on the other hand, from the lateral surface or from a 
 transverse section of the preparation, the organ-current of the 
 length of the columns included between the two leading-off points 
 is always present in the circuit, and any possible electromotive 
 actions of the skin can only be determined by elimination : this 
 can, however, be accomplished in the following manner. 
 
 In Fig. 19 an organ preparation is represented : at Fit is bounded 
 by the ventral skin, and at 
 B by the dorsal. The curves 
 1-6 represent different posi- 
 tions of the galvanometer 
 circuit, the terminals of 
 which are formed by the clay 
 points of the unpolarisable 
 conducting tubes, and in 
 which equilibrium is main- 
 tained by the circular com- 
 pensator for the purpose of 
 measuring the electromotive forces. The point m divides the length of 
 the preparation as nearly as possible in half; the points (l',d-^^ and v\vy^ 
 lie as close as possible to the dorsal and ventral skin respectively^. 
 
 In all the cases, organ-current force is present in the circuit, and 
 besides this, in the cases 2, 4, 6, any electromotive action of the 
 skin that there may be. If the dorsal points are shifted from (l\ d^ 
 to D, or the ventral points from v\ v^ to V, then not only is the 
 skin taken into the circuit, but also the tract of the columns between 
 the points is slightly increased. If the skin is electromotively 
 inactive, then the removal of d\ d^ to J) as well as of v\ v^ to K 
 must result in a slight strengthening of the positive organ-current 
 force, and this must be the same in both cases if the preparation is 
 of regular columnar arrangement, and if the additional-column 
 tracts are the same length. On the other hand, a deviation from this 
 
 * The adjustments mV, mD had been investigated by Eckhard, who did not 
 find electromotive action here any more than between V and D in the organ at rest. 
 (Beitriige zur Anatomic und Physiologie, vol. i, Giessen, 1858, 4, pp. l6lj 162. Comp. 
 i. pp. 204, 210.) 
 
 lis- 
 
484 LIVING TORPEDOS IN BERLIN. 
 
 behaviour under these circumstances is to be attributed to electro- 
 motive action of the skin. 
 
 Such a deviation in fact shows itself with great uniformity, so 
 that on passing- from i to 2 (from md' to mli) on the back, a con- 
 siderable increase of the positive force, and on passing- from 3 to 4 
 (from mv' to m F) on the belly, a less considerable weakening of the 
 positive force is observed. The latter result cannot proceed from 
 increased length of the column tract. On the other hand, both 
 results are fully explained, by assuming that the skin is feebly posi- 
 tive in comparison with the interior. At first sight, one might be 
 inclined to attribute the fact that the enfeeblement on the belly is 
 less in amount than the increase in strength on the back, to differ- 
 ent electromotive character of the pigmented skin of the back and 
 of the pigmentless skin of the belly, which is however not demon- 
 strable with certainty in the case of the frog^. Another ground of 
 explanation, less remote, and which makes no fresh supposition 
 necessary, is that the increase of strength of the organ-current, due 
 to the greater length of the column tract between the points, adds 
 itself to the skin action on the back and deducts itself from that on 
 the belly. 
 
 The first experiments were made with the last summer fish (IV). 
 In a successful experiment I obtained — 
 
 in the curve i (mcT) + 0.0057 Eaoult. 
 
 „ „ 2 (mD) + 0.0103 
 
 r + 0-0049 
 
 ., „ 3(W) ) + °-°°^3 
 
 (Mean + 0-0056 
 
 [ + 0-0034 
 
 » » 4(«»F) ) -^ °-°°^9 
 
 (Mean + 0-0032 
 
 Thus the increase in strength on the back at the removal of md^ 
 to mD amounted to 0-0046, the diminution in strength on the belly 
 at the removal of mv' to m 7, in accordance with what has been said 
 above, amounted to only about the half, 0-0024. Let x be the 
 difference of potential between the skin and the interior of the 
 organ, 8 the force of the column tract between the skin and the 
 points lying next to it, then we have on the back. 
 
 0-0057 +8 + a; = 0-0x03, 
 
 * Untersuchungen; etc., vol. ii. part ii. p. 14. 
 
on the belly 
 which gives 
 
 LIVING TOEPEDOS IN BEKLIN. 485 
 
 0-0056 + 8 — a? = 0-00315, 
 8 = o-ooii, X = 0-0036. 
 
 When I wished to repeat the experiments last winter, on fish V, 
 killed on December 21st, I came across a new kind of phenomenon 
 of the organ current, viz. in several organ preparations I found it 
 positive in the dorsal half agreeably to law, but on the other hand, 
 contrary to law, negative in the ventral half of the columns. As 
 the fish had been removed rapidl}^ from the cold tank of the Aqua- 
 rium to warmer water, a disturbance owing to hydrothermal currents 
 at first occurred to me. But I was compelled to give up this idea, 
 because fish VI, which I killed on December 30th, after I had kept 
 it for more than twenty-four hours in water at 17-5", exhibited the 
 same phenomena, only with the difference, that this time the dorsal 
 half of the columns acted negatively. I now proceeded to make 
 measurements again, and obtained — • 
 
 ,. — 0-0021 Eaoult. 
 
 — 0-00I2 
 
 + 0-0067 
 
 + 0-0057 
 
 , + 0-0029 
 
 + 0-0029 
 
 in the curve 
 
 I {md') 
 
 
 2 (mD) 
 
 
 3 (mv) 
 
 
 4 imV) 
 
 
 5 (^V'l) 
 
 
 6 (VD) 
 
 The first thing to be remarked is, that as in the first series and 
 in this, 7}iF< mv\ but in this series it happened that ml)< mcV, 
 for now the skin current force is deducted from the organ current 
 force on the back also ; further, that the force in curves 5 and 
 6 by reason of the opposite action of the two halves of the columns, 
 was less than that of the stronger half, even if v^ d^ ought to have 
 been somewhat < , and FI) somewhat > 0-0045, instead of both 
 amounting to only 0-0029 without perceptible difference between 
 them. 
 
 In order to determine cr from our numbers, we must not proceed 
 as above, where we could assume that 8, which represents the 
 increase of organ current consequent on the lengthening of the 
 column tracts in the transition from mcV to mD, and mv' to m V, on 
 account of the sensibly equal activity of both halves of the columns, 
 is the same at the belly and the back. But if we assume that the 
 one entire half of the columns acted positively, the other entire half 
 negatively, then we can proceed in the following way. 
 
 Let 8^ denote the increase at the back, 8„ that at the belly, then 
 
486 LIVING TORPEDOS IN BERLIN. 
 
 in ordei' to determine the values of x, 8^, 8^,, we have the three 
 equations, 
 
 — 0-002I — 0^ + ^ = — 0-00I2, 
 
 + 0-0067 +h„ — x = + 0-0057, 
 
 Oi. — — Od^T* 
 
 whence we have 
 
 h,i = 0-00024, 8;. = 0-00131, 
 
 and according' as x is calculated from the first or the second of the 
 equations containing- it, and which do not agree exactly with each 
 other on account of errors of observation, 
 
 X = 0-000924 or = 0-000869, mean = 0-000896. 
 
 It would have been scarcely practicable to give more exact values 
 of the ratio of the lengths of the positively and negatively acting' 
 segments of the columns ; at any rate, the attainment of better 
 numbers, if it is worth the trouble, must be reserved for the future. 
 In the mean time, it seems as though the electromotive force of the 
 skin is considerably less in the winter fish (about fom* times) than 
 in the summer fish ; this is very possible. 
 
 Thus the skin of the torpedo behaves electromotively like that of 
 the gymnotus, in regard to the direction as well as to the magni- 
 tude of the action, and certainly like that of malapterurus in regard 
 to direction ; the force is probably here much the same in amount 
 as in the two other fishes (see above, p. 482). Tkree modes of 
 explaining this action offer themselves. 
 
 The first question to occur is, whether it does not denote a differ- 
 ence of potential between the longitudinal and transverse section of 
 the columns, for perhaps the skin may play the part of an inactive 
 conducting covering over the natural transverse section of the 
 columns, like the tendinous expansions over natural transverse 
 section of muscle. This idea is negatived, because no action is 
 perceived between the lateral sui-faces of the columns and an 
 artificial transverse section, and there is no perceptible change in 
 the organ curi'ent, on moving a clay point which is near a transverse 
 section, to the transverse section itself. 
 
 The second hypothesis is, that in the case of the gymnotus, the 
 skin forms with the material for leading-off" and the organ, a liquid 
 battery in the following way^, — 
 
 Physiological Clay 1 Organ + Organ [ Skin + Skin wetted with river water | Clay; 
 
 * Gesammelte Abhandlungen, etc., vol. ii. p. 265. 
 
LIVING TORPEDOS IN BERLIN. 487 
 
 in the case of the torpedo, — 
 
 Physiological Clay | Organ + Organ | Skin + Skin wetted with sea water | Clay, 
 
 and in the case of the malapterurus, — 
 
 White of egg | Organ + Organ | Skin + Skin wetted with river water | White of egg. 
 
 It may seem surprising" that the interposition of sea water, and 
 still more of white of eg-g" with alkaline reaction, instead of the clay 
 kneaded up with physiological solution of chloride of sodium, does 
 not cause a more considerable difference of action. In order to 
 judge certainly as to this, it would be necessary to study first the 
 chemical reaction of the torpedo's skin, and also to make experi- 
 ments with conducting" liquids of a more decided electro-chemical 
 character. 
 
 The third hypothesis is, that the skin possesses its own inde- 
 pendent electromotive action, like the skin of the naked amphibia, 
 according to my discovery *. This assumption has also much against 
 it. In the first place, the direction of the force would be the 
 opposite, inasmuch as the action of the skin is from without inwards 
 in the amphibia. Secondly, it would be more than eleven times less 
 than in the case just mentioned, where the skin current force is 
 found to be comparable to that of muscle and nerve ^. Thirdly, up 
 to the present time, skin curi'ents have been observed in none of 
 the fishes in vestig'ated. including those without scales, like the eeP. 
 Fourthly, the electromotive activity of the skin seemed to me to 
 outlast that of the organ in the torpedo, and this would be more in 
 accordance with a liquid battery. Fifthly, I did not succeed in 
 observing an electromotive action of determined direction and 
 amount between the external and internal surfaces of a detached 
 piece of skin. I employed the same method in these experiments 
 as Rosenthal for the skin of the frog ^. I made openings with the 
 punch in two mica plates laid one over the other, placed the skin 
 between them, and pressed the mica plates between the clay shields 
 of the leading-in pads, so that on the one side only the ventral 
 sm'face of the skin, and on the other side only its dorsal surface 
 
 * Monatsberichte der Akademie, 1851, p. 380. Untersuchungen, vol. ii. part ii. 
 p. 9f. 
 
 ^ Gesammelte Abhandlungen, etc., vol. ii. p. 261. 
 ^ Untersuchungen, loc. cit. pp. 16, 17. 
 
 * Archiv fiir Anatomic, Physiologic, etc., 1865, p. 309. 
 
488 LIVING TORrEDOS IN BERLIN. 
 
 touched the clay. It should be mentioned that in these experiments 
 the skin was no longer fresh, and might have suffered by dryness 
 as well as from frequent handling* of the fish. The result how- 
 ever, in conjunction with the other reasons, tells strongly against 
 independent electromotive action of the skin. 
 
 Experiments like those by which I determined the electromotive 
 action of the skin of the frog ^ would be necessary, in order to decide 
 between the second and third hypotheses. Unfortunately, there are 
 here great difficulties. The detachment of sufficiently large pieces 
 of uninjured skin, which is facilitated in the frog by the lymph 
 spaces lying underneath the skin, can scarcely be carried out in the 
 torpedo. I tried therefore to observe currents in the two first 
 winter fishes (V and VI), by placing on them, not simultaneously, 
 pads soaked in solution of sodium chloride. In the frog and other 
 naked amphibia, a strong current is always obtained in this way 
 from the more recent to the earlier points of contact, because by the 
 caustic action on the skin, the electromotive force, the direction of 
 which is from the exterior to the interior, is diminished, or even 
 annihilated. I made the experiments on the tail of the Torpedo, 
 having amputated it high up along with a piece of the vertebral 
 column and a slice of the body. Naturally, cm-rents were not 
 wanting, but they had sometimes one, sometimes the other direction, 
 and more often, they did not disappear on a prolonged application 
 of the pads, so that no importance can be attached to them ; all the 
 less because there was a danger of illusive effects, partly arising 
 from the interior of the preparation, partly from the neighbouring 
 cut surfaces. A comparison of the strengths of the currents with 
 those in the frog would have had no meaning ; measurements of force 
 are precluded by the transitoriness of the actions. 
 
 The whole matter is not very important, and has hardly any 
 relation to the electromotive property of the organ of the Torpedo 
 and of the Gymnotus. It is much rather to be supposed that in 
 other fishes also, the skin is feebly positive to internal parts 
 which are without electromotive action. In them, however, parts 
 are wanting, which, like the organ preparations, consist only of 
 skin and a tissue which is inactive or possesses only feeble electro- 
 motive action, in accordance with known laws. Whether the 
 observed positivity of the skin in the Malapterurus may possibly 
 contribute anything in support of Prof. Fritsch's hypothesis, accord- 
 ing to which, the organ of this fish has its origin in the mucus cells 
 * Untersuchungen, loc. cit. 
 
LIVING TORPEDOS IN BERLIN. 489 
 
 of the skin ^, must remain undecided. In any case, the examination 
 of dissimilarities in the skin forms an essential part of the investi- 
 gation of the organ current, and the possibility of separating these 
 two actions from each other, testifies at least to the superiority of 
 our experimental methods, for until recently the existence of any 
 action was regarded as doubtful and even denied. 
 
 3. On Polarisation of the organ by alternating Currents. 
 
 The most important questions in regard to the electrical organ, 
 are for the moment closely connected with its behaviour when 
 homodromous or heterodromous currents are sent through it, first 
 those which concern the polarisations which they leave behind them, 
 secondly, those which concern the apparently irreciprocal conduction 
 of the organ. Is there any relatively positive polarisation of the 
 organ, or is it nothing but the after-effect of a shock which 
 gradually loses itself in the organ current ? Does the organ con- 
 duct irreciprocally, or does only the semblance of such conduction 
 arise, owing to absolutely positive polarisation adding itself to the 
 homodromous current, whatever may be its cause and significance ? 
 
 In order to decide the first question, I had at an earlier date 
 tetanised organ preparations, supposing that if the absolutely 
 positive polarisation were the after-efi'ect of the shock, it must 
 appear with greater strength after a short tetanus. I made use 
 of the sliding inductorium with the ordinary arrangement, and its 
 secondary coil was pushed quite up to- the primary coil, and the 
 latter filled with wires. The terminals of the secondary coil were 
 connected with the two binding screws of the switch which are 
 connected usually with the terminals of the battery circuit^. The 
 spring of the induetorium vibrated, and as soon as the switch 
 had opened the galvanometer circuit and closed the secondary 
 circuit, each in two places, the induced cra'rents reached the pre- 
 paration. On the falling back of the switch, I expected very 
 strong absolutely positive polarisation in any case, but after tetan- 
 ising during 5'', I obtained only feeble polarisation in a direction as 
 if opening shocks alone were present, viz. absolutely and relatively 
 positive polarisation with homodromous, absolutely positive and 
 relatively negative polarisation with heterodromous opening shocks^. 
 
 1 Monatsberichte, etc., 1881, p. 1154. Archiv fiir Physiologie, etc., 1882, p. 66. 
 
 ^ In Fig. 151, Plate vi. of vol. ii. part ii. of the TJntersuchungen, they are the 
 screws with which wires s, and s^ are connected. 
 
 ^ Comp. p. 461. Here, by a slip of the pen [corrected in the translation], the word 
 primary was used instead of secondary. The mode of producing a tetanus of determined 
 
490 LIVING TOKPEDOS IN BERLIN. 
 
 The result could not be considered decisive, because tbe experiments 
 were made in the winter 1883-4 with the last fish and on the second 
 day, when the organ had no longer its full functional activity, and 
 for this very reason, they could not be repeated at that time. 
 
 I resumed the thread of this investigation at the first opportunity, 
 when I received a fresh supply of Torpedos in the following summer. 
 The preparations employed in the following experiments were still 
 not all that could be desired, for they were taken from the first 
 fish which came to hand and which was in a state of opisthotonus 
 (see above, p. 479), and again, 24-27 hours had elapsed since 
 death ; they still gave however a strong organ current, an infallible 
 sign that their functional activity was maintained. 
 
 In order to work under simpler conditions, I began by substituting 
 Helmholtz' arrangement of the inductorium for the usiial one. I 
 did not however obtain closing and opening shocks which were as 
 congruent with each other as I desired, for if great exactitude is 
 required, this is not so easy in practice as in theory. For reasons 
 which may be deduced from my formulas^, the closing discharge 
 was now the shorter and stronger, while the opening discharge was 
 the longer and weaker. In accordance with this, the closing 
 discharge now played the same part as the opening one had done 
 before : if it was homodi-omous, then after tetanus of ^' there 
 ensued strong absolutely and relatively positive polarisation ; if it 
 was heterodromous, feeble absolutely positive, relatively negative 
 polarisation. The former action diminished very rapidly on 
 frequent change of direction. 
 
 But now I replaced the inductorium by Saxton's machine, which 
 I have frequently mentioned and most recently in the treatise on 
 Secondary Electromotive Phenomena (see No. V). This instrument, 
 was constructed by Oertling for Dove, and I acquired it for the 
 physiological institute at the sale of his eflPects after his death ^. 
 The turn-table was removed from the machine, and its axis was 
 connected by a pulley with the spindle of a water-motor, which 
 turned the armature seven times a second. The pachy trope ^ stood 
 
 duration by means of the switch, employed on a former occasion (see No. V of this 
 work, par. x), was excluded here, as is self-evident, because the secondary coil would 
 have furnished a derivation to the galvanometer. 
 
 * Gesammelte Abhandlungen, etc., vol. i. p. 233 f. 
 
 ^ Sitzungsberichte, 1883, vol. i. p. 360. 
 
 ^ [The pachytrope is a contrivance, by which the coils of wire surrounding the two 
 legs of the armature can be connected in two ways, either so that the one is a con- 
 tinuation of the other and they are equivalent to one wire of double length, or so 
 
LIVING TORPEDOS IN BERLIN. 491 
 
 at ' physiological/ With the springs i and l on roller I, 9 on 
 roller II, every revolution between uprights C and D ^ gave two 
 equal and opposite extra currents of unbearable strength when 
 tested subjectively by the handles. The switch was connected with 
 the terminals of the coils of the machine, instead of with the 
 terminals of the induced circuit of the inductorium. As the switch 
 performed its to and fro motion whilst the machine was working, 
 the polarising effect of a series of really congruent alternating 
 currents on the organ could be studied. Although the action of the 
 machine at a distance upon the galvanometer could not be affected 
 by the movement of the switch, inasmuch as the armature was 
 constantly rotating, special care was taken to ascertain that no 
 change actually took place. 
 
 I employed the \" — and the l" — disk of the polarisation switch 
 in succession, and I also tetanised for 30"', 45'', 1 10" by the watch. 
 The result was very uniform. Whether the terminals of the rotating 
 coils were connected with the dorsal or ventral surface of the 
 preparations, and whatever might be the duration of the tetanus in 
 fresh preparations, I could detect nothing but absolutely positive 
 polarisation, of such strength that with 5:000 turns at a distance of 
 30 mm. from the magnet, and with e = n, the scale disappeared from 
 the field of view. It then became more and more feeble, so that 
 with some preparations the action was very weak even from the 
 first. Never was it absolutely negative. The only difference, which 
 seemed to follow on longer duration of the tetanus, was in regard of 
 the persistence of the homodromous polarisation. It disappeared 
 the more quickly as the tetanising was continued for a longer time. 
 
 It was astonishing how the organ preparations sustained such 
 powerful shocks, for no arrangement was made for their gradation ; 
 it seemed as if they were accustomed, in virtue of their function, to 
 impart such shocks to themselves. 
 
 It is not difficult to understand the results on tetanising with 
 congruent alternating currents. First it is clear, that, as soon as 
 congruent shocks succeed each other in alternating direction and at 
 equal intervals, the reversal of the connections of the machine with 
 the preparations is illusory, and in truth produces no change. In 
 both arrangements, one is concerned with a similarly constituted 
 
 tliat both their similar ends are connected together, and they then form a wire of 
 double thickness and half length. In the former case, the physiological effects are 
 the stronger, and the pachytrope is then said to be 'at physiological.' — Tr.] 
 
 * Comp. the drawing in Poggendorff's Annalen, etc., 1842, vol. Ivi. Plate ii. Fig. i. 
 
492 LIVING TORPEDOS IN BERLIN. 
 
 series of alternating" homodromous and heterodromous currents. It 
 is a matter of indiiference for the preparation, whether the homo- 
 dromous cun'ents are produced with a position of the armature in 
 azimuth 90°, and the heterodromous with that in azimuth 270°, or 
 vice versa. According- to our hypotheses, the homodromous and 
 heterodromous currents produce equally strong- relatively negative 
 polarisation, and these polarisations, which moreover on account of 
 the brief duration of the shocks attain no great extent, cancel each 
 other. But besides the relatively negative polarisation, the homo- 
 dromous currents produce absolutely and relatively positive polarisa- 
 tion, while according to our hypotheses the heterodromous produce 
 no relatively positive, absolutely negative polarisation or only 
 a trace, so that under favoui-able circumstances, the absolutely 
 positive polarisation due to the homodromous currents makes its 
 appearance in great force almost or entirely unmixed. The only 
 point which remains obscure, is on what the more rapid falling of the 
 homodromous polarisation after longer tetanising with congruent 
 alternating cm-rents depends. 
 
 4. An attempt is made to compare the negative polarisation 
 due to the closing shocks of an inductorium by themselves, 
 with that due to the opening shocks by themselves. 
 In order to judge of the secondary electromotive phenomena 
 in the electrical organ produced by incongruent alternating 
 currents, it is expedient to answer first the question, which of the 
 currents, due to the same amount of electricity equalizing itself, 
 give the strongest negative polarisation, whether the longer feebler 
 closing shocks, or the shorter stronger opening shocks? This 
 question must permit of a decided answer, by exposing an organ 
 preparation in heterodromous direction to a series of closing shocks 
 of the inductorium and to one of opening shocks alternately. In 
 connexion with this subject, I remarked that in my experiments on 
 internal polarisation of moist, porous conductors in general, I had 
 neglected to perform this experiment, apparently so simple and yet 
 so instructive ; and I therefore determined to fill up this gap 
 immediately. This seemed easy of accomplishment with the aid of 
 Dove's disjunctor. 
 
 This apparatus, which after his death also passed into the 
 possession of the physiological institute^, consists of three pairs of 
 copper wheels on a common glass axis, by rotation on which, they 
 
 ' I had once before worked with the same apparatus, through the kindness of 
 Dove, for a similar object. Untersuchungen, etc., vol. ii. part i. 1849, p. 405. 
 
LIVING TORPEDOS IN BERLIN. 493 
 
 can be set at different positions in relation to each other. The two 
 wheels of each pair are electrically connected with each other ; the 
 three pairs which may be denoted by A, B, C are insulated from 
 each other by the axis. One wheel of each pair is constantly 
 immersed by its amalgamated rim in a mercury trough. The rim 
 of the other has eight copper and eight glass sectors of equal 
 length, and a copper spring presses on its circumference. 
 
 One pair, A, is made to close and interrupt the primary circuit. 
 Another pair, B, is included in the secondary circuit, and it is 
 placed by the breadth of a half-sector in such a position in regard 
 to A, that its spring is against the glass when that of A comes to 
 the glass ; in this way the series of closing shocks alone is trans- 
 mitted. If, however, C is placed in regard to A by the breadth 
 of a half-sector, so that its spring rests upon glass when that of A 
 comes to the metal, then only the series of opening shocks is trans- 
 mitted. The same end may be attained in a simpler way, by using 
 only one of the pairs .S or C in addition to A, and setting the 
 wheels first in one direction, and then in the other. 
 
 In my experiments the machine was driven by a water-motor, the 
 axis of which was connected to the axis of the disjunctor by a pulley. 
 The change in direction of turning was accomplished by crossing 
 or not crossing the cord. The distance of the disjunctor from the 
 water-motor was so considerable, that the crossing of the cord 
 produced no perceptible difference in its tension. The number of 
 shocks was 44 per second. 
 
 The separation of the one series of shocks from the other is 
 indeed effected with the same degree of certainty only in so far as 
 the continuous contact of the spring in its passage can be relied 
 upon. This however seemed sufficient. For when opening shocks 
 were excluded, and by holding handles, I placed myself in the circuit 
 of the secondary coil and brought it to within one or two centimeters 
 of the primary, the sensation I experienced was quite uniform ; if, 
 however, the spring closing the primary circuit had jumped, each 
 opening would have been perceived by me as a violent shock. 
 
 By these means I hoped to learn how the internal polarisation of 
 moist porous bodies and the negative polarisation of the electrical 
 organ, most nearly comparable with it, and perhaps also that of 
 muscles and nerves, depends upon the time occupied in equalisation, 
 when the amount of electricity is constant. The polarisation 
 switch was put into the secondary circuit, and the series of shocks 
 of 5'' duration was led through the usual conducting vessels pro- 
 
494 LIVING TORPEDOS IN BERLIN. 
 
 vided with clay shields, to the object to be polarised. Blotting- 
 paper pads, white beech wood, pumice-stone, roofing tiles were made 
 into bars 50 mm. long and one square centimeter in transverse 
 section, and served for such objects. The blotting-paper was soaked 
 with distilled water, and the wood, pumice-stone and tiles were 
 boiled in it until they sank below the surface^. In the same 
 circuit, secondary in regard to the induction, primary to the polari- 
 sation, was placed the galvanometer P with ^7) turns of wire at a 
 distance of 20 mm. from the mirror. 
 
 A second pair of conducting vessels with wedge-shaped pads, the 
 edges of which, protected with physiological clay, were laid on the 
 object to be polarised, formed as usual the terminals of the gal- 
 vanometer S. 
 
 At first it seemed that the negatively polarising efiect of the 
 closing shocks exceeded that of the opening shocks. It was 
 however soon proved that this could not be relied on. If 
 the experiments went right, the deflection of the galvanometer 
 P, due to the closing shocks, ought to have been sensibly equal 
 to that due to the opening shocks. But it was observed that 
 the closing shocks always produced by far the greater deflec- 
 tions. As at first I employed ordinary inductoria, in which 
 no special measures are adopted for the insulation of the turns of 
 the wire from each other in the secondary coils, I supposed that a 
 channel for sparks had formed itself in the interior of the coils, in 
 which the high tension electricity due to the opening induction, 
 sprang over, instead of making its way through the object to be 
 polarised and the galvanometer P. In order to work under 
 simpler conditions, I abandoned the polarisation object and the 
 conducting vessels in connexion with it, and replaced them by one 
 or two resistance coils out of a plug rheostat, which gave a resist- 
 ance of 60 and 120 Siemens units, respectively. Instead of the 
 ordinary induction apparatus, I took a Ruhmkorff* inductorium by 
 Siemens and Halske, and as its shocks, even with only two Bunsen 
 chromic acid cells in the primary circuit, seemed likely to damage 
 the galvanometer, I drew the primary coil somewhat out of the 
 secondary, so that I gradually lessened the eflect as in the ordinary 
 inductorium. In this apparatus, it is certain that no sparks sprang 
 over in the interior of the secondary coil ; but, on the other hand, it 
 was very difficult to put a stop to all sparks and lateral discharges 
 owing to the length of the conducting wire, which was necessary 
 
 * Comp. Untersuchungen, vol. ii. part ii. p. 430 f. 
 
LIVING TORPEDOS IN BERLIN. 495 
 
 in order that the actions at a distance of the inductorium on the 
 galvanometers should disappear. This aim seemed to be sufficiently 
 attained when all the condvicting" wires were thick and covered with 
 guttapercha. But here also, the opening shocks proved much 
 weaker in their effect on the galvanometer P than the closing 
 shocks, and the difference was so much the greater as the primary- 
 coil was pushed into the secondary ; with a certain strength of the 
 induction indeed, the effect of the opening shock disappeared, and 
 beyond that it was reversed, so that the mirror was deflected in the 
 same direction by it as by the closing shocks, only much more 
 feebly and very irregularly. 
 
 I now suspected that sparks sprang over in the galvanometer 
 coil itself, although I did not understand how inverse deflections 
 should result. However, in order to proceed quite surely, I replaced 
 the usual high resistance bobbin of the galvanometer by a similar 
 one wound with wire covered with guttapercha, but the result 
 remained the very same ; afterwards, just as before, the series of 
 opening shocks produced an infinitely small deflection in comparison 
 with that due to the closing shocks, and with a greater strength of 
 induction, the former deflection was inverted, i.e. the opening 
 shocks acted apparently in the same direction as the closing shocks. 
 That the opening induction continued in full force, was proved 
 by sparks at every interruption at any part, and by lateral dis- 
 charges when opportunity offered at any exposed part of the 
 circuit. 
 
 Single closing and opening discharges, produced in the pri- 
 mary circuit by means of a key while the disjunct or remained at 
 rest, effected quite regular deflections equal in amount on both 
 sides of the zero point ; but the most remarkable fact was, 
 that when Wagner's hammer or Foucault's mercury interi-upter was 
 put in the primary circuit instead of the disjunctor, the closing 
 and opening shocks maintained equilibrium as perfectly as was ex- 
 pected. 
 
 As the mirror was now observed to vibrate in time with the 
 interruptions, the suspicion arose that the disappearance of the 
 effect produced by the opening discharges on the mirror, might be 
 connected with the greater number of interruptions which occurred 
 in using the disjunctor, owing to its being turned by the water- 
 motor. I therefore directed an assistant to turn the disjunctor 
 first slowly, then more and more quickly by means of the pulley 
 connected with it : this proved that in fact the opening shocks 
 
496 LIVING TORPEDOS IN BERLIN. 
 
 acted regularly vdth. slow turning, but more and more feebly with 
 more rapid turning, until their effect was reversed. If it was 
 arranged that both shocks passed, the closing shocks always pre- 
 ponderated, all the more as the turning was more rapid. 
 
 The only question now left was, whether the irregularity which 
 ensued on turning the disjunctor with greater rapidity, was peculiar 
 to this apparatus, or whether it would be found also with a more 
 rapid succession of closing and opening, than can be obtained by the 
 use of Wagner's hammer or Foucault's interrupter. I put into the 
 primary circuit of Ruhmkorff's inductorium, the interrupting spring 
 of an ordinary inductorium which opens the circuit 150-300 times 
 per second, according to a previous determination made by Helm- 
 holtz, and further a C tuning-fork interrupter by Konig of Paris, 
 which performs 256 single \abrations per second, or opens the 
 circuit 128 times. But I obtained regular actions with these 
 means also, as well as with Wagner's hammer and with Foucault's 
 interrupter, although the succession of the closing and opening 
 shocks was much more rapid than with the disjunctor, which gave 
 at most 45 opening shocks per second (see above, p. 493). 
 
 Thus it was proved that in consequence of an unexplained condi- 
 tion, the disjunctor is unable to give series of exclusively closing or 
 of exclusively opening shocks, in which the same quantity of elec- 
 tricity equalizes itself, beyond a certain speed which is difficult to 
 determine. The next problem would be to search out that condition, 
 and to discover what becomes of the large amounts of electricity 
 set in motion at high tension in the opening shocks of the induc- 
 torium, and which seem to disappear and leave no trace. It is 
 essential to ascertain this in order to obtain a basis for the theory 
 and construction of so important an apparatus as the disjunctor ; 
 for this reason, I did not feel justified in withholding this series of 
 experiments although their result is not gratifying. 
 
 It may perhaps be asked, why after failing with the disjunctor, 
 I did not attempt to compare the negative polarisation due to 
 a single closing and opening shock, which, as the above and other 
 observations show, can be obtained. This scheme naturally sug- 
 gests itself, and I intend to revert to it. It will, however, not be 
 at all easy to ensure the closure of the galvanometer circuit at the 
 same time after the polarising shock in the two cases. The end 
 may also be attained by a slowly turned disjunctor, but not with- 
 out further provision for regulating the number of the shocks 
 which reach the polarisation object. It will be as necessary, as in 
 
LIVING TORPEDOS IN BEELIN. 497 
 
 the ease of single shocks, to arrange for a uniform closure of the 
 galvanometer circuit after each individual shock. 
 
 All this demands a special investigation, for which I have not 
 leism'e at present. In regard to the question how the negative 
 polarisation in the electrical organ behaves for closing and opening 
 shocks, we must be content with the simplest assumption in the 
 mean time, viz. that it is proportional to the amount of electricity, 
 and thus is the same for the two shocks. 
 
 5. Continued discussion of the results of polarisation of 
 the organ by alternating currents. 
 
 The result of polarising the organ by the alternating currents of 
 induction apparatus, can be readily explained by means of the last- 
 mentioned hypothesis in conjunction with the indubitable fact, that 
 when the direction is homodromous, strong currents of short dura- 
 tion produce stronger absolutely and relatively positive polarisation 
 than weaker currents of long duration. For here also, as with the 
 congruent alternating currents of Saxton's machine, the relatively 
 negative polarisations in both directions cancel each other : since 
 the heterodromous shocks produce no relatively positive polarisa- 
 tion or only infinitely little, the homodromous absolutely positive 
 polarisations alone remain, and these prove to be stronger with 
 homodromous strong shocks of short dm-ation than with homo- 
 dromous weak shocks of long duration. 
 
 It is another question, how much these experiments have contri- 
 buted to decide whether the homodromous absolutely positive 
 polarisation is only an after-effect of a shock, or an independent 
 phenomenon. This decision is not such as I had expected. I had 
 overlooked the circumstance that the relatively negative polarisa- 
 tions due to both currents may cancel each other, but not, as I had 
 assumed, the relatively positive ones, because, if there are any such at 
 all in the organ, the heterodromous current does not produce them, 
 or only produces traces of them. Thus it was an error, when I sup- 
 posed, that tetanising in itself would leave behind it no polarisation, 
 so that an absolutely positive effect produced by it must be conceived 
 of as solely an after-effect of shocks. If the observed results have 
 any bearing on our question, it is rather in the opposite sense. It 
 has been seen that, to express myself briefly, according to the polari- 
 sation theory the phenomena are not difficult to explain ; whereas 
 the after-effect theory fails to explain on what the weak absolutely 
 positive polarisation with heterodromous direction of the stronger 
 
 Kk 
 
498 LIVING TORPEDOS IN BERLIN. 
 
 shocks of short duration depends. For it is difficult to imagine that 
 when a shock is once discharged, it will not reach a height corre- 
 sponding to the functional acti\aty of the organ, independently of the 
 strength of the excitation ; and it seems just as little probable, that 
 the negative polarisation due to short sharp shocks, should exceed that 
 due to long feeble ones. According to the analogy of the polarisa- 
 tion of metal electrodes, the opposite would rather be expected. 
 This is the point in respect of which the exjjerimental basis sought 
 for in vain in the previous paragrajDhs, is necessary. 
 
 6. Unsuccessful experiments to produce relatively positive 
 polarisation by heterodromous ciu^rents. 
 
 As I reflected upon these questions it occurred to me, that I had 
 not followed a certain path of experiment to its utmost Kmit, even 
 though I had not left it quite untrodden, and that it might possibly 
 lead to the e-oal. We have indeed laid down as a fact founded on 
 experience, that relatively positive polarisation due to the hetero- 
 dromous current, was never observed in the experiments with battery 
 currents on the Malapterurus, Gymnotus and Torpedo ; but we have 
 not hitherto undertaken experiments for the express purpose. Such 
 experiments must be designed so as to act on organ preparations by 
 heterodromous shocks as strong, and at the same time of as short 
 duration, as possible. If we should succeed in observing in this 
 way relatively positive, absolutely negative polarisation, there 
 could be no longer any doubt as to its existence, for there is no 
 reason to assume reversal of the direction of shock of the organ. 
 
 At first, I still adhered to the use of galvanic currents closed for 
 a short time. I had indeed at an earlier date (see p. 473, Series 
 1 9 and 20) sent the current of 50 Groves through organ preparations 
 for a short time in vain ; more often on the contrary, when the 
 direction was homodromous, a negative preliminary shock had 
 shown itself, which it was very difficult to interpret. The period 
 of closing was o''-o629 ; I now chose a period twenty times shorter. 
 
 4. Torp. Fresh. 0C+ 0-0078. 
 Cro".oo3i5. 
 Jj.—S. I -54;-35t +45; + i5 I -14 
 
 t +45; + ii 
 I 21 ; 20-5 
 
 ■P- I 13-5; 13 I 21 ; 20-5 i 15 
 
 The failure could not be plainer. I now placed my hope on induc- 
 tion shocks. 
 
 The Saxton's machine employed before, ofiered facility for such 
 experiments. It has a jiair of T-shaped springs previously de- 
 
LIVING TORPEDOS IN BERLIN. 499 
 
 scribed. These springs change the alternating currents of the 
 machine into currents which are all in the same direction, so 
 that seven revolutions per second gave fourteen extra currents in 
 the same direction. Whilst the machine was in motion, its circuit 
 was closed for a shorter or longer period (o''-03i, o''''098, i"-024) 
 through the organ preparation. When the time of closing was 
 only o'''-03i, it depended on chance whether an extra current took 
 place exactly in this period, but when it was ©'''•098, one current 
 must certainly come into action, and two currents might do so. 
 The experiments were made this winter on the fifth fish on the 
 second day, and on the sixth fish in a comparatively fresh condition. 
 The organ preparations showed mostly reversed organ current 
 between the ventral and dorsal skin ; they were taken from the 
 same organs as those preparations of which the one half acted 
 negatively and the other positively (see above, p. 485). The 
 result of the experiments was very unsatisfactory. There was often 
 no action, as was to be expected with the very short period of 
 closing. With homodromous direction of the cm-rents, there was 
 absolutely and relatively positive weak polarisation, and with 
 heterodromous direction, there was much stronger absolutely posi- 
 tive, relatively negative polarisation. Example : — 
 
 6. Torp. Fresh. 0^ — 0-0027. 2' — P. 
 
 CTd' -oil. ' Cri"-024. 
 
 4 —220 I +18 i —210 I +60 II j — 43°! ?; + 10 i -295. 
 
 The primary current was not observed. 
 
 The Saxton machine was succeeded by an ordinary inductorium of 
 a larger kind, and a battery of two large Bunsen chromic acid cells 
 was put into its primary circuit. The spring of the Wagner hammer 
 was removed, and was replaced by the disjunctor described above. 
 This was arranged so as to allow only the opening shocks to pass, 
 but of these, it gave thirty-seven in one second, of which therefore 
 one or two passed the switch in d'-oo^i. As far as could be judged 
 from the tension in the secondary circuit, the single shocks by far 
 exceeded those of the Saxton machine. The result was essen- 
 tially the same, no relatively positive heterodromous polarisation 
 appeared. 
 
 6. Torp. Fresh. 0C + 0-0065. 
 
 Cro".oo3i. 
 i -500 + a;; -370 t -155; +355. 
 
 In the last case, presumably, two shocks passed through. I re- 
 marked as yet in these experiments, nothing of the disturbances 
 
 K k 3 
 
500 LIVING TORPEDOS IN BERLIN. 
 
 which I obsei*ved in a later series of experiments in using* the 
 disjunctor arranged for opening shocks alone, described above in 
 § 4. But in any case, they would not explain the negative result • 
 obtained here. 
 
 I was unwilling" to accept this rebuff, and determined to try 
 further the effect of a single opening shock of Ruhmkorff's induc- 
 torium at its full strength. The polarisation switch would not 
 answer this purpose, but a special switch had to be constructed, 
 which rendered it possible to open the secondary circuit (which 
 would otherwise have short-circuited the galvanometer) imme- 
 diately after the opening of the primary circuit and the passage of 
 the shock through the organ preparation, and then to close the 
 galvanometer circuit. If the ordinary arrangement had been 
 adopted, lateral discharges might have spread from the wedge-pads 
 placed on the preparation, over the conducting vessels and beyond, 
 and might possibly have left disturbing polarisations. On this 
 account, I did not venture to adopt the plan of closing the galvano- 
 meter circuit at two places situated between the conducting vessels 
 and the galvanometer, but I arranged that the clay points of 
 two non-polarisable conducting tubes connected with the galvano- 
 meter, should be so placed on the preparation, as to take up the 
 polarisation current by the same movement which opened first 
 the primary, and then the secondary current. 
 
 Fom- corks are fixed upon a glass axis, which can be rotated 
 horizontally in wood bearings. Cork I carries a bent bar of 
 copper, which at the initial position of the switch, by dipping into 
 two mercury cups, closes the primary circuit of the inductorium, in 
 which three large I^unsen cells are included. Each of the corks 
 II and III carries a non-polarisable conducting tube, the clay 
 point of which in the first instance is suspended above the organ 
 preparation. Cork IV carries a copper hook, one end of which is 
 the terminal of the secondary coil and dips into a mercury cup. Its 
 other end is connected with the other terminal of the coil perma- 
 nently, but is movable. When the glass axis is turned from its 
 initial position, the primary circuit opens, whilst the secondary is 
 still closed. The shock passes through the preparation, to which 
 it is led by amalgamated zinc plates in solution of zinc sulphate, 
 and by zinc pads protected by clay shields. Immediately after, 
 the secondary circuit opens, and thereupon the clay points touch 
 the preparation. Two marks are made on the milled edge of the 
 wood disk which serves to turn the axis, of which one shows the 
 
LIVING TORPEDOS IN BERLIN. 501 
 
 initial position of the switch, and the other, that position at which 
 the clay points just touch the preparation. 
 
 I convinced myself of course beforehand, that the passage of the 
 shock would not produce any perceptible effect on the galvano- 
 meter, by using a clay model in place of the preparation ; only 
 a trace of negative polarisation could be perceived in the clay. I 
 then compensated any organ current which might appear when the 
 points were placed on the actual preparation, so that no deflection 
 of any importance ensued when the points were replaced after they 
 had been removed. The sparks in the primary circuit were 
 diminished by Fizeau's condensator. The conducting wires of the 
 secondary coil and beyond it consisted of wires covered with gutta- 
 percha throughout. The number of turns was 5000, and the dis- 
 tance from the galvanometer was zero. 
 
 In spite of the risks connected with such considerable movements 
 of electricity in the neighbourhood of a sensitive galvanometer, the 
 experiment proceeded without any disturbance. It was made first 
 on the torpedo which had been kept in the aquarium since October. 
 It gave shocks only when it was strongly excited. But the pre- 
 parations still showed an adequate organ current. The result of the 
 experiment was again negative. I did not succeed in obtaining rela- 
 tively positive polarisation by the heterodromous opening shock of 
 the inductorium, the observed polarisation being relatively negative, 
 absolutely positive. But since the homodromous shock only gave 
 weak and relatively negative polarisation, it was evident that the 
 functional activity of the organ was not sufficiently good for the pur- 
 pose, and that the experiment ought to be repeated with a fresh 
 fish. 
 
 This was effected a short time afterwards with the eighth 
 torpedo, w^hich arrived only in the spring, and which was in a 
 comparatively good condition. But the result remained essentially 
 the same. The homodromous shock now produced absolutely and 
 relatively positive polarisation far beyond the limits of the scale, 
 and the heterodromous absolutely positive, relatively negative 
 polarisation extending also beyond the limits ; effects in the same 
 direction which disappeared very gradually were left behind. In the 
 same measure as the functional activity decreased on more frequent 
 repetition of the experiment, the intensity of the phenomena also 
 diminished, until the ordinary type returned, viz. absolutely and 
 relatively positive deflections due to the homodromous induction 
 currents, and relatively negative, absolutely positive deflections due 
 
502 LIVING TORPEDOS IN BERLIN. 
 
 to the heterodromous induction currents. For instance, I obtained 
 with 5000 turns at a distance of 20 mm. 
 
 i - 85 t + 360 i - 57 t + 135- 
 
 But the heterodromous shocks never produced relatively positive 
 polarisation. 
 
 There was still the possibility that the latter although present, 
 might be masked by relatively negative polarisation. Although 
 the opening shocks of the inductorium, by using the condensator are 
 only of very short duration, I wished also to employ Leyden jar 
 discharges, so that I should have left nothing untried. The same 
 switch served for this as for the Ruhmkorff 's opening shocks. In 
 the first series of experiments on the same seventh Torpedo, Prof. 
 Christiani discharged the jar, charged by Holtz' machine, through 
 the preparation, and I put the clay shields in contact with it as 
 soon as T had heard the sparks. The preparations were perhaps 
 accidentally better than those which had been exposed to the 
 induction shocks ; for instance, we obtained 
 
 1 -448 t + 55 i - II T + 48- 
 It was however clear, that the experiments must be repeated with 
 a fish with greater functional activity. 
 
 These latter and also the renewed experiments with the induc- 
 torium were performed on the eighth torpedo, which had just been 
 killed. The following contrivance was now adopted. A fifth cork 
 at the free end of the glass axis of the switch, carried a strong brass 
 wire bent at right angles, one leg of which formed a prolongation 
 of the axis, and was connected with an amalgamated zinc plate 
 which was to conduct the shock to the organ preparation. The 
 other leg, perpendicular to the axis, terminated in a closed ring. 
 The charged jar was disposed so that its outer coating was con- 
 nected with the other zinc plate. On turning the axis between 
 the marks on the middle edge of the disk, the ring came so near 
 the knob of the inner coating, that the spark sprang over and led 
 the shock to the preparation a short time before the clay points 
 reached it. In this series of experiments, the Leyden jar (with 
 a coating of ^6^ square centimeters) was charged at the conductor 
 of an ordinary electrical machine, for by regulating the number of 
 revolutions of the disk, a greater uniformity and a more certain 
 gradation of chai'ge was ensured than in charging with the influ- 
 ence machine. 
 
 The experiments succeeded extremel}^ well, without however 
 
LIVING TOEPEDOS IN BERLIN. 503 
 
 yielding- the result sought for. At first, I charged the jar as 
 strongly as possible, with fifty revolutions of the disk, and the 
 polarisations produced by the discharge — absolutely and relatively 
 positive by the homodromous, absolutely positive and relatively 
 negative by the heterodromous shock — attained an indescribable 
 intensity : many minutes passed before the end of the scale 
 appeared again in the field of view. Then I decreased the number 
 to five revolutions, to two, and even to one, yet still the com- 
 paratively weak shocks were followed by polarisations of great 
 strength and perfect uniformity. For example, I obtained with 
 five revolutions, 
 
 t + 270 i - 470 t + 125 I - 250. 
 Nothing whatever was to be seen of heterodromons relatively 
 positive polarisation. 
 
 It is imnecessary to state, that all measures were taken to 
 exclude direct actions of the shocks of the jar on the galvanometer, 
 and that a clay model in the place of the preparation, showed at 
 most the very faintest trace of negative polarisation. 
 
 I did not proceed further on this path, and I do not believe that 
 any other result is to be attained here. But (§ 15) in a later 
 experiment, which unfortunately only succeeded once, the hetero- 
 dromous current actually produced relatively positive, absolutely 
 negative polarisation. 
 
 7. The apparent irreciprocity of the conduction in the 
 electrical organ increases with the current density. 
 
 In the first communication, I closed the examination of the 
 nature of the homodromous, absolutely and relatively positive 
 polarisation with the words, ' There is still one method of ex- 
 periment at hand, which under certain conditions might lead 
 to the attainment of the object, that is to observe whether with 
 a longer closing time, the homodromous current maintaiifs its 
 preponderance. If this is the case, positive polarisation cannot 
 be identified with the shock ; for the latter, when the current goes 
 for a long time through the organ, can only add itself to the 
 battery current at the moment of closing. I have in fact seen this 
 ascendancy with a closing time of i^\ 5", or even 20", but ex- 
 periments of this kind will not be conclusive imtil the hypothesis 
 of an irreciprocal resistance can be completely set aside.' 
 
 It is beyond question, that even with a much longer period of 
 
504 LIVING TORPEDOS IN BERLIN. 
 
 closing the homodromous current is seen to preponderate. The result 
 is obscured by two facts, first that the preparation conducts better 
 when warmed, and secondly that the negative polarisation weakens 
 the current ; nevertheless by closing the currents for a minute by 
 the hand, series are also obtained in which, in spite of all 
 hindrances, the preponderance of the homodromous current declares 
 itself plainly though feebly. Series of this nature will be com- 
 municated later (see below, § 12). Dr. Sachs' investigations on the 
 organ of the Gymnotus also speak to the same effect^. 
 
 Thus the question, whether the organ conducts irreciprocally in 
 reality or only in appearance, lies directly across our path, and we 
 must endeavour before all things to clear up this point. 
 
 The first thing that presents itself, is the very striking de- 
 pendence of the phenomenon upon the current density, as is shown 
 in the following experiments. 
 
 I sent the opening shocks of the inductorium, the primary coil 
 of which was filled with wires, through the preparation from one 
 skin surface to the other. The preparation rested in the usual 
 way on the triangular plate between the clay shields of the con- 
 ducting vessels. The galvanometer was in the same circuit. The 
 shocks, produced by opening the mercury key, were by means of 
 a Pohl's reverser, led through the preparation alternately in the 
 homodromous and the heterodromous direction. 
 
 In the following table, C D denotes the distance of the secondary 
 from the primary coil ; the numbers are the deflections of the 
 galvanometer. The great difference of the strengths of the shocks 
 makes it necessary to put the bobbins at different distances from 
 the mirror, and the deflections are therefore reduced to 5000 turns 
 at a distance of 20 mm. 
 
 5. Torp. Fresh. 0C+ 0-0031. 
 
 CD = o t 501 1 215 t 501 i 215 t 453 i 215 t 477 i 191 
 
 „ = 10 cm. i 25 f 28 i 27 I 28 i 27 f 27 
 
 • » = 15 cm. t 7 i 7 T 7 i 7 
 
 » = o 1 453 i 227 
 
 Taking the mean of the numbers of like signification, the ratio of 
 the heterodromous to the homodromous deflections is 
 for CD 
 
 = : 
 
 : 212-3 
 
 477-0 : 
 
 100 
 
 224.7 
 
 — 10 cm. : 
 
 : 26-33 
 
 27-66 : 
 
 • 100 
 
 105-1 
 
 = 15 cm. : 
 
 : 7 
 
 7 ' ■• 
 
 : 100 
 
 1 00-0 
 
 Hence there can l^e no doubt, first, that the phenomenon becomes 
 
 * Untersuchungen, &c., p. 218 ; p. 224 of this volume. 
 
LIVING TOKPEDOS IN BERLIN. 505 
 
 observable by the means employed, when the current strength 
 exceeds a certain limit, and secondly, that the irreciprocity increases 
 with the cuiTent strength. By performing the experiment on 
 preparations of different transverse section, one is convinced, thirdly, 
 that the question here concerns not current strength, but, as was to 
 be expected, current density. Fourthly, it follows from the above 
 numbers, that within its sphere, irreciprocity increases more slowly 
 than current density. If we take (see below, § 13) the difference 
 between the homodromous and heterodromous current strength 
 with the same transverse section, divided by the homodromous 
 current strength, as a measure of the irreciprocity (a method 
 of proceeding which will be justified later), then we find the 
 irreciprocity 
 
 at 10 cm. CD = — ^ = 0-0481, 
 27-66 
 
 at o cm. CD = = 0-15 ^aq, 
 
 477-0 "'^^^' 
 
 instead of = 0-8393, ^^ i^ would be if the irreciprocity were pro- 
 portional to the cm'rent density. There can hardly be a doubt 
 that the irreciprocity, referred to the current density, would approach 
 asymptotically to a fixed limit. 
 
 8. The apparent irreciprocity of conduction in the elec- 
 trical organ has its seat in every transverse lamella of the 
 preparation, and increases with the length of the columnar 
 tract which is traversed by the current. 
 
 I next convinced myself, that the difference in the strength of 
 the currents becomes perceptible, not merely when they are led 
 from the one polar surface of the preparation covered with skin to 
 the other, but also between any points whatever of the lateral 
 surfaces of the preparation, and the difference is the greater the 
 further the points are separated from each other. 
 
 I again led the opening shocks of the inductorium, by opening 
 the mercury key, through the preparation and galvanometer, but 
 this time in accordance with the aim of the experiment, the 
 preparation was not traversed by currents from skin surface to skin 
 surface, but the clay points of the non-polarisable tubes, which 
 took the place of the conducting vessels in the circuit containing 
 the galvanometer and the secondary coil of the inductorium, were 
 brought into contact with the lateral surface. The secondary coil 
 
506 LIVING TORPEDOS IN BERLIN. 
 
 was pushed quite over the primary coil filled with wires. The 
 distance of the bobbin (5000 turns) was 10 cm. from the mirror. 
 
 In Fig-. 20 the curves denote the circuit, including the secondary 
 and galvanometer coil. The preparation was taken from the fifth 
 fish, killed the previous day; the organ current force amounted at 
 first to + 0-0072. 
 
 When the clay points were in position (i), I obtained 
 
 •f 42 i 12 t 42 i 15 sc; 
 in position (2), 
 
 t 38 4 19 t 38 t 20 sc; 
 
 finally, in position (3), 
 
 I 40 i 22 f 40 4 22 sc. 
 
 According to this, there can be no doubt, that irreciprocal con- 
 duction, whatever may be its cause, has nothing to do with the 
 
 skin ; this was indeed already 
 proved when I first observed 
 it in the malapterurus prepar- 
 ations, where the skin consti- 
 tuted only an external closure. 
 It might however still be 
 supposed that its seat is super- 
 ficial, that it depends on a resist- 
 ance which develops itself 
 only at the contacts of the clay 
 points with the lateral surfaces 
 of the preparation, after the manner of external secondary resistance. 
 At first sight this is very improbable, considering that it increases 
 evidently with current density, not with current strength ; that it is 
 observed even with such transient shocks as the opening shocks of 
 an induction apparatus ; that the interposal of the skin makes 
 no change in its mode of appearance ; and finally, that if one of the 
 clay points, the positive or the negative, were the seat of a special 
 resistance, this must be the case for homodromous currents just as 
 much as for heterodromous. 
 
 But this conception would be quite impossible, if it were shown 
 that the irreciprocity increases with the length of the columnar 
 tract traversed by the currents. Now this is in point of fact the 
 case. 
 
 A long tract of organ (4) gave 
 
 t 32 i 12 t 32 i 13 sc; 
 
LIVING TORPEDOS IN BERLIN. 507 
 
 a short (5), 
 
 t 37 i 29 t 37 i 29 sc. 
 
 If we take also the numbers of the tracts of medium length 
 
 (i, 2, 3), then we find, as a mean of all numbers in the same 
 
 category, the ratio of the homodromous to the heterodromous 
 
 currents, in the long tract 
 
 100 : 39-1, 
 
 in the medium-length tracts, 
 
 100 : 45-8, 
 in the short tract, 
 
 100 : 78-4; 
 
 thus the longer the tract traversed, so much the greater is this 
 ratio. The following numbers, which afford one of the most striking 
 instances of irreciprocity, were obtained with the seventh torpedo 
 on the second day, with a short distance (not indicated) of the 
 galvanometer coil : — 
 
 Long tract. 
 I 217 i 28 I 160 I 30. 
 
 Short tract. 
 
 t 176 i 155 t 176 i 155- 
 
 Long tract. 
 
 t 165 i 14 T I'^S i 22. 
 
 The ratio of the current strengths with the long tract is 
 
 100 : 133, 
 with the short, 
 
 100 : 8S-I. 
 
 One might object to this, that with a longer tract, the pre- 
 paration forms a greater part of the total resistance of the circuit, 
 and that this has misled us as regards the relation of greater 
 irreciprocity to greater length. The slight difference between the 
 homodromous current strength for long and short tracts shows that 
 the resistance of the preparation, unquestionable as it may be in com- 
 parison with that of the clay points, forms a by no means large 
 proportion of the total resistance of the circuit. However, it is 
 desirable to establish safe from all doubt, a point so important as 
 that of the increase of irreciprocity with the length of the tract 
 traversed by the current. 
 
 The idea that the irreciprocity of conduction is influenced by the 
 total resistance of the circuit, implies that it can be expressed by an 
 additive term in the denominator of the fraction which represents the 
 heterodromous current strensrth in Ohm's formula. Let us call this 
 
508 LIVING TORPEDOS IN BERLIN. 
 
 term the Irreciprocity-term. According to our numbers, it seems to 
 increase with the length of the tract traversed. If the objection raised 
 against this is well founded, then it should be possible to represent 
 the phenomena in a mathematical form, even when the term is 
 assumed to be independent of the length of the tract traversed. 
 
 First of all, the question arises, in what way the dependence of the 
 irreciprocity upon the current density is to be conceived of. This 
 question cannot be answered with certainty. The best course seems 
 to be to suppose that the irreciprocity increases with the homo- 
 dromous current density, and indeed the small difference abeady 
 pointed out between homodromous cuiTent strengths for longer 
 and shorter tracts, allows us to assume the irreciprocity term to be 
 projDortional to this current strength without any considerable 
 error, rather than to suj)pose that, in accordance with the pre\aous 
 paragraphs, it grows more slowly than it. We may proceed then 
 further in the following way. Let us imagine, in place of the 
 induction shock, that an equivalent portion of the short duration t 
 is cut out of a constant cm-rent, and that H is the electromotive 
 force of the cut-out portion. Next let 
 -S denote the resistance of the circuit from clay point to clay 
 point, containing the secondary coil of the inductorium 
 and the galvanometer ; 
 o" the specific resistance of the organ traversed longitudinally by 
 the current, irrespective of irreciprocal resistance or with a 
 homodromous current ; 
 I the length of the long tract ; 
 k that of the short tract of the organ ; 
 (I the cross section of the preparation^ ; 
 li, I^ the homodromous ; 
 /,;, I,^ the heterodromous current strength for the longer and 
 
 shorter tracts respectively ; 
 a//^, ak/q the resistances of the long and short tracts respectively 
 
 with homodromous current; finally, 
 Iic/q, Ij^c/q the irreciprocity terms for the long and the short 
 tract respectively, in which c is a constant. 
 
 ' In the first communication (p. 454), in a similar consideration, we put the 
 transverse section = i . It would simplify the expressions and change nothing in 
 the result if we did the same here, but the formulas are clearer if the transverse 
 section appears in its proper place. ISIoreover, we neglect the circumstance that 
 in our experiments, the current did not pass through the preparation from transverse 
 section to transverse section, but from one clay point to the other applied to the 
 preparation laterally. 
 
LIVING TOEPEDOS IN BERLIN. 509 
 
 Then we have 
 
 ^ M ^ m 
 
 h = r- » -t J. = 
 
 M 'Et 
 
 '^^ u — t7c' '^^ — u — T'c' ^^^ 
 
 ? ? ? 2' 
 
 The irreciprocal resistance is made here dependent upon the 
 dimensions of the tracts traversed by the current, only in so far 
 as the homodromous current density, which is taken to be pro- 
 portional to the irreciprocity term, depends upon those dimensions. 
 Now according" to our observations, irrespectively of the opposite 
 signs of / and 1, , we must have 
 
 i. e. we must have 
 
 
 
 
 
 » 
 
 
 
 R + 
 
 la 
 
 q 
 
 q 
 
 > B+~ 
 
 q 
 
 + 
 
 V 
 
 7 
 
 R 
 
 + 
 
 Icr 
 
 q 
 
 
 R + 
 
 q 
 
 
 or {qR + ka)Ii>{qR + la)Ij^. (6) 
 
 This is impossible, since for ^ = o, or a mean value of R, the left 
 side, factor for factor, is the smaller, and since, if all resistances are 
 made to disappear in comparison with R, at the most, equality of 
 both sides will be obtained. Consequently we do not succeed, when 
 the irreciprocity term is assumed to be independent of the length 
 of the tract traversed. 
 
 On the other hand, if in (a) we substitute 
 
 for {la + Iic)/q and for {k<T->rIy.c)fq 
 (-+y)V?and(. + ^>/^; 
 
 i. e. if we treat the irreciprocal resistance as an increase of the 
 specific resistance proportional to the current density when the 
 current is a homodromous current, then we get instead of {h) the 
 inequality 
 
 {qR ^Tc<y) 111 > (qR + l(r)M^. . (c) 
 
510 LIVING TORPEDOS IN BERLIN. 
 
 This indeed is still impossible supposing E = o, for then it re- 
 duces itself to Ii > Ij^. But if R has a great value, like that with 
 which we are dealing, it approximates to the actually existing 
 inequality l>k ; and accordingly it may be considered proved that, 
 other things being equal, the irreciprocity increases with the 
 length of the tract of organ traversed. 
 
 9. The question, whether the appearance of irreciprocal 
 conduction in the organ depends upon polarisation or con- 
 duction, cannot be answered by the introduction of additional 
 resistance. 
 
 In the previous paragraph we set out with the hypothesis, that 
 irreciprocal conduction may be expressed by an additive term in the 
 denominator of Ohm^s formula for the heterodromous current 
 strength. This procedure, which has already proved available for 
 oui" purpose, will be justified later ; for the present, we must not 
 forget that the question whether irreciprocal conduction depends 
 upon unequal resistance in both directions, or upon polarisation, is 
 still open. At first sight, the surest way to answer this question 
 seems to be the mode of exj^eriment mentioned already in the 
 previous communication (p. 453), viz. to introduce into the primary 
 circuit, a resistance in addition to that of the organ preparation so 
 great, that the latter shall vanish. In this case, if the difference of 
 the two currents should disappear, that of the polarisations per- 
 sisting, the origin of the former difference would be shown to be 
 the difference of resistance. In an experiment of this nature, in 
 which I used a long narrow tube filled with physiological solution 
 of rock-salt for the additional resistance, the difference of the 
 current strengths in both directions did in fact disappear ; but 
 since, on account of lowered functional activity, the difference of the 
 polarisation currents was also absent, no certain conclusion could be 
 drawn from it. In the experiments of the previous paragraph, the 
 apparent irreciprocity of the conduction was persistent, although 
 the resistance of the preparation was but small in comparison with 
 that of the rest of the circuit. It did not however disappear, and 
 the polarisation was not observed, so that we cannot conclude from 
 these observations that irreciprocity depends upon polarisation. 
 
 I have now carried out several experiments according to the 
 method under consideration, with apparently good result. As liquid 
 resistances are difficult to graduate and to estimate, I replaced the 
 tube containing rock-salt solution by two plug rheostats of Siemens 
 
LIVING TORPEDOS IN BERLIN. 511 
 
 and Halske, each of which has a resistance of 10,000 Siemens 
 
 Units, and the coils of which are wound so as to be free from 
 
 induction, as this mig-ht perhaps not be without importance here. 
 
 By closing and opening" a rubbing- key which bridged the rheostat, 
 
 its resistance was excluded or admitted at pleasure. I so conducted 
 
 the experiment, that I sent through the preparation alternately a 
 
 current of few Groves without the great additional resistance and 
 
 one of many Groves with this resistance, and repeated it in homo- 
 
 dromous and heterodromous direction alternately; the number of 
 
 the Groves and the additional resistance being so arranged, that the 
 
 current without resistance did not too much exceed that with 
 
 resistance. Series such as the following were obtained, in which 
 
 the notation requires no further explanation^. There were 50 turns 
 
 for the galvanometer P, and 5000 turns for the galvanometer S at 
 
 zero distance. 
 
 I. 
 
 2. Torp. Fresh. L. 19mm. OC + o-oi28. 1' — Per. 
 
 CT o".o764. 
 
 X.— /S. I -4io| +155 I -2iot +20 II XXX.— I +59 I -32 t +29 I -10 11 
 
 P. \ 114 I 143 I 115 I 141 II I 75 i 74 I 78 I 77-5 II 
 
 Without additional resistance. With resistance. 
 
 ■ I -63 t +9 I -51 t +8 
 4. 118 I 143 I 118 I 149 
 
 Without resistance. 
 
 II. 
 
 New Preparation. 0C + 0-0224. i' — Per. 
 The same period of closing. 
 
 X.— 5. t +100 I -92 t +141 I -31 ll XXX.— t +80 I -55 t+ 52 I -37 II 
 P. I 187 i 146 I 184 i 148 11 I 82 i 80 I 82 181-5 II 
 
 Without resistance. With resistance. 
 
 X.— t +67 I -57 t - I -34 
 I 184 i 156 I 188 4157.5. 
 
 Without resistance. 
 
 III. 
 
 7. Torp. Second day. OC reversed, — 0-0031. \' — Fer. 
 
 The same period of closing. 
 
 X._^. t +88 I -35t +70 I -65 II XL.-t +39 I -63T +^5 | -77 || 
 
 134 i 189 I 119 i 94 II I 92 i 88 i 90.5 i 89-5 
 
 Without resistance. With resistance. 
 
 X.— t + 26 I -42 t ( -Trace | -38 
 I I I i +7 I 
 
 1 105 i 95-5 I 104-5 i 94-5 
 Without resistance. 
 
 These series fulfil the condition laid down, with sufficient exact- 
 
 * For explanation see Appendix to No. XII. 
 
512 LIVING TORPEDOS IN BERLIN. 
 
 ness. In spite of the large amount of persistent polarisation, which 
 is in favour of the homodromous and opposed to the heterodromous 
 current, the additional resistance almost effaces the difference of the 
 two currents. Accordingly it seems proved that the irreciprocal 
 conduction rests upon resistance. Unfortunately, closer con- 
 sideration showed that this time also, the mode of viewing the 
 matter was not complete, and that it is impossible to decide between 
 polarisation and resistance by this manner of experiment. 
 We will call as before : — 
 I the strength of the homodromous, 
 /, that of the heterodromous current ; 
 
 P the constant, which multiplied into the current density, 
 measures the electromotive force of the homodromous ab- 
 solutely and relatively positive polarisation for the unit length 
 of the tract traversed ; 
 n the con-esponding' constant for the relatively negative polar- 
 isation caused by both the currents ; 
 R the resistance of the circuit containing the battery and the 
 galvanometer, exclusive of the resistance of the battery itself ; 
 01 the small, 
 
 N the large number of Groves ; 
 G the electromotive force, 
 r the resistance of a single cell ; 
 W the additional resistance ; 
 
 o- the specific resistance of the organ when it is traversed longi- 
 tudinally in both directions ; 
 I the length of the tract traversed, i. e. in this case of the pre- 
 paration itself, provided that the resistance of the skin is to 
 be included in R ; finally, 
 q its transverse section. 
 Then, with the small number of Groves and without additional 
 resistance, we have 
 
 nG-l.^{U-P) 
 
 ^= ^—iT-' ^"^^ 
 
 nr + R-\ 
 
 nG-l.^U 
 nr + R-i 
 
LIVING TORPEDOS IN BERLIN. 513 
 
 The two polarisations are here both taken to be proportional to 
 the current density as in the first communication (pp. 455). It is 
 self-evident that they are also proportional to the leng"th of the 
 preparation. 
 
 The question whether we can decide between the two possibilities 
 in this way, is answered in the negative by the very statement of 
 the above formulas. More than fifty years ag-o, when treating- 
 of the similar question as to the existence of a transition resistance 
 besides the polarisation of the electrodes, G. S. Ohm directed atten- 
 tion to the fact that in such expressions as id) and (e) the way in 
 which polarisation and resistances severally enter is only apparent, 
 that polarisation therefore stands in the same relation to an ad- 
 ditional resistance as the constant resistances do, and that it cannot 
 be distinguished from these by the introduction of an additional 
 resistance. 
 
 We find in fact 
 
 nG nG 
 
 1 = ; > 1, = 
 
 I ' -I — I » 
 
 /ir + ^+ -(a + n — P) «r + 7? + -(o- + n) 
 
 and if for sake of brevity we put 
 
 ,ir + 72+ -((r + n) = A„ 
 
 r 
 
 nG nG 
 
 _ nGlP 
 ^-^' - K{qK-lP) ^J^^ 
 
 This expression is finite and jDOsitive, so long as <^A„ >/P; qA^=lP 
 would make /infinite, and ^A„< IF would give the upper hand to 
 the heterodromous cui*rent. By introducing the additional resist- 
 ance and exchanging n for N, [f) becomes 
 
 NGIP 
 1 If — 
 
 {AN+W){q{A^+rF)-lP) 
 
 When W is great, this expression approaches to the value 
 
 NGIP 
 
 ' ~ qjr- ' 
 
 Ll 
 
514 * LIVING TORPEDOS IN BERLIN. 
 
 i. e. to zero, and hence these formulas, in which the irreciprocal 
 resistance is left out and regard had only to polarisation, sufficiently 
 represent the facts of the case, as now presented to us. 
 
 The result comes under the same form, if we put the positive 
 polarisation P independent of / instead of proportional to it ; this 
 corresponds better with the hypothesis, according- to which it is 
 only the after-effect of a shock, for in this case we have simply 
 
 -'■ ^1 A 
 
 without additional resistance, and 
 
 11 
 
 I~l.= 
 
 \n+W 
 
 with such resistance and many Groves. The first expression is 
 positive, the second approximates to zero as W increases. 
 
 Consequently, according to the observations up to the present 
 time, in spite of all appearance, it is not necessary to assume 
 irrecipi'ocal resistance in order to explain how the fact stands. It 
 is so improbable, that the absolutely and relatively positive polari- 
 sation by the homodromous current could effect an increase of 
 this current in comparison with the heterodromous in the ratio of 
 lOO : 13*3 (see above, p. 507), that the decision sought for cannot be 
 arrived at in this way, and we must seek for it in another experi- 
 mental field. 
 
 10. Measurements of the resistance of the electrical organ 
 of the torpedo. 
 
 Quite apart from all other questions attaching to the subject, it 
 was naturally of great interest to ascertain something about the 
 conduction of the organ in comparison with the other tissues. The 
 mode of procedure was given by the experiments which I performed 
 in 1 87 1 in connexion with the resistance of the muscle of the frog, 
 that of physiological rock-salt solution and of clay prepared with it^, 
 in which I made use of the contrivance employed first by Ranke, 
 viz. to enclose the tissue, liquids, etc. in glass tubes of equal 
 dimensions, in order to give them a prismatic form, equal length 
 
 ' Gesammelte Abhandlungen, vol. ii. p. 373 f. 
 
LIVING TORPEDOS IN BERLIN. 
 
 515 
 
 and equal transverse section^. As regards the details of the method, 
 I refer to the description g-iven at that time. But in order to 
 make it more readily understood, and inasmuch as the apparatus 
 in all its simplicity has proved well suited to our object, I add a 
 diagram of it in Fig. 21 in a somewhat more perfect form. The 
 new apparatus differs in some points from my original model. It is 
 unessential that the funnels, which, filled with physiological clay, 
 connect the base of the tube containing the tissue etc. with the 
 leading-in pads, are made of vulcanite instead of gutta-percha, and 
 are supported by brass pillars well lackered ; it is of greater im- 
 portance that the diameter of their apertures very little exceeds the 
 thickness of the pads ; hence the side of each cone is still only 
 13-5 mm. long. Their resistance is thereby considerably diminished, 
 and this is advantageous, in so far as it is desirable that the resist- 
 ance of the rest of the circuit should be as small as possible in 
 
 Fig. 21. 
 
 comparison with that of the conductor of which the resistance is to 
 be determined. In the figure, it is easy to recognise the funnel 
 filled with clay, drawn on the left in elevation and on the right 
 in section, between the leading-in vessels and the resistance tube 
 fixed between the funnels. 
 
 The tubes previously used, 25 mm. in length, with an internal 
 diameter of 4-6 mm. are exactly suited for the M. gracilis of a frog, 
 32 cm. in length from the tip of the nose to the longest toe, a 
 medium size readily found in this country^. I still had the very 
 same tubes in my possession ; it was found that they had also the 
 length and breadth suitable for organ preparations. In order to 
 introduce these into the tubes so as to fill them completely, I drew 
 
 * Joh. Eanke. Tetanus. Eine physiologische Studie. Leipzig, 1865, p. 44. 
 ^ TJeber secundar-elektromotorische Erscheinungen. Sitzungsberichte, 1883, vol. i, 
 P- 355- 
 
 l1 2 
 
516 LIVING TORPEDOS IN BERLIN. 
 
 a thread throug-h the dorsal or ventral skin of the preparation and 
 pulled the preparation into the tube by the thread. Any ends 
 which might happen to project were cut off. I often helped to get 
 it into the tube by sucking. In doing this, it happened that 
 I sucked the preparation quite through the tube into my mouth. 
 To my surprise, I found that the organ was quite insipid, 
 without a trace of the salt taste I had expected in accordance 
 with Boll's statement on the Liquor cerebrospinalis of the Tor- 
 pedo^. 
 
 According to Boll's estimate, the tissues of the Torpedo con- 
 tain a salt solution of 3.5 per cent., so that only a solution of 
 sodium chloride of this strength would be ' ph ysiological ' to them. 
 According to Leon Fredericq the blood of the Octopus vulgaris 
 and Astaais marinus contains about four times more salts than 
 that of mammals. Hence I suspected that the tissues of marine 
 animals, especially of sea fishes, would be richer in salt than those 
 of fresh- water animals, and that they would also conduct better 
 accordingly. They would thus be more strongly affected by the 
 current curves with which a Torpedo shock fills the sea around it, 
 than if they had only the same proportion of salt as fresh-water 
 animals, for these curves are attenuated in a badly conducting body 
 plunged in the sea^. That it would be of service to the Torjjedo if 
 its own organ conducted better, seemed quite obvious. But the 
 state of the case is otherwise. 
 
 First of all, it was found that the tissues of sea fish are not per- 
 ceptibly Salter than those of fresh-water fish. It could not be 
 concluded with certainty from Atvvater's and Konig's tables that 
 sea fish are richer in salt, and Weyl found that the organ of the 
 Torpedo yielded only a little more ash than the muscle flesh of 
 river fish (p. 437)^. But Fredericq had observed what had 
 escaped me, that the characteristic which he had recognized at the 
 date of my first communication in the invertebrate marine animals, 
 is absent in sea fishes — Trachinus spec, Solea vulgaris, G. Aeglejinus 
 (a shark) — and he was thus led to regard the circumstance that the 
 amount of salt contained in the fish is so independent of the 
 medium in which they live, as an indication of their higher organi- 
 
 ^ Monatsberichte der Akademie, 1875, p. 710; Archiv fur Anatomie, Physiologic, 
 1875, p. 463, note. 
 
 ^ Untersuchungen, &c., pp. 133, 415. 
 
 ^ Comp. also Alm^n, in Maly's Jahresbericht iiber die Fortschritte der Thier- 
 Chemie, vol. vii. 1877, p. 308. 
 
LIVING TOKPEDOS IN BEELIN. 517 
 
 sation as compared with that of the invertebrates^. Boll, who 
 judged merely by the taste of the Liquor cerehro spinalis^ was perhaps 
 deceived by the amount of urea discovered by Frerichs and Stadler 
 in the juices of the tissvies of the skate ^. 
 
 However that may be, the organ of the torpedo, far from con- 
 ducting particularly well, does not conduct as well as frog's muscle 
 in the longitudinal direction. In the following statements, i ac- 
 cording to the index attached to it, denotes deflections in scale divi- 
 sions of the aperiodic mirror (with 5000 turns at 100 mm. distance) 
 by the opening shock of the inductorium, when the primary coil was 
 filled with rods and the secondary coil pushed up. The resistances 
 indicated were placed between the clay cones, i^ denotes that re- 
 sistance between the clay cones was nil; //^, «V?oj \w ^re the deflections 
 with the tube full of | per cent, rock-salt solution, full of tap-water 
 and full of sea- water respectively ; i^ is the deflection with the tube 
 full of frog muscle ; finally, ?/,o and /^^ are the deflections with the 
 organ in the tube traversed by the homodromous and heterodro- 
 mous current respectively, v)„^ wj^^, w^^^ w^, wj^,, and Wj^^ are the cor- 
 responding specific resistances refen-ed to the f per cent, rock-salt 
 solution wjg as unity. 
 
 A first series of experiments were performed last summer on the 
 second Torpedo, on the third day, still using the original apparatus 
 which had served for the experiments of 1871 already described. 
 The three organ preparations, the resistance of which was deter- 
 mined with homodromous and heterodromous direction of current, 
 showed the organ current in the right direction and of considerable 
 strength. The result of five determinations in both directions, 
 differing but little from each other, gave a mean of 
 
 io = 667, ijg = 49-2 se. 
 
 In this condition, without tube, or with the tube full of physio- 
 logical rock-salt solution, the apparatus showed no perceptible 
 irreciprocity of conduction. 
 
 The organ preparations gave respectively: 
 
 I. V = 36-0, ...?7,e = SI-O- 
 IL i,,= 3i.2, ...?,^= 23-0. 
 III. /,„=29-o, ...?,, = 8-8. 
 
 * Bulletins de TAcad^mie royale des Sciences . . . de Belgique, 3™® Se'rie, vol. iv. 
 No. 8, AoUt, 1882, p. 209. 
 
 ^ Erdmann's und Werther's Journal fiir praktische Chemie, 1858, vol. Ixxiii. 
 p. 48. 
 
518 LIVING TORPEDOS IN BERLIN. 
 
 Thence we find, putting- wjg = i, according to the formula 
 
 w 
 
 hO 
 
 1 I 
 
 I II 1 
 
 ne ■■ ^0 '■•T~ l'- 
 
 ■ ^he ^0 ' ijg h 
 
 I 7^> = 2-397,., 
 
 -^^'/,. = 3-238. 
 
 II. ?,v = 3-199. • 
 
 ■-'he= 5-342. 
 
 III. Wj^o = 3-^55, ■■■ 'he = I8-50- 
 
 This very surprising result induced me to determine again the 
 resistance of frog muscle in comparison with the physiological rock- 
 salt solution, and on the same occasion, that of the artificial sea- 
 water of the Berlin aquarium, which had a specific gravity of 1.0250 
 at 22-75° C, and was thus apparently somewhat higher than on 
 the occasion of Prof Christiani's determination of the conducting 
 power of sea- water ^. This time 
 
 io = 60.9 sc, ijg = 48-0 sc, ?,„ = 40.0, z,^ = $6-^ sc, 
 
 whence resulted 
 
 ?«m = 1-9383' '^sio = 0.3202. 
 
 I had previously (1871) found w,^ = 1-9045; the difference 
 amounts to -V ; a closer agreement is scarcely to be expected in 
 determinations of the kind. 
 
 I undertook a second similar series of experiments t)n the 
 third Torpedo, on the second day of experimenting. I omit the 
 observed deflections, and give at once the values calculated from 
 them, that is to the specific resistances of three other preparations, 
 through which currents were led homodromously and heterodro- 
 mously. These exhibited powerful organ currents. Their specific 
 resistances are stated below in terms of that of physiological salt 
 solution : — 
 
 IV. w,o = 3-5983, ." 'he = ^3-55^7' 
 
 V. %V = 3-3741, ... w,^ = 6-2752. 
 
 VI. jv^o = 3-7781, ... w^., = 6-2752. 
 
 On the same occasion, I determined the resistance of the tap- 
 water wifp at 37-955. It was interesting to consider how these 
 determinations of sea- and tap-water would agree with those 
 obtained by Prof Christiani by another method, and which are 
 
 ^ Untersuchungen, &c., p. 413. 
 
LIVING TORPEDOS IN BERLIN. 519 
 
 communicated in the book on the gymnotus loc. cit. In Prof. 
 Christiani's experiments the tap- water conducted 126-57 times 
 worse than the sea- water of the Aquarium. Our numbers give 
 
 0-3202 : 37-955 = I •• 118-52. 
 
 A closer agreement is scarcely attainable in such circumstances. 
 
 It is apparent that even in the homodromous direction, in which 
 the organ conducts best, it conducts scarcely half as well as frog 
 muscle parallel to the fibre, and 7-5 to 12 times worse than the sea- 
 water of the Aquarium. The ratio would be still more unfavourable 
 with sea- water from the Mediterranean, which conducts nearly 150 
 times better than tap-water. But in the heterodromous direction, 
 the organ conducts even 30 to 58 times worse than sea- water. 
 
 Our method does not allow of our obtaining corresponding 
 numerical determinations for the transverse conduction of the 
 organ. That the electrical organ does conduct quite reciprocally 
 in a transverse direction, I showed a long time ago in the 
 case of the Malapterurus, where it is easy to cut strips of the 
 organ vertically to the direction of shock. I now endeavoured 
 to ascertain, for the organ of the Torpedo at least, how transverse 
 conduction compares generally with homodromous and hetero- 
 dromous conduction. I cut from the organ square slices as 
 uniformly thick as possible, of which one edge was formed 
 of dorsal skin, the opposite one of ventral skin, and the two 
 lateral edges were formed of lateral surfaces of the columns, and 
 sent induction shocks through them, sometimes from belly to back, 
 sometimes from back to belly, sometimes in transverse direction 
 from one lateral edge to the other. In the homodromous direction, 
 the deflections were, for example, 80-7, in the heterodromous 22-5, 
 in transverse direction each way 99. The suspicion might arise 
 that the semblance of better conduction in the transverse direction, 
 is the result of the skin on the dorsal and ventral edge of the slice 
 increasing the resistance, when it is traversed longitudinally. But 
 when I covered the lateral edges with pieces of skin and sent the 
 current through these, the deflection was still 95 sc. ; after removing 
 the artificial covering of skin the deflection was the same, 
 the homodromous on the other hand being now only 75, the 
 heterodromous 18. It thus seems as if the organ conducts in 
 transverse direction, even better than in homodromous. However, 
 I consider this result as not yet certain, on account of the defective- 
 ness of the mode of experiment. 
 
520 LIVING TORPEDOS IN BERLIN. 
 
 11. Dependence of the irreciprocal resistance and of the resist- 
 ance of the organ in general upon its vital condition. 
 
 The small conducting power of the organ and its irreeiproeity are 
 closely connected with its vitality. If the tube was filled with columns 
 which projected at both ends, and plung-ed in boiling water for 
 several minutes, in consequence of which the organ acquired an acid 
 reaction, it remained filled with organ in spite of the shrivelling of 
 the columns by the boiling. I have already stated that boiled 
 organ conducts better than living organ (p. 458). I am now 
 able to add, first that its resistance becomes exactly the same in 
 both directions, secondly that after cooling it sinks below the 
 minimum resistance in the homodromous direction, and is even 
 smaller than that of physiological salt-solution. Compared to this 
 as unity, I found it to be only = 0-9294. 
 
 The course of events is similar when the organ dies a natural 
 death, only it is evident that a longer period is required. Here, 
 also, homodromous and heterodromous resistance approximate to 
 equality, and homodromous resistance diminishes. Naturally the 
 dead organ, especially in a more advanced stage of decomposition, 
 is ill-fitted for preparations which have to be drawn into the resist- 
 ance-tube. Preparations taken from such an organ, as I remarked 
 long since (p. 439), have a tendency to assume an hour-glass 
 form between the more compact dorsal and ventral skin which 
 covers their polar surfaces. However, it is possible with some 
 patience to obtain determinations like the following. 
 
 The organ of the seventh Torpedo, kept at a low temperature, had 
 an indistinctly acid reaction fifty-five hours after death. It had 
 already a smell of ammonia, and thus the acid was probably 
 neutralized by the ammonium carbonate arising from the decom- 
 position of urea in the tissues, as might be proved by the reaction 
 being acid only here and there in the muscles. The preparations 
 acted feebly in homodromous direction ; however, this could hardly 
 be the organ current, but rather an electromotive action produced 
 by an acid and an alkaline liquid accidentally coming in contact 
 with each other in the interior of the organ. A first preparation 
 gave 
 
 ^'ho= 2-968, ...W^^ = 3-231. 
 
 Thus the irreeiproeity amounted to only eight per cent, of the 
 homodromous current strength. But in another preparation, it was 
 
LIVING TORPEDOS IN BERLIN. 521 
 
 directly afterwards entirely absent ; the resistance in either direction 
 was as follows : — 
 
 «V = «^;,. = 2•3I7• 
 On the following day, seventy-six hours after death, with evident 
 decomposition, no difference of homodromous and heterodromous 
 conduction could be any longer perceived, and I found in two 
 preparations 
 
 «';>o = 'the = 1-0826 = 0-8553. 
 
 Thus in the latter case, the resistance was less than that of physio- 
 logical salt-solution, and even of the boiled preparation. 
 
 It may be asked, whence may the considerable variations in the 
 heterodromous resistance of the organ arise ? It is probable that the 
 largest numbers found for it are the most accurate, and the small- 
 ness of the other values depends upon a double cause. First, it is 
 undoubtedly injurious to the functional activity of the columns, that 
 they are squeezed by drawing them into the resistance -tube. 
 Secondly, it is certain that, in doing this, some if not all of the 
 plates must lie more or less obliquely in the tube, and hence the 
 heterodromous resistance, which is connected with the perpendicular 
 direction of the curi'ent in regard to the plates, is necessarily 
 lowered. 
 
 It would have been desirable to determine the conductivity of 
 other tissues of the Torpedo also, especially of its muscles. But, in 
 the first place, I have not hitherto found in the Torpedo any 
 muscle which is available in any measure as a regular muscle, 
 comparable to certain thigh muscles of the frog, the sartorius of 
 the dog, and others. Secondly, at the commencement of the experi- 
 ments, I had to turn my attention to more important questions in 
 the case of each fish, and in the few cases, when I intended to 
 experiment upon the muscles, I found them already more or less 
 functionally inactive, and indeed with an acid reaction to some 
 extent. This is likewise the reason, why I am still unable to say 
 anything about the electromotive force of the muscles of the 
 Torpedo. 
 
 12. The conduction of the electrical organ when investigated 
 by means of constant currents. 
 
 As regards the electrical organ, it seems impossible to doubt any 
 longer, according to the above, that it conducts irreciprocaUy. If 
 there is no such conduction, we are compelled to attribute to the 
 
522 LIVING TORPEDOS IN BERLIN. 
 
 absolutely and relatively, positive, homodromous polarisation an 
 electromotive force which may amount to that of twenty Groves 
 (pp. 453, 454)- If such a force were the cause of the superiority of 
 the homodromous current, how could it be explained that eveft in 
 homodi-omous direction, the preparation conducts worse than physio- 
 logical salt -solution, or even than frog's muscle? It cannot 
 be maintained that the smallness of the current strength as com- 
 pared with the tremendous homodromous force is dependent on bad 
 conduction, for even the heterodromous conductivity is not bad 
 enough for that. Besides this, if the superiority of the homo- 
 dromous current dej^ended upon polarisation, it would be impossible 
 to conceive how on the death of the organ, or on boiling it, the 
 conductivity for both currents becomes the same and then absolutely 
 increases, so that it even exceeds that of physiological salt-solution. 
 The enfeeblement of the current cannot therefore be a consequence 
 of the falling off of an electromotive force in the same direction 
 with it. 
 
 It seems, as has already been said, impossible to escape the weight 
 of these reasons, and the existence of irreciprocal conduction in the 
 electrical organ is to be regarded as an established fact fi'om hence- 
 forward. However, a considerable change in our conception of this 
 remarkable property of the organ is to be looked forward to. 
 
 I now recollected that I had carried out these experiments, other- 
 wise to all appearance conclusive, with induction shocks only. It 
 occurred to me that in the series of obsei^vations mentioned above, 
 p. ^o^, in which the current of a Grove's battery was closed for a 
 longer period through the preparations, the superiority of the 
 homodromous current continued indeed perceptible, but by no means 
 so considerable as in the experiments with induction shocks, or with 
 battery currents of short duration. I now began to suspect that the 
 irreciprocity of conduction must depend, not only on the density 
 but also on the duration of the current, and this was found to be 
 the case. 
 
 Instead of the secondary coil as in the previous experiments, a 
 Grove's battery was placed directly in the circuit of the gal- 
 vanometer, on the mirror of which, only 45 turns acted from a 
 suitable distance. The battery current was closed through the 
 mercury key by hand for any desired period as timed by the clock. If 
 for a short time only the contrivance mentioned above was employed 
 (p. 481, note). When required, it was exchanged for the secondary 
 coil of the inductorium, which was pushed quite up to the primary 
 
LIVING TOKPEDOS IN BERLIN. 523 
 
 coil filled with rods. The preparations lay upon the triangular 
 g-lass-plate, with ventral and dorsal skin between the clay shields of 
 the leading'-in vessels. In the following series of experiments, 
 ' long- closing- ' signifies that the Lattery was kept closed as long as 
 was possible, without too much interference by heating, polarisation, 
 etc. All the deflections are reduced to the magnitude which they 
 would have had with the 45 turns at zero distance. 
 
 First Series. 
 
 6. Torp. Fresh. 
 
 Grove's Battery. 
 
 Long closing. 
 
 L— t 223 i 223 I 223. 
 
 There is no appearance of any irreciprocity of conduction. 
 
 Long closing, 
 v.— i 1128 t 1159 i 1140 t 1177. 
 Certainly some irreciprocity, even though feeble. 
 
 Long closing. 
 
 XX.— t 4439 i 4658 t 5098 1 4927- 
 — 219 —440 +171. 
 
 A disturbance due to heating of the preparation and of the clay 
 
 shields is observable, so that irreciprocity makes its appearance 
 
 evident only in the series of diflPerences. 
 
 Induction Shocks. 
 
 t 15 4 3; 3 t 15; 15- 
 This shows, that when the preparation is treated with shocks 
 of short duration only, the irreciprocity of conduction is manifested 
 in full extent. 
 
 Grove's Battery. 
 
 Long closing. 
 
 XX.— t 4756 i 4781 t 4903 I 4890. 
 
 -25 -122 +13. 
 
 Again, irreciprocity shows itself in the series of differences 
 
 only. 
 
 C. T. o".03io. 
 XX.- i 86 t 95 i 84 t 95. 
 With short closing, it appears distinctly at once, although less 
 decidedly than with opening shocks. 
 
 Long closing. 
 XX.— i 3927 t 4318 i 4207 t 4634. 
 
 In spite of collateral efi'ects, it manifests itself this time with long 
 closing, the series of differences not being stated. 
 
524 LIVING TOKPEDOS IN BERLIN. 
 
 The preparation was not displaced throughout the whole series of 
 experiments. 
 
 Second Series. 
 
 6. Torp. Second Day. 
 
 Grove's Battery. 
 
 C. T. o".o3io*. Long closing. C. 2^. o"-o3io. 
 
 I-— t 2-55 i 1-45 t 2-6i i 1-34 il t 3° i 29 f 30 i 30.5 || ]• 2.66 i 1.48 t 2.58 f 1-85. 
 
 Thus even with only one Grove, under favourable circumstances, 
 irreciprocity of conduction shows itself for short closing-, whilst for 
 long" closing" it is only very feebly pronounced, so that in the 
 corresponding- experiment of the first series, it was not perceived at 
 all. 
 
 C. T. o"-03io. Long closing. 
 
 v.— t 20 i 17 t 19-5 i 15 II t 878 4 866 t 902 i 866. 
 C. T. o"'03io. Long closing. 
 
 XX.— t 115 i 95 t 120 i 95 II t 5073 i 5415 t 7073 
 
 -342 -1658. 
 
 The heating of the preparation prevented further observation ; 
 the irreciprocal conduction nevertheless makes itself noticeable in 
 the series of differences. 
 
 The last experiment of the appendix of the first communication 
 (25) is allied to these experiments, as being of similar inter- 
 pretation. There, the current strength in the circuit of a battery 
 of twenty cells closed through a preparation, and at the same time 
 in a side circuit, by which the preparation was bridged, was 
 observed. For short closing, the irreciprocity was very great ; 
 if the closing lasted one second, it was already disproportionately 
 feebler. 
 
 13. Of the law and the nature of irreciprocal conduction 
 in the electrical organ. 
 
 In accordance with the preceding, there can be no doubt, that 
 irreciprocity of conduction is disproportionately greater for short 
 closing of a battery and for opening shocks of an indiictorium, than 
 
 ^ Here, instead of 45 turns at zero distance, 5000 turns were used at a distance of 
 30 mm. The deflections are diminished in the ratio 45/5000, and are reduced to 
 the distance zero of the coil (comp. 454). 
 
LIVING TORPEDOS IN BERLIN. 
 
 525 
 
 for longer closing, and that the numerical determinations of § lo, 
 in as far as they refer to heterodromous conduction in the organ, 
 are only correct for opening induction shocks. The two-fold 
 dependence of irreciprocal conduction upon density and duration of 
 current may now be represented graphically in the following 
 way. 
 
 In Fig 11 the A-axis of the current densities runs from the zero 
 point forwards and to the left, and the I'-axis of the times of 
 closing, forwards and to the right. The abscissae independent of 
 
 W^R 
 
 Fig. 22. 
 
 each other on both axes are measured arbitrarily. The homo- 
 dromous resistances W and the heterodromous resistances R are 
 expressed as ordinates on the I'-A-plane. The homodromous re- 
 sistance is independent of current density and duration, and thus 
 the ordinates representing it, have throughout the same arbitrarily 
 chosen height Ow^^ and their tops all lie in the plane w-^, ^, 3, 4 
 parallel to the T-A-plane. The heterodromous resistance is not 
 perceptibly different from the homodromous resistance, below the 
 current density A^ ; the irreciproeity is here apparently zero. 
 Whether it is so in reality, is not yet decided, for I have unfor- 
 
526 LIVING TORPEDOS IN BERLIN. 
 
 Innately postponed the repetition of the experiments with feeble 
 opening" shocks with a galvanometer of greater sensitiveness, and 
 it is possiljle that a difference, too small to be detected by the 
 g-alvanometer of the sensitiveness employed above (p. 517)) might 
 thus be proved to exist. However that may be, it may be assumed 
 for the present, that to the right of and above the line A, (the 
 threshold line) with the means of observation actually used, the 
 tops of the ordinates, representing- the heterodromous resistances, 
 seem to lie in the plane r^, j, 5, 4 parallel to the ^-A-plane. But 
 to the left of this line, the heterodromous resistance becomes so 
 much the greater, as the density is greater and the time of closing 
 is less. Thus with increasing density and decreasing closing- 
 time, the tops of the ordinates r rise above the plane w^^ ^, 3 , ^ , 
 so that they lie in the surface of complex curvature ^'i , 2 > 3 > 4 • I^ is 
 to be assumed that the heterodromous resistance, with increasing 
 current density, approaches a Kmit (see above^ p. 505) ; this is 
 expressed by the course of the curve of heterodromous resistance 
 for minimal duration of closure, which as it nears r^ becomes 
 concave towards the axis. With increasing duration of closing, it 
 approaches an inferior limit, as is expressed by the curve r^, ^, 
 w^hich unites itself asymptotically with iv^,^ ; that is with a line 
 parallel to the time axis ; there will however be somewhat to 
 remark upon this later on. 
 
 This gives upon the whole the main features of the law of the 
 phenomenon. The mode of formulating irreciprocity (pp. 504, 505) 
 seems now to be sufficiently justified. It is at present impossible 
 to form a satisfactory conception as to the cause of the phenomenon. 
 At first the idea may occur, that it is somewhat similar to the 
 internal secondary resistance, which I have described in the case of 
 a porous moist conductor, in which the internal resistance increases 
 with the current density^. The analogy between the two phe- 
 nomena is however scarcely a true one. The internal secondary 
 resistance has this indeed in common with heterodromous re- 
 sistance, that boiling temperature destroys it. But disregarding 
 the fact that hitherto it has been observed in living tissue of 
 plants onlv, it is independent of the direction of the current, it 
 demands longer time of closing for its development, and, when once 
 developed, longer time for its disappearance, whilst, on the contrai-y, 
 heterodromous resistance is j)resent to its greatest extent at the 
 
 * Gesammelte Abhandlungen, vol. i. pp. 90 f., ii6f. 
 
LIVING TORPEDOS IN BERLIN. 527 
 
 first moment, falls with long-er duration of the current, and 
 shows itself again in full force when the current is again closed 
 for a short time. 
 
 It seems rather, that we must suppose that the heterodromous 
 current, as soon as its density surpasses a certain value, the 
 threshold value in the preparation, meets with a special resistance, 
 and that a certain time is required to overcome it. This would 
 explain why irreciprocity of conduction is more strongly developed 
 with current impulses of short duration, while comparatively little 
 is left if the current is of longer duration. 
 
 The small amount of irreciprocity when the flow is continuous, 
 may perhaps not be dependent on the same cause as the extreme 
 iri'eciprocity with currents of minimal duration. For a remark 
 made in the first communication (p. 456) ought not to be forgotten 
 here. Inasmuch as there is no doubt of the existence of homo- 
 dromous positive polarisation, it must be regarded as a concomitant 
 cause of apparently irreciprocal conduction, even though this is 
 actually traced to an unequal resistance in the two directions. 
 But when the time of closing is long, the irreciprocity becomes so 
 small, that we have no longer the same ground for holding it 
 improbable, as at first sight seemed necessary, that it is solely due 
 to polarisation, that ground being the enormous electromotive force, 
 amounting to many Groves, which that explanation obliged us to 
 ascribe to polarisation. When the time of closing is long, the 
 difference of current strengths, so far as the disturbances due to 
 heating etc. allow of its being estimated, amounts to a few per- 
 centages, so that with twenty Groves in the circuit, a polarisation 
 force of less than one Grove would suffice to explain the irreciprocity. 
 If this view of the matter should be confirmed, then irreciprocity 
 of conduction might be entirely connected with short duration 
 of current. The cm-ve ^2,3 in Fig. 23 would then merge into 
 the straight line w^, 3, instead of a line above it and parallel 
 to it. 
 
 It is not necessary to say that the mechanism of irreciprocal 
 conduction in the long direction of the columns remains in com- 
 plete obscurity. "We must not enter here, upon what is already 
 known as to irreciprocal conduction in galvanic and induction 
 circuits, all of which is to be found collected in Prof. Christiani's 
 book^. We may state as probable that it is not nerves, vessels, or 
 
 * Beitriige zur Electricitatslehre. Uber irreciproke Leitung electriscber Strome, 
 &c., Berlin, 1876. 
 
528 LIVING TORPEDOS IN BERLIN. 
 
 connective tissue to which this property belongs, but the electrical 
 plates themselves. It cannot depend upon Boll's striations, because 
 besides the difficalty of understanding how this could be the case, 
 the malapterurus plates are striated on both surfaces ^. Since death 
 and boiling heat put an end to irreciprocity, it seems as if it must 
 be connected in some way with the function of the plates. In the 
 first communication (p. 453) I endeavoured to explain the worse 
 heterodromous conduction, by supposing that the heterodromous 
 current is almost or quite incapable of rotating the hypothetical 
 electromotive molecules in such a way, that its positive pole may 
 face the negative surface, which in the torpedo is the ventral 
 surface of the organ. At first sight, it seems in accordance with 
 this supposition, that the resistances become the same in both 
 directions in the dead organ, in which the hypothetical molecules 
 have ceased to act. But the difficulty arises that they decrease 
 at the same time, so that they become less than the smaller of 
 them was during life. If the electromotive molecules, by virtue of 
 the different electro-chemical nature of their poles, were considered 
 in electrolytic conduction as molecules made up of electro-positive 
 and electro-negative substances according to Grotthuss' theory, then 
 we should expect, on the contrary, that the dead organ would conduct 
 worse in both directions, than the living organ in the hetero- 
 dromous direction. It is, however, no way out of the difficulty to 
 say that the organ by itself, without the aid afforded by the electro- 
 motive molecules, conducts as badly as it does when traversed by 
 heterodromous currents during life, and that owing to alteration in 
 the dead body, formation of acid, etc., it conducts better than when 
 traversed by homodromous cun-ents during life. 
 
 14. Teleology of irreciprocal conduction in the electrical 
 
 organ. 
 
 We have been frequently reminded in the course of these studies, 
 of the difficulty which the older physicists and physiologists, from 
 Volta and Nicholson to Faraday and Valentin, found in explaining 
 the actions of electrical fish, without insulating septa ; these they 
 supposed placed laterally about the organs, or that they came 
 into existence at the moment of the shock. The first fruit of my 
 labours in this department was the removal of this error, more than 
 
 ^ Untersuchungen, &c., pp. 291, 391, 392. 
 
LIVING TORPEDOS IN BERLIN. 529 
 
 forty years ago. For histological reasons, I doubted the existence, 
 and even the possibility, of such investments, and I proved, at first 
 only theoretically, but later by experiments with schemata, that 
 multiplication of the electromotive elementary action in the sur- 
 rounding conducting medium does take place, although a battery 
 so formed must be considered as imperfect^. 
 
 We now learn, that in the electrical organ there exists a sort of 
 conduction of which the counterpart is hitherto nowhere known. 
 Every fresh characteristic of this wonderful structure must excite 
 attention, inasmuch as the hope of attaining a glimpse of its 
 electrical mechanism is bound up with it. But irreciprocal con- 
 duction has a very special interest, inasmuch as it appears to 
 perform for the organ, a function similar to that of insulating 
 coverings, and it would even seem to accomplish this much more 
 completely. The very fact that we succeeded so little in acquiring 
 an understanding of the essence of this property, tends to enhance 
 the gratification which we feel in the discovery, that it plays a not 
 unimportant part in the economy of the shock of the electrical fish. 
 The following consideration will make this intelligible. 
 
 Let us imagine first, a single column of the Torpedo's organ, 
 surrouDded by an unlimited conducting mass. For the present 
 we will disregard all distinctions of ordinary and irreciprocal re- 
 sistance and of polarisation, and picture to ourselves the column as 
 constantly active. It is not difficult to give a general idea of the 
 system of current curves, with which the column fills space. In the 
 first place, it is evident that this system is distributed symmetrically 
 about the axis of the column. Further, it is sj^mmetrical also in 
 regard to the plane which cuts it in half, and which is perpen- 
 dicular to the axis. To make the subject more intelligible, we will 
 call this plane the transverse median plane, the positive dorsal end 
 of the column will be called the superior, and the negative ventral 
 end the inferior. 
 
 Current curves issue from the positive dorsal surface of the 
 column, and also from points of the lateral surface above the 
 median transverse plane. All the curves cut the median transverse 
 plane oo perpendicularly; this plane is in fact the isoelectrical 
 surface of which the tension is o. The curves issuing from the 
 
 ^ Vorlaufiger Abi'iss einer Untersuchung tiber den oogenannten Froschstrom und 
 fiber die elektromotorischen Fische. PoggendorfF's Annalen, 1843, vol. Iviii. 
 p. 25 f. Gesammelte Abhandlungen, vol. ii. p. 678 f. Untersuchungen, &c., pp. 259, 
 284, 300, 
 
 M m 
 
530 
 
 LIVING TOEPEDOS IN BEKLIN. 
 
 lateral surface, re-enter it below the transverse median plane, and the 
 curves starting from points of the dorsal surface, enter at cor- 
 responding points of the ventral surface. The prolongation of the 
 axis is a current curve which loses itself in infinity. In the 
 interior of the column, currents flow from the belly to the back, 
 
 Fig. 23. 
 
 those near the axis being nearly parallel to it, those nearer the 
 lateral surfaces, diverging from each other, so that the curves cut 
 the lateral surfaces. To the left of the figure, distorted for a reason to 
 be mentioned immediately, the diagram represents the intersection 
 of the outer system of curves with the transverse plane of the fish 
 round a single column supposed to be active. 
 
 If we imagine in the unlimited conducting mass, a second exactly 
 similar column by the side of the first column, its axis parallel to 
 that of the first, its dorsal and ventral surface in the same plane, so 
 that the transverse median planes belonging to both columns would 
 coincide, then the second column fills space with a system of curves 
 precisely similar to the first. The cm-rent threads belonging to 
 both the systems, compound themselves in any element of surface 
 whatever of space, according to the parallelopiped of forces, in con- 
 formity with the principle of superposition of currents ^. The case 
 is precisely similar as regards a third, a fourth, or an n^^ column. 
 If the distances of the columns from each other and their mass 
 
 ^ Comp. Helmholtz in PoggendorflF's Annalen, 1853, vol. Ixxxix. p. 212 f. 
 
LIVING TORPEDOS IN BERLIN. 531 
 
 vanish, as compared with the distance of the element of surface 
 under consideration, then the density of the current in the element 
 is doubled by the addition of the second column, trebled by that of 
 the third, and multiplied n times by the addition of the «*^ column. 
 As the distance of the element of surface diminishes^ the density of 
 the current increases in it more slowly with the addition of new 
 columns and in a very complicated manner, but it constantly 
 increases with the number of columns. As each column represents 
 for all other columns, a portion of external space, each one will 
 be met by a portion of the current threads of the others, for the 
 most part heterodromously, and the components of the external 
 currents are deducted in the interior of the column from those of 
 its own current. This is what I call the principle of imperfect 
 pile formation, and thus it is comprehensible, how without insu- 
 lating" coverings, a multiplication of elementary effect is possible in 
 electrical organs. 
 
 Further consideration is simplified by the remark, that since the 
 Torpedo possesses two organs symmetrically placed, it is sufficient 
 to fix the attention on one column symmetrically placed in each 
 organ. The sagittal plane VB of the fish, when extended to 
 infinity, is a surface of flow for the process of flow due to both 
 of these columns. Consequently we can think of it as insulated, 
 without changing anything in the process, or we may also halve 
 the infinite conducting space in this plane, and we require to inves- 
 tigate only what takes place in one half of it. 
 
 It is clear from the above, that all other columns of an organ are 
 traversed heterodromously by the current threads of the one column 
 under consideration, as is shown in Fig. 33 to the left of the sagittal 
 plane. If we fix our attention upon a portion of external space, 
 e.g. the little fish seen in the figure, then the rest of the current 
 threads form a collateral channel of closure for the current threads 
 which meet the little fish. The worse this conducts, so much the 
 stronger is the shock which the fish receives. Consequently, it 
 would receive a stronger shock if the organ were insulated, with 
 the exception of the particular column under consideration. This 
 is represented in the figure, to the right of the sagittal plane, where 
 the shading is intended to signify insulation. Thus, the current 
 threads issuing from the margins of the dorsal surface and the 
 upper half of the lateral surface of the column, can no longer pass 
 through the adjoining columns to the corresponding points below 
 the median transverse plane (which will be almost bent here to a 
 
 w m 2 
 
533 LIVING TORPEDOS IN BERLIN. 
 
 median transverse surface), but are compelled to go round by the 
 edges of the organ. They have therefore to overcome an un- 
 equally greater resistance, and in the same proportion, constitute 
 worse channels of collateral closure, which consequently have less 
 power of weakening the currents which reach the little fish. Thus 
 the insulating character of the organ would tend to increase the 
 action of a conducting and electromotively active column on any 
 part of the external space. The observation that in consequence of 
 the insulating character of the organ the density of the currents 
 flowing round it from the column is increased leads to the same result. 
 This is seen on the right side of the figure, if regard is had to the 
 limitation of the median transverse surface which presents itself 
 as a consequence of insulation. 
 
 The only condition for the legitimacy of these conclusions is, 
 that the resistance of the column should not disappear in comparison 
 with that of external space. This condition in the case of the 
 Torpedo is all the more certainly fulfilled, because, as we saw, 
 the organ even at its best, conducts decidedly worse than sea- 
 water. 
 
 Now the organ is not indeed insulated, but the irreciprocal 
 conduction which we have recognised, performs, as has been already 
 stated, a function similar to complete insulation. The current 
 threads, with the deviation of which into badly conducting channels 
 we are concerned, are all predominantly heterodromous where they 
 meet the organ. Thus they encounter in the organ, a resistance 
 which practically produces the same effect as if the organ were 
 insulated. Each column conducts its own homodromous current 
 comparatively well, but bars the passage to the heterodromous 
 current threads of all the other columns ; and as this is the same 
 for all the columns, the heterodromous current threads are forced 
 to take the circuitous route round the edges of the organ, just as 
 if the organ consisted of a non-conducting substance ; and this 
 remarkable behaviour is the result of irreciprocal conduction. The 
 current of all the columns, and therefore the total current of the 
 organ, increases in density in external space, that is, in physiological 
 effect. 
 
 The increased strength of the shock produced thereby, is at all 
 events more considerable, than that which would have resulted from 
 the insulating property of the fascia covering the organ laterally, 
 which the older investigators imagined to exist, for this would 
 have cut off" only the currents between points of the lateral 
 
LIVING TORPEDOS IN BERLIN. 533 
 
 surfaces of the marg'inal columns, but would not have disturbed 
 the heterodromous current threads in their course through the 
 organ itself. The arrangement which does really exist, is superior 
 even to the insulating property of the cover of the individual 
 columns, for such a property would indeed oblige the currents, 
 between points of the lateral surfaces of all the columns, to take 
 their route along the column, but it would present no obstacle, 
 to the return of the currents from the dorsal to the ventral 
 surface, by the shortest route through the neighbouring columns. 
 On account of the tenuity of the connective tissue sheaths 
 of the columns, the proportion of the flow which equalizes itself 
 through them, instead of through the external conducting mass, 
 need not be taken into account, any more than that which 
 follows the com'se of nerves and blood-vessels. The irreciprocity of 
 conduction must be specially advantageous to the lower marginal 
 columns of the organ, the proper homodromous current of which 
 would otherwise be annihilated by the united heterodromous 
 current threads derived from the median columns, which are about 
 two-thirds higher and proportionally stronger. 
 
 It deserves to be remarked, that on the assumption that the 
 shock is due to the marshalling of dipolar electromotive molecules 
 in the electrical plate, all that has been said about the whole 
 column may be equally applied to the individual molecule. Con- 
 sidering that the transverse section of the course of the homo- 
 dromous cm-rent in the molecule is infinitely small, the resistance, 
 notwithstanding the shortness of this course, is sufficient to 
 justify the application of the above conclusions to the molecule also. 
 
 I endeavoured to give in the first communication an idea of 
 the configuration, in a transverse plane dividing the organs 
 about in half, of the system of current curves surrounding both 
 organs ; it is now necessary to alter this in consequence of irreci- 
 procity of conduction, but I do not venture to say anything more 
 decidedly about it. One thing alone is certain : the part of the 
 animal which lies between the medial edges of the two organs, viz. 
 its central nervous system, is, as was stated, in the course of the 
 strongest flow ; it will be traversed by even denser current curves 
 than was before supposed. The immunity of the Torpedo in regard 
 to electrical shocks (p. 432, 433), now an established fact, makes 
 the difference unimportant. 
 
 What has been said here of the Torpedo, applies also essentially 
 to the Gymnotus and the Malapterurus. It was in studying the 
 
534 LIVING TORPEDOS IN BEELIN. 
 
 Malapterurus, that I first recog-nised the irreciprocal conduction, 
 and traces of it are not wanting in Dr. Sachs' experiments on the 
 Gymnotus. In these animals also, the impenetrability of the 
 organ to heterodromous currents will tell in the same way in 
 strengthening its action on external points. This will however 
 be of special service to the Torpedo, for this fish is at a dis- 
 advantage, as compared with electrical fresh-water fishes, in con- 
 sequence of sea water being so good a conductor. 
 
 I have on former occasions brought under notice, in electrical 
 fishes, surprising" instances of that organic adaptedness, which is 
 ever a new source of wonder even to the strict adherent of 
 mechanical causality. I showed how the construction of the 
 electrical organ of the Torpedo on the one hand, of the Gymnotus 
 and the Malapterurus on the other, is adapted to the different 
 conducting- power of sea and fresh water ^ ; how with their growth, 
 the organs of the Gymnotus and of the Torpedo change precisely 
 in the way in which fresh and sea water demand^; how also 
 the electromotive forces of both fish, according to all measurement, 
 are in the same ratio to each other as the resistances of fresh and 
 sea water ^ ; how the circumstance that the organ is positively 
 polarised by shocks of extremely short duration, tends to strengthen 
 the shock, and how in consequence of the negative polarisability of 
 the organ when tetanised, it is brought about that a fresh discharge 
 starts again almost from zero*. I showed also the indispensable 
 immunity of electrical fishes and their progeny in reg-ard to 
 electrical shocks. Finally, the remarkable mode of terminal 
 branching", described by Wagner in the organ of the Torpedo, 
 seems to be adapted to ensure the simultaneous coming- into action 
 of all electrical plates, innervated by the same primitive-nerve 
 fibres ^. Bvit in one point only we failed to understand the arrang-e- 
 ment of the Torpedo organ : it seemed to us, that the region which 
 is above the back of the animal, in the neighbourhood of the 
 sagittal plane, was, to borrow an expression from fortification, a 
 dead angle, although it is precisely most in need of protection 
 (434, 435). Now this drawback is undoubtedly to a certain 
 extent remedied by irreciprocal conduction. 
 
 * Gesammelte Abhandlungen, vol. ii. p. 696. 
 
 * Untersuchungen, &c., p. 14 f. ; Archiv fiir Physiologie, 1883, p. 252. 
 ^ Untersuchungen, &c., p. 41 1 f. 
 
 * Untersuchungen, &c., p. 320. 
 ' Untersuchungen, &c., p. 393. 
 
LIVING TORPEDOS IN BERLIN. 535 
 
 However that may be, the instance of internal adaptation to 
 which we have been now led, far surpasses all earlier ones in 
 ingenuity. It would certainly have demanded the profoundest 
 reflection of a clever brain, to hit upon the idea of making* each 
 column as good a conductor for its own shock as any other animal 
 tissue can be, but comparatively a non-conductor for the current of 
 all other columns. In connexion with the extremely transitory 
 nature of the shock, it is not a little remarkable, that it is only 
 currents of extremely short duration that the organ conducts 
 irreciprocally. Of what service would it have been to the fish, if it 
 had become a perfect non-conductor for continuous heterodromous 
 currents also ^ ? 
 
 15. Positive polarisation in its dependence upon the density of 
 the polarising current. A case of relatively positive polar- 
 isation by the heterodromous cvirrent. 
 
 By the establishment of irrecij^rocal conduction, a foundation 
 hitherto wanting, has been gained for the investigation of the 
 phenomena of polarisation in the electrical organ, in one direction 
 at least. We no longer require to attribute a clearly quite incon- 
 ceivable strength to the relatively positive polarisation produced by 
 the homodromous current, nor an equally unaccountable transitoriness 
 in the first moments after opening the jjolarising circuit, which is 
 entirely at variance with its later persistence. From the circum- 
 stance that when the period of closure is lengthened, the homodro- 
 mous current is stronger than the heterodromous^ we cannot draw the 
 conclusion which we held to be good in the first communication 
 (p. 456), on the assumption that there is no such thing as 
 irreciprocal conduction. Such conduction does exist, consequently 
 that superiority may be the result of it ; and as a further conse- 
 quence, it does not prove that positive polarisation and the shock 
 are different processes. 
 
 * It should be remembered, in regard to the history of the subject, that irreciprocal 
 conduction has been brought into connexion with the shock of electrical fishes once 
 before in a remarkable way. P. Erman's unipolar conductors, flame, soap, white of 
 egg, and other animal substances, in a certain degree of dryness presented the 
 earliest known instance of irreciprocal conduction. ' Noch frage man mich nicht,' 
 he says in his somewhat curious language, ' ob ich wohl am Ende zu glauben vermag 
 dass die erwahnte isolirende Eigenschaft der unipolaren Leiter . . . mit dem Me- 
 ohanismus der Spontaneitat der vorzugsweise sogenannten elektrischen Thiere einen 
 denkbaren Zusammenhang habe, Ich kann diese Frage zur Zeit nicht beantworten, 
 aber geetehen will ich dass ihrer Losung mein Augenmerk bei dieser Untersuchuno- 
 war.' Gilbert's Annalen der Physik, 1S06, vol. xxii. pp. 44, 45. 
 
536 
 
 LIVING TORPEDOS IN BERLIN. 
 
 The foremost duty enjoined seemed to be to establish more com- 
 pletely the dependence of absolutely positive polarisation upon the 
 current density, but at the close of the first communication, I still 
 needed systematic experiments on this subject. I have now made 
 such experiments in such a manner, that I sent a momentary 
 current of always the same duration through the same preparation, 
 but the current was produced by a Grove's battery with a number 
 of cells increasing" from I. up to L. and decreasing- again to I. 
 The following numbers give a picture of the results : — 
 
 I. 
 
 4. Torp. Fresh. OC + 0.0102. I' — P. 
 CT o"-o764. 
 I. V, X. XX. XXX. L. XXX. XX. X. V. I. 
 
 iS". t + 500 + 0- II 500 + 3- II 250 II 250 II — II 315 II 125 II 41 II 10 II —3 11 o II 
 ■P- 1 3-5 II 20 II 38 II 87 II 148 II 322 II 152 II 78 II 35 II 18 II 2.5 II 
 
 XX. XXX. L. 
 
 S. t +60 Ij +43 II +173 
 
 -P- I 78 II 173 II 319. 
 
 The preparation was very vigorous as the strong organ current 
 shows ; it shocks perceptibly the two first times, when excited by a 
 current of only one Grove, and shows from thenceforward absolutely 
 and relatively positive polarisation rising and falling regularly 
 with the current density. A disturbance occurred with thirty 
 cells, but it did not detract otherwise from the usefulness of the 
 
 whole series. 
 
 II. 
 All as in Series I, except OC +0-0341. 
 I. V. X. XX. XXX. I. X. 
 
 ^■t+3;+3i-i2 I I- 5;- 3 I-13 1-7 j-13 j-i7t-<-238 j -12 
 
 I I \+i9; + io \ + 22 \— \+ 9 \+i5| I 
 
 P-\ 3-5 i 2.5 I 15-5; 15 32 64 97 178 I 267 i 34 
 
 I have given the second series, chiefly because it presents the 
 example of absolutely negative, relatively positive polarisation by 
 the heterodromous cm-rent ; this is the instance mentioned above on 
 p. 503, and though apparently incontestable, it is hitherto unique. 
 The occurrence of this polarisation is all the more significant, as 
 a relatively negative preliminary deflection precedes the relatively 
 positive principal deflection. Thence we may infer, that the relatively 
 positive heterodromous polarisation is always present, and is only 
 concealed by the homodromous, relatively negative polarisation. 
 
 The first series shows in a very pronounced way, the phenomena 
 which were described in general language in the first communi- 
 cation. It appears to support the assumption of an absolutely and 
 
LIVING TORPEDOS IN BERLIN. 537 
 
 relatively positive polarisation, different from the shock and due 
 to the homodromous current. A completely reliable proof of the 
 correctness of the one or the other conception is wanting", such 
 as would he afforded by a more frequent recurrence of the 
 behaviour exhibited in the second series, as a set-off for the failure 
 in obtaining" the result sought after in § 6. 
 
 16. On conduction in both directions in the electrical 
 nerves of the Torpedo. 
 
 In the experiments of the first communication, one question 
 among" others remained unanswered, which has indeed been decided 
 long ago, but can never be too frequently made the object of 
 renewed examination, viz. the question as to conduction in both 
 directions [doppeUinnige LeUimg) in the electrical nerves, the purely 
 centrifugal character of which gives them inestimable value for 
 this purpose. The first experiment which I made on the first fish 
 (the one in a condition of opisthotonus) last summer, was aimed at 
 this question. 
 
 With io,ooo turns at zero, the feeble current of the electrical 
 nerves could be observed very well : the stronger negativity of the 
 peripheral transverse section, upon which stress was laid in the 
 first communication, could always be clearly made out. Thus I 
 obtained for the nerves of the first Torpedo, which was unhealthy, 
 the following values in Raoults : — 
 
 I (left). II. I (right). 
 
 With peripheral... ) Transverse J + 0-00718 + 0-00822 + 000568 
 
 ■:! 
 
 With central \ section... ( — 0-00508 — 0-00568 — 0-00409 
 
 Difference + 0-00209 + 0-00254 + 0-00159 
 
 These numbers are in complete conformity with those in the 
 first communication. On a later occasion, I tried to prove also that 
 the electromotive force of the current ascending between two 
 transverse sections of the electrical nerves — it may be called the 
 axial current — increases in proportion to the length of the por- 
 tion of nerve, but I did not succeed in this with the uniformity 
 I could have wished. In observing the axial current between 
 points in the longitudinal section of the uninjured nerve, which 
 I also attempted (comp. p. 465), I met with the difficulty, that 
 since the electrical lobes had been punched out, the negativity of 
 the upper transverse section chiefly asserted itself. It will be 
 
538 LIVING TORPEDOS IN BERLIN. 
 
 necessary to make the experiment on a Torpedo not divested of its 
 electrical lobes, but it will then be difficult to prevent disturbances 
 arising- from voluntary or reflex shocks^. 
 
 I now tetanised the nerves of the first Torpedo, from which the 
 numbers given above were obtained on this occasion, with Helm- 
 holtz' arrangement of the sliding" inductorium. The primary coil 
 was filled with rods, and the shocks were led into the nerve by the 
 clay points of the unpolarisable tubes. It was necessary to bring 
 the secondary coil up to 30 mm. in order to obtain a good variation 
 to the amount of 7-10 sc. The variation was observed repeatedly in 
 the same nerve, with descending as well as with ascending direction 
 of excitation, and thus this important gap has been filled up. 
 
 The disturbances which had shown themselves in the correspond- 
 ing experiments of the first communication, now only occurred once. 
 Ewald Hering recently made similar observations on the sciatic 
 nerve of the frog^, so that the fact has nothing whatever to do with 
 the function of electrical nerves. It is attributable to the cii'cum- 
 stance, that by the new methods, many minute details can now be 
 recognised in phenomena which were formerly only sketched in 
 outUne. When the nerve was replaced by a clay model, there was 
 no such effect on tetanising. 
 
 The next step was to observe electrotonus in the electrical 
 nerves of the same fish. The experiments were conducted precisely 
 as on the first occasion (p. 466). I also now saw anelectrotonic 
 and kathelectrotonic increments spread in both directions, as Prof. 
 Christiani had already done, but I had neglected this observation 
 the first time, because my attention was taken up with disturbances 
 which had hitherto been unexplained, and which presented them- 
 selves to me, as previously to Prof. Christiani. They were now also 
 again noticeable. Electrotonic increments presented themselves in 
 the wrong direction ; after cutting through the nerve, there re- 
 mained actions in the right direction and in considerable strength. 
 The significance of these irregularities must be decided by further 
 experiment. 
 
 * Very shortly, a paper worked out in the physical department of the Physiological 
 Institute by Dr. M. Mendelssohn will appear in the Archiv fiir Physiologie ; the 
 axial current is here traced in different nerves acting wholly centrifugally or centri- 
 petally. 
 
 ^ Beitrage zur allgemeinen Nerven- und Muskelphysiologie, 15 Mittheilung. 
 "Uber positive Nachschwankung des Nervenstromes nach elektrischer Reizung. In 
 den Wiener Sitzungsberichte, 1884, vol. Ixxxix. p. 137 f. 
 
LIVING TORPEDOS IN BERLIN. 539 
 
 17. On the secondary electromotive phenomena in the 
 electrical nerves of the Torpedo. 
 
 I endeavoured to answer the question as to the secondary electro- 
 motive behaviour of the electrical nerves in the second fish. It will 
 be remembered that in the paper on the secondary electromotive 
 phenomena of muscles and nerves, it is stated that positive polarisa- 
 tion by the ascending* current, exceeds very uniformly that due to 
 the descending- current in the posterior roots of the spinal nerves of 
 the frog-. If here also we denote the physiological direction of 
 action of the roots as homodromous, and the opposite as hetero- 
 dromous, then positive polarisation in the homodromous direction 
 is strong-er than in the heterodromous. On the other hand, the 
 corres]3onding difference in the case of the anterior roots is not so 
 clear, and if at first it is seen, it is soon effaced; i.e. positive 
 polarisation in the homodromous, here the descending direction, does 
 not exceed perceptibly or not permanently that in the heterodro- 
 mous, here the ascending direction^. On account of the great 
 difficulty which attends these experiments on the roots, owing to 
 their shortness and delicacy, it appeared extremely desirable to 
 repeat these experiments on the electrical nerves, which, according 
 to general opinion, can be obtained almost or wholly centrifugal, 
 and of a length and thickness, such as is not to be met with in the 
 animals which have hitherto been available for physiological expe- 
 riment. I stated in the first communication (p. 463) that they can 
 be easily found unbranched of a length of 3-4 cm., but I have since 
 had them as long as ^-^ cm. with a thickness of 3-5 mm. 
 
 The nerves were laid in a groove, which was cut with a three- 
 cornered file in a piece of cork afterwards varnished. The clay 
 shields of the leading-in pads of the battery circuit were pat upon 
 the two transverse sections. The clay points of the unpolarisable 
 tube electrodes, as terminals of the galvanometer circuit, were placed 
 on symmetrical points of longitudinal section near both transverse 
 sections. With this arrangement, the axial current made itself 
 uniformly perceptible with a force of o«ooi35 to O'Ooaia, corre- 
 sponding as closely as possible to the difference recorded on p. ^^y. 
 The galvanometer ^S" had 10,000 turns at zero distance, and the 
 galvanometer P 50 turns at 20 mm. distance. I obtained the fol- 
 lowing numbers : — 
 
 1 Gesammelte Abhandlungen, 18S3, vol. i. pp. 382-387. 
 
540 
 
 LIVING TORPEDOS IN BERLIN. 
 
 2. Torp. 
 CTo".oi54. 
 
 Fresh. I'-P. 
 
 CT o".o764. 
 
 ■^- V ! - X ! +^ XXX ! ^4"5t +9 I +5t +7 II ! +i2.5t +1° I +9t +^5 
 S. ^-ii-s ^-i 5 ^^^- j i8 I 18-51 18 1 i8 || i 82 | 83 | 83 | 82.5 
 
 Only half the length of nerve. 
 
 I + 14 I + 10; 6 f + 10; 9-5 
 I 162 \ 164; 162-5 I ^^2 ; 163-5 
 
 CT i"-024. 
 
 I — 20| —17 I — 20| —20 
 i II7I I 1079! II71 I 1063. 
 
 As the streng-th of the primary current seemed to be very nearly 
 doubled with half the leno-th of nerve (i62'8 instead of 165-2), the 
 current density was not (as one mig-ht have suj^posed) on account' 
 of the large transverse section of the nerve, less than in the experi- 
 ments on sensory roots of the frog. A more decided negative to 
 the question proposed could not be given by the experiment. Still 
 I was not deterred thereby from renewed endeavours, which had 
 however no other result. This time 20,000 turns were used with 
 the galvanometer 8 at zero distance. 
 
 4. Torp. 
 
 Fresh. 
 
 Cro"-oo3i. 
 ^9._X. t +5 I -4r +5 i 
 
 2 I 
 
 2 I 2 4. 
 S'.— XX. 
 
 p. 
 
 First nerve. 
 
 XX.- t +5 I -2 
 
 I 3-5i 3 
 Cro"-o3io. 
 
 I'-p. 
 
 Cro"-oi23. 
 
 t-7|/-i.-i t|-?;-o-5 
 
 I I I + 3 ' + 4-5 i I + .S +6 
 
 10 i 10; 10 
 
 10; 10 
 
 i 
 
 -? t 
 + 14 I 
 
 2-2 
 
 _ ? 
 + 10 
 
 I S ~ '; — ' I 
 
 I I + IO-5 + 7 I 
 
 i 21-5; 21-5 1 
 
 Second nerve. 
 
 -0-5 
 + 12 
 21-5 
 
 CT o"-o3io. 
 
 R-XXX. I +15 t +16 I +7-5 t +17 
 P. i 25-5 I 24 i 24 I 24 
 
 After completing this series of experiments, it was discovered that 
 in one place of the battery current, the conduction was imperfect ; 
 it was the same fault which had caused the derangement recorded 
 above (p. 536). Still it does not seem that this would tend to 
 tliminish the significance of the negative result obtained, nor when 
 we consider the variety of combinations of current densities and 
 times of closing which were tried, is it to be expected that any later 
 result will make any change in the conclusion now arrived at. We 
 should however do wrong to rest perfectly satisfied, until the ex- 
 amination has been extended to other nerves of which the consti- 
 tuent fibres are physiologically homogeneous. 
 
ADDENDUM. 
 
 During the period which has been occupied in preparing- the 
 foregoing- Memoirs for the press, the nature of the secondary 
 electromotive chang-es of nerve and muscle has continued to occupy 
 the attention of physiologists. In order to render this series 
 of memoirs more representative than it would otherwise be of the 
 actual state of knowledge, without delaying its publication, it 
 seems desirable to give a short account of one or two experimental 
 researches which have very recently appeared relating to a subject 
 dealt with in these memoirs. 
 
 I. Intrapolar Polarisation of Nerve. 
 
 One of the questions to which the investigations of du Bois- 
 Reymond, Hering and Hermann, have given special interest is 
 that of the changes which take place in a nerve during the trans- 
 mission through it of a voltaic current in the tract which intervenes 
 between anode and cathode. Can it be demonstrated that an 
 electromotive force which did not before exist, comes into existence 
 in either direction in the intrapolar tract and, if so, what is its 
 relation to the intrapolar after effects observed in the various 
 modifications of Peltier's experiment ? 
 
 The direct investigation of the intrapolar tract during the 
 passage of a current is beset with difficulties, of which the most 
 fundamental is that of discriminating between the galvanometric 
 results of change of resistance and those of change of electromotive 
 force. The first observations of any importance bearing on this 
 subject were those of Professor Hermann, made some years ago, 
 who found that when compared with the resistance of a living 
 nerve, that of the same nerve when devitalised was considerably 
 greater. Prof. v. Fleischl has recently attempted to get over 
 the difficulties with the aid of the capillary electrometer^. He 
 found that when he introduced a capillary electrometer into the 
 
 ^ E. V. Fleischl, Studien iiber den Electrotonus, ' Arch. f. Anat. u. Physiol.' 1885, 
 p. 490. 
 
542 ADDENDUM. 
 
 circuit of a thermopile, which circuit also included a ' currentless ' 
 nerve (i. e. a nerve to which the leading off electrodes were applied 
 at equipotential points) the hydrostatic pressure required to zero 
 the electrometer, i. e. to bring back the column of mercury to the 
 position which it occupied when ' short circuited,' accurately 
 indicated the electromotive force of the pile. In other words, 
 it made no difference as regards the reading of the electrometer 
 whether the polarised nerve were in or out of the circuit. The form 
 of Fleischl's experiment, of which the result is regarded by him as 
 inconsistent with all that has been advanced on the subject of 
 nerve-polarisation on either side, may be described as follows : 
 
 ■~\ 
 
 The line e k E /(• n ^' k' represents a circuit in which are included 
 at n a nerve (by equij)otential contacts) and at e a capillary 
 electrometer, furnished with an arrangement by which the pressure 
 of mercury in the capillary can be regulated and measured, k k' 
 and k k' represent keys or bridges by which either nerve or 
 electrometer can be short-circuited at will. At E an electromotive 
 source is introduced into the circuit, which, as above stated, was 
 in Professor Fleischl's experiments a thermopile, of which one 
 set of junctions was heated by a small Bunsen. The experiment 
 consisted in first measuring the electromotive force of E with 
 k k' closed and k k' open, then closing k k' and oj)ening ^ i' so as 
 to lead the current of E through the nerve, and finally opening 
 k k' so as to obtain a second measurement. It was found that the 
 two measurements were identical. It was therefore evident that 
 in the intrapolar tract, the effect of the polarisation of the nerve 
 was not such as to balance the opposed polarisation of the electro- 
 meter. When the experiment was varied by substituting for the 
 nei've a non-living polarisable arrangement — a voltameter of the 
 kind known as ' Wollaston's points ' (platinum wires fused into 
 glass tubes so as to cover all but their tips) — the polarisation of the 
 points evidenced itself by a diminution of the reading of the 
 electrometer which persisted for some hours. 
 
 It is evident therefore that the polarisation of a nerve is not 
 quite the same as the polarisation of a voltameter. As to what 
 the difference consists in, Hermann has sought to enlighten us 
 
ADDENDUM. 543 
 
 in a very recent paper which purports to be a theoretical and 
 experimental criticism of Fleischl's work ^. A polarised nerve, 
 he says, is not comparable to a voltameter, because in nerve, 
 the polarised surface is not in the unbranched circuit but in a 
 collateral branch [Zweigkreise), and that consequently the case of 
 the nerve is not strictly represented by substituting for it the 
 WoUaston's points at w in the diagram, unless at the same time 
 a permanent derivation is introduced as under, in which r 
 
 r 
 
 e 
 
 V 
 
 ^ E 
 
 
 r \\ 
 
 represents a second bridge which in the experiment remains 
 closed when k k' is opened, and of which the resistance though 
 considerable in itself, is small as compared either with that of 
 w or with that of the unbranched part of the circuit. The reason 
 why a current in passing along a nerve is divided into two 
 channels is to be found, according to Hermann, in the fact that 
 every nerve fibre consists of an axial core and of an envelope of 
 difierent material, and that, whereas the latter conducts without 
 polarisation, whatever proportion of the current enters the core 
 from the envelope and then leaves the core to reenter the envelope, 
 produces polarisation as it passes from the one to the other and 
 vice versa. The grounds on which Professor Hermann regards a 
 nerve, so long as it is alive, as a conductor of this kind are so well 
 known that it is scarcely needful to refer to them here. It will be 
 sufficient to remind the reader that the electrotonic currents of 
 nerve are reproduced or imitated in a model {Brahtmodell or 
 Kernleiter) which, in accordance with the principle above referred 
 to, consists of a core of wire sheathed in a moist envelope. When 
 in Fleischl's experiment a model of this kind is substituted for 
 the nerve at n (in the first diagram) it is found to behave exactly 
 as a living nerve does. The introduction of the polarised model 
 into the circuit is without influence as regards the reading of the 
 electrometer, i. e. as regards the electromotive force indicated by 
 it. The difference, therefore, between the negative polarisation 
 of a nerve and that of any such polarisable arrangement as a 
 voltameter is that whereas the nerve loses its charge in an open 
 
 ^ Hermann, Ueber die Ursache des Electrotonus, Pfluger's ' Archiv,' vol. xxxviii. 
 P- 153- 
 
544 ADDENDUM. 
 
 circuit very rapidly indeed (according to Hermann by derivation 
 [Algleichung) througli the envelope) the latter retains its charge 
 under the same conditions for a considerable time. That this is 
 actually the case, Hermann has shown by a separate series of 
 experiments, in which he investigated with the galvanometer 
 the state of polarisation of various polarisable arrangements 
 (voltameter, Wollaston's points, wire model and living nerve) at 
 various periods after the opening of a polarising current led through 
 them, the result being that the nerve and wire model contrasted 
 with the others in respect of the rapidity of their depolarisation. 
 
 In the introduction to his paper the subject is treated mathe- 
 matically : it is shown that on the hypothesis that a living nerve 
 possesses as regards polarisation the properties of the ' Kernleiter' 
 the result of Fleischl's experiment must be as it actually is ^. 
 
 If the ^ Kernleiier' theory could be accepted as proved, the ex- 
 planation might be regarded as satisfactory. But for the present 
 a serious difficulty exists, in the fact that the extrapolar currents 
 which are regarded as the characteristic phenomena of ' electro- 
 tonus' in the sense in which the word was first used by du Bois- 
 Heymond, do not, as they ought to do if they are directly dependent 
 on physical polarisation betw^een the sheath and core of the nerve 
 fibres, appear at the instant that the polarising current is closed. 
 Not long ago an experiment was made by Helmholtz for the pur- 
 pose of proving this ^. It consisted in comparing the time which 
 elapses when an excitatory wave is propagated through two nerve 
 channels, of which one is an undivided nerve, the other consists of 
 two nerves a and b of which the ends are opposed to each other in 
 such a way that when a current of short duration is led through 
 a where it is not in contact with h, h is excited by the electrotonus 
 induced in the applied part of a. It being found that if the whole 
 length of the channel of conduction was the same, the time lost in 
 transmission was the same, it was concluded by Helmholtz that the 
 rate of propagation of electrotonus did not. materially differ from 
 that of excitation. It has now been, so far as can be seen, conclu- 
 sively proved by a new series of experiments conducted by Prof. 
 Bernstein ^ with his repeating rheotome, that the electrotonic effect 
 
 1 The answer which is expected from Prof. Fleischl has not yet appeared. Editor, 
 April, 1887. 
 
 ^ Helmholtz, ' Monatsberichte der Berl. Acad.' 15 .Tuli, 1885. 
 
 ^ J. Bernstein, Ueber das Entstehen u. Vei'schwiuden der electrotonischen 
 StriJme im Nerven, ' Archiv. f. Anat. u. Physiol.' 1886, p. 197. 
 
ADDENDUM. 
 
 545 
 
 is propagated from cathode or anode as the case may be, at a rate of 
 from 10 to 15 metres per second. As explanatory of the result of 
 his very extended series of experiments, he gives the following 
 diagram. 
 
 In the diagram, <S' on the time axis expresses the duration of 
 the closure of the polarising current, which in most of the experi- 
 ments was about xf o'^- When the distance between the intrapolar 
 tract and the leading off electrodes (which were applied respectively 
 to the natural and end surfaces of the nerve) was about 15 millims. 
 the electrotonic effect began between j^j/^ and f^o'^ after the 
 closure of the polarising current, the interval of time being always 
 approximately proportional to the distance. When the current was 
 so directed that the cathode was next the leading off electrodes, the 
 excitatory wave due to the excitation at the cathode was strong 
 enough to reverse the previous negativity of the end surface of 
 the nerve, as shown in the diagram, where the difference of 
 potential between the two leading off contacts is indicated by the 
 distance S y, and the excitatory variation by the loop n (/ a. The 
 cathelectrotonic effect which is of course in the same direction as the 
 variation, does not begin until the latter has entirely ceased, as 
 indicated by the line k t e. When the anode was next the leading 
 off electrodes, there was also a feeble excitatory effect due to the 
 same cause as before, but in this case its sign was opposed to that of 
 the relatively much stronger anelectrotonic current a t' e. In both 
 
 N n 
 
546 ADDENDUM. 
 
 cases the electrotonus ceased abruptly about ^J^" after the polarisino- 
 current was broken. 
 
 Prof. Bernstein carefully avoids drawing any theoretical conclu- 
 sions from his experiments, which must be regarded as a continuation 
 of those embodied in his well-known ' Researches on the excitatory 
 process in Muscle and Nerve,' published in 1871. He regards 
 extrapolar electrotonic currents as indications of a ' state of polar- 
 isation peculiar to living nerve, which is propagated in nerve fibre 
 from section to section,' but declines to express any opinion as to its 
 relation to ordinary physical polarisation. 
 
 2. time-eelations op the excitatory process in muscle and 
 
 Nerve. 
 
 An investigation recently conducted in Prof. Hering's laboratory 
 by Mr. Head of Cambridge^ is of sufficient novelty in relation to 
 the e:^citatory process in nerve to make it desirable that the reader 
 should be informed on the subject. 
 
 Our knowledge of the time-relations of the process have been 
 hitherto exclusively founded on the experiments of Bernstein with 
 the repeating rheotome, recorded in his well-known monograph ' On 
 the excitatory process in Muscle and Nerve.' Mr. Head's experi- 
 ments on the subject have been made with an improved form of 
 instrument, the construction of which renders it possible to subject 
 a nerve to excitations of a frequency much greater than was attain- 
 able with the original rheotome of Bernstein, and thereby, as well 
 as by employing" a galvanometer better adapted to the purpose, to 
 obtain indications of much greater delicacy than those afibrded by 
 Bernstein's method. As regards the duration of the negative 
 variation, due to a single excitation, the new determinations show 
 that Bernstein's estimate, which has up to the j)resent moment 
 been universally accepted, is very much too short, that the period 
 of negativity lasts, even in the nerves of summer frogs, for more 
 than a hundredth of a second, and in winter frogs twice as long. 
 But this point, however interesting and important, was not the 
 main object of the enquiry. Mr. Head's efforts were chiefly directed 
 to obtaining determinations of the conditions which lead to the 
 positive after effect which, as Hering showed in his paper ' On 
 tetanic excitation of Nerve/ succeeds the concomitant diminution 
 
 ' Henry Head, Ueber die negativen u. positiven Schwankungen des Nerven- 
 systems. Pflfiger's ' Archiv,' vol. 40. p. 207, 1887. 
 
ADDENDUM. 547 
 
 or reversal of the ' demarcation current ' at the moment that the 
 excitation ceases. The main facts ascertained by Hering were 
 that effects appeared even when the tetanisation lasted only a frag- 
 ment of a second, and that when the excitation was prolonged in 
 successive experiments, the positive after effect increased up to 
 a duration of 30'', but subsequently diminished. 
 
 To these facts Mr. Head has added the following. He has first 
 shown that the positive after effect cannot be regarded as a second 
 phase of the excitatory effect produced by each single excitation, 
 for his rheotome observations show that the negative phase is fol- 
 lowed by a period of complete restoration of the previously existing 
 electrical state, which lasts for several hundredths of a second. 
 Secondly, he has sho^vn that in short tetanisation lasting not more 
 than a few seconds, the excitatory effect, i. e. diminution of the 
 demarcation current, is nearly constant — a result which could not 
 happen unless the positive effect were deferred. Thirdly, that in 
 comparing nerves prepared with the utmost possible care and pro- 
 tected from evaporation, with nerves in a less favourable state, it is 
 found that although the latter may show no inferiority or sign of 
 impaired functional activity as regards the negative variation, they 
 are inferior as regards the positive after effect, so that, in repeated 
 observations, the former remains unimpaired while the latter rapidly 
 diminishes. 
 
 These facts he interprets to mean that the positive after effect is 
 in the nature of a reaction of the nerve, consequent on excitation, 
 i. e. a change in the opposite direction to the excitatory change, by 
 which it not merely retm'ns to the normal, but passes into a condi- 
 tion on the other side of the normal, so that the longitudinal 
 surface to which the electrode is applied becomes more positive 
 than it was before excitation. This view he considers to be con- 
 firmed by the observation that the effect in question is only observed 
 when the leading off contact on the longitudinal surface is at a 
 considerable distance from the injured end of the nerve. 
 
 N n :i 
 
INDEX. 
 
 Action - current, Hering's definition, 
 p. 230; Hermann's theory as to its re- 
 lation to du Bois' positive polarisation, 
 286, 321, 
 
 After-currents, defined by du Bois, 164 ; 
 in nerve and muscle produced by break- 
 excitation, 319; combinationof polarisa- 
 tion and excitation after-currents, 320. 
 
 Anaphoric actions of current, accord- 
 ing to Jiirgensen and Quincke, 225. 
 
 Anodic and cathodic polarisation, 
 Hering's definition, 230. 
 
 Anodic negative polarisation, ex- 
 plained by inhibition of excitation, 332, 
 
 Anodonta, experiments on, in relation 
 to inhibition, 332 ; in relation to muscle 
 poisoned by veratrine, 361. 
 
 Bernstein, experiments in confirmation 
 of Pfiiiger's doctrine of electrotonus, 
 1 2 ; on rate of propagation of electro- 
 tonus, 544. 
 
 V. Bezold, on effects of veratrine on 
 striated muscle, 333. 
 
 Biedermann, on break-excitation, 55 ; 
 on break-tetanus, 74 ; on compensation, 
 94, 95 ; on currents which pass out of 
 transverse section of nerve, 135, of 
 muscle, 136; on make- and break-con- 
 tractions, 246, 250 ; on changes of form 
 of muscle poisoned by veratrine, 333 ; 
 the secondary electromotive phenomena 
 of veratrine muscle, 343 ; positive 
 cathodic polarisation in normal striated 
 muscles, 352; experiments on positive 
 cathodic polarisation, 352 ; discusses 
 du Bois' internal positive polarisability 
 of muscle, 361. 
 
 Bilharz, experiments confirming Pfiiiger's 
 doctrine of electrotonus, 8, 10 ; on struc- 
 ture of electrical organs, 370. 
 
 Boiling, effect of on resistance of muscle, 
 190, of nerve, 220; on polarisation of 
 electrical organ, 214, 219. 
 
 du Bois-Eeymond, his law of excita- 
 tion, 49; on break-contraction, 52; 
 explains the differences in amount of 
 contraction on opening and closing 
 main circuit and nerve circuit by ex- 
 
 ternal polarisation-current, 98 ; proves 
 that in a nerve or muscle through which 
 a current has been passed there is after- 
 wards a current in an opposite direc- 
 tion, 102 ; observation of positive po- 
 larisation, 105 ; on Galvani's funda- 
 mental experiments in the electro- 
 physics of nerve, 140; his 'paradoxical' 
 contraction discussed by Hering, 153- 
 157; on secondary electromotive phe- 
 nomena in muscles, nerves, and elec- 
 trical organs, 163 ; observations and 
 experiments on Malapterurus brought 
 to Berlin alive, 369 ; Living Torpedos 
 in Berlin, first paper, 417; Living 
 Torpedos in Berlin, second paper, 
 
 479- 
 
 Break-contraction, contribution to the 
 theory of, by Tigerstedt, 49 ; Peltier's 
 views and du Bois-Beymond's criticism 
 of them, 51, 52 ; Pfiiiger's andMunk's 
 theories, 53 ; Tigerstedt's experiments 
 to show that its amplitude varies with 
 the intensity of negative polarisation 
 of a motor nerve, 54-58 ; experiments 
 on effect of changes in polarisability, 
 58-64; Tigerstedt's discussion of Griitz- 
 ner's experiments, 71-74 ; Griitzner's 
 discussion of Tigerstedt's experiments, 
 106-109 ; break-contractions appear 
 earlier at break of the main circuit than 
 of the nerve circuit, 1 1 1 ; discussion of 
 Hermann's conclusions on this subject, 
 112-134. 
 
 Break-excitations in many cases raake- 
 excitations, 93 ; called neuro-electric, 
 or additive, by Griitzner, 94 ; interfer- 
 ence of the nerve-current with the 
 make- and break-excitation of nerves, 
 131-135; break-excitation after-cur- 
 rents, 319. 
 
 Break-tetanus, Engelmann's theory dis- 
 cussed by Tigerstedt, 70-74 ; Her- 
 mann's experiments, 322-325. 
 
 Cavendish, observations on shock of 
 Torpedo discussed by du Bois-Eeymond, 
 
 427-434- 
 Chauveau, theory of the physiological 
 action of electricity, 53. 
 
550 
 
 INDEX. 
 
 Chemical excitation of nerves, Eck- 
 hard's experiments, 143. 
 
 Christiani, on irreciprocnl conduction, 
 1S6 ; experiments as to relation be- 
 tween electromotive force of GNmnotus 
 and Torpedo to the electrical resistance 
 of fresh and sea water, 442 ; on elec- 
 trical nerves of Torpedo, 463 ; on elec- 
 trotonus-currents in them, 467. 
 
 Circmt, effects of breaking according to 
 Volta, Ritter, Pfliiger, 116. 
 
 Cold, influence of in producinif liability 
 to tetanus in frogs, 136 ; Hering dis- 
 cusses the tetanus produced by section 
 of motor nerves under the influence of 
 cold, 139-142. 
 
 Colladon, on distribution of electricity 
 in Torpedo when discharging, 428 ; 
 currents between dorsal and ventral 
 surfaces of Torpedo called by du Bois- 
 Reymond ' Colladon's currents,' 428. 
 
 Compensation of nerve- or muscle- 
 cxirrent, 95. 
 
 Conducting axis schema, Hermann's, 
 described, 311. 
 
 Conduction, experiments as to irreci- 
 procal conduction, 186 ; conduction in 
 both directions, 537, 538. 
 
 Contraction, Ritter's law of contraction, 
 no; secondary, 147; ' paraduxical,' 
 
 153-157- 
 
 Critical time, du Bois-Reymond's ex- 
 planation of, 179; Hering' s discussion 
 of, 249. 
 
 Crusell, experiment on the effects of vol- 
 taic-current on the tissues, 50. 
 
 Czermak, experiments on excitation of 
 nerve by contact with moist conductor, 
 131- 
 
 Demarcation -ciirrent, discussed by 
 Griitzner. 96, 97. 
 
 Dove's disjunctor, described and criti- 
 cised, 492-496. 
 
 Eckhard, experiments on chemical exci- 
 tation of nerves, 143 ; reference to ex- 
 periments on Torpedo, 444. 
 
 Electrical organs, their secondary elec- 
 tromotive phenomena, 211-217 ; effect 
 of boiling on resistance, 214; BiLharz 
 on the structure of, 370 ; organ of 
 Gymnotus, 371, 379; of Malapterurus, 
 375; of Toi-pedo, 433; pseudo-electrical 
 organs of Mormyri and Skate, 388 ; 
 discussion concerning currents in organ 
 of Torpedo, 427-434 ; en secondary 
 electromotive actions of the organ, 445 ; 
 on homodromous and heterodromous 
 currents, 451 ; actions of the organ de- 
 pendent on its vitality, 457 ; experi- 
 mental results, 469-475 ; polarisation 
 
 of organ of Torpedo, 489-492 ; ex- 
 periments on polarisation, 498 ; irre- 
 ciprocal conduction of organ, 503 ; 
 measurements of resistance, 514. 
 
 Electrical nerves, their number in Tor- 
 pedo, Gymnotus, and Malapterurus, 
 379 ; experiments on that of Mala- 
 pterurus, 410; of Torpedo, 462; cur- 
 rent of rest, 463 ; negative variation 
 of current, 466 ; conduction in Torpedo 
 nerves, 537 ; secondary electromotive 
 phenomena of, 539. 
 
 Electricity of electrical fishes, identical 
 with ordinary electricity, 390-397 ; 
 essential identity of electrical phe- 
 nomena in all electrical fishes, 449, 
 
 459- 
 
 Electrolysis by the Torpedo shock, 435. 
 
 Electromotive phenomena, du Bois- 
 Reymond's three classes of, in muscle 
 and nerve, 163 ; arranged later into 
 two classes, 203 ; secondary electromo- 
 tive phenomena investigated by du 
 Bois-Reymond, 170-177 ; in regard to 
 direction of primarj' current, 181 ; in- 
 vestigation of secondary electromotive 
 phenomena in nerves, 204 ; ditto in 
 electrical nerves of Torpedo, 539 ; ditto 
 in electrical organ, 211 ; du Bois-Rey- 
 mond's experiments on and theory of 
 electromotive phenomena criticised by 
 Hering, 229, 248 ; Hermann's explana- 
 tion of secondary electromotive phe- 
 nomena, 320 ; du Bois' discussion of, 
 in Torpedo, 458. 
 
 Electro-physics of nerve, Galvani's ex- 
 periment discussed by du Bois-Rey- 
 mond, 140. 
 
 Electro -physiology, Hermann's propo- 
 sitions relating to, 3 2 8. 
 
 Electrotonus, Pfliiger's doctrine of, 5 ; 
 electrotonus in man, 7 ; Bilharz' ex- 
 periments on, 8 ; Nasse's ditto, 8 ; 
 Tigerstedt's method of experimenting, 
 16 ; Hermann's theory discussed by 
 du Bois-Reymond, 199 ; du Bois-Rey- 
 mond's doctrine criticised by Hermann, 
 280, 327 ; Pfliiger's researches on, re- 
 feried to, 273 ; verified by Hermann, 
 32S ; electrotonus currents in electrical 
 nerves, 467 ; rate of propagation of, 544. 
 
 Engelmann, theory of break-tetanus, 
 
 70, 74- 
 
 Excitability, changes of, produced by 
 ascending cuiTcnt, 21 ; by descending 
 current, 31 ; physiologically different 
 in different groups of muscles, no; 
 conditions of, in nerve near a transverse 
 section, 146. 
 
 Excitation of nerve, propositions relat- 
 ing to, 49 ; du Bois' observations on, 
 102 ; Bering's experiments on, 127. 
 
INDEX. 
 
 551 
 
 Extrapolar after-cvirrents, observa- 
 tions by Fick and Hermann, 304. 
 
 Faraday, references to his observations 
 on Gyamotus, 370, 381, 391, 398. 
 
 Fiek, on extrapolar after-currents, 304 ; 
 on effects of veratrine on striated 
 muscle, 333, 342. 
 
 Fleisclil, on increase of excitability of 
 nerves when treated with weak alkaline 
 solutions, 50; on intrapolar polarisation 
 of nerve, 541. 
 
 Fritsch, on origin of organ of ala- 
 pterurus, 488. 
 
 Frog alarum, and Frog interrupter, 
 384 ; use of in experiments on Mala- 
 pterurus, 403, and on Torpedo, 426. 
 
 Griinhagen, experiments confirming 
 Pfltiger's doctrine of electrotonus, 14; 
 on increase of strength of a current 
 passed through a nerve by tetanising 
 the nerve, 210. 
 
 Grutzner, on break-contraction, 60 ; on 
 influence of nerve-current as an auxil- 
 iary cause of break-contraction, 71 ; 
 phenomena of ' hiatus,' 93 ; replies to 
 objections to his theory, 96 ; criticises 
 Hering, 100; experiments confirming 
 Tigerstedt's results, 103 ; states condi- 
 tions favourable to break-contractions, 
 107; criticism of Hermann, 1 1 2 ; dis- 
 cusses conditions of 'incomplete' break- 
 contractions, 119, and Hermann's 
 theory in connection therewith, 123 ; 
 on importance of nerve-current to the 
 excitability of the nerve, 131. 
 
 Gymnotus, Sachs' researches on, 216; 
 the organ of, 371 ; as adapted for experi- 
 ment, 379 ; its electricity identical with 
 ordinary electricity, 391 ; Faraday's 
 experiments on, 420, 435 ; its organ- 
 current, 438, 441 ; discussion of Sachs' 
 results as to polarisation, 448 ; on ho- 
 modromous and heterodromouscun-ent, 
 457; on electromotive behaviour of 
 skin, 482 ; traces of irreciprocal con- 
 duction, 534. 
 
 Head, on the time relation of the exci- 
 tatory process in muscle and nerve, 546. 
 
 Helmholtz, on the rate of propagation of 
 electrotonus, 544. 
 
 Hering, his conclusions as to break-exci- 
 tation similar to Grtitzner's, 94 ; experi- 
 mentson excitation of a muscleoranerve 
 by its own current, 99; on nerve-ex- 
 citation by the nerve-current, 127 ; his 
 method of producing tetanus without 
 metals, 129; on tetanus of motor nerves 
 under the influence of cold, 138 ; on 
 chemical excitation of nerves, 143; on 
 
 true secondary contraction, 147 ; on 
 du Bois-Reymond's researches on se- 
 condary electromotive phenomena of 
 muscle, 229; influence of tendinous 
 intersections on polarisation of muscle, 
 231 ; experiments on the sartorius of 
 the frog, 241 ; on the positive after- 
 variation of the nerve-current conse- 
 quent on electrical stimulation, 255 ; 
 criticism of du Bois-Reymond's experi- 
 ments on negative variation, 267-273. 
 
 Hermann, experiments as to Pfliiger's 
 doctrine of electrotonus criticised by 
 Tigerstedt, 10; criticism of his theory 
 of electrotonus, 44 ; on absence of break- 
 excitation when a current exactly com- 
 pensates nerve or muscle-current, 95 ; 
 on excitatory effects of 'like' current, 
 loi ; his theory of polarisation criticised 
 by du Bois-Reymond, 199 ; on so-called 
 electromotive phenomena in muscle and 
 nerve, 277; criticism of du Bois' views 
 on positive polarisation, 279; theory as 
 to action-current, 2 86 ; experiments 
 with conducting axis schema, 312 ; ex- 
 periments on Ritter's break-tetanus, 
 322 ; his four propositions to explain 
 phenomena of electro-physiology, 328 ; 
 on intrapolar polarisation of nerve, 541. 
 
 Heterodromous and Homodromous 
 currents defined, 446. 
 
 Hiatus, description of phenomena so 
 called, 93 ; why it is sometimes absent, 
 107. 
 
 Indiflference point, verification of Pflii- 
 ger's statement as to the movement of, 
 286, 328. 
 
 Inhibition, experiments on, in Ano- 
 donta, 332 ; in veratrine muscle, 339. 
 
 Intrapolar after-current, Hermann's hy- 
 pothesis in relation to, 284 ; experi- 
 ments relating to it, 2 8 7-301 ,310,322; 
 intrapolar polarisation of nerve, 541. 
 
 Irreciprocal conduction, preliminary 
 experiments, 186 ; of electrical organ 
 of Torpedo, 503 ; discussion of, and 
 hypothesis of electromotive molecules, 
 525 ; forms a foundation for investi- 
 gation of polarisation phenomena in 
 electrical organs, 535. 
 
 Irreciprocal resistance, discussion of, 
 and experiments relating to, 453. 
 
 Jiirgensen, discovery of the anaphoric 
 actions of the current referred to by 
 du Bois-Reymond, 225. 
 
 Kiihne, observations of the excitation of 
 a nerve by its own current referred to 
 by Hering, 127. 
 
55^ 
 
 INDEX. 
 
 Leading-ofF cover, for experiments on 
 electrical fish, described, 380 ; modifi- 
 cations of, 399, 401 ; further descrip- 
 tion of, 420. 
 
 Malapterurus, secondary electromotive 
 phenomena of its organ discussed, 
 211; described by Bilharz, 370; habits 
 in confinement, 3 74 ; experiments on 
 its shock, 390-397 ; its immunity from 
 electric shocks, 404; experiments on 
 electrical nerve 410; chemical reaction 
 of orcjan, 41 2 ; leading-off cover for 
 experiments on, 420 ; no organ-current 
 was found by du Bois in Malapterurus, 
 481 ; irreciprocal conduction in, 534. 
 
 Matteucci, experiments on nerve polar- 
 isation, 197; error as to origin of 
 electricity in brain of Torpedo, 42S. 
 
 Meissner, vindication of Pfliiger, 6. 
 
 Munk, his error in his criticism of Pfliiger, 
 9 ; theory of break-contraction, 53. 
 
 Negative polarisation, experiments by 
 means of Dove's disjunctor, 492. 
 
 Nemerowsky, investigation of electro- 
 motive force of nerve-current, 97. 
 
 Nerve-current, its interference with 
 make- and break-excitation of nerves, 
 131 ; Kiihne on contractions produced 
 by nerve-current, 141 ; Hering's first 
 ^ method of observing positive after-vari- 
 ation of 255; second and third methods, 
 259, 261 ; effect of unipolar actions on, 
 269 ; du Bois-Reymond's investigation 
 of, criticised by Hering, 270-273. 
 
 Obernier, addition to Pfltiger's experi- 
 ments on changes of excitability, 6. 
 
 Organ-current, definition of, 438 ; 
 investigation of, in Torpedo, 440 ; 
 present in imperfect electrical oigans 
 of Skate, 445. 
 
 Pacini, on direction of shock in electrical 
 organ, 371. 
 
 Peltier, experiments on polarisation of 
 muscles, 5 1 ; on polarisation of a frog's 
 limb, 164. 
 
 Pfliiger, doctrine of electrotonus, 5 ; 
 experiments in confirmation of, by 
 Bilharz and others, 8, 10; criticism of, 
 by Munk, Hermann, Wundt, Tiger- 
 stedt, 9-15 ; his principle as to excita- 
 tion of nerve, 49 ; Tigerstedt's simpli- 
 fication of Pfliiger's law of contraction, 
 65 ; attempt to explain influence of 
 cold on the liability of frogs to tetanic 
 excitation, 136; on Kitter's break- 
 tetanus in nerve, 319. 
 
 Polarisability of nerve, discussed by 
 Tigerstedt, 58-64 ; of muscle, how 
 
 modified and destroyed, 190; affected 
 by tetanus, 193 ; of electrical organ, 
 affected by boiling, 214; du Bois- 
 Eeymond's views on internal positive 
 polarisability of muscle criticised by 
 Biedermann, 361. 
 
 Polarisation, internal of nerve and 
 muscle, 77; Tigerstedt on, 79-90 ; ex- 
 ternal/ polarisation-current explains, 
 according to du Bois-Reymond, differ- 
 ences in amount of contraction on 
 opening and closing main circuit and 
 nerve circuit, 98 ; rate of disappearance 
 ofpolarisati on- current, 105; Hermann's 
 theory of polarisation increment criti- 
 cised by Grutzner, 121; internal polar- 
 isation of moist porous bodies, 165 ; 
 experiments on positive and negative 
 polarisation of muscle as depending on 
 density and duration of current, 1 73 ; 
 du Bois-Reymond's results of experi- 
 ments on polarisation of nerves, similar 
 to those in the case of muscle, 197 ; 
 polarisation of nerve affected by boil- 
 ing, and by loss of vitality, 209 ; 
 Hermann on polarisation increment by 
 tetanus, 211; Hering's definition of 
 anodic and cathodic polarisation, 230; 
 du Bois-Reymond's doctrine of polari- 
 sation criticised by Hermann, 278- 
 284; Hermann's experiments on polar- 
 isation after-currents, in order to eluci- 
 date facts of after-currents in nerve and 
 muscle, 312 ; polarisation after-current 
 in muscle and nerve opposed to polaris- 
 ing current in the intra-polar, in the same 
 direction in the extra-polar region, 319. 
 
 Positive polarisation, predominates in 
 muscle in the direction from the 
 equator to the two ends, 188 ; in nerve 
 probably in the direction of the physio- 
 logical innervation wave, 206 ; in 
 electrical organ in direction of shock, 
 212 ; dependent on the integi-it}' of the 
 anodic spots, 244 ; du Bois-Reymond's 
 views on positive polarisation criticised 
 by Hermann, 325 ; positive cathodic 
 polarisation in adductor muscle of 
 Anodonta, 331 ; effect on, of the mode 
 of killing muscle, 360 ; discussion of 
 homodromous absolutely positive polar- 
 isation, 497 ; dependence of positive 
 polarisation on density of polarising 
 current, 536. 
 
 Quincke, explanation of anaphoric ac- 
 tions of current, 225. 
 
 Eanke, on diminution of resistance of 
 muscle by boiling, 190 ; on non-effect of 
 boiling or death on resistance of nerve, 
 220. 
 
INDEX. 
 
 553 
 
 Ranzi, on direction of shock in Mala- 
 pterurus, 398. 
 
 Reciprocal conduction in electrical 
 organ of Malapterurus, 519, 
 
 Relaxation, phenomena of, on electrica 1 
 excitation of cardiac muscle, 332 ; of 
 veratrine muscle, 335 ; is analogous to 
 that of systolically contracted cardiac 
 muscle and tonically shortened ad- 
 ductor muscle of Anodonta, 342. 
 
 Resistance, secondary, of moist porous 
 bodies, 168. 
 
 Ritter, on law of contraction for the 
 thigh of the frog, no; his opinion as 
 to the nature of the change which 
 
 - gives rise to the phenomena of break- 
 excitations, 116; on break tetanus in 
 nerve, 319. 
 
 Sachs, researches on Gymnotus, 216; on 
 organ-current in Gymnotus, 438 ; on 
 tetanus of organ, 445 ; on secondary 
 electromotive actions of Gymnotus, 448 ; 
 on com]:)arative strength of homodro- 
 mous and heterodromous currents, 457 > 
 on electromotive behaviour of skin, 482. 
 
 Schultze, Max, as to nerves of electrical 
 organ of Malapterurus, 389. 
 
 Secondary electromotive actions, of 
 the organ of Torpedo, 445 ; phenomena 
 of, essentially the same in the three 
 electrical fishes, 449 ; dependence of 
 these effects on the vitality of the 
 organ, 457 ; results of experiments on 
 these actions in Torpedo, 458-462. 
 
 Tetanus, Hering's method of producing 
 it without metals, 130; Hering's 
 secondary tetanus, 147 ; its effect on 
 polarisability of muscle, 193; its in- 
 fluence on internal positive polarisation 
 of nerve, 210 ; Hermann on ' polarisa- 
 tion increment ' by tetanus, 211. 
 
 Tigerstedt, changes of excitability pro- 
 duced in nerves by a constant current, 
 5 ; investigates special points in Pflti- 
 ger's doctrine of electrotonus, 14; his 
 method of experimenting, 16 ; his in- 
 strument for mechanical stimulation of 
 nerves, 1 8 ; experiments as to changes 
 of excitability produced by ascending 
 and descending currents, 21-42 ; criti- 
 cism of Hermann's theory of electro- 
 tonus, 44 ; contribution to the theory 
 of the break-contraction, 49 ; his ex- 
 perimental proof that the amplitude of 
 the break-contraction varies with the 
 
 intensity of negative polarisation of a 
 motor nerve and is influenced by its 
 polarisability, 54-64 ; proposed simpli- 
 fication of Pfliiger's law of contraction, 
 66 ; discussion of his theory, 66-70 ; 
 on the internal polarisation of nerves, 
 77; methods of experiment on in- 
 ternal polarisation of nerve, 79 ; ex- 
 periments showing the dependence of 
 polarisation on the strength of the 
 polarising current, 83; and on the 
 duration of its closure, 86 ; his con- 
 clusions as to break-excitation the 
 same as Griitzner's, 102 ; on increase 
 of polarisation-current with duration 
 of polarising current, 105 ; on stimu- 
 lus of weak and strong current, 114. 
 
 Torpedo, construction of organ, 371 ; 
 identity of its electricity with ordinary 
 electricity, 391 ; chemical reaction of 
 organ, 412 ; method of leading oS" the 
 shock, 423; effect of shock on man 
 compared with that of Malapterurus, 
 426 ; on iodide of potassium electro- 
 lysis by shock of Torpedo, 435 ; in- 
 vestigation of organ-current, 438 ; de- 
 pendence of the conductivity of the 
 organ on its vitality, 457 ; on electro- 
 motive actions of the electrical nerves, 
 462 ; experiments on electromotive be- 
 haviour of skin, 482 ; polarisation of 
 organ by alternating currents, 497. 
 
 Transmission time, defined, 172. 
 
 Tschirjew, comparison of his experi- 
 ments on polarisation of muscle with 
 those of du Bois-Eeymond, 247. 
 
 Unipolar action, Hering's discussion of, 
 267. 
 
 Veratrine, experiments on muscle 
 poisoned with it, 333; forms of con- 
 traction and relaxation on electrical 
 excitation of such muscle, 335 ; how to 
 poison muscle with it for experiments 
 on secondary electromotive phenomena, 
 343 ; discussion of experiments, 349- 
 
 352. . . 
 
 Vitality, loss of, its effect on polarisation 
 of nerve, 209. 
 
 Volta, his theory as to bi-eaking a con- 
 ducting circuit, 116; Tigerstedt's ex- 
 planation of voltaic alternatives, 70 ; 
 Griitzner discusses their influence, 100. 
 
 "Wundt, criticism of Pfliiger's doctrine 
 of electrotonus, 13, 
 
 THE END. 
 
 O 
 
April, 1887. 
 
 (ttlatentron iPtess, ©xfottr 
 
 A SELECTION OF 
 
 BOOKS 
 
 PUBLISHED FOR THE UNIVERSITY BY 
 
 HENRY PROWDE, 
 
 AT THE OXFORD UNIVERSITY PRESS WAREHOUSE, 
 AMEN CORNER, LONDON. 
 
 ALSO TO BE HAD AT THE 
 
 CLARENDON PRESS DEPOSITORY, OXFORD. 
 [^Every book is bound in cloth, ujtless otherwise described.'] 
 
 LEXICONS, GRAMMARS, ORIENTAL WORKS, &c. 
 
 Anglo-Saxon. — An Anglo-Saxon Dictionary, based on the 
 
 MS. Collections of the late Joseph Bosworth, D.D., Professor of Anglo-Saxon, 
 Oxford. Edited and enlarged by Prof. T. N. Toller, M.A. (To be completed 
 in four parts.) Parts I and II. A— HWISTLIAN. 4to. 15J. each. 
 
 Chinese. — A Handbook of the Chinese Language. By James 
 
 Summers. 1863. Svo. half bound, i/. 8j. 
 
 A Record of Buddhistic Kingdoms, by the Chinese Monk 
 
 FA-HIEN. Translated and annotated by James Legge, M.A., LL.D. Crown 
 4to. cloth back, \os. 6d. 
 
 English. — A New English Dictionary, on Historical Prin- 
 ciples : founded mainly on the materials collected by the Philological Society. 
 Edited by James A. H. Murray, LL.D., with the assistance of many Scholars 
 and men of Science. Part I. A— ANT. Part II, ANT— BATTEN. 
 Part m. BATTER— BOZ. Imperial 4to. 12 J. 6^. each. 
 
 An Etymological Dictionary of the English Language. 
 
 By W. W. Skeat, M.A. Second Edition. 1884. 4to. 2/. 4J. 
 
 Supplement to the First Edition of the above. 4to. 2s. 6d. 
 
 A Concise Etymological Dictionary of the English Lan- 
 guage. By W. W. Skeat, M.A. Second Edition. 1885. Crown 8vo. ^s.6d. 
 
 [9] 
 
CLARENDON PRESS, OXFORD. 
 
 Greek. — A Greek-English Lexicon, by Henry George 
 
 Liddell, D.D., and Robert Scott, D.D. Seventh Edition, Revised and Aug- 
 mented throughout. 1883. 4to. \l. i6s. 
 
 A Greek-English Lexicon, abridged from Liddell and 
 
 Scott's 4to. edition, chiefly for the use of Schools. Twenty-first Edition. 
 1884. Square 1 2 mo. ']s.6d. 
 
 A copious Greek-English Vocabulary, compiled from 
 
 the best authorities. 1850. 24mo. 
 
 A Practical Introduction to Greek Accentuation, by H. 
 
 \V. Chandler, M.A. Second Edition. 18S1. 8vo. \os.(>d. 
 
 Hebrew. — The Book of Hcbrezv Roots, by Abu '1-Walid 
 
 Marwan ibn Janah, otherwise called Rabbi Yonah. Now first edited, with an 
 Appendix, by Ad. Neubauer. 1875. 4to. 2/. 7j. 6^. 
 
 A Treatise on the use of the Tenses in Hcbrezv. By 
 
 S. R. Driver, D.D. Second Edition. 1 881. Extra fcap. 8vo. is.dd. 
 
 Hebrew Accentuation of Psalms, Proverbs, and Job. 
 
 By William Wickes, D.D. 1881. Demy 8vo. stiff covers, 5^. 
 
 Icelandic.—^;? Icelandic-EnglisJi Dictionary, based on the 
 
 MS. collections of the late Richard Cleasby. Enlarged and completed by 
 G. Vigfusson, M.A. With an Introduction, and Life of Richard Cleasby, by 
 G. Webbe Dasent, D.C.L. 1874. 4to. 3/. is. 
 
 A List of English Words the Etymology of which is 
 
 ilhisirated by comparison with Icelandic. Prepared in the form of an 
 Appendix to the above. By W. W. Skeat, M.A. 1876. stitched, 2s. 
 
 An Icelandic Primer, with Grammar, Notes, and 
 
 Glossary. By Henry Sweet, M.A. Extra fcap. 8vo. 3^-. 6d. 
 
 An Icelandic Prose Reader, with Notes, Grammar and 
 
 Glossary, by Dr. Gudbrand Vigfusson and F. York Powell, M.A. i87( 
 Extra fcap. Svo. 10s. dd. 
 
 Latin. — A Latin Dictionary, founded on Andrews' edition 
 
 of Freund's Latin Dictionary, revised, enlarged, and in great part rewritten 
 by Charlton T. Lewis, Ph.D., and Charles Short, LL.D. 1879. 4to. \l. 5J. 
 
 Melanesian. — The Melafiesian Languages. By R. H. 
 
 Codrington, D.D., of the Melanesian Mission. 8vo. i8j. 
 
 Sanskrit. — A Practical Grammar of the Sanskrit Language^ 
 
 arranged with reference to the Classical Languages of Europe, for the use of 
 English Students, by Sir M. Monier-Williams, M.A. Fourth Edition. Svo. 15^. 
 
 A Sanskrit-English Dictionary, Etymologically and 
 
 Philologically arranged, with special reference to Greek, Latin, German, Anglo- 
 Saxon, English, and other cognate Indo-European Languages. By Sir M. 
 Monier-Williams, M.A. 1872. 4I0. 4/. 14J. 61/. 
 
CLARENDON PRESS, OXFORD. 
 
 Sanskrit. — Nalopdkhydnam. Story of Nala, an Episode of 
 
 the Maha-Bharata : the Sanskrit text, with a copious Vocabulary, and an 
 improved version of Dean Milman's Translation, by Sir M. Monier- Williams, 
 M.A. Second Edition, Revised and Improved. 1879. 8vo. 15^. 
 
 Saktmtald. A Sanskrit Drama, in Seven Acts. Edited 
 
 by Sir M. Monier-Williams, M.A. Second Edition, 1876. 8vo. 2\s. 
 
 Syriac. — TJiesaiiriis Syriaais : collegerunt Quatremere, Bern- 
 stein, Lorsbach, Arnoldi. Agrell, Field, Roediger: edidit R. Payne Smith, 
 S.T.P. Ease. I-VI. 1868-83. sm. fol. each, i/. is. Ease. VII. \l.\\s.(>d. 
 Vol. I, containing Ease. I-V, sm. fol. 5/. ^s. 
 
 The Book ofKalilah and Dimnah. Translated from Arabic 
 
 into Syriac. Edited by W. Wright, LL.D. 1884. Svo. 21^. 
 
 GREEK CLASSICS, &c. 
 
 Aristophanes: A Complete Concordance to the Comedies 
 
 and Eragments. By Henry Dunbar, M.D. 4to. 1/. \s. 
 
 Aristotle: The Politics, with Introduction, Notes, etc., by 
 
 W. L. Newman, M.A., Eellow of Balliol College, Oxford. Vols. I. and II. 
 Nearly ready. 
 
 Aristotle : The Politics, translated into English, with Intro- 
 duction, Marginal Analysis, Notes, and Indices, by B. Jowett, M.A. Medium 
 8vo. 2 vols. 2 1 J. 
 
 Catalogus Codicnm Graecormn Sijiaiticoriim. Scripsit V. 
 
 Gardthausen Lipsiensis. With six pages of Facsimiles. 8vo. linen, 2^s. 
 
 Heracliti EpJiesii Reliquiae. Recensuit I. Bywater, M.A. 
 
 Appendicis loco additae sunt Diogenis Laertii Vita Heracliti, Particulae Hip- 
 pocratei De Diaeta Libri Primi, Epistolae Heracliteae. 1877. 8vo. 6j. 
 
 Herctilanensiiiin Vohmiimmi Partes II. 1824. 8vo. \os. 
 Fraginenta Herculanensia. A Descriptive Catalogue of the 
 
 Oxford copies of the Herculanean Rolls, together with the texts of several 
 papyri, accompanied by facsimiles. Edited by Walter Scott, M.A., Eellow 
 of Merton College, Oxford. Royal 8vo. cloth, 2\s. 
 
 Homer: A Complete Concordance to the Odyssey and 
 
 Hymns of Homer ; to which is added a Concordance to the Parallel Passages 
 in the Iliad, Odyssey, and Hymns. By Henry Dunbar, M.D. 18S0. 410. \l.is. 
 
 Scholia Graeca in Iliadeni. Edited by Professor W. 
 
 Dindorf, after a new collation of the Venetian MSS. by D. B. Monro, M.A., 
 Provost of Oriel College. 4 vols. 8vo. 2/. \qs. Vols. V and VI. In the Press. 
 
 Scholia Graeca in Odysseam. Edidit Guil. Dindorfius. 
 
 Tomi II. 1855. 8vo. 15s. 6d. 
 
 B 2 
 
CLARENDON PRESS, OXFORD. 
 
 Plato : Apology, with a revised Text and English Notes, and 
 
 a Digest of Platonic Idioms, by James Riddell, M.A. 1S78. 8vo. 8j. 6d. 
 
 P/iilebus, with a revised Text and Enghsh Notes, by 
 
 Edward Poste, M.A. i860. 8vo. 7^. 6d. 
 
 Sophistes and Poliiicus, with a revised Text and Enghsh 
 
 Notes, by L. Campbell, M.A. 1867. 8vo. iSj. 
 
 — Theaetetus, with a revised Text and Enghsh Notes, 
 
 by L. Campbell, M.A. Second Edition. 8vo. loj. dd. 
 
 — The Dialogues, translated into English, with Analyses 
 
 and Introductions, by B. Jowett, M.A. A new Edition in 5 volumes, medium 
 Svo. 1875. 3/. \os. 
 
 The Republic^ translated into English, with an Analysis 
 
 and Introduction, by B. Jowett, M.A. Medium Svo. \2s. 6d. 
 
 Thucydides : Translated into English, with Introduction, 
 
 Marginal Analysis, Notes, and Indices. By B. Jowett, M.A. 2 vols. 188 1. 
 Medium Svo. i/. 12s. 
 
 THE HOLY SCRIPTURES, &c. 
 
 Studia Biblica. — Essays in Biblical Archaeology and Criti- 
 cism, and kindred subjects. By Members of the University of Oxford. Svo. 
 10 J. 6d. 
 
 English. — The Holy Bible in the earliest English Versions, 
 
 made from the Latin Vulgate by John Wycliffe and his followers : edited by 
 the Rev. J. Forshall and Sir F. Madden. 4 vols. 1S50. Royal 4to. 3/. 3J. 
 
 [Also reprinted from the above, with Introduction and Glossary 
 by W. W. Skeat, M.A. 
 
 The Books of Job, Psalms, Proverbs, Ecclesiastes, and the 
 
 Song of Solomon : according to the Wycliffite Version made by Nicholas 
 de Hereford, about A.D. 1381, and Revised by John Purvey, about A.D. 1388. 
 Extra fcap. Svo. 3J. 6d. 
 
 The New Testament in English, according to the Version 
 
 by John Wycliffe, about A.D. 1380, and Revised by John Purvey, about a.d. 
 1388. Extra fcap. Svo. 6j.] 
 
CLARENDON PRESS, OXFORD. 
 
 English. — The Holy Bible: an exact reprint, page for page, 
 
 of the Authorised Version published in the year i6i i. Demy 4to. half bound, 
 
 l/. IS. 
 
 The Psalter, or Psalms of David, and certain Canticles, 
 
 with a Translation and Exposition in English, by Richard Rolle of Hampole. 
 Edited by H. R. Bramley, M.A., Fellow of S. M. Magdalen College, Oxford. 
 With an Introduction and Glossary. Demy 8vo. il. \s. 
 
 Lectures on Ecclesiastes. Delivered in Westminster 
 
 Abbey by the Very Rev. George Granville Bradley, D.D., Dean of West- 
 minster. Crown 8vo. 4^. dd. 
 
 Gothic. — The Gospel of St. Mark in Gothic, according to 
 
 the translation made by Wulfila in the Fourth Century. Edited with a 
 Grammatical Introduction and Glossarial Index by W. W. Skeat, M.A. 
 Extra fcap. 8vo. 4^. 
 
 Greek. — Vetus Testame^itnm ex Versione Septuaginta Inter- 
 
 pretum secundum exemplar Vaticauum Romae editum. Accedit potior varietas 
 Codicis Alexandrini. Tomi III. Editio Altera. iSmo. iSj-. 
 
 Origenis Hexaploriim quae supersunt ; sive, Veterum 
 
 Interpretum Graecorum in totum Vetus Testamentum Fragmenta. Edidil 
 Fridericus Field, A.M. 2 vols. 1875. 4^0- 5^- S-f- 
 
 The Book of Wisdom: the Greek Text, the Latin 
 
 Vulgate, and the Authorised English Version ; with an Introduction, Critical 
 Apparatus, and a Commentary. By William J. Deane, M.A. Small4to. 12s. 6d. 
 
 Novum Testamentum Graece. Antiquissimorum Codicum 
 
 Textus in ordine parallelo dispositi. Accedit coUalio Codicis Sinaitici. Edidit 
 E. H. Hansell, S.T. B. Tomilll. 1864. 8vo. half morocco. Price reduced 
 to 24J. 
 
 Novum Testamentum Graece. Accedunt parallela S. 
 
 Scripturae loca, etc, Edidit Carolus Lloyd, S.T.P.R. i8mo. 3J-. 
 On writing paper, with wide margin, i os. 
 
 Novum Testamentum Graece juxta Exemplar Millianum. 
 
 i8mo. 2 J. 6^. On writing paper, with wide margin, 9J. 
 
 Evangelia Sacra Graece. Fcap. 8vo. limp, \s. 6d. 
 
 The Greek Testament, with the Readings adopted by 
 
 the Revisers of the Authorised Version: — 
 
 (1) Pica type, with Marginal References. Demy 8vo. loj. dd. 
 
 (2) Long Primer type. Fcap. Svo. 4s. 6rf. 
 
 (3) The same, on writing paper, with wide margin, 15s. 
 
 The Parallel New Testament, Greek and English ; being 
 
 the Authorised Version, 1611 ; the Revised Version, 1881; and the Greek 
 Text followed in the Revised Version. Svo. \2s. 6d. 
 
 The Revised Version is the joint property of the Universities of Oxford attd Cambridge. 
 
CLARENDON PRESS, OXFORD. 
 
 Greek. — Canon Mtiratorianus : the earliest Catalogue of the 
 
 Books of the New Testament. Edited with Notes and a Facsimile of the 
 MS. in the Ambrosian Library at Milan, by S. P. Tregelles, LL.D. 1S67. 
 4to. loj. dd. 
 
 Outlines of Textual Criticism applied to the New Testa- 
 
 7nent. By C. E. Hammond, M.A. Fourth Edition. Extra fcap. 8vo. 3J. dd. 
 
 Hebrew, etc. — The Psalms in Hebrew without points . 1879. 
 
 Crown 8vo. 3J. dd. 
 
 ■ A Commentary on the Book of Proverbs. Attributed 
 
 to Abraham Ibn Ezra. Edited from a MS. in the Bodleian Library by 
 S. R. Driver, M.A. Crown 8vo. paper covers, ^s. 6d. 
 
 The Book of Tobit. A Chaldee Text, from a unique 
 
 MS. in the Bodleian Library ; with other Rabbinical Texts, English Transla- 
 tions, and the Itala. Edited by Ad. Neubauer, M.A. 1878. Crown 8vo. 6s. 
 
 ■ Horae Hebraicae et Talmudicae, a J. Lightfoot. A new 
 
 Edition, by R. Gandell, M.A. 4 vols. 1859. 8vo. i/. is. 
 
 Latin. — Libri Psalmorum Versio antiqua Latina, cum Para- 
 
 phrasi Anglo-Saxonica. Edidit B. Thorpe, F.A.S. 1835. ^^°' ^°-''- ^^- 
 
 Old-Latin Biblical Texts: No. I. The Gospel according 
 
 to St. Matthew from the St. Germain MS. (g,). Edited with Introduction 
 and Appendices by John Wordsworth, D.D. Small 4to., stiff covers, 6s. 
 
 Old-Latin Biblical Texts : No. IL. Portions of the Gospels 
 
 according to St. Mark and St. Matthew, from the Bobbio MS. (k), &c. 
 Edited by John Wordsworth, D.D., W. Sanday, M.A., D.D., and H. J. White, 
 M.A. Small 4to., stiff covers, 2\s. 
 
 Old-French. — Libri Psalmorum Versio antiqua Gallica e 
 
 Cod. MS. in Bibl. Bodleiana adservato, una cum Versione Metrica aliisque 
 Monumentis pervetustis. Nunc primum descripsit et edidit Franciscus Michel, 
 Phil. Doc. i860. 8vo. \os. 6d. 
 
 FATHERS OP THE CHURCH, &c. 
 
 St.Athanasius: Historical Writings, according to the Bene- 
 dictine Text. With an Introduction by William Bright, D.D. 1881. Crown 
 8vo. loj. 6d. 
 
 Orations against the Arians. With an Account of his 
 
 Life by William Bright, D.D. 1873. Crown 8vo. pj-. 
 
 St. Augustine: Select Anti-Pelagian Treatises, and the Acts 
 
 of the Second Council of Orange. With an Introduction by William Bright, 
 D.D. Crown 8vo. 9^. 
 
CLARENDON PRESS, OXFORD. 7 
 
 Canons of the First Four General Councils of Nicaea, Con- 
 stantinople, Ephesus, and Chalcedon. 1877. Crown 8vo. 2s. 6c?, 
 
 Notes on the Canons of the First Four General Councils. 
 
 By William Bright, D.D. 1882. Crown 8vo. 5^. 6(/. 
 
 Cyrilli Archiepiscopi Alexandrini in XII Prophetas. Edidit 
 
 P. E. Pusey, A.M. Tomi II. 1868. 8vo. cloth. 2/. 2s. 
 
 in D. Joannis Evangelium. Accedunt Fragmenta varia 
 
 necnon Tractatus ad Tiberium Diaconum duo. Edidit post Aubertum 
 P. E. Pusey, A.M. Tomi III. 1872. 8vo. 2/. 5J. 
 
 Commentarii in Liicae Evangelium quae supers unt 
 
 Syriace. E MSS. apud Mus. Britan. edidit R. Payne Smith, A.M. 1858. 
 4to. l/. 2S. 
 
 Translated by R. Payne Smith, M.A. 2 vols. 1859. 
 
 8vo. 1A,S. 
 
 Ephraemi Syri^ Rabulae Episcopi Edesseni, Balaei, aliorum- 
 
 que Opera Selecta. E Codd. Syriacis MSS. in Museo Britannico et Bibliotheca 
 Bodleianaasservatis primus edidit J. J. Overbeck. 1865. 8vo. li, is. 
 
 Eusedius' Ecclesiastical History, according to the text of 
 
 Burton, with an Introduction by William Bright, D.D, 1881. Crown 8vo. 
 8j. 6(/. 
 
 Irenaetts : The Third Book of St. Ireiiaeus, Bishop of Lyons, 
 
 against Heresies. W'ith short Notes and a Glossary by H. Deane, B.D. 
 1874. Crown 8vo, 5J. 6d. 
 
 Patrum Apostolicoruni, S, Clementis Romani, S. Ignatii, 
 
 S. Polycarpi, quae supersunt, Edidit Guil, Jacobson, S,T,P.R, Tomi II. 
 Fourth Edition, 1863. 8vo. i/. is. 
 
 Socrates' Ecclesiastical History, according to the Text of 
 
 Hussey, with an Introduction by William Bright, D.D. 1878. Crown 8 vo. 
 7j-. ()d. 
 
 ECCLESIASTICAL HISTORY, BIOGRAPHY, &c. 
 
 Ancient Liturgy of the Church of England, according to the 
 
 uses of Sarum, York, Hereford, and Bangor, and the Roman Liturgy arranged 
 in parallel columns, with preface and notes. By William Maskell, M.A. 
 Third Edition. 1882. 8vo. 15^. 
 
 Baedae Historia Ecclesiastica. Edited, with English Notes, 
 
 by G. H. Moberly, M.A. 18S1. Crown 8vo. los. f)d. 
 
CLARENDON PRESS, OXFORD. 
 
 Bright ( W.). Chapters of Early English Church History. 
 
 1878. 8vo. \2S. 
 
 Burnefs History of the Reformation of the Church of England. 
 
 A new Edition. Carefully revised, and the Records collated with the originals, 
 by N. Pocock, M.A. 7 vols. 1865. 8vo. Price reduced to \l. \os. 
 
 Councils and Ecclesiastical Documents relating to Great Britain 
 
 and Ireland. Edited, after Spelman and Wilkins, by A. W. Haddan, B.D., 
 and W, Stubbs, M.A. Vols. I. and III. 1869-71. Medium 8vo. each i/. \s. 
 
 Vol. II. Part I. 1873. Medium Svo. loj. 6c/. 
 
 Vol.11. Part II. 1878. Church of Ireland; Memorials of St. Patrick. 
 Stiff covers, y. 6d. 
 
 Hamilton [jfohji, Archbishop of St. Andrews^, The Catechism 
 
 of. Edited, with Introduction and Glossary, by Thomas Graves Law. With 
 a Preface by the Right Hon. W. E. Gladstone. Svo. 1 2s. 6d. 
 
 HammoJtd {C. E.). Liturgies, Eastern aJid IVestern. Edited, 
 
 with Introduction, Notes, and Liturgical Glossary. 1878. Crown 8vo. 10s. 6d. 
 An Appendix to the above. 1879. Crown Svo. paper covers, is. 6d. 
 
 John, Bishop of Ephesus. The Third Part of his Eccle- 
 siastical History. [In Syriac] Now first edited by William Cureton, M.A. 
 
 1853. 4to. l/. 12S. 
 
 Translated by R. Payne Smith, M.A. i860. Svo. 10s. 
 
 Leofric Missal, The, as used in the Cathedral of Exeter 
 
 during the Episcopate of its first Bishop, a.d. 1050-1072 ; together with some 
 Account of the Red Book of Derby, the Missal of Robert of Jumieges, and a 
 few other early MS. Service Books of the English Church. Edited, with In- 
 troduction and Notes, by F. E. Warren, B.D. 4to. half morocco, 35J. 
 
 Monumenta Ritualia Ecclesiae Anglicanae. The occasional 
 
 Offices of the Church of England according to the old use of Salisbury, the 
 Prymer in English, and other prayers and forms, with dissertations and notes. 
 By William Maskell, M.A. Second Edition. 1S82. 3 vols. Svo. 2I. los. 
 
 Records of the Reformation. The Divorce, 1527-1533. Mostly 
 
 now for the first time printed from MSS. in the British Museum and other libra- 
 ries. Collected and arranged by N. Pocock, M.A. 1870. 2 vols. Svo. i/. i6j. 
 
 Shirley ( W. W.). Some Account of the Church in the Apostolic 
 
 Age. Second Edition, 1874. Fcap. Svo. 3s. 6rf. 
 
 Stubbs ( W.). Registrum Sacrum Anglicanum. An attempt 
 to exhibit the course of Episcopal Succession in England. 1858. Small 4to. 
 8 J. dd. 
 
 Warren {F. E.). Liturgy and Ritual of the Celtic Church. 
 
 1 88 1. Svo. 14J. 
 
CLARENDON PRESS, OXFORD 
 
 ENGLISH THEOLOGY. 
 
 Bampton Lectures, 1886, The Christian Platonists of Alex- 
 andria. By Charles Bigg, D.D. 8vo. los. 6d. 
 
 Butler's Works, with an Index to the Analogy. 2 vols. 1874. 
 
 8vo. IIS. 
 
 Also separately. 
 
 Sermons, ^s. 6d. Analogy of Religion, ^s. 6d 
 
 GreswelF s Harmonia Evangelica. Fifth Edition. 8vo. 1855. 
 
 9^. 6d. 
 
 LLeurtlefs Harmonia Symbolica: Creeds of the Western 
 
 Church. 1858. 8vo. 6j. (id. 
 
 Liomilies appointed to he read in Churches. Edited by 
 
 J. Griffiths, M.A. 1859. 8vo. 'js. 6d. 
 
 Hooker s Works, with his life by Walton, arranged by John 
 
 Keble, M.A. Sixth Edition, 1874. 3 vols. 8vo. i/. lu. 6d. 
 
 the text as arranged by John Keble, M.A. a vols. 
 
 1875. 8vo. lis. 
 
 Jewel's Works. Edited by R. W. Jelf, D.D. 8 vols. 1848. 
 
 8vo. i/. los. 
 
 Pearson's Exposition of the Creed. Revised and corrected by 
 
 E. Burton, D.D. Sixth Edition, 1877. 8vo. loj. 6d. 
 
 Waterland's Review of the Doctrine of the Eticharist, with 
 
 a Preface by the late Bishop of London. Crown 8vo. 6j. dd. 
 
 Works, with Life, by Bp. Van Mildert. A new Edition, 
 
 with copious Indexes. 6 vols. 1856. 8vo. 2/. lu. 
 
 Wheatlys Illustration of the Book of Common Prayer. A new 
 
 Edition, 1846. 8vo. 5J. 
 
 Wyclif. A Catalogue of the Original Works of John Wyclif 
 
 by V^. W. Shirley, D.D. 1865. 8vo. 3^-. 6d. 
 
 Select English Works. By T. Arnold, M.A. 3 vols. 
 
 1869-1871. 8vo. il. IS. 
 
 Trialogus. With the Supplement now first edited. 
 
 By Gotthard Lechler. 1869. 8vo. 7j-. 
 
lo CLARENDON PRESS, OXFORD. 
 
 HISTORICAL AND DOCUMENTARY WORKS. 
 
 British Barrozus, a Record of the Examination of Sepulchral 
 
 Mounds in various parts of Enc;land. By \Yilliam Greenwell, M.A., F.S.A. 
 Together with Description of Figures of Skulls, General Remarks on Pre- 
 historic Crania, and an Appendix by George Rolleston, M.D., F.R.S. 1877. 
 Medium 8vo. 25^-. 
 
 Britton. A Treatise upon the Common Law of England, 
 
 composed by order of King Edward I. The French Text carefully revised, 
 with an English Translation, Introduction, and Notes, by F. M.Nichols, M. A. 
 2 vols. 1S65. Royal Svo. i/. i6.f. 
 
 Clarendoii s History of the Rebellion and Civil Wars in 
 
 England. 7 vols. 1839. i8mo. i/. \s. 
 
 Clarendons Llistoiy of the Rebellion and Civil Wars in 
 
 England. Also his Life, written by himself, in which is included a Con- 
 tinuation of his History of the Grand Rebellion. With copious Indexes. 
 In one volume, royal Svo. 1842. i/.2s. 
 
 Clinton'' s Epitome of the Fasti Hellenici. 1 85 1 . Svo. 6s. 6d. 
 
 Epitome of the Fasti Romani. 1854. Svo. 'js. 
 
 Corpvs Poeticvm Boreale. The Poetry of the Old Northern 
 
 Tongue, from the Earliest Times to the Thirteenth Century. Edited, clas- 
 sified, and translated, with Introduction, Excursus, and Notes, by Gudbrand 
 Vigfusson, M.A., and F. York Powell, M.A. 2 vols. 1883. Svo. 42^-. 
 
 Freeman [E. A.). History of the Norman Conquest of Eng- 
 land; its Causes and Results. In Six Volumes. Svo. 5/. ^s. 6d. 
 
 The Reign of William Rnfus and the Accession of 
 
 Henry the First. 2 vols. Svo. i/. ids. 
 
 Gascoignes Theological Dictionary ("Liber Veritatum"): 
 
 Selected Passages, illustrating the condition of Church and State, 1403-1458. 
 With an Introduction by James E. Thorold Rogers, M.A. Small 4to, 
 loj. 6^. 
 
 Magna Carta, a careful Reprint. Edited by W. Stubbs, D.D. 
 
 1879. 4to. stitched, IJ. 
 
 Passio et Miracula Beati Olaiii. Edited from a Twelfth- 
 Century MS. in the Library of Corpus Christi College, Oxford, with an 
 Introduction and Notes, by Frederick Metcalfe, M.A. Small 4to. stiff 
 covers. 6^. 
 
CLARENDON PRESS, OXFORD. ir 
 
 Protests of the Lords, including those which have been ex- 
 punged, from 1624101874; with Historical Introductions. Edited by James 
 E. Thorold Rogers, M. A. 1875. 3 vols. Svo. 2/. 2j. 
 
 Rogers [y. E. T.). History of Agriculture and Prices in 
 
 England, A.D. 1 259-1 793. 
 
 Vols. I and II (i 259-1400). 1866. 8vo. 2l.2s. 
 Vols. IIIandIV(i40i-i582). 1882. 8vo. 2L10S. 
 
 Saxon Chronicles [Two of the) parallel, with Supplementary- 
 Extracts from the Others. Edited, with Introduction, Notes, and a Glos- 
 sarial Index, by J. Earle, M.A. 1865. 8vo. i6j. 
 
 Sttibbs (W., D.D.). Seventeen Lectures on the Study of 
 
 Medieval and Modern History, &c., delivered at Oxford 1867-1884. Demy 
 Svo. half-bound, 105. 6d. 
 
 Sturlunga 'Saga, inckiding the Islendinga Saga of Lawman 
 
 Sturla Thordsson and other works. Edited by Dr. Gudbrand Vigfusson. 
 In 2 vols. 1878. Svo, 2/. 2s. 
 
 York Plays. The Plays performed by the Crafts or Mysteries 
 
 of York on the day of Corpus Christi in the 14th, 15th, and i6th centuries. 
 Now first printed from the unique MS. in the Library of Lord Ashburnham. 
 Edited with Introduction and Glossary by Lucy Toulmin Smith. Svo, 2\s. 
 
 Statutes made for the University of Oxford, and for the Colleges 
 
 and Hails therein, by the University of Oxford Commissioners. 1882. Svo. 
 I2J. dd. 
 
 Statuta Universitatis Oxoniensis. i885. Svo. $s. 
 
 The Examination Statutes for the Degrees of B.A., B. Mtis., 
 
 B.C.L., and B.M. Revised to Hilary Term, 1887. Svo. sewed, \s. 
 
 The Students Handbook to the University and Colleges of 
 
 Oxford. Extra fcap. Svo. 2S. 6d. 
 
 The Oxford University Calendar for the year 1887. Crown 
 
 Svo, 4J. 6d. 
 The present Edition includes all Class Lists and other University distinctions 
 for the seven years ending with 1886. 
 
 Also, supplementary to the above, price 5s. (pp. 606), 
 
 The Honours Register of the University of Oxford. A complete 
 
 Record of University Honours, Officers, Distinctions, and Class Lists; of the 
 Heads of Colleges, &c., &c., from the Thirteentli Century to 1883. 
 
I a CLARENDON PRESS, OXFORD. 
 
 MATHEMATICS, PHYSICAL SCIENCE, &c. 
 
 Acland{H. W., M.D., F.R.S.). Synopsis of the Pathological 
 
 Series in the Oxford Museum. 1867. 8vo. 2s.6d. 
 
 De Bary {Dr. A.). Coviparative Anatomy of the Vegetative 
 
 Organs of the Phanerogams and Ferns. Translated and Annotated by F. O. 
 Bower, M.A., F.L.S., and D. H. Scott, M.A., Ph.D., F.L.S. With 241 
 woodcuts and an Index. Royal Svo., half morocco, i/. 2s. 6d. 
 
 De Bary [Dr. A). Comparative Morphology and Biology of 
 
 the Fungi, Mycetozoa and Bacteria. Authorised English Translation by 
 Henry E. F. Gamsey, M.A. Revised by Isaac Bayley Balfour, M.A., M.D., 
 F.R.S. With 19S Woodcuts. Royal Svo., half morocco, i/. 2s. 6d. 
 
 Goebel {Dr. K.). Outlines of Classification and Sp'ecial Mor- 
 phology of Plants. A New Edition of Sachs' Text Book of Botany, Book II. 
 English Translation by H. E. F. Gamsey, M.A. Revised by T. Bayley Balfour, 
 M.A., M.D., F.R.S. With 407 Woodcuts. Royal Svo. half Morocco, 2\s. 
 
 Sachs {Jidins voji). Lectures on the Physiology of Plants. 
 
 Translated by H. Marshall Ward, M.A. With 445 Woodcuts. Royal Svo. 
 Just ready. 
 
 Mailer (y.). On certain Variations in the Vocal Organs of 
 
 the Passeres that have hitherto escaped notice. Translated by F. J. Bell, B.A., 
 and edited, with an Appendix, by A. H. Garrod. M.A., F.R.S. With Plates. 
 1878. 4to. paper covers, 7^. dd. 
 
 Price [Bartholomew^ M.A., F.R.S.). Treatise on Infinitesimal 
 
 Calculus. 
 
 Vol. I. Differential Calculus. Second Edition. Svo. 14^.6^. 
 
 Vol. II. Integral Calculus, Calculus of Variations, and Differential Equations. 
 Second Edition, 1865. Svo. i8j. 
 
 Vol. III. Statics, including Attractions; Dynamics of a Material Particle. 
 Second Edition, 1868. Svo. i6j. 
 
 Vol. IV. Dynamics of Material Systems; together with a chapter on Theo- 
 retical Dynamics, by W. F. Donkin, M.A., F.R.S. 1862. Svo. i6j. 
 
 Pritchard [C., D.D., F.R.S.). Uranometria Nova Oxoniensis. 
 
 A Photometric determination of the magnitudes of all Stars visible to the naked 
 eye, from the Pole to ten degrees south of the Equator. 1885. Royal Svo. Sj.6«/. 
 
 Astronomical Observations made at the University 
 
 Observatory, Oxford, under the direction of C. Pritchard, D.D. No. i. 
 1S78. Royal Svo. paper covers, 3J. M. 
 
CLARENDON PRESS, OXFORD. 13 
 
 Rigaud's Correspondence of Scientific Me7i of the I'jtk Century, 
 
 with Table of Contents by A. de Morgan, and Index by the Rev. J. Rigaud, 
 M.A. 2 vols. 1841-1862. 8vo. i8j. 6^. 
 
 Rolleston [George, M.D., F.R.S.). Scientific Papers and Ad- 
 dresses. Arranged and Edited by William Turner, M.B., F.R.S. With a 
 Biographical Sketch by Edward Tylor, F.R.S. With Portrait, Plates, and 
 Woodcuts. 2 vols. 8vo. il. 4?. 
 
 Westwood {jf. O., M.A., F.R.S.). Thesaurus Entomologicus 
 
 Hopeianus, or a Description of the rarest Insects in the Collection given to 
 the University by the Rev. William Hope. With 40 Plates. 1874. Small 
 folio, half morocco, 7/. iQs. 
 
 W^t Sbamlr 23oofes of tSe l^ast. 
 
 Translated by various Oriental Scholars, and edited by 
 F. Max Mxjller. 
 
 [Demy 8vo. cloth.] 
 
 Vol. I. The Upanishads. Translated by F. Max Miiller. 
 
 Part I. The A'^andogya-upanishad, The Talavakara-upanishad, The Aitareya- 
 ara«yaka, The Kaushitaki-brahmawa-upanishad, and The Va^saneyi-saw^hita- 
 upanishad. loj. dd. 
 
 Vol. II. The Sacred Laws of the Aryas, as taught in the 
 
 Schools of Apastamba, Gautama, Vasish^/^a, and Baudh3,yana. Translated by 
 Prof. Georg Biihler. Part I. Apastamba and Gautama. \os. 6d. 
 
 Vol. III. The Sacred Books of China. The Texts of Con- 
 fucianism. Translated by James Legge. Part I. The Shfi King, The Reli- 
 gious portions of the Shih King, and The Hsiao King. 1 2s. 6d. 
 
 Vol. IV. The Zend-Avesta. Translated by James Darme- 
 
 steter. Part I. The Vendidad. 10s. 6d. 
 
 Vol. V. The Pahlavi Texts. Translated by E. W. West. 
 
 Part I. The Bundahij-, Bahman Yaj't, and Shayast la-shayast. 12s. 6d. 
 
 Vols. VI and IX. The Qur'an. Parts I and II. Translated 
 
 by E. H. Palmer. 2ij. 
 
 Vol. VII. The Institutes of Vish«u. Translated by Julius 
 
 Jolly. loj. 6d, 
 
14 CLARENDON PRESS, OXFORD. 
 
 Vol. VIII. The Bhagavadgita, with The Sanatsu^atiya, and 
 
 The Anugita. Translated by Kashinath Trimbak Telang. loj-. dd. 
 
 Vol. X. The Dhammapada, translated from Pali by F. Max 
 
 Miiller; and The Sutta-Nipata, translated from Pali by V. Fausboll; being 
 Canonical Books of the Buddhists. \os. 6d. 
 
 Vol. XI. Buddhist Suttas. Translated from Pali by T. W. 
 
 Rhys Davids, i. The Mahaparinibbana Suttanta ; 2. The Dhamma-/('akka- 
 ppavattana Sutta ; 3. The Tevifo-a Suttanta; 4. The Akahkheyya Sutta ; 
 5. The Aetokhila Sutta; 6. The Maha-sudassana Suttanta ; 7. The Sabbasava 
 Sutta. los. 6d. 
 
 Vol. XII. The 5atapatha-Brahma«a, according to the Text 
 
 of the Madhyandina School. Translated by Julius Eggeling. Part I, 
 Books I and II. 12s. 6d. 
 
 Vol. XIII. Vinaya Texts. Translated from the Pali by 
 
 T. W. Rhys Davids and Hermann Oldenberg. Part I. The Patimokkha. 
 The Mahavagga, I-IV. 10s. 6d. 
 
 Vol. XIV. The Sacred Laws of the Aryas, as taught in the 
 
 Schools of Apastamba, Gautama, Vasish;'//a and Baudhayana. Translated 
 by Georg Biihler. Part II. Vasish/Zia and Baudhayana. loj. 6(/. 
 
 Vol. XV. The Upanishads. Translated by F. Max Muller. 
 
 Part II. The Ka//;a-upanishad, The Mu«(/aka-upanishad, The Taittiriyaka- 
 upanishad, The Br/Tiadarawyaka-upanishad, The 6'vetajvatara-upanishad, The 
 Praj-^a-upanishad, and The Maitraya;za-Brahma«a-upanishad. los. dd. 
 
 Vol. XVI. The Sacred Books of China. The Texts of Con- 
 fucianism. Translated by James Legge. Part II. The Yi King. ioj. 6^. 
 
 Vol. XVII. Vinaya Texts. Translated from the Pali by 
 
 T. W. Rhys Davids and Hermann Oldenberg. Part II. The Mahavagga, 
 V-X. The AuUavagga, I-III. ioj-. (>d. 
 
 Vol. XVIII. Pahlavi Texts. Translated by E. W. West. 
 
 Part II. The Da^istan-i Dinik and The Epistles of Manu^-z^ihar. 1 2s, dd. 
 
 Vol. XIX. The Fo-sho-hing-tsan-king. A Life of Buddha 
 
 by Ajvaghosha Bodhisattva, translated from Sanskrit into Chinese by Dhar- 
 maraksha, a.d. 420, and from Chinese into English by Samuel Beal. ioj. dd. 
 
 Vol. XX. Vinaya Texts. Translated from the Pali by T. W. 
 Rhys Davids and Hermann Oldenberg. Part III. The A'uUavagga, IV-XII. 
 IOJ-. dd. 
 
CLARENDON PRESS, OXFORD. 15 
 
 Vol. XXI. The Saddharma-pu;?^arika ; or, the Lotus of the 
 
 True Law. Translated by H. Kern. 12s. 6d. 
 
 Vol. XXII. G^aina-Sutras. Translated from Prakrit by Her- 
 mann Jacobi. Part I. The AMranga-Sutra. The Kalpa-Sutra. loj. 6d. 
 
 Vol. XXIII. The Zend-Avesta. Translated by James Dar- 
 
 Efiesteter. Part II. The Sirozahs, Ya^ts, and Nyayij. 10s. dd. 
 
 Vol. XXIV. Pahlavi Texts. Translated by E. W. West. 
 
 Part III. Dina-i Mainog-i Khirad, 6'ikand-gumanik, and Sad-Dar. 10s. 6d. 
 
 Second Series. 
 Vol. XXV. Manu. Translated by Georg Biihler. 'Zis. 
 Vol. XXVI. The wSatapatha-Brahma«a. Translated by 
 
 Julius Eggeling. Part II. 12s. 6d. 
 
 Vols. XXVII and XXVIII. The Sacred Books of China. 
 
 The Texts of Confucianism. Translated by James Legge. Parts III and IV. 
 The Li K\, or Collection of Treatises on the Rules of Propriety, or Ceremonial 
 Usages. 2 5 J. 
 
 Vols. XXIX and XXX. The GrzTiya-Sutras, Rules of Vedic 
 
 Domestic Ceremonies. Translated by Hermann Oldenberg. 
 
 Part I (Vol, XXIX), \2s. (,d. Jtist Published. 
 Part II (Vol. XXX). In the Press. 
 
 The following Volumes are in the Press:— 
 
 Vol. XXXI. ^ The Zend-Avesta. Part III. The Yasna, 
 
 Visparad, Afrinagan, and Gahs. Translated by L. H, Mills. Jtist ready. 
 
 Vol. XXXII. Vedic Hymns. Translated by F. Max Muller. 
 
 Parti. 
 
 Vol. XXXIII. Narada, and some Minor Law-books. 
 
 Translated by Julius Jolly. [^Preparing.'] 
 
 Vol. XXXIV. The Vedanta-Sutras, with 6'ahkara's Com- 
 mentary. Translated by G. Thibaut. [Preparing.l 
 
 *^ 7he Second Series will consist of Twenty-Four Volumes, 
 
1 6 CLARENDON PRESS, OXFORD. 
 
 Cfarmtr0n %xt^% Snijes 
 
 I. ENGLISH, &c. 
 
 A First Reading Book. By Marie Eichens of Berlin ; and 
 
 edited by Anne J. Clough. Extra fcap. 8vo. stiff covers, 4c/. 
 
 Oxford Reading Book, Part I. For Little Children. Extra 
 
 fcap. 8vo. stiff covers, 6d. 
 
 Oxford Reading Book, Part II. For Junior Classes. Extra 
 
 fcap. 8vo. stiff covers, dd. 
 
 An Elementary Efiglish Grammar and Exercise Book- By 
 
 O. W. Tancock, M.A. Second Edition. Extra fcap. 8vo. is. 6d. 
 Afi English Grammar and Reading Book, for Lower Forms 
 
 in Classical Schools. By O. W. Tancock, M.A. Fourth Edition. Extra 
 fcap. 8vo. 3^. (id. 
 
 Typical Selections from the best English Writers, with Intro- 
 ductory Notices. Second Edition. In 2 vols. Extra fcap. 8vo. 3^. dd. each. 
 Vol. I. Latimer to Berkeley. Vol. II. Pope to Macaiilay. 
 
 Shairp {J. C., LL.D.). Aspects of Poetry ; being Lectures 
 
 delivered at Oxford. Crown 8vo. loj. 6d. 
 
 A Book for the Beginner in Anglo-Saxon. By John Earle, 
 
 M.A. Third Edition. Extra fcap. 8vo. 2S. 6d. 
 
 An Anglo-Saxon Reader. In Prose and Verse. With Gram- 
 matical Introduction, Notes, and Glossary. By Henry Sweet, M.A. Fourth 
 Edition, Revised and Enlarged. Extra fcap. 8vo. Ss. 6d. 
 
 A Second Anglo-Saxon Reader. By the same Author. Extra 
 
 fcap. 8vo. Nearly ready. 
 
 An Anglo- Saxo7i Primer, with Grammar , Notes, ajid Glossary. 
 
 By the same Author. Second Edition. Extra fcap. 8vo. 2j. 6d. 
 
 Old Ejiglish Reading Primers ; edited by Henry Sweet, M.A. 
 
 I. Selected Homilies of .(Elfric. Extra fcap. 8vo., stifT covers, \s. 6d. 
 
 II. Extracts from Alfred's Orosius. Extra fcap. 8vo., stiff covers, is. 6d. 
 
 First Middle E7iglish Primer, with Grammar and Glossary. 
 
 By the same Author. Extra fcap. Svo. 2s. 
 
 Second Middle English Primer. Extracts from Chaucer, 
 
 with Grammar and Glossary. By the same Author. Extra fcap. Svo. 2s. 
 
 Principles of English Etymology. First Series. By W. W. 
 Skeat, Litt.D. Crown Svo. Nearly ready. 
 
CLARENDON PRESS, OXFORD. 17 
 
 The Philology of the English Tongue. By J. Earle, M.A. 
 
 Third Edition. Extra fcap. 8vo. 1$. dd. 
 
 An Icelandic Pj-imer, with Grammar, Notes, and Glossary. 
 
 By the same Author. Extra fcap. 8vo. 3J. ()d. 
 
 An Icelandic Prose Reader, with Notes, Grammar, and Glossary. 
 
 By G. Vigfiisson, M.A., and F. York Powell, M.A. Ext. fcap. Svo. 10s. 6d. 
 
 A Handbook of Phonetics, including a Popular Exposition of 
 
 the Principles of Spelling Reform. By H. Sweet, M.A. Extra fcap. Svo. 4^. 6d. 
 
 Elemcntarbuch des Gesprochenen Englisch. Grammatik, 
 
 Texte und Glossar. Von Henry Sweet. Extra fcap. 8vo., stiff covers, 2s. 6d. 
 
 The Ormulnm ; with the Notes and Glossary of Dr. R. M, 
 
 White. Edited by R. Holt, M.A. 1878. 2 vols. Extra fcap. Svo. 2 u. 
 
 Specimens of Early English. A New and Revised Edition. 
 
 With Introduction, Notes, and Glossarial Index. By R. Morris, LL.D., and 
 W. W. Skeat, M.A. 
 
 Part I. From Old English Homilies to King Horn (A.D. nsoto A.D. 1300). 
 
 Second Edition. Extra fcap. Svo. 9J. 
 Part II. From Robert of Gloucester to Gower (A.D. 1298 to A.D. 1393). 
 Second Edition. Extra fcap. Svo. 7^. dd. 
 
 Specimens of English Literature, from the ' Ploughmans 
 
 Crede' to the ' Shepheardes Calender' (a.d. 1394 to a.d. 1579). With Intro- 
 duction, Notes, and Glossarial Index. By W. W. Skeat, M.A. Extra fcap. 
 Svo. 1$. 6d. 
 
 The Vision of William concerning Piers the Plowman, in three 
 
 Parallel Texts ; together with Richard the Redeless. By William Langland 
 (about 1362-1399 A.D.). Edited from numerous Manuscripts, with Preface, 
 Notes, and a Glossary, by W. W. Skeat, Litt.D. 2 vols. Svo. 31^-. 6^. 
 
 The Vision of William concerning Piers the Plowman, by 
 
 William Langland. Edited, with Notes, by W. W. Skeat, M.A. Third 
 Edition. Extra fcap. Svo . 4^.6^. 
 
 Chancer. I. The Prologue to the Cajtterbury Tales; the 
 
 Knightes Tale ; The Noniie Prestes Tale. Edited by R. Morris, Editor of 
 Specimens of Early English, &c., &c. Extra fcap. Svo. 2s. f^d. 
 
 II. The Prioresses Tale; Sir Thopas ; The Monkes 
 
 Tale ; The Clerkes Tale ; The Squieres Tale, &c. Edited by W. W. Skeat, 
 M.A. Second Edition, Extra fcap. Svo. \s. 6d. 
 
 III. The Tale of the Man of Lawe ; The Pardoneres 
 
 Tale ; The Second Nonnes Tale ; The Chanouns Yemannes Tale. By the 
 same Editor. Second Edition. Extra fcap. Svo. 4.;. dd. 
 
 Gamelyn, The Tale of. Edited with Notes, Glossary, &c., by 
 
 W. W. Skeat, M.A. Extra fcap. Svo. Stiff covers, \s. 6di 
 
 Minot [Latirence). Poems. Edited, with Introduction and 
 
 Notes, by Joseph Hall, M.A. Extra fcap. Svo. Nearly ready. 
 
18 CLARENDON PRESS, OXFORD. 
 
 Spenser's Faery Queene. Books I and II. Designed chiefly 
 
 for the use of Schools. With Introduction, Notes, and Glossary. By G. W. 
 Kitchin, D.D. Extra fcap. 8vo. 2 J. dd. each. 
 
 Hooker. Ecclesiastical Polity^ Book I. Edited by R. W. 
 
 Church, M.A. Second Edition. Extra fcap. 8vo. 2s. 
 
 OLD ENGLISH DRAMA. 
 The Pilgrimage to Parnassus with TJie Two Parts of the 
 
 Retiirfi frojji Parnassus. Three Comedies performed in St. John's College, 
 Cambridfje, A.D. MDXCVil-MDCi. Edited from MSS. by the Rev. W. D. 
 Macray, M.A., F.S.A. Medium 8vo. Bevelled Boards, Gilt top, 8j. dd. 
 
 Marlowe and Greene. Marlozve^s Tragical History of Dr. 
 
 Faustus, and Greene's Honourable History of Friar Bacoft and Friar Bungay. 
 Edited by A. W. Ward, M.A. New and Enlarged Edition. Extra fcap. 
 Svo. 6j. dd. 
 
 Marlotve. Edivard IT. With Introduction, Notes, &c. By 
 
 O. W. Tan cock, M.A. Extra fcap. Svo. 3J. 
 
 SHA KESPEARE. 
 Shakespeare. Select Plays. Edited by W. G. Clark, M.A,, 
 
 and W. Aldis Wright, M.A. Extra fcap. Svo. stiff covers. 
 The Merchant of Venice. \s. Macbeth, xs.dd. 
 
 Richard the Second. i.f. dd. Hamlet, is. 
 
 Edited by W. Aldis Wright, M.A. 
 The Tempest, is. 6d. Midsummer Night's Dream, n. 6d. 
 
 As You Like It. is. 6d. Coriolanus. 2s. 6d. 
 
 Julius Caesar. 2s. Henry the Fifth. 2s. 
 
 Richard the Third. 2s. 6d. Twelfth Night, is. 6d. 
 
 King Lear. is. 6d. King John. is. 6d. 
 
 Shakespeare as a Dramatic Artist ; a popular Illustration of 
 
 the Principles of Scientific Criticism. By R. G. Moulton, M.A. Crown Svo. ^s. 
 
 Bacon. I. Advancement of Learning. Edited by W. Aldis 
 
 Wright, M.A. Second Edition. Extra fcap. Svo. 4^-. dd. 
 
 II. The Essays. With Introduction and Notes. By 
 
 S. H. Reynolds, M.A., late Fellow of Brasenose College. In Preparation. 
 
 Milton. I. Areopagitica. With Introduction and Notes. By 
 
 John W. Hales, M.A. Third Edition. Extra fcap. Svo. 3J. 
 
 II. Poems. Edited by R. C. Browne, M.A. 2 vols. 
 
 Fifth Edition. Extra fcap. Svo. 6j. 6^^. Soldseparately, Vol. 1. 4J.; Vol.11. 3^. 
 
 In paper covers : — 
 
 Lycidas, ^d. L'AIlegro, id. II Penseroso, 4</. Comus, dd. 
 
 Samson Agonistes, dd. 
 
 III. Samson Agonistes. Edited with Introduction and 
 
 Notes by John Churton Collins. Extra fcap. Svo. stiff covers, is. 
 
CLARENDON PRESS, OXFORD. 19 
 
 Bunyan. I. The Pilgrinis Progress, Grace Abounding, Rela- 
 tion of the Imprisonment of Mr. John Bunyan. Edited, with Biographical 
 Introduction and Notes, by E. Venables, M.A. 1879. Extra fcap. 8vo. 5J. 
 In ornamental Parchment, 6^. 
 
 II. Holy War, &-c. Edited by E. Venables, M.A. 
 
 In the Press. 
 
 Clarendon. History of the Rebellion. Book VI. Edited 
 
 by T. Arnold, M.A. Extra fcap. 8vo. 4J. 6d. 
 
 Drydeii. Select Poems. Stanzas on the Death of Oliver 
 
 Cromwell ; Astrrea Redux ; Annus Mirabilis ; Absalom and Achitophel ; 
 Religio Laici ; The Hind and the Panther. Edited by W. D. Christie, M.A. 
 Second Edition. Extra fcap. 8vo. 3^^. ^d. 
 
 Locke's Conduct of the Understanding. Edited, with Intro- 
 duction, Notes, &c., by T. Fowler, M.A. Second Edition. Extra fcap. 8vo. 2s. 
 
 Addison. Selections from Papers in the Spectator. With Notes. 
 
 By T. Arnold, M.A. Extra fcap. 8vo. 4^.6^. In ornamental Parchment, 6j-. 
 
 Steele. Selections from the Tatler, Spectator, and Guardian. 
 
 Edited by Austin Dobson. Extra fcap. Svo. 4^-. dd. In white Parchment, ^s. 6d. 
 
 Pope. With Introduction and Notes. By Mark Pattison, B.D. 
 
 I. Essay on Man. Extra fcap. Svo. 'is. 6d. 
 
 II. Satires atid Epistles. Extra fcap. Svo. 2s. 
 
 Par7iell. The Hermit. Paper covers, 2d. 
 
 Gray. Selected Poems. Edited by Edmund Gosse. Extra 
 
 fcap. Svo. Stiff covers, is. 6d. In white Parchment, ^s, 
 
 Elegy and Ode on Eton College. Paper covers, 2d. 
 
 Goldsmith. The Deserted Village. Paper covers, 2d. 
 JoJmson. I. Rasselas ; Lives of Dryden ajid Pope. Edited 
 
 by Alfred Milnes, M.A. (London). Extra fcap. Svo. ^s. 6d., or Lives of 
 Dryden and Pope only, stiff covers, 2s. dd. 
 
 II. Vafiity of Human Wishes. With NoteS; by E. J. 
 
 Payne, M.A. Paper covers, 4^. 
 
 Boswell's Life of Johnson. With the Journal of a Tour to 
 
 the Hebrides. Edited, with copious Notes, Appendices, and Index, by G. 
 Birkbeck Hill, D.C.L., Pembroke College. With Portraits and Facsimiles. 
 6 vols. Medium Svo. Nearly ready. 
 
 Cowper. Edited, with Life, Introductions, and Notes, by 
 
 H. T. Griffith, B.A. 
 
 I. The Didactic Poems <?/" 1782, with Selections from the 
 
 Minor Pieces, A.D. 1779-1783. Extra fcap. Svo. 3J. 
 
 II. The Task, with Tirocinium, and Selections from the 
 
 Minor Poems, A.D. 1 7S4-1 799. Second Edition. Extra fcap. Svo. 3J. 
 
 C 2 
 
CLARENDON PRESS, OXFORD. 
 
 Burke. Select Works. Edited, with Introduction and Notes, 
 
 by E. J. Payne, M.A. 
 
 ■ I. TJiotights on the Present Discontents ; the two Speeches 
 
 on America- Second Edition. Extra fcap. Svo. ^s. (^d. 
 
 II. Reflections on the French Revolution. Second Edition. 
 
 Extra fcap. Svo. 5J. 
 
 III. Four Letters on the Proposals for Peace with the 
 
 Regicide Directory of France. Second Edition. Extra fcap. Svo. 5^. 
 
 Keats. Hyperion, Book I. With Notes by W. T. Arnold, B.A. 
 
 Paper covers, 4^. 
 
 Byron. Childe Harold. Edited, with Introduction and Notes, 
 
 by H. F. Tozer, M.A. Extra fcap. Svo. 3^. 6a'. In white Parchment, 5^-. 
 
 Scott. Lay of the Last Minstrel. Edited with Preface and 
 
 Notes by W. Minto, M.A. With Map. Extra fcap. Svo. Stiff covers, 2S. 
 Ornamental Parchment, 3^. 6d. 
 
 Lay of the Last Minstrel. Introduction and Canto I, 
 
 with Preface and Notes, by the same Editor. 6(f. 
 
 II. LATIN. 
 
 Rudimenta Latina. Comprising Accidence, and Exercises of 
 
 a very Elementary Character, for the use of Beginners. By John Barrow 
 Allen, M.A. Extra fcap. Svo. 2J. 
 
 An Elementary Latin Grammar. By the same Author. 
 
 Forty-second Thousand. Extra fcap. Svo. 2S.(>d. 
 
 A First Latin Exercise Book. By the same Author. Fourth 
 
 Edition. Extra fcap. Svo. 2s. 6d. 
 
 A Second Latin Exercise Book. By the same Author. Extra 
 
 fcap. Svo. y. Gd. 
 Reddenda Minora, or Easy Passages, Latin and Greek, for 
 
 Unseen Translation. For the use of Lower Forms. Composed and selected 
 by C. S. Jerram, M.A. Extra fcap. Svo. \s. 6d. 
 
 Anglice Reddenda, or Easy Extracts, Latin and Greek, for 
 
 Unseen Translation. By C. S. Jerram, M.A. Third Edition, Revised and 
 Enlarged, Extra fcap. Svo. 2s. dd. 
 
 Anglice Reddcjida. Second Series. By the same Author. 
 
 Extra fcap. Svo. 3^. Just Published. 
 
 Passages for Translation into Latin. For the use of Passmen 
 
 and others. Selected by J. Y. Sargent, M.A. Fifth Edition. Extra fcap. 
 Svo. 2 J. 6</. 
 
CLARENDON PRESS, OXFORD. 21 
 
 Exercises in Latin Prose Composition ; with Introduction, 
 
 Notes, and Passages of Graduated Difficulty for Translation into Latin. By 
 G. G. Ramsay, M.A., LL.D. Second Edition. Extra fcap. 8vo. 4J. 6</, 
 
 Hints and Helps for Latin Elegiacs, By H. Lee- Warner, M. A. 
 
 Extra fcap. 8vo. },s. 6d. 
 
 First Latin Reader. By T. J. Nunns, M.A. Third Edition. 
 
 Extra fcap. Svo. 2j. 
 
 Caesar. The Commentaries (for Schools). With Notes and 
 
 Maps. By Charles E. Moberly, M.A. 
 
 Part I. The Gallic War. Second Edition. Extra fcap. Svo. 4J. dd. 
 
 Part II. The Civil War. Extra fcap. Svo, 3s. 6 J. 
 
 The Civil War. Book I, Second Edition. Extra fcap. Svo. 2j. 
 
 Cicero. Speeches against Catilina. By E. A. Upcott, M.A., 
 
 Assistant Master in Wellington College. In the Press. 
 
 Cicero. Selection of interesting and descriptive passages. With 
 
 Notes. By Henry Walford, M-A. In three Parts. Extra fcap. Svo. 4J. 6^. 
 Each Part separately, limp, \s. 6d. 
 
 Part I. Anecdotes from Grecian and Roman History. Third Edition. 
 Part II. Omens and Dreams: Beauties of Nature. Third Edition. 
 Part III. Rome's Rule of her Provinces. Third Edition. 
 
 Cicero. De Senectiite. Edited, with Introduction and Notes, 
 
 by L. Hu.xley, M.A. In one or two Parts. Extra fcap. Svo. 2s. 
 
 Cicero. Selected Letters (for Schools). With Notes. By the 
 
 late C. E. Prichard, M.A., and E. R. Bernard, M.A. Second Edition. 
 Extra fcap. Svo. 3^. 
 
 Cicero. Select Orations (for Schools). In Verrem I. De 
 
 Imperio Gn. Pompeii. Pro Archia. Philippica IX. With Introduction and 
 Notes by J. R. King, M.A. Second Edition. Extra fcap. Svo. 2s. 6d. 
 
 Cornelius Nepos. With Notes. By Oscar Browning, M.A. 
 
 Second Edition. Extra fcap. Svo. 2s. 6d. 
 
 Horace. Selected Odes. With Notes for the use of a Fifth 
 
 Form. By E. C. Wickham, M.A. In one or tw^o Parts. Extra fcap. Svo. 
 cloth, 2S. 
 
 Livy. Selections (for Schools). With Notes and Maps. By 
 
 H. Lee-Warner, M.A. Extra fcap. Svo. In Parts, limp, each \s. 6d. 
 Part I. The Caudine Disaster. Part II. Hannibal's Campaign 
 in Italy. Part III. The Macedonian War. 
 
 Livy. Books V— VII. With Introduction and Notes. By 
 
 A. R. Cluer, B.A. Second Edition. Revised by P. E. Matheson, M.A. 
 (In one or two vols.) Extra fcap. Svo. 5^. 
 
 Livy. Books XXI, XXII, and XXIII. With Introduction 
 
 and Notes. By M. T. Tatham, M, A. Extra fcap. Svo. 45. 6d. 
 
2 2 CLARENDON PRESS, OXFORD. 
 
 Ovid. Selections for the use of Schools. With Introductions 
 
 and Notes, and an Appendix on the Roman Calendar. By W. Ramsay, M.A. 
 Edited by G. G. Ramsay, M.A. Third Edition. Extra fcap. 8vo. c,s. 6d. 
 
 Ovid. Tristia. Book I. The Text revised, with an Intro- 
 duction and Notes. By S. G. Owen, B.A. Extra fcap. Svo. y. 6d. 
 
 Plaiitus. Captivi. Edited by W. M. Lindsay, M.A. Extra 
 
 fcap. Svo. (In one or two Parts). 2s. 6d. Just Published. 
 
 Plaiitus. The Trinmnmtis. With Notes and Introductions. 
 
 Intended for the Higher Forms of Public Schools. By C. E. Freeman, M.A., 
 and A. Sloman, M.A. Extra fcap. Svo. 3J, 
 
 Pliny. Selected Letters (for Schools). With Notes. By the 
 
 late C. E. Prichard, M.A., and E. R. Bernard, M.A. Extra fcap. Svo. 3^. 
 
 Salhist. With Introduction and Notes. By W. W. Capes, 
 
 M.A. Extra fcap. Svo. 4.?. dd. 
 
 Tacitus. The Annals. Books I-IV. Edited, with Introduc- 
 tion and Notes for the use of Schools and Junior Students, by H. Fumeaux, 
 M.A. Extra fcap. Svo. 5^. 
 
 Terence. Andria. With Notes and Introductions. By C. 
 
 E. Freeman, M.A., and A. Sloman, M.A. Extra fcap. Svo. 3J. 
 
 Adelphi. With Notes and Introductions. Intended for 
 
 the Higher Forms of Public Schools. By A. Sloman, M.A. Extra fcap. 
 Svo. 3J-. 
 
 Tibidhis and Proper tins. Selections. Edited by G. G. Ramsay, 
 
 M.A. Extra fcap. Svo. (In one or two vols.) 6s. /ust Puhiisked. 
 
 Virgil. With Introduction and Notes. By T. L. Papillon, 
 
 M.A. Two vols. Crown Svo. ios.6d. The Text separately, 4J. 6^. 
 
 Virgil. The Eclogues. Edited by C. S. Jerram, M.A. In 
 
 two Parts. Crown Svo. Nearly ready. 
 
 Catulli Veronensis Liber. Iterum recognovit, apparatum cri- 
 
 ticum prolegomena appendices addidit, Robinson Ellis, A.M. 187S. Demy 
 Svo. i6s. 
 
 A Commentary on Catidhis. By Robinson Ellis, M.A. 
 
 1S76. Demy Svo. i6j. 
 
 Catidli Veronensis Carmina Selecta, secundum recognitionem 
 
 Robinson Ellis, A.M. Extra fcap. Svo. is. 6d. 
 
 Cicero de Oratore. With Introduction and Notes. By A. S. 
 
 Wilkins, M.A. 
 
 Book I. 1879. Svo. 6j. Book II. iSSi. Svo. 5J. 
 
 Philippic Orations. With Notes. By J. R. King, M.A. 
 
 Second Edition. 1S79. Svo. loj-. 6d. 
 
CLARENDON PRESS, OXFORD. 23 
 
 Cicero. Select Letters. With English Introductions, Notes, 
 
 and Appendices. By Albert Watson, M.A. Third Edition. Demy 8vo. i8s. 
 
 ■ Select Letters. Text. By the same Editor. Second 
 
 Edition. Extra fcap. 8vo. 4J. 
 
 pro Cluentio. With Introduction and Notes. By W. 
 
 Ramsay, M.A. Edited by G. G. Ramsay, M.A. 2nd Ed. Ext. fcap. Svo. 3J. 6flr. 
 
 Horace. With a Commentary. Volume I. The Odes, Carmen 
 
 Seculare, and Epodes. By Edward C. Wickham, M.A. Second Edition. 
 1877. Demy 8vo. \2s. 
 
 A reprint of the above, in a size suitable for the use 
 
 of Schools. In one or two Parts. Extra fcap. 8vo. 6j. 
 
 Livy, Book I. With Introduction, Historical Examination, 
 
 and Notes. By J. R. Seeley, M.A. Second Edition. 1881. 8vo. 6^. 
 
 Ovid. P. Ovidii Nasonis Ibis. Ex Novis Codicibus edidit, 
 
 Scholia Vetera Commentarium cum Prolegomenis Appendice Indice addidit, 
 R. Ellis, A.M. 8vo. ioj-. U. 
 
 Persius. The Satires. With a Translation and Commentary. 
 
 By John Conington, M.A. Edited by Henry Nettleship, M.A. Second 
 Edition. 1874. 8vo. 7^. dd. 
 
 jfuvenal. XIII Satires. Edited, with Introduction and 
 
 Notes, by C. H. Pearson, M.A., and Herbert A. Strong, M.A., LL.D., Professor 
 of Latin in Liverpool University College, Victoria University. In two Parts. 
 Crown 8vo. Complete, ds. 
 Also separately, Part I. Introduction, Text, etc., 3^. Part II. Notes, 3^. dd. 
 
 Tacitus. The Annals. Books I- VI. Edited, with Intro- 
 duction and Notes, by H, Furneaux, M.A. 8vo. i8j. 
 
 Nettleship (//., M.A.). Lectures and Essays on Subjects con- 
 nected with Latin Scholarship and Literature. Crown 8vo. is. 6d. 
 
 The Roman Satura : its original form in connection with 
 
 its literary development. 8vo. sewed, u. 
 
 Ancient Lives of Vergil. With an Essay on the Poems 
 
 of Vergil, in connection with his Life and Times. 8vo. sewed, 2s. 
 
 Papillon [T. L., M.A.). A Ma7iual of Comparative Philology. 
 
 Third Edition, Revised and Corrected. 1882. Crown 8vo. 6j. 
 
 Pinder {North, M.A.). Selections from the less known Latin 
 
 Poets. 1869. 8vo. 15J. 
 
 Sellar{W. Y., M.A.). Roman Poets of the Augnstan Age. 
 
 Virgil. New Edition. 1883. Crown 8vo. 9J. 
 
 Roman Poets of the Republic. New Edition, Revised 
 
 and Enlarged. 1881. Svo. 14J. 
 
 Wordsworth [J., M.A.). Fragments and Specimens of Early 
 
 Latin. With Introductions and Notes. 1874. Svo. \%s. 
 
2 4 CLARENDON PRESS, OXFORD. 
 
 III. greek:. 
 
 A Greek Primer^ for the use of beginners in that Language. 
 
 By the Right Rev. Charles Wordsworth, D.C.L. Seventh Edition. Extra fcap. 
 8vo. IS. 6d. 
 
 Easy Greek Reader. By Evelyn Abbott, M.A. In one or 
 
 two Parts. Extra fcap. 8vo. 3^-. 
 
 Graecae Grammaticae Rudimenta in usum Scholarum. Auc- 
 
 tore Carolo Wordsworth, D.C.L. Nineteenth Edition, 1882, i2mo. 4^. 
 
 A Greek-English Lexicon, abridged from Liddell and Scott's 
 
 4to. edition, chiefly for the use of Schools. Twenty-first Edition. 1886. 
 Square izmo. 'js. ()d. 
 
 Greek Verbs, Irregtdar ajid Defective; their forms, meaning, 
 
 and quantity; embracing all the Tenses used by Greek writers, with references 
 to the passages in which they are found. By W. Veitch. Fourth Edition. 
 Crown 8vo. los. 6d. 
 
 The Elements of Greek Accenttiatioji (for Schools) : abridged 
 
 from his larger work by H. W. Chandler, M.A. Extra fcap. 8vo. 7s. 6d. 
 
 A Series of Graduated Greek Readers: — 
 
 Eirst Greek Reader. By W. G. Rushbrooke, M.L. Second 
 
 Edition. Extra fcap. 8vo. 2s. ()d. 
 
 Second Greek Reader. By A. M. Bell, M.A. Extra fcap. 
 Svo. 3J-. 6a?. 
 
 Eourth Greek Reader ; being Specimens of Greek Dialects. 
 
 With Introductions, etc. By W. W. Merry, M.A. Extra fcap. Svo. 4J. dd. 
 
 Fifth Greek Reader. Selections from Greek Epic and 
 
 Dramatic Poetry, with Introductions and Notes. By Evelyn Abbott, M.A. 
 Extra fcap. Svo. \s. 6d. 
 
 The Golden Treasitry of Ancient Greek Poetry: being a Col- 
 lection of the finest passages in the Greek Classic Poets, with Introductory 
 Notices and Notes. By R. S. Wright. M.A. Extra fcap. Svo. 8j. dd. 
 
 A Golden Treasury of Greek Prose, being a Collection of the 
 
 finest passages in the principal Greek Prose Writers, with Introductory Notices 
 and Notes. By R. S. Wright, M.A., and J. E. L. Shadwell, M.A. Extra fcap. 
 Svo. 4J. 6(/. 
 
 Aeschylus. Prometheus Bound (for Schools). With Introduc- 
 tion and Notes, by A. O.Prickard, M.A. Second Edition. Extra fcap. Svo. 2s. 
 
 Agamemnon. With Introduction and Notes, by Arthur 
 
 Sidgwick, M.A. Second Edition. Extra fcap. Svo. 3^. 
 
 Choephoroi. With Introduction and Notes by the same 
 
 Editor. Extra fcap. Svo. is. 
 
CLARENDON PRESS, OXFORD. 25 
 
 Aristophanes. In Single Plays. Edited, with English Notes, 
 
 Introductions, &c., by W. W. Merry, M.A. Extra fcap. 8vo. 
 
 I. The Clouds, Second Edition, 2s. II. The Acharnians, 2s. 
 
 III. The Frogs, Second Edition. In one or two parts, is. 
 
 Cebes. Tabula. With Introduction and Notes. By C. S. 
 
 Jerram, M.A. Extra fcap. 8vo. 2^. dd. 
 
 Demosthenes. Olynthiaes and Philippics. Edited by Evelyn 
 
 Abbott, M.A. Extra fcap. 8vo. In two Parts. In the Press. 
 
 Euripides. Alcestis (for Schools). By C. S. Jerram, M.A. 
 
 Extra fcap. Svo. 2s. dd. 
 
 Helena. Edited, with Introduction, Notes, etc., for 
 
 Upper and Middle Forms. By C. S. Jerram, M.A. Extra fcap. Svo. 2,s. 
 
 — Iphigenia in Tanris. Edited, with Introduction, Notes, 
 
 etc., for Upper and Middle Forms. By C. S. Jerram, M.A. Extra fcap. Svo. 
 cloth, 3J-. 
 
 Medea. By C. B. Heberden, M.A. In one or two Parts. 
 
 Extra fcap. Svo. 2s. 
 
 Herodotus^ Selections from. Edited, with Introduction, Notes, 
 
 and a Map, by W. W. Merry, M.A. Extra fcap. Svo. 2s. 6d. 
 
 Homer. Odyssey, Books I-XII (for Schools). By W. W. 
 
 Merry, M.A. Twenty-seventh Thousand. Extra fcap. Svo. 4J. 6d. 
 Book II, separately, \s. 6d. 
 
 Odyssey, Books XIII-XXIV (for Schools). By the 
 
 same Editor. Second Edition. Extra fcap. Svo. 5^. 
 
 Iliad, Book I (for Schools). By D. B. Monro, M.A. 
 
 Second Edition. Extra fcap. Svo. 2s. 
 
 Iliad, Books I-XII (for Schools). With an Introduction, 
 
 a brief Homeric Grammar, and Notes. By D. B. Monro, M.A. Second 
 Edition. Extra fcap. Svo. 6s. 
 
 Iliad, Books VI and XXI. With Introduction and 
 
 Notes. By Herbert Hailstone, M.A. Extra fcap. Svo. is. 6d. each. 
 
 Lucian. Vera Historia (for Schools). By C. S. Jerram, 
 
 M.A. Second Edition. Extra fcap. Svo. \s. 6d. 
 Plato. Selections from the Dialogues [including the whole of 
 
 the Apology and Crito\ With Introduction and Notes by John Purves, M.A., 
 and a Preface by the Rev. B. Jowett, M.A. Extra fcap. Svo. 6j, 6^. 
 
 C 3 
 
26 CLARENDON PRESS, OXFORD. 
 
 Sophocles. For the use of Schools. Edited with Intro- 
 ductions and English Notes. By Le\vis Campbell, M.A., and Evelyn Abbott, 
 M.A. Neiv and Revised Edition. 2 Vols. Extra fcap. 8vo. xos.dd. 
 Sold separately, Vol. I, Text, 4J. ()d. ; Vol. II, Explanatory Notes, ds. 
 
 SopJwcles. In Single Plays, with English Notes, &c. By- 
 Lewis Campbell, M.A., and Evelyn Abbott, M.A. Extra fcap. 8vo. limp. 
 Oedipus Tyrannus, Philoctetes. New and Revised Edition, 2 j. each. 
 Oedipus Coloneus, Antigone, \s, <^d. each. 
 
 Ajax, Electra, Trachiniae, 2J. each. 
 
 Oedipus Rex: Dindorfs Text, with Notes by the 
 
 present Bishop of St. David's. Extra fcap. 8 vo. limp, \s. 6d. 
 
 TheocriUis (for Schools). With Notes. By H. Kynaston, 
 
 D.D. (late Snow). Third Edition. Extra fcap. 8vo. 4J. 6^, 
 
 Xenophon. Easy Selections (for Junior Classes). With a 
 
 Vocabulary, Notes, and Map. By J. S. Phillpotts, B.C.L., and C. S. Jerram, 
 M.A. Third Edition, Extra fcap. 8vo. y. 6d. 
 
 • Selections (for Schools). With Notes and Maps. By 
 
 J. S. Phillpotts, B.C.L. Fourth Edition. Extra fcap. 8 vo. 3J. (id. 
 
 Anabasis, Book I. Edited for the use of Junior Classes 
 
 and Private Students. With Introduction, Notes, etc. By J. Marshall, M.A., 
 Rector of the Royal High School, Edinburgh. Extra fcap. 8vo. 2^. 6d. 
 
 Anabasis, Book II. With Notes and Map. By C. S. 
 
 Jerram, M.A. Extra fcap. 8 vo. 2S. 
 
 Cyropaedia, Books IV and V. With Introduction and 
 
 Notes by C. Bigg, D.D. E.xtra fcap. 8vo. 2s. 6d. 
 
 Aristotle's Politics. By W. L.Newman, M.A. \In the Press ?^ 
 Aristotelian Studies. I. On the Structure of the Seventh 
 
 Book of the Nicomachean Ethics. By J. C. Wilson, M.A. 8vo. stiff, 5J. 
 
 Aristotelis Ethica Nicomachea, ex recensione Immanuelis 
 
 Bekkeri. Crown 8vo. ^s. 
 
 Demosthenes and Aeschines. The Orations of Demosthenes 
 
 and .(^schines on the Crown. With Introductory Essays and Notes. By 
 G. A. Simcox, M.A., and W. H. Simcox, M.A. 1872. 8vo. \2S. 
 
 Head {Barclay V.). Historia Numorum: A Manual of Greek 
 
 Numismatics. Royal 8vo. half-bound. 2/. 2s. 
 
 Hicks [E. L., M.A.). A Manual of Greek Historical Inscrip- 
 tions. Demy 8vo. xos, ()d. 
 
CLARENDON PRESS, OXFORD. 27 
 
 Homer. Odyssey, Books I-XII. Edited with English Notes, 
 
 Appendices, etc. By W. W. Merry, M.A., and the late James Riddell, M.A. 
 1886. Second Edition. Demy 8vo. i6j. 
 
 Homer. A Grammar of the Homeric Dialect. By D. B. Monro, 
 
 M.A. Demy 8vo. \os. 6d. 
 
 Sophocles. The Plays and Fragments. With English Notes 
 
 and Introductions, by Lewis Campbell, M.A. 2 vols. 
 
 Vol.1. Oedipus Tyrannus. Oedipus Coloneus. Antigone. Svo. i6j. 
 Vol. II. Ajax. Electra. Trachiniae. Philoctetes. Fragments. Svo. 16s. 
 
 IV. FRENCH AND ITALIAN. 
 
 Bracket's Etymological Dictionary of the French Language, 
 
 with a Preface on the Principles of French Etymology. Translated into 
 English by G. W. Kitchin, D.D. Third Edition. Crown Svo. ^s. 6d. 
 
 Historical Grammar of the French Language. Trans- 
 lated into English by G. W, Kitchin, D.D. Fourth Edition. Extra fcap, 
 Svo. zs. 6d. 
 
 "Works by GEORGE SAINTSBURY, M.A. 
 
 Primer of French Literatiire. Extra fcap. Svo. 'Xs. 
 
 Short History of French Literature. Crown 8vo. ios.6d. 
 
 Specimens of French Literature, from Villon to Hiigo. Crown 
 Svo. gj. 
 
 MASTERPIECES OF THE FRENCH DRAMA. 
 Corneille's Horace. Edited, with Introduction and Notes, by 
 
 George Saintsbury, M.A. Extra fcap. Svo. 2j. 6d. 
 
 Molihe's Les Prkieuses Ridicules. Edited, with Introduction 
 
 and Notes, by Andrew Lang, M.A. Extra fcap. Svo. \s. 6d. 
 
 Racine's Esther. Edited, with Introduction and Notes, by 
 
 George Saintsbury, M.A, Extra fcap. Svo. 2s. 
 
 Beau7narchais' LeBarbier de Seville. Edited, with Introduction 
 
 and Notes, by Austin Dobson. Extra fcap. Svo. 2s. 6d. 
 
 Voltaire's M^rope. Edited, with Introduction and Notes, by 
 
 George Saintsbury. Extra fcap. Svo. cloth, 2s. 
 
 Mussefs On ne badine pas avec V Amour, and Fantasia. Edited, 
 
 with Prolegomena, Notes, etc., by Walter Herries Pollock. Extra fcap. 
 
 Svo. 2J. 
 
 The above six Plays may be had in ornamental case, and bound 
 in Imitation Parchment, price 12S. 6d. 
 
28 CLARENDON PRESS, OXFORD. 
 
 Sainte-Beuve. Selections from the Causeries du Lundi. Edited 
 
 by George Saintsbury. Extra fcap. 8vo. 2s. 
 
 Qiiinefs Lettres a sa Mere. Selected and edited by George 
 
 Saintsbury. Extra fcap. 8vo. 2J. 
 
 Gautier, Thhphile. Scenes of Travel. Selected and Edited 
 
 by George Saintsbury. Extra fcap. 8vo. 2J. 
 
 V Eloquence de la Chaire et de la Tribune Frangaises. Edited 
 by Paul Blouet, B.A. (Univ. Gallic). Vol. I. French Sacred Oratory. 
 Extra fcap. 8vo. 2s. 6d. 
 
 Edited by GtrSTAVE MASSON, B.A. 
 
 Corneille^s Cinna. With Notes, Glossary, etc. Extra fcap. 8vo. 
 
 cloth, 2s. Stiff covers, i j. 6d. 
 
 Louis XIV and his Contemporaries ; as described in Extracts 
 
 from the best Memoirs of the Seventeenth Century. With English Notes, 
 Genealogical Tables, &c. Extra fcap. Svo. 2s. 6d. 
 
 Maistre, Xavier de. Voyage autojir de ma Chambre. Ourika, 
 
 by Madame de Duras; Le Vieux Tailleur, by MM. Ercktiiann-Chatrian; 
 La Veillee de Vincennes, by Alfred de Vigny ;^ Les Jumeaux de I'Hotel 
 Cornei\le,hy £dmond A doul ; Mesaventures d'un Ecolier, hy Rodolphe Topffer, 
 Third Edition, Revised and Corrected, Extra fcap. Svo. 2s. 6d. 
 
 Molih'e^s Les Fourberies de Scapin, and Racine s Athalie. 
 
 With Voltaire's Life of Moliere. Extra fcap. Svo. 2s. 6d. 
 
 Molihe's Les Fourberies de Scapin. With Voltaire's Life of 
 
 Moliere. Extra fcap. Svo. stiff covers, \s. 6d. 
 
 Molihe's Les Femmes Savantes. With Notes, Glossary, etc. 
 
 Extra fcap. Svo. cloth, 2S. Stiff covers, is. 6d. 
 
 Racine^s Androniaque, and Corneille's Le Menteur. With 
 
 Louis Racine's Life of his Father. Extra fcap. Svo. 2s. 6d. 
 
 Regnards Le Joueur, and Brueys and Palaprafs Le Grondcur. 
 
 Extra fcap. Svo. 2s. 6d. 
 
 SevigJi/, Madame de, and her chief Contemporaries, Selections 
 
 from the Correspondence of Intended more especially for Girls' Schools. 
 Extra fcap. Svo. 3J. 
 
 Dante. Selections from the Inferno. With Introduction and 
 
 Notes. By H. B. Cotterill, B.A. Extra fcap. Svo. 4^. 6d. 
 
 Tasso. La Gerusalemme Liberata. Cantos i, ii. With In- 
 troduction and Notes. By the same Editor. Extra fcap. Svo. is. 6d. 
 
CLARENDON PRESS, OXFORD. 29 
 
 V. GERMAN. 
 Scherer ( W^. A History of German Literature. Translated 
 
 from the Third German Edition by Mrs. F. Conybeare. Edited by F. Max 
 Miiller. 2 vols. 8vo. 2IJ. 
 
 Max Mailer. The German Classics, from the Fourth to the 
 
 Nineteenth Century. With Biographical Notices, Translations into Modern 
 German, and Notes. By F. Max Miiller, M.A. A New Edition, Revised, 
 Enlarged, and Adapted to Wilhelm Scherer's ' History of German Literature,* 
 by F. Lichtenstein. 2 vols, crown 8vo. 21 J. 
 
 GERMAN COURSE. By HERMANN LANGE. 
 
 The Germans at Home ; a Practical Introduction to German 
 
 Conversation, with an Appendix containing the Essentials of German Grammar. 
 Second Edition. 8vo. 2s. 6d. 
 
 The German Manual ; a German Grammar, Reading Book, 
 
 and a Handbook of German Conversation. 8vo. Is. 6d. 
 
 Grammar of the German Language. 8vo. y. 6d. 
 
 German Composition ; A Theoretical and Practical Guide to 
 
 the Art of Translating English Prose into German. 8vo. 4^. 6^. 
 
 Gerinan Spelling ; A Synopsis of the Changes which it has 
 
 undergone through the Government Regulations of 1880. Paper covers, (id. 
 
 Lessing's Laokoon. With Introduction, English Notes, etc. 
 
 By A. Hamann, Phil. Doc, M.A. Extra fcap, Svo, 4J. 6d. 
 
 Schiller s Wilhelm Tell. Translated into English Verse by 
 E. Massie, M.A. Extra fcap. Svo. 5J. 
 
 Also, Edited by C. A. BUCHHEIM, PhU. Doc. 
 
 Becker s Friedrich der Grosse. Extra fcap. 8vo. In the Press. 
 Goethe's Egmont. With a Life of Goethe, &c. Third Edition. 
 
 Extra fcap. 8vo, 3J. 
 
 —— Iphigenie auf Tauris. A Drama. With a Critical In- 
 troduction and Notes. Second Edition. Extra fcap, 8vo. 3J. 
 
 Heine's Prosa, being Selections from his Prose Works. With 
 
 English Notes, etc. Extra fcap. Svo. 4J. 6d. 
 
 Heine's Harzreise. With Life of Heine, Descriptive Sketch 
 
 of the Harz, and Index. Extra fcap. Svo. paper covers, is. 6d. ; cloth, 2s. 6d. 
 
 Lessing's Minna von Barnhelm. A Comedy. With a Life 
 
 of Lessing, Critical Analysis, etc. Extra fcap. Svo. 3J. 6d. 
 
 • Nathan der Weise. With Introduction, Notes, etc. 
 
 Extra fcap. Svo. 4J. 6</. 
 
36 CLARENDON PRESS, OXFORD. 
 
 Schiller'' s Historische Skizzen ; Egmonfs Leben nnd Tod, and 
 
 Belagefung von Antwerpen. With a Map. Extra fcap. 8vo. 2s. 6d. 
 
 WilJiclm Tell With a Life of Schiller; an his- 
 torical and critical Introduction, Arguments, and a complete Commentary, 
 and Map. Sixth Edition. Extra fcap. 8vo. 3J. dd. 
 
 Wilhclm Tell. School Edition. With Map. ^s. 
 
 Modern German Reader. A Graduated Collection of Ex- 
 tracts in Prose and Poetry from Modern German writers : — 
 Part I. With English Notes, a Grammatical Appendix, and a complete 
 
 Vocabulary. Fourth Edition. Extra fcap. 8vo. 2s.6d. 
 Part II. With English Notes and an Index. Extra fcap. 8vo. 2s. 6d. 
 
 Niebuhrs GriecJiische Heroen-Geschichten. Tales of Greek 
 
 Heroes. Edited with English Notes and a Vocabulary, by Emma S. Buchheim. 
 School Edition. Extra fcap. Svc, cloth, 2s. Stiff covers, \s. 6d. 
 
 VI. MATHEMATICS, PHYSICAL SCIENCE, &e. 
 
 By LEWIS HEKTSLET, M.A. 
 
 Figures made Easy : a first Arithmetic Book. Crown 8vo. 6d. 
 Answers to the Examples in Figures made Easy, together 
 
 with two thousand additional Examples, with Answers. Crown 8vo. \s. 
 
 The Scholar's Arithmetic. Crown 8vo. is. 6d. 
 
 Answers to the Examples in the Scholar s Arithmetic. Crown 
 
 8vo. \s. 6d. 
 
 The Scholar s Algebra. Crown 8vo. is. 6d. 
 
 Aldis [W. S., M.A.). A Text-Book of Algebra. Crown 8vo. 
 
 Nearly ready. 
 
 Baynes {R. E., M.A.). Lessons on Thermodynamics. 1878. 
 
 Crown 8vo. ']s. 6d. 
 
 Chambers {G. F., F.R.A.S.). A Handbook of Descriptive 
 
 Astrojioiny. Third Edition. 1877. Demy 8vo. 2Sj. 
 
 Clarke {Col. A. R.,C.B.,R.E.). Geodesy. 1880. 8vo. 11s. 6d. 
 
 Cremona {Ltiigi). Eleme7its of Projective Geometry. Trans- 
 lated by C. Leudesdorf, M.A. 8vo. \2s.(id. 
 
 Donkin. Acoustics. Second Edition. Crown 8vo. ']s. 6d. 
 
 Euclid Revised. Containing the Essentials of the Elements 
 
 of Plane Geometry as given by Euclid in his first Six Books. Edited by 
 R. C. J. Nixon, M.A. Crown 8vo. "js. 6d. 
 
 Sold separately as follows, 
 
 Books'I-IV. ^s. 6d. Books I, II. is. 6d. 
 
 Book I. IS. 
 
CLARENDON PRESS, OXFORD. 31 
 
 Galto7i [Douglas, C.B., F.R.S.). The Construction of Healthy 
 
 Dwellings. Demy 8vo. loJ. dd, 
 
 Hamilton {Sir R. G. C), and J. Ball. Book-keeping. New 
 
 and enlarged Edition. Extra fcap. 8vo. limp cloth, 2S. 
 
 Ruled Exercise books adapted to the above may be had, price 2f. 
 
 Harconrt {A. G. Vernon, M.A.), and H G. Madan, M.A. 
 
 Exercises in Practical Chemistry. Vol. I. Elementary Exercises. Third 
 Edition. Crown 8vo. qj. 
 
 Maclaren [Archibald). A System of Physical Education : 
 
 Theoretical and Practical. Extra fcap. 8vo. "js. 6d. 
 
 Madan [H. G., M.A.). Tables of Qualitative Analysis. 
 
 Large 4to. paper, 4^. (td. 
 
 Maxzvell [J. Clerk, M.A., F.R.S.). A Treatise on Electricity 
 
 and Magnetism. Second Edition. 2 vols. Demy 8vo. i/. lis. 6d. 
 
 A7t Elementary Treatise on Electricity. Edited by- 
 William Garnett, M.A. Demy 8vo. 7^. ^d. 
 
 Minchin [G. M., M.A). A Treatise on Statics with Applica- 
 tions to Physics. Third Edition, Corrected and Enlarged. Vol. I. Equili- 
 brium of Coplanar Forces. Svo. 9^. Vol. II. Statics. 8vo. ids. 
 
 Uniplanar Kinematics of Solids and Fluids. Crown 
 
 8vo. 'js. 6d. 
 
 Phillips [John, M.A., F.R.S.). Geology of Oxford and the 
 
 Valley of the Thames. 1S71. Svo. 21s. 
 
 ■ Vesuvius. 1869. Crown Svo. \os. 6d. 
 
 Prestwich [Joseph, M.A., F.R.S.). Geology, Chemical, Physical^ 
 
 and Stratigraphical. Vol. I. Chemical and Physical. Royal 8vo. 255. 
 
 Roach [T., M.A.). Elementary Trigonometry. Crown Svo. 
 
 Nearly ready. 
 
 Rollestons Forms of Animal Life. Illustrated by Descriptions 
 
 and Drawings of Dissections. New Edition. (^Nearly ready ^ 
 
 Smyth. A Cycle of Celestial Objects. Observed, Reduced, 
 
 and Discussed by Admiral W. H. Smyth, R.N. Revised, condensed, and 
 greatly enlarged by G. F. Chambers, F.R.A.S. i88i. Svo. Price reduced 
 to 12s. 
 
 Stewart [Balfour, LL.D., F.R.S.). A Treatise on Heat, with 
 
 numerous Woodcuts and Diagrams. Fourth Edition. Extra fcap. Svo. 
 
 7J. 6d, 
 
32 CLARENDON PRESS, OXFORD. 
 
 Vernon- Harcourt {L. F., M.A.). A Treatise on Rivers and 
 
 Canals, relating to the Control and Improvement of Rivers, and the Design, 
 Construction, and Development of Canals. 2 vols. (Vol. I, Text. Vol. II, 
 Plates.) 8vo. 21s. 
 
 • Harbours and Docks ; their Physical Features, History, 
 
 Construction, Equipment, and Maintenance ; with Statistics as to their Com- 
 mercial Development. 2 vols. 8vo. 25J. 
 
 Watson [H. IV., M.A.). A Treatise on the Kitietic TJieory 
 
 of Gases. 1S76. 8vo. is.()d. 
 
 Watson [H. W., D. Sc., F.R.S.), and S. H. Btirbury, M.A. 
 
 I. A Treatise on the Application of Getieralised Coordiriates to the Kinetics of 
 
 a Material Sy Steffi. 1879. 8vo. 6j. 
 
 II. The Mathematical Theory of Electricity and Mag7tetism. Vol. I. Electro- 
 statics. 8vo. los. 6d. 
 
 Williamson {A. W., Phil. Doc, F.R.S.). Chemistry for 
 
 Students. A new Edition, with Solutions. 1873. Extra fcap. 8vo. 8j, 6d. 
 
 VII. HISTORY. 
 
 Bliintschli {J. K.). The Theory of the State. By J. K. 
 
 Bluntschli, late Professor of Political Sciences in the University of Heidel- 
 berg. Authorised English Translation from the Sixth German Edition. 
 Demy 8vo. half bound, 12s. 6d. 
 
 Finlay {George, LL.D.). A History of Greece from its Con- 
 quest by the Romans to the present time, B.C. 146 to A.D. 1864. A new 
 Edition, revised throughout, and in part re-written, with considerable ad- 
 ditions, by the Author, and edited by H. F. Tozer, M.A. 7 vols. Svo. 3/. io.f. 
 
 Fortescue {Sir John, Kt.). The Governance of England: 
 
 otherwise called The Difference between an Absolute and a Limited Mon- 
 archy. A Revised Text. Edited, with Introduction, Notes, and Appendices, 
 by Charles Plummer, M.A. Svo. half bound, 12^-. (^d. 
 
 Freeman {E.A., D.C.L.). A Short History of the Norman 
 
 Conquest of England. Second Edition. Extra fcap. Svo. 2s. 6d. 
 
 George {H. B.,M.A .). Genealogical Tables illustrative of Modern 
 
 History. Third Edition, Revised and Enlarged. Small 4to. 1 2s. 
 
 Hodgkin {T.). Italy and her Invaders. Illustrated with 
 
 Plates and Maps. Vols. I— IV., A.D. 376-553. Svo. 3/. 8j. 
 
 Kitchin {G. W.,D.D.). A History of France. With numerous 
 
 Maps, Plans, and Tables. In Three Volumes. Second Edition. Crown Svo. 
 each los. dd. 
 
 Vol. I. Down to the Year 1453. 
 
 Vol.2. From 1453-1624. Vol. 3. From 1624-1793. 
 
CLARENDON PRESS, OXFORD. 33 
 
 Payne {E. J., M.A.). A History of the United States of 
 
 America. In the Press. 
 
 Ranke [L. von). A History of England, principally in the 
 
 Seventeenth Century. Translated by Resident Members of the University of 
 Oxford, under the superintendence of G. W. Kitchin, D.D., and C. W. Boase, 
 M.A. 1875. 6 vols. 8vo. 3/. IS. 
 
 Razvlinson [George, M.A.). A Manual of Aneient History, 
 
 Second Edition. Demy 8vo. 14^-. 
 
 Rogers [J. E. Thorold, M.A.). The First Nine Years of the 
 
 Bank of England. 8vo. 8j. ()d. 
 
 Select Charters and other Illustrations of English Constittitional 
 
 History, from the Earliest Times to the Reign of Edward I. Arranged and 
 edited by W. Stubbs, D.D. Fifth Edition. '1883. Crown 8vo. 8j. 6^. 
 
 Stiihbs ( W., D.D.). The Constitutional History of England^ 
 
 in its Origin and Development. Library Edition. 3 vols, demy 8vo. 2/. 8j. 
 Also in 3 vols, crown 8vo. price 12s. each. 
 
 Seventeen Lectures on the Study of Medieval and 
 
 Modern History, 8cc., delivered at Oxford 1867-1884. Demy 8vo. half-bound, 
 los. 6d. 
 
 Wellesley. A Selection from the Despatches, Treaties, and 
 
 other Papers of the Marquess Wellesley, K.G., during his Government 
 of India. Edited by S. J. Owen, M.A. 1877. 8vo. 1/. 4J. 
 
 Wellington. A Selection from the Despatches, Treaties, and 
 
 other Papers relating to India of Field-Marshal the Duke of Wellington, K.G. 
 Edited by S. J. Owen, M.A, 1880. 8vo. 24J. 
 
 A Llistory of British India. By S. J. Owen, M.A., Reader 
 
 in Indian History in the University of Oxford. In preparation. 
 
 VIII. LAW. 
 
 Alberici Gentilis, I. CD., I.C., De lure Belli Libri Tres. 
 
 EdiditT. E. Holland, LCD. 1877. Small 4to. half morocco, 21s. 
 
 Anson (Sir William R., Bart., D.C.L.). Pri^tciples of the 
 
 English Law of Contract, and of Agency in its Relation to Contract. Fourth 
 Edition. Demy 8vo. ioj. f>d. 
 
 Law and Custom of the Constitution. Part I. Parlia- 
 ment. Demy 8vo. \os. 6d. 
 
 Bent ham [Jeremy). An Introduction to the Principles of 
 
 Morals and Legislation. Crown 8vo. 6j. dd. 
 
 Digby [Kenehn E., M.A.). An Introductiott to the History of 
 
 the Law of Real Property. Third Edition. Demy 8vo. Jos. 6d. 
 
 Gaii Institutiomiin jfttris Civilis Commentarii Quattuor ; or, 
 
 Elements of Roman Law by Gains. With a Translation and Commentary 
 by Edward Poste, M.A. Second Edition. 1875. 8vo. iSj. 
 
34 CLARENDON PRESS, OXFORD. 
 
 Hall ( W. E., M.A.). International Law. Second Ed. 8vo. 2ij. 
 Holland [T. E., D.C.L.). The Elements of Jurisprudence. 
 
 Third Edition. Demy 8vo. \os. dd. 
 
 The Enropeati Concert in the Eastern Qtiestion, a Col- 
 lection of Treaties and other Public Acts. Edited, with Introductions and 
 Notes, by Thomas Erskine Holland, D.C.L. 8vo. 12s. 6d. 
 
 Imperatoris lustiniani Institntionum Libri Quatiuor ; with 
 
 Introductions, Commentary, Excursus and Translation. By J.E. Moyle, B.C.L., 
 M.A. 2 vols. Demy 8vo. 21^. 
 
 jfustinian, The Institutes of, edited as a recension of the 
 
 Institutes of Gaius, by Thomas Erskine Holland, D.C.L. Second Edition, 
 1881. Extra fcap. 8vo. ^s. 
 
 jfustinian, Select Titles from the Digest of. By T. E. Holland, 
 D.C.L., and C. L. Shadwell, B.C.L. 8vo. 14J. 
 
 Also sold in Parts, in paper covers, as follows : — 
 
 Part I. Introductory Titles. 2s. 6c?. Part 11. Family Law. is. 
 
 Part HI. Property Law. 2s. 6d. Part IV. Law of Obligations (No. i). 3s, 6d 
 
 Part IV. Law of Obligations (No. 2). 4s. 6d. 
 
 Lex Aquilia. The Roman Law of Damage to Property : 
 
 being a Commentary on the Title of the Digest ' Ad Legem Aquiliam ' (ix. 2). 
 With an Introduction to the Study of the Corpus luris Civilis. By Erwin 
 Grueber, Dr. Jur., M.A. Demy 8vo. \os. 6d. 
 
 Markby ( W., D.C.L.). Elements of Laiv considered with refer- 
 ence to Principlesof General Jurisprudence. Third Edition. Demy 8vo. 12^.6^ 
 
 Twiss [Sir Tr avers, D.C.L.). The Law of Nations considered 
 
 as Independent Political Communities. 
 Part I. On the Rights and Duties of Nations in time of Peace. A new Edition, 
 
 Revised and Enlarged. 1884. Demy Svo. 15J. 
 Part II. On the Rights and Duties of Nations in Time of War. Second Edition, 
 
 Revised. 1875. Demy 8vo. 2ij. 
 
 IX. MENTAL AND MORAL PHILOSOPHY, &c. 
 
 Bacon^s Novum Orgamim. Edited, with English Notes, by 
 G. W. Kitchin, D.D. 1855. Svo. 9J. 6^. 
 
 Translated by G. W. Kitchin, D.D. 1855. Svo. 9.?. 6d. 
 
 Berkeley. The Works of George Berkeley, D.D., formerly 
 
 Bishop of Cloyne ; including many of his writings hitherto unpublished. 
 With Prefaces. Annotations, and an Account of his Life and Philosophy, 
 by Alexander Campbell Eraser, M.A. 4 vols. 1871, Svo. 2/. i^s. 
 The Life, Letters, &c. i vol. iGs, 
 
CLARENDON PRESS, OXFORD. 35 
 
 Berkeley. Selections from. With an Introduction and Notes. 
 
 For the use of Students in the Universities. By Alexander Campbell Fraser, 
 LL.D. Second Edition. Crown 8vo. 'js. 6d. 
 
 Fowler ( T., D.D.). The Elements of Deductive Logic, designed 
 
 mainly for the use of Junior Students in the Universities. Eighth Edition, 
 . with a Collection of Examples. Extra fcap. 8vo. 3^.6^. 
 
 The Elemetits of Inductive Logic, designed mainly for 
 
 the use of Students in the Universities. Fourth Edition. Extra fcap. 8 vo. 6j. 
 
 and Wilson {J. M., B.D.). The Principles of Morals 
 
 (Introductory Chapters). 8vo. boards, y. 6d. 
 
 The Principles of Morals. Part II. (Being the Body 
 
 of the Work.) 8vo. \os. 6d. 
 
 Edited by T. FCWLER, D.D. 
 
 Bacon. Novum Orga7ium. With Introduction, Notes, &c. 
 
 1878. 8vo. 14^. 
 
 Locke's Conduct of the Understanding. Second Edition. 
 
 Extra fcap. 8vo. 2s. 
 
 Danson {y. T). The Wealth of Households. Crown 8vo. ^s. 
 
 Green [T. H., M.A.). Prolegomena to Ethics. Edited by 
 A. C. Bradley, M. A. Demy Svo. 1 2j. 6(/. 
 
 Hegel. The Logic of Hegel ; translated from the Encyclo- 
 paedia of the Philosophical Sciences. With Prolegomena by William 
 Wallace, M.A. 1874. 8vo. 14J. 
 
 Lotzes Logic, in Three Books ; of Thought, of Investigation, 
 
 and of Knowledge. English Translation; Edited by B. Bosanquet, M.A., 
 Fellow of University College, Oxford. Svo. cloth, 12s. 6d. 
 
 Metaphysic, in Three Books; Ontology, Cosmology, 
 
 and Psychology. English Translation ; Edited by B. Bosanquet, M.A. 
 Svo. cloth, 1 2J. ()d. 
 
 Martineau [James, D.D.) . Types of Ethical Theory. Second 
 
 Edition. 2 vols. Crown Svo. 15J. 
 
 Rogers{y.E. Thorold,M.A.). A Manual of Political Economy^ 
 
 for the use of Schools. Third Edition. Extra fcap. Svo. 4J. 60'. 
 
 Smith's Wealth of Nations. A new Edition, with Notes, by 
 
 J. E. Thorold Rogers, M.A. 2 vols. Svo. 1880. 21s. 
 
36 CLARENDON PRESS, OXFORD. 
 
 X. PINE ART. 
 
 Butler [A. J., M.A., F.S.A.) The Ancient Coptic Chirches of 
 
 Egypt. 2 vols. 8vo. 30J. 
 
 Head [Barclay K). Historia Ntimornm. A Manual of Greek 
 
 Numismatics. Royal 8vo. half morocco, 42 j. 
 
 Hullah [John). The Cultivation of the Speaking Voice, 
 
 Second Edition. Extra fcap. 8vo. 2s. 6d. 
 Jackson [T. C, M.A.). Dalmatia, the Qnarnero, ajid Istria ; 
 
 with Cettigne in Montenegro and the Island of Grado. By T. G. Jackson, 
 M.A., Author of ' Modem Gothic Architecture.' In 3 vols. With many 
 Plates and Illustrations. Just ready. 
 
 Ouseley [Sir F. A. Gore, Bart.). A Treatise on Harmony. 
 
 Third Edition. 4to. ioj. 
 
 A Treatise on Coujiterpoint, Canon, and Fugue, based 
 
 upon that of Cherubini. Second Edition. 4to. i6j. 
 
 A Treatise on Musical Form and General Composition. 
 
 Second Edition. 4to. los. 
 
 Robinson[y.C., F.S.A. ). A Critical Account of the Drawings 
 
 by Michel A^igelo and Ra.ffaello in the University Galleries, Oxford. 1870. 
 Crown 8vo. 4^. 
 
 Troutheck [J., M.A.) and R. F. Dale, M.A. A Mttsic Primer 
 
 (for Schools). Second Edition. Crown 8vo. u. 6^. 
 
 Tyrwhitt[R. St. J., M.A). A Handbook of Pictorial Art. 
 
 With coloured Illustrations, Photographs, and a chapter on Perspective by 
 A. Macdonald. Second Edition. 1875. 8vo. half morocco, i8j. 
 
 Upcott [L. E., M.A). An Introduction to Greek Sculpture. 
 
 Crown 8vo. 4J. dd. 
 
 LONDON: HENRY FROWDE, 
 Oxford University Press Warehouse, Amen Corner, 
 
 OXFORD: CLARENDON PRESS DEPOSITORY, 
 116 High Street. 
 
 ^f The Delegates of the Press invite suggestions and advice from all persons 
 interested in education; and will be thankful for hints, ^c. addressed to the 
 Secretary to the Delegates, Clarendon Press, Oxford. 
 
COLUMBIA UNIVERSITY LIBRARY 
 
 This book is due on the date indicated below, or at the 
 expiration of a definite period after the date of borrowing, 
 as provided by the rules of the Library or by special ar- 
 rangement with the Librarian in charge. 
 
 DATE BORROWED 
 
 DATE DUE 
 
 DATE BORROWED 
 
 DATE DUE 
 
 
 W Z-'^ 
 
 
 
 
 ^f^v " " 'ifB 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C£8(239)MI00