COLUMBIA LIBRARIES OFFSITE 
 
 HEALTH SCIENCES STANDARD 
 
 HX641 72929 
 QP190 .M42 Physiology of secret 
 
 THE PHYSIOLOGY 
 
 RECAP 
 
 SECRETION. 
 
 ALBERT P. MATHEWS. 
 
 Submitted in Partial Fulfilment of the Requirements for 
 
 THE Degree of Doctor of Philosophy in the Faculty 
 
 of Pure Science, Columbia University. 
 
 [Reprinted from Annat.s N. Y. Acad. Sci., XI, No. 14, pp. 293-368.] 
 
 LANCASTER, PA. 
 The New Era Printing Comp.\ny, 
 ", 1898. 
 
 THE i.ienAK. 
 
CIPI9D 
 
 M42 
 
 CoUege of ^ijj'sticiang anij g)urseonj{ 
 
 l^ibrarp 
 
Digitized by tine Internet Arciiive 
 
 in 2010 witii funding from 
 Columbia University Libraries 
 
 http://www.archive.org/details/physiologyofsecr02math 
 
THE PHYSIOLOGY 
 
 SECRETION. 
 
 ALBERT P. MATHEWS. 
 
 Submitted in Partial Fulfilment of the Requirements for 
 
 THE Degree of Doctor of Philosophy in the Faculty 
 
 OF Pure Science, Columbia University. 
 
 [Reprinted from Annals N. Y. Acad. Sci., XI, No. 14, pp. 293-368.] 
 
 NO LONaCR THE PROPtHTY Of 
 
 LANCASTER, PA. _^, «rr<*#^ 
 
 THI LIBRARY OF WNWiy 
 
 The New Era Printing Company, tMEOto»<c*L mtrnt^ny, N y 
 i8q8. - ' ■ . 
 
 THEOLOGIC^l SEMINARY, 
 

 ^ ALBERT PRESCOTT MATHEWS. 
 
 S. B., Massachusetts Institute of Technology, 1892. 
 Assistant in Biology, Massachusetts Institute of Technology, 1892-3. 
 Fellow in Biology, Columbia University, 1893-5. 
 Student at Cambridge University, England; and Marburg, Ger- 
 many, 1895-7. 
 Student at Columbia University, 1897-8. 
 Assistant in Physiology, Harvard Medical School, Boston, 1898-9. 
 
PUBLICATIONS. 
 
 1. On Wurtz's Method for the Differentiation of Bacillus typhi ab- 
 
 dominalis from Bacillus coli communis. Technology Quar- 
 terly, 1894. 
 
 2. Maturation, Fertilization and Polarity in the Echinoderm Egg. 
 
 New Light on the " Quadrille of the Centers." With E. B. 
 Wilson. Journal of Morphology, Vol. X., 1895. 
 
 3. The Origin of Uric Acid in the Organism. A Review. Ne7v 
 
 York Medical Journal, 1896. 
 
 4. Internal Secretions Considered in Relation to Variation and 
 
 Development. Science, Vol. V., 1897. 
 
 5. The Scope and Present Position of Bio-chemistry. American 
 
 Naturalist, Vol. XXXI., 1897. 
 
 6. Zur Chemie der Spermatozoon. Zeitschft. filr physiol. Chemie, 
 
 Bd. XXIII. , 1897. 
 
 7. Die Verdanungsprodukte des Protamins. With A. Kossel. 
 
 Zeitschft. filr physiol. Chetnie, Bd. XXIV., 1898. 
 
 8. A Contribution to the Chemistry of Cytological Staining. 
 
 Aviej'ican Journal of Physiology, Vol. I., 1898. 
 
 9. The Determining Causes of Cell Division. New York Medical 
 
 Journal, 1898. 
 [o. The Life and Work of Felix Hoppe-Seyler. Popular Science 
 Monthly, 1898. 
 
 [I. Structural Changes in the Pancreas Cell, together with some 
 General Considerations on Cell Metabolism. Submitted as an 
 Alternative Thesis to Columbia University for the Degree of 
 Doctor of Philosophy. To appear in the Journal of Mor- 
 phology. 
 
[Annals N. Y. Acad. Sci., XL, No. 14, pp. 293 to 368, September 12, 1898.] 
 
 THE PHYSIOLOGY OF SECRETION. 
 
 Albert Mathews. 
 
 (Read April ii, 1898.) 
 
 I. Introduction : ' Page. 
 Criticism of the secretory nerve theory . 294 
 
 II. Sympathetic Salivary Secretion : . . 303 
 
 a. The rate of sympathetic secretion 304 
 
 /'. The decrease in amount of saUva obtainable upon several successive stimu- 
 lations 309 
 
 c. The augmentation of sympathetic saliva 311 
 
 d. Paralysis of the sympathetic by emptying the ducts and its restoral to power 
 
 by injection of fluid into the ducts 314 
 
 e. The character of sympathetic saliva 320 
 
 f. Further evidence of the muscular nature of the mechanism of sympathetic 
 
 secretion 324 
 
 g. The location and nature of the contractile substance in the gland .... 328 
 
 //. Changes in the gland cells on sympathetic stimulation 328 
 
 ?'. Summaiy and conclusion 329 
 
 III. Other Secretions due to Muscle-Action : 331 
 
 IV. Salivary Secretion Ensuing on Stimulation of the Vaso-Dil.ator 
 
 Nerves : 
 
 a. The increase in the percentage of organic constituents coincident with an 
 
 increased rate of secretion 332 
 
 b. The post-mortem chorda secretion 337 
 
 c. The nature of the action of atropine and pilocarpine 349 
 
 d. The action of quinine and nicotine 355 
 
 c. Evidence of the osmotic character of the salivary secretions which are ac- 
 companied by vaso-dilation 356 
 
 f. Conclusion. The physiology of salivary secretion 358 
 
 V. Some Other Secretions : 359 
 
 a. The secretion of sweat 359 
 
 b. The secretion of the pancreas 360 
 
 VI. General Conclusion and Summary : 361 
 
 VII. Literature : 364 
 
 Annals N. Y. Acad. Sci., XI, September 12, 1898 — 20. 
 
 (293) 
 
294 MATHEWS. 
 
 I. INTRODUCTION. 
 
 A Criticism of the Secretory-nerve Theory. 
 
 Nearly fifty years ago it was suggested by Ludwig*^ that se- 
 cretion was a function of the gland cells controlled by the ac- 
 tivity of special nerve fibres. Upon the gland cell, thus em- 
 phasized as the prime factor in secretion, and upon its relation 
 to nerve action, most of the subsequent study of the physiology 
 of secretion has been focussed. This study has unearthed such 
 evidences of the truth of Ludwig's hypothesis that to-day few 
 theories of physiology rest upon a foundation apparently firmer, 
 or are more widely accepted, than the hypothesis of secretory 
 nerves. Indeed, the recent discovery,*" by means of the Golgi 
 and Ehrlich methylen-blue methods, of the remarkably rich 
 distribution of nerves to glands, and of the endings of these 
 nerves about the gland cells, has seemed the final convincing 
 demonstration of the truth of the theory which so many years 
 ago foretold their existence. 
 
 The theory of secretory nerves did not long remain in the 
 simple form suggested by Ludwig, for it soon received, at the 
 hands of Heidenhain, a more complete and definite shape. 
 First seriously worked out by him in 1868^^ the theory Avas 
 further developed in 1878^" and took its final form in his great 
 treatise on secretion embodied in Hermann's Handbuch der 
 Physiologic in 1880.^^ The Ludwig-Heidenhain theory, thus 
 crystallized by Heidenhain, has been the lens through which 
 the facts of secretion accumulated from 1868 to the present 
 time, have been viewed. This theory may be briefly stated as 
 follows : 
 
 Secretion is a specific function of the gland cells controlled by 
 special secretory nerve fibres, acting directly upon these cells. 
 There are two kinds of these nerve fibres : trophic fibres, which 
 render the cell contents soluble ; and secretory fibres, which 
 diminish the resistance to filtration offered by the lumen end of 
 the cell. In consequence of this decreased resistance, the con- 
 tents of the cell, which are under high endosmotic pressure, 
 escape into the lumen. At the same time the cell imbibes liquid 
 from the lymph space. 
 
SECRETION PHYSIOLOGY. 295 
 
 Heideiihain, R. Ueber secretorische ziud trophische Drihennerven, PJli'iger'' s 
 Archv. f. d. gesam. Physiologie. Bd. XVII, iSjS, pp. 60 and following : "The 
 cell is normally under high endosmotic pressure. On nerve stimulation a 
 molecular rearrangement takes place at the lumen end of the cell, so that the re- 
 sistance to filtration is diminished and water flows out. This flow may be hastened 
 by contractions of the protoplasm, as Klihne observed in the rabbit's pancreas under 
 the microscope. The tension of the water within the cell being thus diminished, 
 water begins to flow out of the lymph and capillaries into the cell. At the end of 
 stimulation molecules are rearranged, the loss of water by the cell ceases, and se- 
 cretion stops." " The attractive pull on the water comes from the protoplasm of 
 the outer zone." 
 
 Before proceeding with the discussion of the evidence upon 
 which this theory rests, it will make the matter clearer to recall 
 the conception of secretion which the Ludwig-Heidenhain 
 theory supplanted. For some of the facts brought forward by 
 these authors are of value, not as direct evidence of the exist- 
 ence of secretory nerves, but because they disprove an alterna- 
 tive earlier conception. The prevalent conception of secretion, 
 before Ludwig's time, was that liquid driven by intra-capil- 
 lary pressure filtered out through the gland. ^'^ The chorda 
 tympani was the principal secretory nerve then known, and it 
 was believed to cause secretion by greatly increasing intra-ca- 
 pillary pressure by contraction of the veins or arterioles. The 
 discovery of the vaso-dilator function of this nerve shortly 
 thereafter by Claude Bernard re-emphasized the possibility of 
 a high intra-capillary pressure being an essential cause of secre- 
 tion. It is not surprising that many physiologists of that day 
 believed that this striking correspondence between vaso-dilation 
 and secretion could not be accidental, and it was natural for 
 them to refer the secretory power of the nerve to its action on 
 the blood vessels. 
 
 The first blows against the theory that the vascular system 
 stood necessarily in a causal relation to secretion were dealt by 
 Ludwig and his pupils. They discovered that stimulation of the 
 upper end of the cut cervical sympathetic nerve caused a secretion 
 from the submaxillary gland of the dog,^'^ but this secretion, un- 
 like that due to the chorda, was afterwards found to be accom- 
 panied by a pronounced vaso-constriction instead of dilation. 
 They found that the pressure capable of being generated by the 
 saliva flowing from Wharton's duct might considerably surpass 
 
296 MATHEWS. 
 
 the pressure of the blood even in the carotid artery. They 
 thus demohshed, once and for all, the filtration theory. They 
 found, further, that the temperature of the sahva secreted from 
 the dog's submaxillary might surpass by i.5°C., the temperature 
 of the blood in the carotid artery,"*^ and as final evidence that the 
 chorda tympani could induce secretion independent of the vaso- 
 motor action, they brought forward the observation that stimu- 
 lation of this nerve still caused a secretion, some minutes after 
 the heart ceased to beat.^^ It is not strange that, in the face of 
 such facts, Ludwig should have felt compelled to assume the 
 secretory activity of the gland cell. 
 
 Heidenhain soon added other facts pointing in the same di- 
 rection. He found that if the blood supply be cut off from the 
 submaxillary gland by compression of the artery the chorda 
 still caused a secretion analogous to the post-mortem secretion 
 after the heart ceases to beat.'^ Giannuzzi^^ discovered that by 
 the injection of sodium carbonate or a dilute solution of hydro- 
 chloric acid into Wharton's duct a pronounced vaso-dilation en- 
 sued, on stimulation of tKe chorda, but no secretion. Heiden- 
 hain^^ found that quinine sulphate injected into the duct had a 
 similar action, and that atropine"^ effectually paralyzed secretion, 
 while leaving the vaso-dilator power of the nerve unaltered. 
 Heidenhain^" also discovered, and Langley confirmed his obser- 
 vation, that after the chorda tympani had been paralyzed by the 
 action of nicotine, either injected subcutaneously or applied di- 
 rectly to the submaxillary ganglion, the chorda tympani recov- 
 ered its secretory function before its dilator function. He 
 observed, also, that after the chorda had been cut and allowed 
 to degenerate for 2-3 days stimulation of the nerve still caused 
 an increase in secretion, without an increase in the flow of blood 
 from the gland's vein. This evidence showed that vaso-dilation 
 might ensue without a secretion, that secretion might take 
 place unaccompanied by vaso-dilation, and that secretion might 
 be caused by stimulating dilator nerves after cutting off the 
 blood supply. If these facts were true vaso-dilation could not 
 be the cause of secretion, and hence that cause must be sought 
 in some other gland element than the blood vessels. 
 
SECRETION PHYSIOLOGY. 297 
 
 Evidence of a more positive kind of the direct action of nerves 
 upon the gland cells was not long lacking. Heidenhain showed 
 that stimulation of secretory nerves caused well-marked changes 
 in the structure of the gland cells.-' He discovered that the 
 specific constituents of the secretion were accumulated in the cell 
 during glandular rest, and discharged from the cell during secre- 
 tion. That these substances were not simply dissolved from the 
 cells by the water stream passing through them he endeavored 
 to show by the fact that on passing from a weak to a stronger 
 stimulation of the chorda tympani, or other dilator secretory 
 nerve, not only the rate, but also the concentration of the secre- 
 tion increased. Apparently the more rapidly secreted saliva, 
 although in contact with the cell contents for a briefer time, never- 
 theless dissolved more of them than that more slowly secreted. 
 This obviously would have been impossible if the contents of the 
 cell had not been rendered more soluble by the action of the 
 nerve during the stronger stimulus. He brought forward, also, 
 still more convincing evidence. ^^ In the dog's parotid gland 
 stimulation of the cervical sympathetic causes, generally, no 
 secretion, but if this nerve be irritated coincident with the dila- 
 tor secretory nerve the saliva secreted under the influence of 
 both nerves is more concentrated than that secreted during irri- 
 tation of the dilator nerve alone. Apparently the sympathetic, 
 though causing no secretion, must, nevertheless, act on the cells, 
 so as to render their contents more soluble. That this effect of 
 the sympathetic could not be due to any possible action of the 
 nerve on contractile tissue of the gland, as suggested by Schiff,'^'' 
 Eckhard^'^ and others, Heidenhain believed von Wittich" had 
 conclusively demonstrated. That the well-known high concen- 
 tration of the sympathetic saliva could not be referred to the 
 nerve's vaso-constrictor action Heidenhain^- showed by the 
 fact that, if the gland artery be almost totally compressed, the 
 following chorda saliva was not rendered more concentrated. 
 
 These facts undoubtedly furnish strong evidence that the 
 sympathetic and other nerves act on the gland cells, not only 
 increasing the flow of water through them, but also rendering 
 their contents more soluble. 
 
298 MATHEWS. 
 
 Most of these facts, brought out chiefly in the saHvary glands, 
 have been found to be true for other glands. The independ- 
 ence of blood pressure and secretion, the inhibitory action of 
 \ atropine, and an increase in concentration of the secretion coinci- 
 'dent with a more rapid flow, have been observed by Afanassiew 
 arid Pawlow,^ Gottlieb, ^^ Pawlow and S.^^Simonoskajain-the pan- 
 creas, stomach and other glands, in which secretion is normally 
 accompanied by vaso-dilation. Sweat may be secreted during 
 vaso-constriction or vaso-dilation, and in the cat's foot, twenty 
 minutes after ligaturing the artery or cutting the leg from the 
 body/" The skin glands of amphibia can secrete in the total 
 absence of blood supply.'^ Moreover, of recent years, the im- 
 portance of the condition of the secreting cells, as a factor of 
 secretion, has been clearly realized. The quick paralysis of 
 some secretions during dyspnoea or by the action of drugs has 
 emphasized this factor of secretion. . Even in the kidney, where 
 secretion apparently more nearly approaches a filtration, it has 
 been shown that the condition of the capillary, or glomerular 
 epithelium, and the character of the blood, exerts an influence 
 on the secretion.^ The possibility at once suggests itself that 
 if the condition of the cells is so readily affected by external 
 agents it maybe modified by direct nerve action. The very 
 rich nerve supply of many glands and the intimate association 
 of nerve end and gland-cell undoubtedly bring strong confirma- 
 tion to this supposition. 
 
 From this brief outline the extreme complexity of the problem 
 of secretion will be manifest. Some secretions are accompanied 
 by vaso-dilation ; others by vaso-constriction. Some may per- 
 sist twenty minutes after cutting off the blood supply ; others 
 are paralyzed within two or three minutes. Some are paralyzed 
 by atropine and quinine ; others are not. In the same gland 
 stimulation of one nerve may cause the secretion of a large 
 amount of watery secretion, while stimulation of another nerve 
 causes the secretion of a small amount of exceedingly viscid 
 secretion. There seems, in fact, to be no general rule of secre- 
 tion true for all glands. The great difference between the phe- 
 nomena of different secretions suggests that the mechanisms of 
 
SECRETION PHYSIOLOGY. 299 
 
 those secretions may be different in different cases. However 
 probable it may seem, a priori, that there is everywhere one 
 fundamental mechanism underlying all these secretions, a de- 
 cent regard for truth forbids one accepting so far reaching a con- 
 clusion, unless it be supported by very strong evidence. 
 
 In the present paper, therefore, I wish to reopen the question 
 whether all secretions are due to the activity of the gland cells, 
 and to re-examine the evidence of the existence of nerves act- 
 ing on those cells. The great theoretical and practical im- 
 portance of Ludwig's conception is a sufficient excuse for a 
 critical and experimental review, in the light of the physiology 
 of the present day, of the evidence upon which that theory rests. 
 Since the publication of Ludwig's and Heidenhain's Avork on se- 
 cretion knowledge has been acquired of vaso-motor changes, 
 osmosis, lymph formation as well as secretion proper, which 
 might, possibly, cause even Heidenhain or Ludwig, if consider- 
 ing the subject at this time, to adopt a somewhat different 
 interpretation of much of this evidence from that heretofore pro- 
 posed. Such a review seems the more necessary for the reason 
 that special applications of the theory have been, from time to 
 time, questioned, and because, as will be apparent in the course 
 of the following discussion, some of Heidenhain's inferences are 
 unsound, owing to his having neglected to consider possibilities 
 now known to be of importance. His recent extension of the 
 theory to lymph formation, for example, has been seriously dis- 
 puted by Starling,*^^ Cohnheim and others. Starling especially 
 has shown the uselessness of assuming any such secretory 
 mechanism in certain special cases, and has thus thrown doubt 
 upon the truth of the theory as a whole. Langley^' has ques- 
 tioned the necessity for assuming distinct "trophic" fibres to 
 explain salivary secretion, and for the kidney secretion special 
 inferences of Heidenhain have been challenged by Senator, 
 Adami^ and v. Sobiranski.^'' The difference in pressure be- 
 tween blood and secretion observed by Ludwig may be readily 
 accounted for on the basis of osmosis quite apart from any cell- 
 activity.^^ "The difference in temperature between saliva and 
 blood has been denied by Bayliss and Hill,'^ working with bet- 
 
300 MATHEWS. 
 
 ter methods. For some of the facts, also, errors of method 
 greatly diminish the value of the testimony they offer, and 
 some of that evidence depends upon the assumption that all se- 
 cretions are probably due to the same cause. Hence, whether 
 the theory of secretory nerves is true or not, it must be admit- 
 ted, I believe, that little of the evidence which has hitherto been 
 presented in support of that hypothesis can be accepted as it 
 stands. 
 
 While fully aware, therefore, of the strong a priori probability 
 that nerves may act on gland cells so as to affect osmosis through 
 them, and while appreciating the strength of the evidence that 
 they do so act, I feel myself compelled, for the reasons presented 
 in the following criticisms of that evidence, to question whether 
 secretion is really' controlled in this manner. 
 
 But not only is the evidence upon which the secretory nerve 
 theory rests inconclusive; there are also certain weaknesses in the 
 theory itself which deserve more attention than they have 
 hitherto received. It is by no means easy to understand how 
 the nerve can affect the cell in such a way as to cause a secre- 
 tion. The mere discharge of liquid from the cells into the gland 
 lumen would, as pointed out elsewhere, lead to no secretion 
 from the gland ducts. To obviate this difficulty Heidenhain 
 supposed that, while the secretory nerve diminished the resist- 
 ance of the inner end of the cell, the outer zone imbibed water 
 from the lymph and capillary. The outer zone exerted an at- 
 tractive pull upon the lymph. By the imbibition of this lymph 
 the secretion was forced along the ducts. This explanation leads 
 at once to difficulties. Not only is the explanation exceedingly 
 hypothetical, but it is difficult to see why, if the pull on the lymph 
 comes from the outer zone, secretion should be slowest after 
 long stimulation, or during paralytic secretion, when the outer 
 zone is at its greatest development, and how secretion can take 
 place at all, or with any rapidity, in glands in which the outer 
 zone has almost, or completely, disappeared, as in mucous sal- 
 ivary glands, the stomach or pancreas, after a long rest. It is 
 also difficult to understand sympathetic secretion, which takes 
 place during a period of vascular constriction. Nor can we ig- 
 
SECRETION PHYSIOLOGY. 301 
 
 nore the extreme complexity of the theory. The assumption 
 that each, or any, cell of the sub-maxillary gland has acting 
 upon it four totally different nerve ends is, in itself, highly im- 
 probable. A further difficulty is encountered when we critically 
 examine Heidenhain's assumption that the trophic and secretory 
 fibres are unequally distributed to the chorda tympani and sym- 
 pathetic. It seems simple enough to refer the small secretion 
 ensuing on sympathetic stimulation to the presence of a small 
 number of secretory fibres in this nerve, but if it be asked 
 whether these fibres innervate all the cells, or only a portion of 
 them, we are at once plunged into a maze from which there is 
 no way out. If they innervate all the cells we may ask why, 
 if a few fibres suffice, more should be present in the chorda, 
 and why the secretion should not be as copious as the chorda's. 
 If they innervate a part of the cells only, new assumptions must 
 be made to understand why stimulation of the sympathetic 
 should exhaust the constituents of the whole gland. If we 
 abandon the trophic fibres and postulate one sort of fibre only, 
 the secretory, acting on the cell, Heidenhain's facts become 
 largely inexplicable. Furthermore, when Heidenhain^'^ assumed 
 secretory nerves to the capillaries he undermined much of the 
 evidence accumulated by him of secretory nerves to glands. For 
 many of the facts of gland physiology might be understood by 
 reference to these capillary nerves. Atropine, for instance, might 
 conceivably prevent secretion by paralyzing the ends of the se- 
 cretory nerves of the capillaries, thus inhibiting the production 
 of lymph and fluid necessary for secretion. 
 
 In the present paper I have considered chiefly the physiology 
 of secretion in the salivary glands. The experimental work has 
 been devoted chiefly to studying the exceptional features of that 
 secretion which have seemed difficult of comprehension on 
 any other than the cellular theory of secretion. I have ven- 
 tured, however, to bring some other secretions into relation 
 with the conclusions concerning the mechanisms of salivary 
 secretion. 
 
302 MATHEWS. 
 
 It may prevent confusion and reconcile what might appear to 
 be contradictory statements, to give here the chief conclusion 
 drawn in the present paper. This is, that there is no single 
 mechanism of secretion. In some glands the stored metabolic 
 products are driven out of the cells by the action of muscle, as 
 in Amphibian skin glands and sudoriferous glands ; in others 
 they are removed by currents of lymph, which are probably the 
 result of osmosis, as in the pancreas, stomach, salivary glands ; 
 in some cases the cells imbibe water until they burst, and their 
 contents rush into the gland-lumen, as in the intestinal cells of 
 Ptychoptera larvae ; in others the inner end of the cell crumbles 
 to pieces, as in the mammalian milk glands. Two, or more, of 
 these mechanisms may coexist in one gland, and it is this which 
 has rendered the physiology of such glands as the salivary so 
 confusing. In the submaxillary gland, for example, I believe 
 we have a muscular mechanism, innervated by the sympathetic ; 
 and an osmotic mechanism, innervated by the chorda. The 
 sympathetic, in other words, causes secretion as Eckhard,^^ 
 Schiff,* ^' and others^" have maintained, by its action on contrac- 
 tile tissue in the gland body, thus mechanically compressing the 
 ducts and alveoli and squeezing out the secretion. The chorda 
 probably causes secretion, by its dilator action on the blood ves- 
 sels. The following pages present the evidence for these con- 
 clusions. 
 
 Before proceeding farther it is necessary to define the sense 
 in which the word "secretion" is here used. At present the 
 word has no very definite significance, as it refers to different 
 processes. For the sake of clearness it would be better to 
 designate these various processes by different names. I suggest 
 that, in the future, the word secretion be used to indicate the 
 process of extruding subtances from cells into the lumen of the 
 gland, the process of expulsion from the ducts, and the substances 
 secreted by the gland. By this use of the word cellular secre- 
 tion will be generally coincident in time with glandular. For the 
 
 *Schifif, loc. cit., p. 304, I. " It is probable that the great sympathetic which 
 causes constriction of the parotid vessels causes, at the same time, the tissue of the 
 gland to contract, and that by this contraction the gland empties itself of its con- 
 tents formed independent of nerve action." 
 
SECRETION PHYSIOLOGY. 303 
 
 process of the formation of substance by the gland cell — a dif- 
 ferent process, but one at present included under secretion — I 
 propose the name " Hylogenesis " (Gr. bhj matter and yivtatz 
 generation), and for the substances formed the name " Hylogens." 
 Thus trypinogen, mucinogen, pepsinogen are hylogens. The 
 secretions consist of the hylogens plus water, salts and other 
 substances derived unchanged from the blood. The present 
 paper deals solely with secretion proper. Hylogenesis is con- 
 sidered elsewhere.* This word seems to me preferable to that 
 of " Mesastates," suggested by Mr. J. N. Langley. Ranvier''^ 
 and Van Gehuchten^'^ wish to call the process here named hylo- 
 genesis, " secretion." This seems to me inadvisable, as thereby 
 cellular secretion would correspond with glandular rest. 
 
 The experimental work embodied in this paper has been 
 carried on chiefly in the Physiological Laboratory of Columbia 
 University, and I am particularly indebted to Professor Curtis 
 and Professor Lee both for extending to me facilities of the 
 laboratories and for suggestive criticism. A portion of the 
 work was done in the physiological laboratories of Cam- 
 bridge University, England, and Marburg University, Germany. 
 I desire to express my hearty appreciation of the courtesy of 
 Professor Michael Foster and Professor Kossel in placing the 
 facilities of their laboratories at my disposal. To Mr. J. N. 
 Langley I am indebted for critical suggestions. 
 
 IL SYMPATHETIC SALIVARY SECRETION. 
 
 Stimulation of the upper end of the divided cervical sympa- 
 thetic nerve of the cat, horse, dog, sheep or rabbit generally causes 
 a secretion from the salivary glands. This secretion has every- 
 wheref the same characteristic features, indicating that it is pro- 
 duced in all salivary glands in the same manner. These com- 
 mon features are the following : The saliva reaches its maximum 
 rate of flow in the first lo or 20 seconds, and then generally 
 ceases, although stimulation lasts for several minutes. If sev- 
 
 * Shortly to appear in the Jonr)ial of MorpJwlogy. 
 
 I Except in the resting parotid and submaxillary glands of the dog. See next 
 page. 
 
304 MATHEWS. 
 
 eral stimulations follow closely, one upon the other, the amount 
 of saliva secreted at each stimulation rapidly diminishes and 
 often becomes nothing. Stimulation becomes then again effec- 
 tive if the gland be allowed to rest, if the chorda be irritated, or 
 if liquid be injected into the gland duct. Finally, sympathetic 
 secretion is invariably accompanied by vascular constriction, and 
 the saliva, with the doubtful exception of that of the cat,^^ con - 
 tains more organic matter than that secreted from the same 
 gland under the influence of the dilator nerve. 
 
 That there are deviations from the typical course of a sympathetic secretion just 
 sketched need hardly be said. Such deviations are probably due (seep. 309) to 
 the changing fluidity of the saliva. When the saliva is thin, as in the horse, rabbit, cat 
 or sheep, the secretion follows a very typical course ; if the saliva be viscous, as in 
 the resting salivary glands of the dog, the latent period is longer, and the secretion 
 persists longer. These variations shed a not unimportant light on the mechanism 
 of secretion. 
 
 To explain these typical phenomena, assuming the secretory 
 activity of the gland cell, Heidenhain supposed that the sympa- 
 thetic nerve carried three kinds of fibres : trophic, secretory and 
 vaso-constrictor. The trophic fibres converted large quantities 
 of mucinogen (submaxillary) into soluble mucin, making the 
 juice rich in organic bodies ; the secretory fibres caused secretion ; 
 the constrictor neutralized the secretory action and stopped 
 secretion. The quick failure of the nerve on successive stimu- 
 lations was referred to the exhaustion of nerve, nerve end, or 
 gland cell. 
 
 The general features of syiupatlietic secretion seem to me, hozo- 
 ever.^ plainly to suggest that the secretion has been driven from the 
 gland by a compression of the ducts and alveoli by some contractile 
 tissue. I wish to consider these features separately, from this 
 point of view, together with experiments bearing on their proper 
 interpretation. 
 
 a. The Rate of Sympathetic Secretion. 
 
 Experiments I. and II. 
 
 Cat and dog. Submaxillary. Animals under ether. Canula 
 in Wharton's duct, connected with a narrow tube graduated in 
 millimeters, 250 mm. = 0.82 cc. Reading's every ten seconds 
 
SECRETION PHYSIOLOGY. 305 
 
 in mm. Chorda-lingual divided in each case. Cervical sympa- 
 thetic divided and stimulated by tetanic shocks, secondary coil 
 180-100 mm. The chorda was first stimulated intermittently 
 for an hour, so that the glands were secreting watery saliva. 
 
 Cat. Dog. 
 
 I II in I II 
 
 1st 10 seconds of sympathetic stimulation ... 10 .. 9 10 ... 25 .. 17 
 
 2d " " " " " . . . 9 . . 5 . . 6 . . . 4 . . 2 
 
 3d " " " " " . . . o . . o . . o . . . 3 . . 2 
 
 4th " " " " " ...0..0..0. .2.. I 
 
 5th " " " " " ...0..0..0...2..2 
 
 off off off 
 
 6th " " " " " , . , o . , o . . o . . . 8 . . I 
 
 off off 
 
 By inspection of these figures, It is seen that on stimulation 
 the secretion comes suddenly, reaches its maximum rate of flow 
 in the first few seconds, and then quickly subsides. In the cat, 
 it abruptly ceases after 20 seconds. In the dog, probably owing 
 to the greater viscidity of the saliva and the resistance offered to 
 its passage by the fine gland-tubules, it persists slightly through- 
 out the stimulation. 
 
 Heidenhain attributes the abrupt cessation of secretion, after 
 a few seconds, to the vaso-constrictor action of the nerve, in 
 consequence of which the secretory mechanism is, as it were, 
 suffocated."^ That this explanation is incorrect may readily be 
 shown by cutting off the blood by compressing the gland's 
 artery, or by decapitation. In such cases, as the following ex- 
 periments show, a perfectly typical secretion may ensue on 
 stimulation of the sympathetic, ten or more minutes after Hga- 
 turing the artery, or decapitation. 
 
 Experiment Va. 
 
 (A full account of this experiment is given on page 343.) 
 Large dog, which had received 3CC. i % morphine sulphate 
 subcutaneously. Ether given through tracheal tube. Sub- 
 maxillary dissected free, and remained attached only at the 
 hilus and by its veins. Chorda-lingual and sympathetic cut. 
 Canula connected with tube graduated in millimeters in Whar- 
 ton's duct. Gland's artery exposed by extirpation of the 
 digastric muscle. Tetanic shocks. Secondary coil at 150. The 
 
306 
 
 MATHEWS. 
 
 secretion of the sympathetic is given in mm. at ten second in- 
 tervals, 250 mm. = 0.82. cc. 
 
 25 
 
 3 25 
 
 3 
 
 30 
 
 3 
 
 32 
 
 3 
 
 35 
 
 3 
 
 37 
 
 3 
 
 40 
 
 3 42 
 
 Time. 
 h 
 
 3 30 
 
 4 
 
 C7 
 
 30 
 
 
 
 4 
 
 07 
 
 30 
 
 - 4 
 
 08 
 
 4 
 
 08 
 
 
 - 4 
 
 09 
 
 4 
 
 09 
 
 
 - 4 
 
 II 
 
 4 
 
 12 
 
 
 - 4 
 
 13 
 
 4 
 
 13 
 
 
 - 4 
 
 14 
 
 4 
 
 15 
 
 
 - 4 
 
 17 
 
 4 
 
 17 
 
 30 
 
 - 4 
 
 18 
 
 4 
 
 20 
 
 
 
 
 4 
 
 25 
 
 
 
 
 4 
 
 26 
 
 
 - 4 
 
 27 
 
 4 
 
 29 
 
 30 
 
 
 
 4 
 
 45 
 
 30 
 
 
 
 4 
 
 46 
 
 30 - 
 
 - 4 
 
 47 
 
 4 
 
 •48 
 
 30 - 
 
 - 4 
 
 49 
 
 4 
 
 50 
 
 - 
 
 - 4 
 
 51 
 
 4 
 
 51 
 
 30 - 
 
 - 4 
 
 52 
 
 4 
 
 53 
 
 
 
 
 4 
 
 53 
 
 - 
 
 - 4 
 
 54 
 
 4 
 
 54 
 
 - 
 
 - 4 
 
 55 
 
 4 
 
 55 
 
 30 - 
 
 - 4 
 
 56 
 
 4 
 
 57 
 
 - 
 
 - 4 
 
 58 
 
 5 
 
 02 
 
 
 
 
 5 
 
 03 
 
 - 
 
 - 5 
 
 09 
 
 5 
 
 09 
 
 - 
 
 - 5 
 
 10 
 
 Nerve Stimulated. 
 
 The artery going to the gland 
 was clamped close to the 
 
 hilus. 
 
 Chorda (intermittent 
 
 Secretion. 
 
 30 
 
 15 
 
 39 
 
 Sympathetic 
 
 Copious at first, it gradually 
 ceases. 
 
 16, 3, 2, 2, 0, 0, off. 
 
 O, O, O, I, o, o, off. 
 
 o, o, o, o, o, o, off. 
 
 Interval (see page 317). 
 Artery undamped. Chorda 
 stimulated intermittently for 
 several minutes. 
 
 Artery clamped. 
 
 Chorda 155 
 
 " (10 sec. int. ) 30 
 
 16 
 
 Sympathetic. 
 
 Chorda 
 Sympathetic. 
 
 Interval (see page 317). 
 Sympathetic. 
 Chorda. , 
 Interval (see page 317). 
 Artery undamped. The 
 gland secretes spontane- 
 ously. Chorda stimulated 
 intermittently. 
 
 Artery clamped. 
 
 n, 4, 2, 2, 0, off. 
 
 o 
 10, 4, 0, 0. 
 
 o, o, o. 
 
 o, o, o. 
 
 Chorda 
 
 175 
 
 " 
 
 30 
 
 " 
 
 10 
 
 (< 
 
 2 
 
 
 
 
 Sympathetic 
 
 8, 2, 1, 0. 
 
 Chorda 
 
 0, 0, 0, 0. 
 
 Sympathetic 
 
 
 
 0, 4, 3, 0, 0. 
 
 Artery undamped. 
 
 
 Spontaneous secretion. 
 
 
 Sympathetic. 
 
 9, 3, 2, 6, 0. 
 
SECRE TION PHYSIOL OGV. 
 
 307 
 
 Experiment V. 
 
 Large dog under morphine and chloroform. Right submax- 
 illary gland prepared. Chorda lingual and sympathetic cut. 
 Each nerve causes a good secretion. Readings as in previous 
 experiments. Canula in Wharton's duct. Secondary coil 150. 
 Tetanic shocks. 
 
 h. m. s. 
 5 49 30 
 
 5 
 
 50 
 
 30 
 
 5 
 
 55 
 
 
 5 
 5 
 
 57 
 58 
 
 
 
 6 
 
 10 
 
 55 
 
 - 6 
 
 NERVE STIMULATED. 
 
 Head cut oft' as rapidly as possible. 
 Spinal cord and vertebral column 
 not severed. 
 Chorda (intennittent) 
 
 " (coil 70) 
 Sympathetic (coil 7) 
 No stimulation. 
 Sympathetic 
 
 SECRETION. 
 175 
 
 o 
 40, 20, 6, 2, 0. 
 
 % 5, 2, 0, 0. 
 
 
 TIME. 
 
 
 m. 
 
 S. 
 
 h. 
 
 m. 
 
 4 
 
 30 
 
 
 
 !i 
 
 40 
 
 - 4 
 
 35 
 
 4 
 
 35 
 
 - 4 
 
 38 
 
 4 
 
 38 
 
 
 
 Experiment VI. 
 
 Dog. Conditions of experiment the same as in Experiment 
 V. Submaxillary. Both nerves active. 
 
 NERVE. SALIVA SECRETED IN MM. 
 
 Head completely severed from body. 
 
 Chorda intermittent 65 
 
 Chorda. o 
 
 Sympathetic. 14, 3, 2, 2, 0. 
 
 The foregoing experiments, demonstrating that a sympathetic 
 secretion may be- obtained ten minutes after all fluid and oxygen 
 have been cut off from the gland shows, I think, that Heidenhain 
 was wrong in ascribing the quick normal cessation of secretion 
 during sympathetic stimulation to the nerve's action on the 
 blood vessels. It is obvious that vascular constriction can have 
 nothing to do with such cessation, because the changes produced 
 in a normal gland by vascular constriction, namely, diminution 
 of water and oxygen, have existed in all three experiments at 
 least seven minutes before the nerve was stimulated, and con- 
 tinue during that stimulation without in any way affecting the 
 course of the secretion. 
 
 Even a normal gland secreting a very viscous saliva furnishes 
 evidence against the truth of Heidenhain's explanation. In the 
 
308 MATHEWS. 
 
 resting submaxillary of the dog the sympathetic secretion may 
 have a latent period of many seconds and persist for minutes. 
 An instance of such a kind is the following : 
 
 Experiment III. 
 
 Large morphinized dog, receiving chloroform. Both chorda 
 lingual and sympathetic cut. The submaxillary has not pre- 
 viously been secreting. Sympathetic stimulated by tetanic 
 shocks. Secondary coil 15. Readings every 10 seconds in 
 millimeters as before. The saliva was extraordinarily viscid. 
 Total stimulation 2 minutes, 40 seconds. Latent period ^5 
 seconds. 
 
 Amount of secretion : o, o, o, o, 5, 7, 7, 5, 5, 5, 4, 5, 5, 4,4, 
 3 ; off, 3, I, o. 
 
 If secretion can begin after 42 seconds, and endure for two 
 minutes, during a period of vascular constriction, as was the 
 case in this experiment, it can hardly be assumed that vaso- 
 constriction is the cause of the normal failure of that secretion 
 within twenty seconds. 
 
 Heidenhain seems to have overlooked the fact that a sympa- 
 thetic secretion may be obtained after cutting off the blood 
 supply, at least five minutes after the chorda becomes inopera- 
 tive. He referred the quick loss of the chorda's power in these 
 experiments, to the suffocation of the gland cell.* If the loss 
 of the chorda's secretory power is due to the paralysis of the 
 gland cell by suffocation, the sympathetic must cause secretion 
 in some other way than action on the cell, since this nerve causes 
 a normal secretion long after the chorda has been paralyzed. 
 
 The quick gush of saliva and its abrupt cessation, as well as 
 the anomalous cases represented by Experiment III, clearly indi- 
 cate a muscular mechanism of secretion. They are probably to 
 be explained as follows : On sympathetic stimulation the ducts 
 
 * Heidenhain, R. Hermann's Handbuch der Physiologic V, p. 46: "Die 
 Ursache der Verlangsamung der Absonderung bei hochgradiger Gefassverengerung 
 Oder Gefassverschluss liegt nichtin dem Sinken des Capillardruckes, sondern in der, 
 mit der kiinstliche Anamie der Driise verbundenen Verlangsamung des Blutstromes, 
 bei welcher sich das Secretions Material, und namentlich der Sauerstoff fiir die 
 Driisenzellen allmalig erschopft so dass der secretorische Apparat erstickt. ' ' 
 
SECRETION PHYSIOLOGY. 309 
 
 and alveoli are compressed and the liquid in them ejected. If 
 that liquid is thin and runs readily, as in most albuminous 
 glands, for example the parotid and submaxillary of the rabbit, 
 sheep and horse, and the cat's submaxillary, or in mucous 
 glands after long stimulation, the latent period is short, and 
 the saliva is all expelled in from 10—20 seconds. Thereafter, 
 although contraction persists, no more secretion escapes. If, 
 on the other hand, the saliva is viscid, as in the first stim- 
 ulation of a previously resting mucous gland (submaxillary and 
 parotid of dog), it offers a great resistance in passing through 
 the fine ducts and consequently requires a greater pressure and 
 a longer time to start and to expel. Consequently the latent 
 period is long and the secretion persists for some time. This 
 explains the anomalous cases represented by Experiment III. 
 In cases of very great viscidity, as in the parotid gland of the 
 dog, the resistance may even be too great to be overcome by the 
 compressing strength of the tissues. In this gland stimulation 
 of the symyathetic either causes no secretion at all or very lit- 
 tle, unless the saliva in the gland be previously diluted by the 
 action of the dilator nerve. The muscular theory, too, readily 
 explains why a typical sympathetic secretion can ensue in the 
 total absence of blood supply. 
 
 b. The Decrease in the Amount of Saliva Obtainable upon 
 Several Successive Stimulations. 
 
 If one sympathetic stimulation be followed by several others 
 the amount of saliva obtainable on the second, or followinsf 
 stimulations, is much less than the first, and may be nothing at 
 all.* If, however, the gland be allowed to rest, or if the chorda 
 be stimulated, the nerve again produces a copious secretion upon 
 sympathetic stimulation. This is shown in the following ex- 
 cerpts from experiments on the dog's and cat's submaxillary. 
 Readings in mm. Stimulation in each case for thirty seconds. 
 It is also clearly seen in Experiment VII, p. 311. 
 
 *This phenomenon has, of course, been often described. See among others 
 Langley.39 
 
 Annals N. Y. Acad. Sci., XI, September 12, 189S — 21. 
 
310 MATHEWS. 
 
 Cat. Cat. Dog. 
 
 I. ' II. 
 
 Atnoimt. Amount. 
 
 1st stimulation . .20 16 36 
 
 Rest . . . . 25 seconds I minute 2 minutes. . . . 
 
 2d stimulation . . . . o ......... 6 25 
 
 Rest .... 3 minutes 2 minutes .... I minute .... 
 
 3d stimulation .... H iO-5 ^^ 
 
 j^ggl- 2 minutes . . . 
 
 4th stimulation ^° 
 
 j^est I minute .... 
 
 5th stimulation ^^ .... 
 
 Rest . . . • I minute .... 
 
 6th stimulation 2.5 .... 
 
 Rest Chorda stimulated . 
 
 7th stimulation 25 .... 
 
 Rest ^ 2 minutes . . . 
 
 8th stimulation 6 .... 
 
 Rest I min. 40 sec. . . 
 
 9th stimulation 4 • • ' • 
 
 The great decrease in the amount of saliva obtainable on a 
 second stimulation, closely following a first, even though a min- 
 ute's intreval of rest elapse, might be explained on Heidenhain's 
 theory, by assuming an exhaustion of secretory fibres, nerve 
 ends or gland cells. Such an assumption is highly improbable. 
 There is, I believe, no other example of a nerve end, or fibre, 
 becoming exhausted by a weak stimulus of a minute's duration. 
 That the secretory fibres of the chorda, their nerve ends and 
 the gland cells are not exhausted or suffocated is shown by the 
 fact that the following chorda stimulation is little, if at all, al- 
 tered. The phenomena are clearly explicable, on the other 
 hand, if the Sympathetic causes secretion by compression of the 
 ducts and alveoli. By the first stimulation the gland is largely 
 emptied of its saliva. If no time be given for the ducts to be 
 refilled, the following stimulation finds less available saliva, or 
 none at all. The nerve appears, in fact, to have become inoper- 
 ative until, through the resting of the gland, or the action of the 
 chorda, the ducts be again filled. The exhausted element of 
 the gland inferred by Heidenhain is the fluid in the ducts and 
 alveoli. 
 
SECRETION PHYSIOLOGY. 311 
 
 c. The Augmentation of Sympathetic Saliva. 
 
 That the small amount of sympathetic secretion, in the cases 
 just cited, is due to the presence of a small amount of fluid in 
 the ducts and alveoli is indicated by the abnormally large sym- 
 pathetic secretion, when the amount of liquid sahva in the gland 
 is rendered abnormally large by stimulation of the chorda, or by 
 the action of pilocarpine, nicotine and other drugs. 
 
 Langley^^ first observed the augmentation of sympathetic 
 saliva by an immediately preceeding stimulation of the dilator 
 nerve in the dog's parotid and submaxillary and the cat's sub- 
 maxillary. The following experiments confirming Langley illus- 
 trates this augmentation. 
 
 Experiment VII. 
 
 Dog under morphine and chloroform, sympathetic and chorda 
 cut. Canula in Wharton's duct. Secretion in mm. is given 
 above the line for every lo seconds, 250 mm. =0.82 cc. 
 Below the line is indicated the nerve stimulated ; s, is the sym- 
 pathetic ; c, the chorda. If no letter is written, it indicates that 
 at these intervals there was no stimulation. 
 
 10, 35. 3i> 2, 25, 4, 3, 2, 2, 8, 6, 4, 17, 2, 2, I, 2,, I, 3, o 
 
 c sssss ssssss 
 
 3, 4, i>^, 1^, I, 2, I, 4, 4, i> 1= I. I. !> 2.j[/^> 3, I5 I, o, 5, I, 1,0 
 
 sssss ssss sss sss 
 
 M!..°^.M^./^^.^^.^°A.3?A.^:.^^*:^^3> I. 3. 3> 2, 2, 3, 2, 4, ^^ 
 
 s cc ssss sss 
 2, I, I, 2, I, 45, 30, 3, 25, 2, 4, 2, 3, 2, 3, I, I, I, 2, 
 
 sss c ssss ssss c. c. c. 
 
 %7; 3; 3;..2, >^, 4,2. 2:,J^, I, ^.^!^.4> 0>, O;. 
 
 ssssss sss Chorda i minute ssss 
 
 4, 3, 2, 2, I, I, 3, etc., 
 
 sssss 
 
 It will be noticed, in this experiment, that the first secretion of 
 the sympathetic, immediately following the chorda stimulation, 
 is abnormally large, but that the augmentation effect rapidly 
 passes off. The augmented saliva, as Langley pointed out, is 
 
312 MATHEWS. 
 
 more watery than normal and has a shorter latent period. It 
 resembles chorda saliva. A similar watery and copious sym- 
 pathetic saliva occurs after the injection of nicotine,^* or pilo- 
 carpine,^" and during paralytic secretion.^^ 
 
 This augmented saliva may be explained, assuming that the 
 nerve acts on the gland cell, as follows : If the chorda and sym- 
 pathetic act as the same gland cells (Heidenhain) it may be said 
 that stimulation of the chorda renders the cells more responsive 
 to a sympathetic stimulation immediately following. If, on the 
 other hand, the chorda and sympathetic innervate different gland 
 cells (Langley), we are forced to the assumption that nerve im- 
 pulses traverse glands outside of the nerve tracts. " When 
 either nerve is stimulated," Langley says, "there is an irradia- 
 tion of impulses of less intensily to the cells in the neighborhood 
 of those directly affected ; that on stimulation of the chorda 
 tympani the cells connected with it are left for a time in a state 
 of weak excitation, so that irridiation of impulses reaching the 
 gland by the sympathetic is much greater than normal, and these 
 irradiating impulses being weak lead to a more fluid secre- 
 tion."^^ It can hardly be said, I think, that either of these ex- 
 planations is satisfactory. That irritability of the gland cells 
 probably has nothing to do with this augmentation, but that it 
 is the simple result of the presence of an abnormally large 
 amount of fluid saliva in the gland is shown by the injection of 
 innocuous fluid into Wharton's duct. By this means we pass- 
 ively distend the ducts and aveoli, without the intervention of cell 
 activities. Following stimulation, of the sympathetic causes an 
 augmented secretion. I have tried such experiments only in the 
 case of the dog's submaxillary, a somewhat unsatisfactory gland, 
 owing to the viscidity of the saliva. The experiment, particu- 
 larly if tried on a fresh gland full of viscous saliva, is not always 
 successful. The cause of the failures has not been investigated, 
 but I suppose they are due to the unavoidable driving into the 
 gland of the viscous saliva and partly to the use of too great 
 pressure in such cases, causing an over-distension of the ducts 
 and a consequent injury to the nerves. The positive results are, 
 however, sufficiently conclusive. 
 
SECRETION PHYSIOLOGY. 313 
 
 Experiment VIII. 
 
 Small dog under morphine and chloroform. Left submaxil- 
 lary duct and nerves prepared. Nerves cut. The chorda is 
 first stimulated intermittently for an hour. The sympathetic is 
 stimulated each time for 30 seconds. Secondary coil 70. Se- 
 cretion in mm. as before. 
 
 Time. Nerve Secretion. 
 
 h. m. s. 
 3 30 Sympathetic 10 
 
 3 32 " 4 
 
 Inject '3CC. 0.6% NaCl solution into Wharton's duct. 
 3 34 Sympathetic 15 
 
 3 36 
 
 o 
 
 o 
 
 II 
 
 3 41 
 
 4 10 
 
 4 II 
 
 4 12 30 
 
 Inject y^ cc. 0.5% NaCl into duct. 
 
 4 14 Sympathetic 
 
 4 15 
 
 Experiment IX. 
 
 Conditions of experiment as in 8. Dog larger. Sympathetic 
 30 seconds stimulation, unless otherwise indicated. 
 
 Secretion in mm. 
 
 40 
 
 15 
 20 
 10 
 20 
 18 
 
 40 
 
 7 
 o 
 
 n 
 
 2 
 
 o 
 
 11 
 
 Time. 
 
 Nerve. 
 
 1. m. s 
 
 
 5 20 
 
 Sympathetic 
 
 5 22 
 
 " 
 
 5 24 
 
 -5 25 
 
 5 26 
 
 " 
 
 5 27 
 
 -5 27 40 
 
 5 28 
 
 -5 28 40 " 
 
 Inject 
 
 .4 cc. 0.6% NaCl into duct 
 
 5 30 
 
 Sympathetic 
 
 5 31 
 
 " 
 
 5 32 
 
 " 
 
 Inject 
 
 .3 cc. 0.6% NaCl. 
 
 5 34 
 
 Sympathetic 
 
 5 35 
 
 (C 
 
 5 36 
 
 " 
 
 Inject 
 
 3 cc. 0.6% NaCl. 
 
 538 
 
 Sympathetic 
 
314 MATHEWS. 
 
 The results of these experiments, in conjunction with those 
 following, are most readily explicable, I believe, on the muscu- 
 lar theory. The augmented saliva, in whatever manner pro- 
 duced, gives fairly conclusive evidence that the nerve causes 
 secretion by compression of the ducts and alveoli. If these are 
 filled with an unusually large amount of fluid saliva an unusu- 
 ally large secretion, characterized by its short latent period and 
 watery character, is secreted. If there be little saHva present, 
 or if it be very viscous, we obtain a small secretion of long latent 
 period and lasting for some time. 
 
 {d) Paralysis of the Sympathetic by Emptying the Ducts 
 
 AND ITS ReSTORAL TO PoWER BY INJECTION OF 
 
 Fluid into the Ducts. 
 
 Further strong evidence of the muscular action of the sympa- 
 thetic may be obtained by preventing the passage of fluid into the 
 gland and stimulating the nerve until all available saliva in the 
 ducts has presumably been expelled. The nerve then appears 
 to have lost its action, but it may be shown to be still active by 
 the injection of fluid into the ducts. The passage of fluid into 
 the gland may be prevented either by the use of quinine or by 
 compression of the gland artery. 
 
 Heidenhain * showed that if quinine sulphate be injected into 
 Wharton's duct the secretory action of the chorda is ultimately 
 paralyzed, but the gland becomes oedematous. This indicates 
 that, although liquid is present in the lymph spaces, it is pre- 
 vented in some way from passing through the cell. If, after 
 paralysis of the chorda, the sympathetic be stimulated, a copious 
 secretion is obtained. After a few stimulations, however, the 
 nerve appears to be paralyzed. If that paralysis is only apparent, 
 due to the emptiness of the gland's ducts, we should be able to 
 obtain a secretion on sympathetic stimulation, by the injection 
 Into the duct of more quinine sulphate. The following experi- 
 ment proves this to be the case. 
 
 * Heidenhain, Studien aus Breslau, IV, 1868. 
 
SECRETION PHYSIOLOGY. 315 
 
 Experiment X. 
 
 Large dog. Operation as in other experiments. Secretion 
 in mm. 250 mm. = 0.82 cc, s=sympathetic ; c=chorda. 
 
 Time. Nerve. Coil in cm. Secretion in mm. 
 
 h. m. s. 
 
 12 24 s . 15 72 
 
 12 25 s 15 12 
 
 Chorda stimulated for several minutes, then .5 cc. of saturated solution of qui- 
 nine sulphate injected slowly into Wharton's duct. 
 
 12 37 c 13 o 
 
 12 38 C II o 
 
 12 39 s II 50 
 
 12 40 ...... s II 27 
 
 12 44 s • ... 15 12 
 
 12 45 s 15 9 
 
 12 47 c II o 
 
 12 48 c 7 o 
 
 12 50 s 14 15 
 
 12 54 c . . II o 
 
 12 55 s 14 7 
 
 12 57 s 13 10 
 
 12 59 s 13 ........ 3 
 
 Inject mixture equal parts 0.6% NaCl and sat. quinine sulphate. 
 
 I 03 s . 13 24 
 
 I OS s 13 4 
 
 I 09 s 12 o 
 
 I 10 s 9 I 
 
 III S ID O 
 
 Neither nerve produces a secretion, though stimulated from time to time. 
 
 4 00 c 8 o 
 
 4 01 s 8 o 
 
 Inject 0.5 % NaCl into duct. 
 
 4 02 s 6.5 32 
 
 4 03 s 6 14 
 
 4 06 s 6 3 
 
 4 09 Inject HCl 0.5% into duct. 
 
 4 10 s 6 9 
 
 Chorda ineffective at any strength. 
 
 In the foregoing experiment the chorda became completely 
 ineffective at 12:30. The gland, however, was abnormally full 
 of quinine fluid, and the first sympathetic stimulation after the 
 
316 MATHEWS. 
 
 injection consequently gave a greatly augmented secretion at 
 12:39. Thereafter each stimulation yielded less and less, and 
 finally at 12:59 o^^X 3 "^"^- '^^'^''^ secreted. The ducts may be 
 assumed to be practically empty. Quinine solution was now 
 again injected, and the next sympathetic stimulation yielded 
 again a greatly augmented secretion. Finally at i:i i the sym- 
 pathetic failed to yield any secretion, and from then until 
 4 p. M. was totally ineffective. It would be said, at first sight, 
 that the nerve was paralyzed. Such, however, was not the 
 case, its seeming paralysis being due to the Emptiness of the 
 gland. This was shown by the injection of .5 % NaCl solution 
 into the duct. The following stimulation of the sympathetic at 
 4:02 yielded a very large secretion. 
 
 This experiment in two ways furnishes very strong evidence 
 of the muscular nature of the sympathetic secretion. The fact 
 that sympathetic secretion may be obtained long after paralysis 
 of the chorda is very suggestive. Heidenhain* maintains that 
 the chorda secretion is paralyzed by the action of the drug on 
 the gland cells. If this be true, and I see no reason to doubt 
 it, it furnishes very strong evidence that the sympathetic pro- 
 duces its secretion in sonic other manner than action on the gland 
 cell, for the sympathetic secretion is not materially affected long 
 after the gland cells have been completely paralyzed. The fact 
 that the nerve's effect soon passes away, but may be restored 
 by the simple injection of more quinine solution or other fluid 
 into the duct, I believe to be susceptible of but one explanation, 
 z. c., that the nerve causes this secretion by compression of the 
 ducts and alveoli. 
 
 A similar phenomenon is witnessed if the gland artery be 
 compressed and fluid thus cut off from the gland. A few stimu- 
 lations of the sympathetic sufiflce to render the nerve inoperative, 
 but by injection of fluid into the duct the nerve is shown to be 
 still active. 
 
 '" Heidenhain, Studien aus Breslau, IV, i868, p. 85, "so wird die Erregbarkeit 
 der absondernden Elemente bald herabgesetzt und nach kurzer Zeit ganz vernichtet. ' ' 
 
SECRETION PHYSIOLOGY. 317 
 
 Experiment Va (Continued; see p. 305). 
 
 Time. Nerve. Secretion in mm. 
 
 h. m. s. 
 
 3 25 Artery clamped close by the hilus. 
 
 3 30 Chorda . . O 
 
 3 35 SjTiipathetic 23 
 
 3 37 Sympathetic o 
 
 3 40 Sympathetic o 
 
 0.2 cc, .5 % NaCl solution injected into duct. 
 3 41 Sympathetic • • 1? 
 
 3 42 Artery undamped 
 
 4 07 30 Artery clamped 
 
 4 12 Chorda o 
 
 4 13 Sympathetic 25 
 
 4 15-4 17 ... . Chorda o 
 
 4 17 30-4 18 15 . Sympathetic 14 
 
 4 20 Sympathetic ...... o 
 
 4 23 • -3 cc., .5 % NaCl injected into duct 
 
 4 24 S}Tnpathetic .... 13 
 
 4 25 Sympathetic o 
 
 4 26-4 27 .... Sympathetic o 
 
 4 28 2 cc, .5 % NaCl injected 
 
 4 29 Sympathetic 8 
 
 In this experiment the sympathetic appeared paralyzed at 
 3:40, 4:20 and 4:26, but the injection of normal salt solution 
 into the duct was followed by a secretion little less than normal, 
 on the next stimulation. In one case twenty minutes after the 
 artery had been clamped, the sympathetic was thus shown still 
 to be active. Heidenhain attributes the loss of the chorda's 
 power to the suffocation and consequent paralysis of the gland 
 cell. (See footnote, p. 308.) As already pointed out (p. 316) this 
 would, if true, show that the sympathetic produces its secretion 
 in some other way than by action on the cell. The fact that 
 the nerve's power may be restored by the injection of innocuous 
 fluid into the ducts is readily explicable on the muscular theory 
 of secretion, but, with difficulty, on the cellular theory. 
 
 I found that a similar phenomenon may, at times, be seen in 
 the cat's submaxillary, which has been paralyzed by just suffi- 
 cient atropin to prevent chorda secretion. As was first pointed 
 out by Langley, atropin paralyzes the sympathetic in the cat, 
 but more atropin is required than to paralyze the chorda. The 
 
 THfOLOGlCAl Sew,NARY 
 
318 
 
 MATHEWS. 
 
 sympathetic may appear paralyzed, wholly or in part, before it 
 actually is. In this condition gently forcing the secreted saliva 
 back into the gland restores the nerve's power. 
 
 Experiment XII. 
 
 Cat etherized. Canula in duct of left submaxillary. Both 
 chorda and cervical sympathetic cut. Both nerves active. In- 
 ject .1^ solution of atropin carefully into femoral vein until 
 chorda just paralyzed. Sympathetic stimulated 30 seconds 
 each time. 
 
 Time. 
 
 Nerve. 
 
 
 
 Secretion in cc 
 
 h. m. 
 
 s. 
 
 
 
 
 
 
 3 50 
 
 Chorda 
 
 
 
 
 0. 
 
 
 3 51 
 
 Sympathetic 
 
 
 
 
 
 
 I 
 
 3 52 
 
 " 
 
 
 
 
 
 
 I 
 
 3 53 
 
 (( 
 
 
 
 
 
 
 I 
 
 3 54 
 
 (f 
 
 
 
 
 
 
 05 
 
 3 55 
 
 " 
 
 
 
 
 
 
 05 
 
 3 56 
 
 C( 
 
 
 
 
 
 
 03 
 
 
 Blew the secretion gently back 
 
 into 
 
 gland. 
 
 
 
 3 57 
 
 Sympathetic 
 
 
 
 
 0.13 
 
 4 00 
 
 li 
 
 
 
 
 
 
 15 
 
 4 06 
 
 Inject .1 cc. atropin into femoral 
 
 vein 
 
 
 
 
 4 07 
 
 Sympathetic 
 
 
 
 
 
 
 10 
 
 4 08 
 
 " 
 
 
 
 
 
 
 10 
 
 4 09 
 
 a 
 
 
 
 
 
 
 10 
 
 4 10. 
 
 Sympathetic 
 
 Inject .2 cc. atropin 
 
 
 • • 
 
 • • • 
 
 • • 
 
 .10 
 
 4 13 
 
 Sympathetic . 
 
 
 
 . . . 
 
 
 .07 
 
 4 14 
 
 ££ 
 
 
 
 
 
 .01 
 
 4 15 
 
 ,i 
 
 
 
 
 
 .0^ 
 
 Blew saliva into gland. 
 
 
 
 
 
 
 4 16 
 
 Sympathetic . 
 
 
 . . 
 
 . . . 
 
 
 .25 
 
 4 17 
 
 a 
 
 
 
 
 
 .05 
 
 4 18 
 
 (.i 
 
 
 
 
 
 .04 
 
 4 19 
 
 iC 
 
 
 
 
 
 .02 
 
 
 Blew . I cc. saliva back into 
 
 gland. 
 
 
 
 
 4 20 
 
 Sympathetic . 
 
 
 . . 
 
 . . • 
 
 . . 
 
 .12 
 
 4 21 
 
 " 
 
 
 
 
 
 .04 
 
 4 22 
 
 << 
 
 
 
 
 
 •03 
 
 Blew . I cc. saliva back into gland. 
 
SECRETION PHYSIOLOGY. 319 
 
 4 23 Sympathetic 14 
 
 4 24 " 02 
 
 4 25 " .04 
 
 4 26 " 02 
 
 Blew . I cc. saliva back into gland. 
 
 427 Sympathetic 13 
 
 4 28 " 01 
 
 4 29 " 06 
 
 4 30 " 02 
 
 4 31 " 03 
 
 Blew . I cc. saliva back into gland. 
 
 4 32 ,.,... . Sjmpathetic 10 
 
 4 33 " 02 
 
 4 34 " 05 
 
 4 35 " -04 
 
 26 
 
 Blew back .1 cc. saliva. 
 
 437 . Sympathetic 12 
 
 4 38 " 04 
 
 4 39 " 01 
 
 4 40 , . " 04 
 
 4 41 " 03 
 
 Blew back .1 cc. saliva. 
 
 4 42 Sympathetic 09 
 
 4 43 " 03 
 
 4 44 " 04 
 
 4 45 " 02 
 
 Blew back . i cc. saliva. 
 
 4 46 Sympathetic 10 
 
 4 47 " 05 
 
 4 48 " 03 
 
 4 49 " 04 
 
 4 50 " 02 
 
 Blew back . i cc. saliva. 
 
 4 51 Sympathetic 11 
 
 4 52 " 02 
 
 4 53 " 04 
 
 4 54 ..." 025 
 
 Blew back .1 cc. 
 
 4 55 S 075 
 
 4 56 S 025 
 
 4 57 S 04 &c. 
 
 The most probable explanation of the apparent failure, partial 
 or total of the sympathetic, in all the immediately preceding 
 experiments, appears to me to be this : That by the injection of 
 
320 MATHEWS. 
 
 quinine, or atropin, or compression of the gland's artery, liquid 
 is prevented from entering the gland. A few stimulations of 
 the sympathetic suffice to expell all, or most, of the available 
 saliva in the gland, and the nerve thereafter appears paralyzed. 
 If, now, the ducts and alveoli be passively redistended by the 
 injection of liquid into the duct the nerve again causes a 
 compression of the duct, and the fluid is again expelled and gives 
 a secretion. This renewed secretion cannot, however, be re- 
 ferred to the action of the gland cell, because the latter has been 
 in one case paralyzed by the action of quinine, and in the other 
 case by suffocation. Nor could it be referred to the action of 
 the cell, even were the latter not paralyzed, for the mere hypo- 
 thetical taking-up of fluid into the cell from the duct, and its 
 discharge again into the latter, would in no way alter the bulk 
 of fluid in the ducts plus the bulk of the cell. There would, 
 hence, be no pressure to drive the secretion from the gland. 
 
 e. The Character of Sympathetic Saliva. 
 
 Evidence that the sympathetic nerve innervates the gland cell 
 has been derived from the character of the sympathetic saliva. 
 This, as is well known, is richer in organic matters than the 
 saliva secreted under the influence of the gland's dilator nerve. 
 This greater richness Heidenhain attributes to the predominance 
 in this nerve of so-called " trophic " fibres, the function of which 
 is to render the stored-up metabolic products of the cell (hylo- 
 gens) more soluble, and the juice consequently more concen- 
 trated. This assumption involves such consequences that by 
 common consent it has been considered the most unsatisfactory 
 part of the Heidenhain theory. It is, however, practically the 
 only probable explanation, with one exception, which has been 
 offered. The exception is the view suggested by Schiff, dis- 
 cussed below. 
 
 If the sympathetic simply drives out the saliva already present 
 in the gland the sympathetic saliva must be of the character of 
 that present in the ducts and alveoli at the moment of stimulation. 
 There is evidence that this is the case. That the saliva in the 
 ducts of the dog's parotid is very viscid has been shown by 
 
SECRETION PHYSIOLOGY. 321 
 
 Langley.''^ Sections show the ducts plugged with a viscous 
 looking mass, and Langley suggests that the saliva is here too 
 thick to be expelled. In one experiment Langley found a dog's 
 parotid which secreted under the influence of the sympathetic 
 1.3 cc. Concerning this saliva Langley says •}'^ 
 
 " The saliva was of the most remarkable nature ; it formed a 
 thick jelly-like mass ; if allowed to collect at all in the canula 
 it could be drawn out as a continuous clot. During the experi- 
 ment the duct was frequently emptied by pressure to prevent 
 its being stopped up." The saliva contained 7.8 % of organic 
 solids. We can, moreover, artificially alter the fluidity of the 
 saliva in the ducts, rendering it more dilute, by the action of the 
 chorda tympani or pilocarpine. In such cases, as we have seen 
 in speaking of the augmented secretion, sympathetic saliva is 
 almost as thin as chorda saliva. By long stimulation of the 
 chorda, moreover, we may exhaust the soluble constituents of 
 the gland. In such cases it may be presumed that the gland 
 saliva is thinner than normal. It is known that under such cir- 
 cumstances the sympathetic saliva may fall within the limits of 
 density of chorda saliva.* A similar change occurs in paralytic 
 secretions following division of the chorda. The gland then 
 secretes a very thin saliva, and sections show the cells practic- 
 ally exhausted of their mucous. The sympathetic in these 
 causes a very abundant and very watery secretion. 
 
 We may obtain still further evidence of the character of the 
 saliva normally present in the ducts of the resting gland by a 
 sudden, strong stimulation of the chorda tympani. The rapid 
 inflow of fluid from the capillaries about the alveoli, taking place 
 under the influence of that nerve, drives out the saliva in the 
 ducts before it has time to become diluted. If we examine this 
 saliva first appearing on chorda stimulation we find it in all re- 
 spects typical sympathetic saliva. From this Schiff concludedf 
 that sympathetic saliva was nothing more than the saliva nor- 
 mally present in the ducts, formed during glandular rest. 
 
 * Heidenhain, Studien aus Breslau, IV, 1868. After long sympathetic stimula- 
 tions the saliva becomes " diinnflussig, hell, und dadurch dem chorda Speichel 
 ganz und gar ahnlich." 
 
 t Schiff. LeQons sur la Digestion. Tome I., p. 296, 1867 ; also p. 304. 
 
322 MATHEWS. 
 
 Schiff found that if the sympathetic nerve of the horse be 
 stimulated the parotid secreted quickly 8— lo volumes of white 
 saliva, and then, as in the cat's submaxillary, secretion ceased. 
 If the horse be fed there ensued a copious, clear secretion of 
 watery cerebral saliva. The gland was now, presumably, full 
 of such saliva. If it be allowed to rest for twenty min- 
 utes without secretion on again feeding the horse the first 
 saliva (8-io volumes) zvas typical, thick, ivhitc syinpatlietic 
 saliva. This was followed by the clear cerebral saliva. Schiff 
 repeated this many times, thus showing that in the interval of rest 
 the gland, uninfluenced by the sympathetic, converts the clear 
 cerebral saliva into typical so-called sympathetic saliva. A sim- 
 ilar phenomenon has been described, with a somewhat different 
 interpretation for the dog's submaxillary, by Heidenhain.* I 
 have repeated Schiff's experiment on the dog's submaxillary, 
 fully confirming him. This is shown in the following exper- 
 iment. 
 
 Experiment XIII. 
 
 Large dog, morphine and ether. At 10:30 a. m. canula in 
 right Wharton's duct. Sympathetic and chorda-lingual cut. 
 On the first stimulation of the chorda the first saliva was viscid, 
 whitish and filled with corpuscles. The chorda Avas stimu- 
 lated until 2 cc. of saliva were secreted. This saliva was thin, 
 clear, typical chorda saliva. Gland rested without secretion 
 until 11:30. Stimulated chorda. The first saliva zvas thick, 
 viscid, white saliva. The gland then secreted i cc, clear 
 chorda saliva. Rested until 2:30 P.M. Stimulated the chorda. 
 A very large amount of typical, sympathetic saliva appeared first, 
 followed by 2 cc. of watery chorda sahva. Gland rested until 
 4 P. M. Stimulated chorda. The first saliva was viscid and 
 contained many salivary corpuscles. Secreted afterward i cc. 
 clear saliva. Rested until 5 p. m. Stimulated the chorda. 
 The first saliva zvas again viscid, zvhitish saliva, filled zvitli sali- 
 vary corpuscles and lumps. 
 
 * Heidenhain. Studien aus Breslau, IV, 1868, p. 52. "Die erste Speichel por- 
 tion war sehr dick, fast gallertartig, reich an Schleimballen wie sie sonst im Sympa- 
 thicus Speichel vorkommen, und ebenso an Speichelkorperchen die haufenweise 
 bei einander lagen." 
 
SECRETION PHYSIOLOGY. 323 
 
 This experiment proves that after each stimulation of the 
 chorda, the thin, chorda saHva filHng the gland ducts is quickly 
 converted, even in the absence of sympathetic influence, into 
 typical viscid, sympathetic saliva.* It shows, also, that the ducts 
 of the normal, resting mucous gland are filled with saliva, sup- 
 posed to be characteristic of the sympathetic's action. This 
 observation seems to me to render Heidenhain's assumption of 
 special "trophic" nerve fibres to account for the character of 
 such saliva, superfluous ; and, also, to give additional evidence 
 that sympathetic saliva is nothing more than this " saliva of 
 rest," expelled by compression of ducts and alveoli. The cor- 
 rectness of the latter view is, in my opinion, strongly confirmed 
 by the great variation in character of sympathetic saliva, with a 
 variation of character of the saliva within the gland. 
 
 I wish to point out, also, that the influence of sympathetic 
 stimulation upon the composition of the saliva secreted during 
 coincident stimulation of the dilator nerve, upon which special 
 stress has been laid by Heidenhain, is also readily understood on 
 this hypothesis of the nature of sympathetic action. Langley's 
 discovery'^^ that the sympathetic produces a secretion from the 
 dog's parotid unless the saliva be too thick for expulsion make 
 Heidenhain's results clear." 
 
 Heidenhain found, in harmony with all other observers, that 
 stimulation of the sympathetic usually causes no secretion from 
 the dog's parotid. He concluded from this that the nerve 
 carried no, or few, secretory fibres. f He discovered, however, 
 that if Jacobson's nerve be irritated so as to cause a secretion, 
 and during this irritation the sympathetic be stimulated, the 
 saliva secreted during simultaneous irritation of both nerves was 
 far richer in organic solids than that secreted under the influ- 
 ence of Jacobson's nerve alone.;}; Denying that the sympathetic 
 
 ■ * This is a pretty conclusive reply to the statement of Heidenhain that the simple 
 contact of the water with thehylogens is not sufficient to dissolve them 
 We have here a demonstration that it is sufficient in the total absence of nerve in- 
 fluence. 
 
 t Heidenhain. Hermann's Handbuch d. Phys. V, p. 55. " Der Sympathicus 
 des Hundes enthalt fiir die Parotis nur trophische, fiir die submaxillaris daneben 
 wenige secretorische Fasern." 
 
 j Heidenhain, Hermann's Handbuch d. Phys. V, p. 55. 
 
824 MATHEWS. 
 
 exerted a secretory effect upon the gland, he considered the 
 secretion to be due to Jacobson's nerve alone. He concluded, 
 therefore, that stimulation of the sympathetic enormously in- 
 creased the content of organic solids in the cerebral saliva. The 
 sympathetic must hence act on the gland cells so as to render 
 their contents far more soluble. From Langley's results, how- 
 ever, we can safely conclude that the saliva, secreted when both 
 nerves are stimulated, is not pure cerebral saliva, but largely, if 
 not wholly, augmented sympathetic saliva. Like all sympa- 
 thetic saliva, it is more concentrated than the saliva secreted 
 under the influence of the dilator nerve, because it is expelled 
 without dilution. 
 
 f. Other Evidence of the Muscular Nature of the 
 Mechanism of Sympathetic Secretion. 
 
 Very clear evidence, also, has been brought forward by Eck- 
 hard,^^ von Wittich"'' and Heidenhain^^ himself that the sympa- 
 thetic causes at least the major part of its secretion, by a com- 
 pression of the ducts and alveoli. The parotid gland of the 
 sheep is an albuminous gland, capable of secreting against a 
 pressure of 400-500 m. m. of water (Eckhard). If while secret- 
 ing against a somewhat lower pressure (200—300 mm.) the 
 cervical sympathetic be stimulated, the water rises suddenly in 
 the manometer for some distance (30-100 mm.). On ceasing 
 stimulation the secretion riislies hack at once into the gland nearly, 
 tho?igh never quite, to its former level. The higher the pressure 
 the more sudden the flow backward. The quick rise at the 
 beginning of stimulation and the abrupt back flow of the secretion 
 at the end plainly suggest that the nerve caused compression 
 of the ducts and alveoli, and thus pressed out the secretion. 
 On ceasing stimulation these structures dilated, and the secre- 
 tion, being under pressure, rushed back into the gland. I see 
 no other explanation for the back flow, as it takes place too 
 suddenly and at too low a pressure (200 mm. water) to be due 
 to back filtration. 
 
 Heidenhain's observation is less striking, but it is similar to 
 
SECRETION PHYSIOLOGY. 325 
 
 the above. (Breslau Studien, p. 69, IV.) In taking the secre- 
 tory pressure of the dog's submaxillary he stimulated the 
 chorda until the pressure in the ducts was 271 mm. Hg. On 
 ceasing stimulation the manometer gradually fell. On stinnila- 
 tiiig the syTiipatJictic the sinking became unicJi slcnver, and the ma- 
 nometer remained stationary at 160 mm. On breaking the 
 stimulation the manometer sank gradually to 100. On stimu- 
 lating the sympathetic it rose to 107, and on chorda stimulation 
 to 271. It gradually fell during following sympathetic stimula- 
 tion, but on bi^eaking the stinmlation it fell with striking rapidity 
 (Auftalig beschleunigtes Sinken). Heidenhain thus records for 
 the dog's submaxillary the same sudden back flow on breaking 
 the stimulation of the sympathetic as Eckhard and von Wittich 
 describe in the sheep. 
 
 Paradoxical though it may seem, the experiments just quoted 
 of von Wittich and Eckhard have been cited by Heidenhain as 
 conclusive evidence that the sympathetic does not simply drive 
 out the secretion already in the gland. And it is this con- 
 viction which led Heidenhain, in the discussion of all experi- 
 ments involving the sympathetic, to ignore the possibility of its 
 having such an action. Heidenhain believed von Wittich was 
 right in contending that the failure of the manometer to return 
 to its former level on breaking stimulation proved that the 
 amount of saliva in the gland had been increased. It will be 
 instructive to consider von Wittich's explanation of the phe- 
 nomena of this secretion, von Wittich'''' suggests that the back 
 flow of the saliva is due to the saliva being pushed back into the 
 cells. Let us examine this more closely. von Wittich and 
 Heidenhain assumed that the cells, on stimulation, discharge 
 their stored products into the lumen. Such a process, it need 
 hardly be said, would lead to no secretion from the ducts, as 
 the bulk of the cell would diminish to just the extent that the 
 bulk of fluid in the ducts increases. Hence the bulk of cell 
 plus liquid would remain unaltered. We must, therefore, make 
 either one of two farther assumptions : First, that the alveoli are 
 greatly distended owing to the turgor of the cells. Stimulation 
 of the nerve might conceivably diminish the resisting power of 
 
 Annals N. Y. Acad. Sci., XI, September 13, 1898 — 22. 
 
326 MATHEWS. 
 
 the inner end of the cell, and the secretion be expelled from the 
 cell by intra-cellular tension, and from the ducts by the elastic 
 tension of the distended alveolar wall. Or, second, it must be 
 assumed that, as the fluid flows from the cell, new fluid enters 
 the cell from the rear, so that the cell does not diminish in bulk 
 to an extent aqual to the bulk of secretion it has lost. Either 
 of these assumptions lands us at once in difficulties. If the first 
 be true we cannot understand why the sympathetic secretion 
 should be abnormally large, just in those cases, such as par- 
 alytic secretions, or after long-continued chorda secretion, in 
 which the alveoli are not distended and are not presumably 
 under pressure. The second assumption, besides being wholly 
 imaginary, has to explain whence comes the fluid flowing into 
 the cell, and why it should flow in during sympathetic stimu- 
 lation at a time when there is a pronounced vaso-constriction. 
 With this difficulty of understanding how the nerve could cause 
 a secretion by action on the cell, let us see how the sudden back 
 flow could be understood. According to von Wittich and 
 Heidenhain the diameter of the alveoli has remained constant. 
 The secretion, manifestly, cannot upon this assumption return 
 into the gland, unless there be a diminution in the combined 
 bulk of the secretion in the ducts and the cells. There will be 
 no such alteration in bulk, however, by the secretion passing into 
 the cell as von Wittich assumes, for the cell will grow to just 
 the amount that the secretion in the lumen diminishes. The 
 only way a diminution in bulk could be brought about is by a 
 back filtration. The fall is, however, much too sudden for this, 
 and takes place at a pressure much less than the gland can sus- 
 tain without becoming oedematous. It is also impossible to see 
 why on ceasing stimulation the permeability of the gland to back 
 filtration should suddenly increase. Easy though it seems at 
 first sight, therefore, to ascribe such a back flow to a reabsorp- 
 tion under pressure of saliva by the cell, closer inquiry shows 
 that it is impossible to account for this back flow except on the 
 assumption either of a back filtration or that there has been an 
 alteration in the diameter of the alveoli. I maintain with Eck- 
 hard that a back filtration is highly improbable, and there re- 
 
SECRETION PHYSIOLOGl. 327 
 
 mains only the alternative of an increase in the diameter of the 
 alveoli, probably following an active compression. 
 
 But if the saliva is simply pressed out, why is it that it does 
 not return to its former level on ceasing stimulation ? This was 
 supposed by von Wittich to prove that the nerve increased the 
 amount of saliva in the gland. I fully agree with von Wittich 
 in this contention, but I disagree with him entirely in 
 referring- the increase to the action of the nerve on the cell 
 This increase maybe readily understood on the muscular theory, 
 without any assumption of nerve activity on the gland cell, as 
 follows : On breaking sympathetic stimulation of considerable 
 duration a temporary vaso-dilation occurs and the ducts and 
 alveoli relax. It takes an appreciable time for the saliva to pass 
 back into the fine tubules, and during this time the cells are ab- 
 sorbing water from the lymph and capillaries. Hence their 
 bulk and the amount of saliva is increased and the saliva is 
 never able to return to its former level. The proof of this is 
 sufficiently clear. That vaso-dilation does occur temporarily on 
 ceasing stimulation of constrictor nerves has often been re- 
 marked. I have myself often seen it in" the rabbit's ear and in 
 the cat's submaxillary. In the dog's submaxillary I have often 
 seen, also, that coincident with this vaso-dilation a slight secre- 
 tion may actually ensue (See Expt. VII, p. 3 1 1). It is, also, well 
 established that the cells do imbibe fluid and food during or after 
 sympathetic stimulation and thus increase the bulk of undifferen- 
 tiated protoplasm. 
 
 In view of these facts, I believe that von Wittich's and Eck- 
 hard's experiments, instead of proving that sympathetic stimu- 
 tion can not possibly be due to compression of the ducts and 
 alveoli, demonstrates that it must be due to such compression ; 
 that it is impossible to account for the back flow on any other 
 probable hypothesis, and that the fact that the saliva does not 
 reach its former level is readily understood by reference to the 
 nerve's constrictor action and the temporary vaso-dilation ensuing 
 on breaking simulation. I do not believe that von Wittich ever 
 endeavored to analyze in detail his own explanation, or he must 
 have perceived its impossibility. 
 
328 MATHEWS. 
 
 g. The Location and Nature of the Contractile Substance 
 
 IN the Gland. 
 
 The contractile tissue, responsible for the sympathetic secre- 
 tion, resides neither in the gland capsule nor in the capillaries. 
 Glands dissected free from the capsules secrete normally. The 
 capillaries cannot be held responsible, as Vierheller ''^ supposed, 
 because, as one may readily see in the cat's submaxillary, the 
 nerve may be still active on the blood vessels while producing 
 no secretion, and von Wittich''^ records that after curare, the 
 rabbit's sympathetic loses its secretory activity while still active 
 on the blood vessels of the ear. Unna^° has suggested that the 
 basement membrane is contractile, and this may possibly be the 
 case. There is, however, no evidence of it. That there is 
 smooth muscle about some of the principal ducts of the salivary 
 glands is well-known, but most histologists have failed to find 
 any between or about the alveoli. However, Pfliiger''" and 
 Schluter'''' have each described isolated fibres, and strands of 
 smooth muscle lying between the alveoli, distinct from the 
 blood vessels, "so that the stroma is not entirely lacking in 
 contractility." 
 
 Whether the contractile tissues thus far recognized histo- 
 logically in the gland are those active in the production of this 
 secretion appears to be doubtful. The physiological evidence 
 is of itself so strong, however, that I believe we can safely as- 
 sume the existence of such a tissue, even had we no histolog- 
 ical evidence of its presence. 
 
 Ji. The Changes in Gland Cells upon Sympathetic Stimu- 
 lation. 
 
 The changes in gland cells, induced by stimulation of the 
 sympathetic nerve, are most clearly seen in the rabbit's parotid,**^ 
 less clearly in the dog's parotid, where the nerve causes normally 
 little or no secretion. The changes consist in the diminution in 
 the size of the cell, the discharge of the mucous or secretory 
 products, the formation of new undifferentiated protoplasm and 
 
SECRETION PHYSIOLOGY. 329 
 
 in the nucleus becoming round and moving toward the center 
 of the cell. These changes are identical in kind with, though 
 taking place generally more slowly than, those following stimu- 
 lation of the dilator nerve or the injection of pilocarpine. Do 
 they indicate the direct action of the nerve on the cell ? Al- 
 though they might be so interpreted, they may be readily under- 
 stood without any such assumption, as follows : Stimulation of 
 the nerve causes a compression of the cells and thus expels from 
 them their stored-up metabolic products and liquid. By this 
 means the cells discharge their products. On ceasing stimula- 
 tion the alveoli and ducts relax, and the cells take up water and 
 food from the lymph. The latter process is hastened probably 
 by a temporary vaso-dilation ensuing when the sympathetic 
 stimulation is broken. In virtue of the food, oxygen and lymph 
 thus brought to them the cells form new undifferentiated proto- 
 plasm. On several successive stimulations the accumulated 
 metabolic products are largely discharged, the cells become 
 smaller and the nuclei, relieved from pressure, become round 
 and move toward the center of the cells. The same explanation 
 holds also for the changes following stimulation of the dilator 
 secretory nerve, with the exception that the stored products are 
 dissolved out of the cell, instead of being squeezed out, and as 
 vaso-dilation accompanies this secretion the changes take place 
 at a more rapid rate. These changes are discussed more at 
 length in my paper on the Pancreas Cell.* 
 
 i. Summary and Conclusion. 
 
 The phenomena of sympathetic secretion, which have been con- 
 sidered, could hardly indicate more clearly, I think, the muscular 
 mechanism of that secretion. The sudden gush of saliva; its 
 sudden cessation, however prolonged the stimulation ; the dim- 
 inution in the amount of saliva secreted when the stimulations 
 are rapidly repeated ; the apparent paralysis of the nerve when 
 the ducts are empty and its restoral to power if the ducts be 
 passively redistended ; the augmentation in volume of the secre- 
 tion, when the ducts are abnormally full of fluid saliva, and the 
 
 * Shortly to appear in the Journal of Morphology. 
 
330 MATHEWS. 
 
 diminution in amount of secretion when there is little saliva 
 present ; the dependence of the character of the sympathetic 
 saliva upon that present in the gland at the moment of stimu- 
 lation ; the back flow of saliva into the gland on stopping 
 stimulation when the gland is secreting against pressure ; the 
 presence of smooth muscle in the ducts and between the alveoli — 
 these facts point unmistakably in one direction. A stronger 
 chain of circumstantial and direct evidence that this secretion is 
 caused by compression of the ducts and alveoli by contractile 
 tissue would be hard to imagine. If some of these phenomena 
 are susceptible of explanation upon the hypothesis that the 
 secretion is due to gland cell activity, others of them, /. c, the 
 augmented salivary secretion, the back flow of saliva on break- 
 ing stimulation, the paralysis of the nerve when the ducts are 
 empty, and its restoral to power if the ducts be redistended, are 
 explicable, if at all, by that theory, only by means of improba- 
 ble and unproven assumptions. 
 
 The surprisingly ready acceptance of the Ludwig-Heidenhain 
 theory of secretory nerves, acting on gland cells, as an explana- 
 tion of the sympathetic salivary secretion in the face of unmis- 
 takable indications of a muscular mechanism, has been due, 
 largely, I believe, to the generally prevalent belief that there is 
 but one mechanism of secretion. That this belief is erroneous, 
 there has long been, I believe, many indications. For there is 
 direct evidence in many glands, such as the poison glands of 
 snakes, the skin glands of amphibia, many unicellular glands, 
 sebaceous and sweat glands, that many secretions are due to 
 muscular action. And in many other glands the phenomena of 
 secretion have shown as clearly that here the mechanism was 
 some other than muscular. There must evidently be at least 
 two different mechanisms, a muscular and some other one. 
 Once the idea that there is but one mechanism of secretion is 
 abandoned, the salivary secretions will be found, I believe, to 
 lose much of their puzzling character. 
 
 The facts which Heidenhain urges as showing that the sym- 
 pathetic produces secretion by action on the gland cell are 
 readily accounted for if the sympathetic cause compression of 
 the ducts and alveoli and vaso-constriction. 
 
SECRETION PHYSIOLOGY. 331 
 
 III. OTHER SECRETIONS DUE TO MUSCLE 
 ACTION. 
 
 Probably many other secretions are due to muscle action. ' 
 
 The unicellular glands of the carp-louse, Argulus foliaceus, 
 are surrounded by muscle fibres. Nussbaum,'" observing the liv- 
 ing glands, states that they are emptied by the contraction of this 
 musculature. Muscle surrounds the unicellular glands in the 
 mantel of Aplysia,'^ and the glandular pedicellaria of the Echino- 
 derms.'^^ The gasteropod liver* possesses, beneath the serosa, 
 an incomplete musculature, the contraction of which has been 
 watched in the living gland. A similar sheath is found in the 
 livers of Crustacea, land and water Isopods, Amphipods and 
 Decapods."'* 
 
 The poison glands of spiders have their alveoli enclosed in a 
 tunic of spirally arranged muscular fibres.^* In the salivary 
 glands of Cephalopods''^ the cells rest on connective tissue, which 
 is, in turn, surrounded by muscle fibres. An examination of the 
 physiology of these glands leaves little doubt that the secretion is 
 due to muscular action.^* The amphibian skin glands are sur- 
 rounded by a muscular sheath lying between the cells and the 
 basement membrane. There is no doubt from observations on 
 the living glands (Engelmann,**^ Drasch," Ranvier''^) that this 
 muscle at times contracts, compresses the gland and thus causes 
 a secretion. A similar muscular mechanism prevails in the 
 mucous glands of Petromyzon, in which the cells are bodily 
 extruded. 
 
 The poison glands of amphibia and reptilia and others of the 
 salivary glands'^" are provided with their own musculature, or are 
 emptied by surrounding skeletal muscles. Many anal and 
 cloacal glands,*' sweat^^ and sebaceous glands are provided with 
 a musculature lying between the basement membrane and the 
 cells. There is little doubt that the secretion of sebum is pro- 
 duced by the action of this muscle. The same can be said for 
 the secretion of the oil gland of birds. Probably the most in- 
 teresting secretion due to muscular action, outside of the sali- 
 vary glands, is found in the mammalian sweat glands. From 
 
332 MATHEWS. 
 
 the observations of Ranvier,*"^ Joseph"^ and others certain secre- 
 tions of sweat are probably due to the compression of the gland 
 by this muscle. Probably the post-mortem sweat secretions, 
 secretion after closing the artery, or the injection of strychnia 
 are due to this cause. (There is, however, a second sweat 
 mechanism associated with vaso-dilation.) 
 
 Many more examples of the muscular mechanism of secretion 
 might be given, but these suffice to indicate the very wide dis- 
 tribution of such a mechanism. Muscular mechanisms are, pos- 
 sibly, more common among the invertebrates, but they play, 
 also, a not inconsiderable part in vertebrate secretions. The 
 vertebrate, however, with its delicately coordinated, closed vas- 
 cular system, develops a second mechanism, that of osmosis, 
 which we will now consider. 
 
 IV. SALIVARY SECRETION ENSUING UPON STIM- 
 ULATION OF THE VASO-DILATOR NERVE. 
 
 That the general features of chorda secretion coincide with 
 the phenomena of osmosis, regulated by the nerve's dilator action, 
 is pointed out briefly on p. 356. I wish here to consider more 
 particularly those facts which have hitherto been irreconcilable 
 with such a theory, and have been generally considered evidence 
 of a special action of the nerve on the gland cell. These facts 
 are the most important evidences of a secretory nerves and so 
 warrant a careful consideration. They are : {a) the increase in 
 the percentage of organic solids of a secretion coincident with an 
 increased rate of secretion ; (1!^) the action of atropine ; {c) the 
 chorda-secretion after clamping the artery ; {d^ the action of 
 nicotine. 
 
 a. The Increase in the Percentage of Organic Constitu- 
 ents COINCIDENT with AN INCREASED RaTE OF SeCRETION. 
 
 Heidenhain * observed that on passing from a weak to a strong 
 stimulation of the dilator nerve in the fresh submaxillary and 
 
 * Heidenhain. Hermann's Handbuch der Physiologie V. p. 50. Studien aus 
 Breslan IV, 1868, p. 32. 
 
SECRETION PHYSIOLOGY. 
 
 333 
 
 parotid gland of the dog, not only was the rate of secretion in- 
 creased, but also the percentage of solids. He obtained a simi- 
 
 No. of 
 
 
 
 
 ^§ 
 
 Rate of 
 
 
 
 •~ 1) 
 
 Stimula- 
 
 
 Time. 
 
 Coil. 
 
 f 01 
 
 a s 
 
 <; D 
 
 Secretion 
 
 Solids. 
 
 Salts. 
 
 5 B 
 
 tion. 
 
 
 
 
 in I min. 
 
 
 
 6g. 
 
 
 h. 
 
 m. m. 
 
 
 
 
 
 
 
 I 
 
 9 
 
 20-45 
 
 315—288 
 
 3-5 
 
 0.14 
 
 0.74 
 
 0.22 
 
 0.52 
 
 2 
 
 9 
 
 47-51 
 
 160—130 
 
 3-5 
 
 .87 
 
 2.10 
 
 .56 
 
 1^54 
 
 3 
 
 lO 
 
 54-5-59 
 
 100 — 60 
 
 30 
 
 .66 
 
 2.08 
 
 •45 
 
 1.63 
 
 4 
 
 lO 
 
 19-40 
 
 264—245 
 
 2.8 
 
 .11 
 
 1.44 
 
 •36 
 
 1.07 
 
 5 
 
 lO 
 
 45-48 
 
 160 — 130 
 
 3-0 
 
 1. 00 
 
 I.41 
 
 •49 
 
 0.91 
 
 6 
 
 lO 
 
 50-56 
 
 80— 65 
 
 3.0 
 
 •50 
 
 1. 16 
 
 •39 
 
 0.76 
 
 7 
 
 II 
 
 9-27 
 
 270 — 250 
 
 2.5 
 
 •13 
 
 0.78 
 
 • 30 
 
 0.48 
 
 8 
 
 II 
 
 30-34 
 
 150 — 120 
 
 3-1 
 
 •77 
 
 0.90 
 
 •38 
 
 0.51 
 
 9 
 
 II 
 
 35-44 
 
 80 — 30 
 
 2.8 
 
 31 
 
 0.79 
 
 •3b 
 
 0.42 
 
 lar result in the dog's pancreas, Gottlieb^^ in the rabbit's pan- 
 creas, and Pawlow and Schumowa-Simanowskaja''^ in the dog's 
 stomach. In the sheep's submaxillary, on the other hand, there 
 was little or no increase in the per cent, of solids on increasing the 
 stimulus. 
 
 Heidenhain believed that this increase in solids meant that the 
 cerebal nerve, besides quickening the flow of water through the 
 cells, rendered the cell contents more soluble. How otherwise 
 shall we explain the fact, he asks, that although given a shorter 
 time of contact with these solids, the water passing through the 
 cells, nevertheless dissolves more than during slow secretion. 
 " Die blosse Beriihrung mit der aus dem Blute ausgeschiedenen 
 Fliissigkeit ist zur Uberfuhrung des Schleimes in das Secret 
 nicht ausreichend, denn sonst musste das Secret um so reicher 
 daran sein, je langer die Fliissigkeit in den Driisenraumen ver- 
 weilt, d. h. je langsamer die Secretion vor sich geht.""' He 
 further assumes that the trophic fibers require a stronger stimu- 
 lus than the secretory. " Das cerebrale Secret wird, so lange 
 die Driise unermiidet ist, bei Reizverstarkung reicher an or- 
 ganischen Bestandtheilen, weil der Umsatz der organischen Sub- 
 stanzen in den Zellen unter den Einflusse der starker gereizten 
 trophischen Fasern schneller steigt, als der Wasserstrom unter 
 dem Einflusse der starker gereizten secretorischen Fasern."""* 
 
334 MATHEWS. ' 
 
 There are two possible fallacies in Heidenhain's argument. 
 One fallacy probably lies in his tacit assumption that the gland 
 secretes as a whole ; that the secretion following a strong stimu- 
 lus is derived from the same alveoli as the secretion following a 
 weak stimulus. The other fallacy is the assumption that all of 
 the organic constituents of saliva secreted from a fresh gland 
 upon a strong stimulus are in solution. The true reason why 
 the dilator-secretory nerve may cause an increase in the organic 
 matter present in a secretion, coincident with an increased rate 
 of flow, in passing from a weak to a strong stimulus, may be 
 the following : 
 
 If a very weak stimulus be used, only a portion of the alveoli 
 are aroused to activity. The supply of stored up products 
 (hylogens) in these, becomes soon exhausted and the secre- 
 tion derived from them is poor in organic constituents. On 
 passing to a strong stimulus, the previously resting alveoli are 
 thrown into activity and the secretion derived from them is rich 
 in organic constituents. It is the secretion from these fresh 
 alveoli, which increases the percentage of organic constituents 
 in the whole secretion. On passing from a long continued 
 weak to a strong stimulus in a fresh gland, one is really pass- 
 ing from an exhausted to a fresh portion of the gland. 
 
 Moreover, in Heidenhain's observation there is a second 
 source of error which he has overlooked. Heidenhain treats all 
 of the organic constituents of the rapidly secreted saliva as if 
 they were in solution and considers that the liquid derived from 
 the blood is in contact with the materials to be dissolved, only 
 during the time of its passage through the cell. There .can be 
 little question, however, that saliva, and particularly the rapidly 
 secreted saliva of a fresh gland, cannot be considered a true 
 solution, for it contains many bodies in suspension. Heidenhain 
 himself has been one of those to describe the microscopical 
 appearance of the lumps of mucous matter, salivary corpuscles 
 and occasional leucocytes found in this secretion. The presence 
 of these bodies in saliva indicates that the rapidly secreted saliva 
 carries out of the cell not only substances in solution, but vis- 
 cous masses of mucous matter not in solution. Its swift cur- 
 
SECRETION PHYSIOLOGY. 335 
 
 rent is able to transport these masses, while a more slowly 
 flowing secretion is not. Furthermore, in all probability the 
 saliva keeps on dissolving them as it carries them along and 
 hence becomes actually more concentrated, because it is in con- 
 tact with them really for a longer time than the more slowly 
 secreted saliva and not for a shorter time as Heidenhain thought. 
 Heidenhain made no endeavor to distinguish between the mat- 
 ters in suspension and those in solution. 
 
 That any gland functions as a whole, as Heidenhain tacitly 
 assumes in his explanation, can not be maintained. 
 
 The whole surface of the stomach, for instance, may be con- 
 sidered as one large gland. It has long been known that se- 
 cretion can ensue in one spot, and not in another. Heidenhain 
 himself, has called special attention to the marked differences in 
 the condition of the various alveoli in the salivary glands. Even 
 in the resting gland, here and there alveoli will be found posses- 
 sing the structural features of secretory activity." In the stomach 
 he remarks that some glands show changes on stimulation before 
 others,^^ and I have, myself, repeatedly observed glands in the 
 Newt's stomach close together in very different stages of activity. 
 Kljhne and Lea^^ have observed this in the living rabbit's 
 pancreas, a portion only of the gland being normally active. 
 After pilocarpine all the alveoli passed into a condition of activity. 
 In the kidney the independence of the various tubules in se- 
 cretion has been remarked for the bird's kidney by von Wittich, 
 and for the mammalian kidney by Ribbert,-* and by Dr. Herter 
 in conjunction with the author. Finally, in the case of the sali- 
 vary glands, Langley says that even on prolonged activity of the 
 chorda many alveoli show no change. " This is due, in some 
 cases, to fibres escaping stimulation, fibres which leave the 
 lingual later than usual." This histological evidence appears 
 to me to be conclusive with reference to the idea that the gland 
 does not function as a whole, but that the individual alveoli in 
 the secreting gland may be here active, there passive. 
 
 The physiological evidence that the foregoing is the true ex- 
 planation of Heidenhain's observation is hardly less conclusive. 
 We can easily obtain evidence that the secretion obtained 
 
336 MATHEWS. 
 
 during a weak stimulus is derived from a portion of the gland 
 only in the following manner : Let us stimulate the chorda nerve 
 carefully with a very weak current, until a large amount of se- 
 cretion has been obtained. If this secretion has been derived 
 from the whole gland a stronger stimulus should yield a se- 
 cretion much less concentrated than a stimulus of equal strength 
 before the weak stimulus. The glands should show, in other 
 words, a considerable exhaustion of the gland products. If, on 
 the contrary, the whole of this secretion has been derived from 
 a portion only of the gland the rest of the alveoli must remain 
 practically unaltered, and a stronger stimulus arousing these 
 should yield a juice, little, if any, poorer in organic matters than 
 was yielded by a stronger stimulus before the weak. 
 
 Werther''^ has unintentionally tried this experiment and found 
 the latter possibility to be what actually occurs. A very weak 
 stimulus, with the secondary coil at 300—240 mm., was em- 
 ployed for over three hours, and more than 20 cc. of saliva 
 were secreted. The percentage of organic solids secreted in the 
 slowly flowing saliva steadily fell, but the percentage of such 
 bodies in the saliva secreted on a succeeding stronger stimulus 
 was little if any less, after this long secretion, than it was with 
 an equally strong stimulus before. If, however, a somewhat 
 stronger stimulus was employed, the secretion from a still 
 stronger stimulus was much poorer in organic solids, than the 
 similar stimulus before the weak. 
 
 The fact that rapidly secreted saliva is not a pure solution, 
 and the considerations just presented concerning the independ- 
 ence of the alveoli of the gland render this observatoin of Hei- 
 denhain of doubtful value as evidence of the existence of se- 
 cretory nerves. 
 
 Moreover, there is good reason for doubting the truth of 
 Heidenhain's statement, in the quotation on page 333, that the 
 liquid derived from the blood is incapable of dissolving the con- 
 stituents of the cells in the absence of nerve influence. As has 
 already been pointed out, in treating of sympathetic saliva, 
 (page 322), if the thin chorda saliva be simply left in the gland 
 for twenty minutes, or more, it is converted into a dense, vis- 
 
SECRETION PHYSIOLOGY. 337 
 
 cous fluid having all the characteristics of sympathetic saliva. 
 This conversion takes place with equal readiness whether the 
 gland nerves be intact or divided. 
 
 Heidenhain's own explanation, also, will be found on an- 
 alysis, I believe, to involve such assumptions as to arouse seri- 
 ous doubt of its truth. To explain this phenomenon on the 
 basis of secretor}^ cell activity, he assumed separate "trophic" 
 nerve fibers acting on the cells. He thus necessitated the im- 
 probable conclusion, that at least many of the cells of the sub- 
 maxillary gland received at least four different nerve ends, /. c, 
 trophic and secretory of the sympathetic, and trophic and secre- 
 tory of the chorda ; and at least two entirely different nerve 
 impulses, /. c, trophic and secretory. That such a conse- 
 quence should not have aroused suspicion in his own mind of 
 the truth of his explanation is difficult to understand. 
 
 /'. Post-mortem Chorda Salivary Secretion. 
 
 Another strong argument that the chorda does not produce 
 its secretion by its dilator action on the blood vessels, but by di- 
 rect action on the gland cell, has been derived from the so-called 
 post-mortem chorda secretion. Ludwig and Heidenhain found 
 that if the gland's artery be completely closed, or if the head be 
 rapidly cut off, and the chorda at once stimulated, a fairly copious 
 secretion ensued. This secretion was most abundant in the first 
 minute after section, and thereafter rapidly diminished, but a lit- 
 tle could still be obtained four, and in some cases five, minutes 
 after decapitation, or compression of the artery. Thereafter the 
 nerve was ineffective. Heidenhain beheved this secretion to be 
 due to the action of the nerve on the gland cell, and its rapid fail- 
 ure to lack of oxygen and water. Both Ludwig and Heidenhain 
 believed that by the conditions of the experiment they entirely 
 eliminated the factor of the nerve's vaso-motor action, and hence 
 thought it demonstrative evidence that the secretory and dilator 
 functions of the nerve were independent. 
 
 I think it may be questioned, however, whether the condi- 
 tions of the experiment do entirely obviate the vaso-motor action 
 of the nerve, and whether it is not still possible that this dila- 
 
338 MATHEWS. - 
 
 tion may cause the secretion. It is conceivable that this post- 
 mortem secretion might be due to the flow of blood from the 
 veins and arterioles into the capillaries, owing to the active dila- 
 tion of the latter during chorda stimulation. This explanation, 
 it is true, necessitates the assumptions that the chorda tympani 
 causes, on stimulation, an active dilation of the capillaries, or 
 veins, as well as of the arterioles, and that that dilation in some 
 manner makes it easier for the liquid to pass out into the secre- 
 tion. Both of these assumptions are difficult of proof, and in 
 the limited time at my disposal I have not been able to get 
 demonstrative evidence, either of their truth or error. There is 
 some reason to believe, however, that they may possibly be true. 
 That liquid passes out of the capillaries into the secretion of 
 the submaxillary gland because of an attractive pull exerted 
 upon it by some constituents of the gland cells, has been sug- 
 gested both by Ludwig and Heidenhain. To the evidence pre- 
 sented in favor of such a view by Heidenhain, I have nothing to 
 add, and in the normal condition of the capillary and gland 
 wall, I presume that the hypothesis is true. Ludwig supposed 
 that during chorda stimulation the attractive pull of the cell was 
 increased, owing to the formation of substances in the cell pos- 
 sessed of a higher endosmotic equivalent. Heidenhain believed 
 that the attraction of the cell for the liquid in the blood was 
 constant, but that on stimulating the chorda, the turgor of the 
 cell diminished owing to the passage of liquid into the gland 
 lumen, and water was thus enabled to enter the cell from the 
 blood. Both of these explanations, as will be noticed, assume 
 that in some manner the effectiveness of the attractive pull of 
 the cell is increased during nerve stimulation and water enters 
 the cells independent of the state of the vascular system. The 
 question which confronts us and which it was supposed this 
 post-mortem secretion settled is this : Does stimulation of the 
 nerve cause secretion by increasing in some manner the attrac- 
 tive pull exerted by the gland cells on the liquid of the blood, 
 or does it indirectly render effective by vaso-dilation an attrac- 
 tion which is constantly exerted by the cell on this liquid ? 
 This is a very difficult point to determine. The endeavor 
 
SECRETION PHYSIOLOGY. 339 
 
 has been made to answer this question indirectly by showing 
 that vaso-dilation may ensue without secretion, and secretion 
 without vaso-dilation. But all the evidence which has hitherto 
 been offered, that vaso-dilation may ensue without secretion, 
 and that it alone is incapable of causing secretion, is invalidated 
 by the fact that the conditions of such experiments produce an 
 abnormal gland, or capillary wall, both factors which research 
 on lymph formation have shown to be of importance. Quinine, 
 hydrochloric acid, sodium carbonate, or atropine, drugs which 
 enable vaso-dilation to ensue without secretion, probably alter 
 the permeability of the capillary, or gland cell. So that infer- 
 ences can be drawn from such experiments as to processes oc- 
 curring in the normal gland only with the greatest caution. The 
 evidence with the exception of the post-mortem secretion, that 
 the chorda may cause a secretion without vaso-dilation is also 
 unsatisfactory, as pointed out on p. 355. Attention may now be 
 directed, hence, to this post-mortem chorda secretion. 
 
 It is probable from the considerations presented on page l^'i, 
 that the liquid causing this secretion is derived from the blood. 
 Can the chorda tympani act on the blood vessels in the absence 
 of circulation, in such a manner as to facilitate the passage of 
 that liquid from the capillaries to the gland cells ? The only 
 possible way in which it might so act, I believe, is by causing 
 an active dilation of the capillaries or veins, as well as of the 
 arterioles. Is there any evidence that the chorda has such an 
 action ? 
 
 Tiegerstedt'^"^'- states that the capillaries are contractile but that 
 they have not hitherto been shown to be under nerve control. 
 Roy and Brown have brought forward strong evidence that the 
 capillaries are normally in a state of tonic contraction and that 
 they may actively expand independent of the blood pressure. 
 They observed in the capillaries of the web of the frog's foot 
 that, although blood pressure might be diminished almost to 
 atmospheric pressure, the application for an instant of chloroform 
 to the web caused an enormous expansion of the capillaries. 
 Interesting, also, in this connection, are the observations of von 
 Frey. v. Frey^'' examined microscopically the capillaries of the 
 
340 MATHEWS. 
 
 frog's tongue. He found that on stimulation of the dilator, 
 hypoglossal nerve, a dilation of the capillaries ensued even after 
 the blood supply had been cut off. If the artery be clamped, 
 he observed that the blood streamed out of the capillaries both 
 into the arteries and veins. If, now, the hypoglossal be stimu- 
 lated the capillaries dilate and blood streams into them from the 
 arterioles and veins. This movement persisted for from one to 
 two minutes after clamping the artery. Furthermore, in ex- 
 perimenting on the blood flow from the veins of the submaxil- 
 lary gland of the dog during stimulation of the chorda, v. Frey 
 often observed that stimulation of the chorda was followed by a 
 temporary decrease in the rate of flow of blood from the vein, 
 before the ordinary increase. He suggests that this would seem 
 to indicate a widening of the capillary area leading to a back 
 flow of blood from the veins were it not more probable that the 
 increased flow from the dilated arterioles would be more than 
 sufficient to offset this. 
 
 These facts justify the conclusion, I believe, that on stimu- 
 lating the chorda tympani in the severed head, the capillaries of 
 the gland probably dilate, and that blood enters them from the 
 veins. 
 
 How such a vaso-dilation might lead to a secretion is not 
 clear, but two possibilities suggest themselves : (i) that the 
 capillaries are thus brought into closer relation with the alveoli, 
 and the constant attraction exerted by the gland contents for 
 the water of the blood is thus rendered effective ; or (2) that 
 vaso-dilation may in some way increase the permeability of the 
 capillary wall. The post-mortem chorda secretion can not, I 
 believe, be accepted unconditionally as illustrative of a secre- 
 tion independent of vaso-dilation, until these possibilities have 
 been shown to be non-existent, or non-essential. 
 
 If it shall be found that vaso-dilation of itself is a cause of 
 secretion in the normal gland, and that the gland cell is not the 
 secretory agent, the facts of secretion in the submaxillary gland 
 will probably necessitate the following conclusions, which are 
 not without interest for those studying the physiology of the 
 circulation: (i) That stimulation of the chorda causes an ac- 
 
SECRETION PHYSIOLOGY. 341 
 
 tive dilation of the capillaries, as well as a dilation of the arte- 
 rioles. (2) That the sympathetic is able to overcome the 
 chorda's action on the arterioles, but not its action on the capil- 
 laries. This is shown by the following fact : If, during strong 
 stimulation of the sympathetic, the chorda be irritated by a cur- 
 rent which by itself is barely able to arouse a secretion, a secre- 
 tion ensues which is certainly as large, if not somewhat larger, 
 than the chorda alone would cause. Such a weak stimulus of 
 the chorda is, however, unable to neutralize the sympathetic's 
 constrictor action on the arterioles, as shown by the observa- 
 tions of v. Frey. It will be necessary to assume, hence, that 
 the arterioles have remained contracted, while the capillaries 
 have dilated and blood has entered them from the veins produc- 
 ing a secretion analogous to the post-mortem chorda secretion. 
 
 I endeavored, in a variety of ways, to obviate with certainty 
 all possibility of the chorda's dilator action. By the injection 
 of supra-renal extract into the circulation I hoped to cause 
 such an intense peripheral constriction as to neutralize the di- 
 lator action of the nerve. I am indebted to Dr. R. H. Cunning- 
 ham for this suggestion. After division of the chorda I injected 
 into the jugular vein the whole of a normal salt extract of two 
 powdered supra-renal capsules of another dog. I found, how- 
 ever, that the injection was followed by a slow constant secre- 
 tion of what appeared to be sympathetic saliva, and that this 
 secretion Avas increased at all times by a very weak stimulation 
 of the chorda. Indeed, the chorda caused a larger secretion 
 after the injection than before, probably due to the vaso- con- 
 striction in other areas of the vascular system. This result was 
 so discouraging that I did not attempt to repeat it. 
 
 Heidenhain remarks that large doses of physostigmin cause 
 such an intense constriction of the arterioles of the gland after 
 division of the chorda that stimulation of the latter nerve is un- 
 able to cause either a vaso-dilation, or secretion. Unfortunately, 
 Heidenhain does not give a full account of the experiment. 
 Were it true that the drug produces this effect within three or 
 four minutes of its injection, it would be, I believe, conclusive 
 evidence that secretion can not ensue in the absence of vaso- 
 
 Annals N. Y. Acad. Sci., XI, September 13, 1898 — 23. 
 
342 MATHEWS. 
 
 dilation, and that the nerve does not cause secretion by. action 
 on the gland cells ; for it is known that the drug does not 
 directly paralyze the hypothetical secretory fibers, or the gland 
 cell. To obtain the details of the drug's action, I injected into 
 the jugular vein of a medium-sized dog o. i gr. of physostigmin 
 sulphate. But although the chorda was divided, a spontaneous 
 secretion began which stimulation of the chorda considerably 
 increased. This discrepancy from Heidenhain's results is prob- 
 ably due, I believe, to the impure calabar extract he used. 
 
 I endeavored to ascertain whether the presence of blood in 
 the capillaries was an essential condition of the post-mortem se- 
 cretion by forcing the blood out with air. After ligaturing the 
 carotid artery and placing in it a canula directed headwards I 
 rapidly cut off the head and allowed air to pass into the carotid 
 under a pressure of loo mm. of Hg. The first experiment gave 
 a positive result. On stimulating the chorda a brief, scanty se- 
 cretion was obtained which quickly ceased. Examination of the 
 gland showed it to be practically bloodless. In two other simi- 
 lar experiments the post-mortem secretion was greatly reduced 
 in amount and ceased after i to 3 minutes, instead of lasting for 
 from 3 to 5 minutes, as normally. The glands in these experi- 
 ments still contained blood in the veins. The experiments indi- 
 cate, I believe, that the presence of blood in the capillaries is an 
 essential condition of this secretion. I regret not having been 
 able to bring my experiments to a more satisfactory con- 
 clusion, but it is to be hoped that the important bearing of this 
 post-mortem saliva upon the theory of secretion may lead to 
 its being made the subject of careful investigation. 
 
 From the following experiments the following conclusions 
 may be drawn relative to this post-mortem secretion : 
 
 I. After clamping the gland artery, or cutting off the head, 
 a secretion may be obtained from the submaxillary gland on stim- 
 ulating the chorda. This secretion is most abundant in the first 
 minutes, and thereafter rapidly diminishes. After four or five 
 minutes no more secretion can be obtained. The total amount 
 of saliva secreted varies from 0.3 to 1.5 cc. (Experiments 
 XVIII, XXII and LXIV.) 
 
SECRETION PHYSIOLOGY. 343 
 
 2. If the gland be left without stimulation for a minute after 
 decapitation the total amount of saliva obtainable is considerably- 
 reduced. 
 
 3. If the gland be not stimulated until 3 or 4 minutes have 
 passed a small secretion may be obtained 6 minutes after decapi- 
 tation. (Experiment XVIII.) 
 
 4. If air be blown into the carotid artery, after cutting off 
 the head, the secretion of saliva is reduced in amount and se- 
 cretion ceases, either abruptly or after 2 to 3 minutes. (Experi- 
 ments LXIII, LXVI and LXVII.) 
 
 5. If defibrinated blood be run under small pressure into the 
 vein of the gland a small secretion may be obtained 20 to 30 
 minutes after clamping the gland artery. 
 
 6. If the blood supply be cut off for 30 minutes, on read- 
 mitting blood the arterioles dilate, arterial colored blood issues 
 from the vein at a rapid rate and a spontaneous secretion begins. 
 The rate of this secretion is not changed by stimulation of the 
 chorda in the first minute. (Experiment Va.) 
 
 Experiment Va. 
 
 Large dog. 3 cc. i % morphine sulph. subcut. Tracheot- 
 omy. Ether. Canuls in both submaxillary ducts. Both 
 chordo-linguals and both sympathetics cut. The left vagus sub- 
 sequently divided also. The right gland is stimulated from 
 time to time. See p. 305. The left is freed from its tunic and 
 is attached only by the hilum. The vein on the upper surface is 
 open and flows continuously. The only blood vessel coming 
 to the gland is the hilum artery. The other artery was tied and 
 cut. 
 
 Readings computed in cc. 
 
 Nerve. Amount of Secretion in cc. 
 
 Clamped artery going to gland. 
 
 c Gradually less. 
 
 c None. 
 
 
 
 Time. 
 
 h 
 
 m 
 
 s h 
 
 3 
 
 25 
 
 
 3 
 
 25 
 
 ~ 3 
 
 3 
 
 30 
 
 
 3 
 
 32 
 
 
 3 
 
 35 
 
 
 .07 
 
344 MATHEWS. 
 
 3 
 
 37 
 
 
 
 
 
 s 
 
 .00 
 
 3 
 
 40 
 
 
 
 
 
 s 
 
 .00 
 
 
 
 
 
 
 
 Inject 5 cc. .5% NaCl into duct. 
 
 3 
 
 41 
 
 
 
 
 
 S 
 
 .05 
 
 3 
 
 42 
 
 
 
 
 
 Undamped artery. 
 
 
 3 
 
 43 
 
 30 
 
 
 
 
 c 
 
 Active secretion. 
 
 3 
 
 44 
 
 
 
 
 
 Gland secretions spontaneously .17 cc. per minute. 
 
 
 
 
 
 
 
 Cut left vagus. 
 
 
 4 
 
 07 
 
 30 
 
 
 
 
 Clamped artery again. 
 
 
 4 
 
 07 
 
 30 
 
 - 4 
 
 08 
 
 
 Chorda ( intermittent ) . 
 
 •50 
 
 4 
 
 08 
 
 
 - 4 
 
 09 
 
 
 c 
 
 .18 
 
 4 
 
 09 
 
 
 - 4 
 
 II 
 
 30 
 
 c 
 
 .07 
 
 4 
 
 12 
 
 
 
 
 
 c 
 
 .00 
 
 4 
 
 13 
 
 
 - 4 
 
 14 
 
 
 S 
 
 .08 
 
 4 
 
 15 
 
 
 - 4 
 
 17 
 
 
 c-coil 12 
 
 .00 
 
 4 
 
 17 
 
 30 
 
 - 4 
 
 18 
 
 15 
 
 S 
 
 .05 (very viscid) 
 
 4 
 
 20 
 
 
 
 
 
 s 
 
 .00 
 
 4 
 
 23 
 
 
 
 
 
 Inject NaCl. 5% into duct. 
 
 
 4 
 
 24 
 
 
 
 
 
 s 30 sec. 
 
 .04 
 
 4 
 
 25 
 
 
 
 
 
 c 
 
 .00 
 
 4 
 
 26 
 
 
 - 4 
 
 27 
 
 
 s 
 
 .00 
 
 4 
 
 28 
 
 
 
 
 
 Inject y^, cc. fluid into duct, 
 stimulation. 
 
 Most of it runs out before 
 
 4 
 
 29 
 
 
 
 
 
 S 
 
 .025 
 
 4 
 
 29 
 
 30 
 
 
 
 
 Unclamp artery (red blood rushes out of vein). 
 
 4 
 
 30 
 
 
 - 4 
 
 31 
 
 
 Gland secretes spontaneously. 
 
 .1 cc. 
 
 4 
 
 31 
 
 
 - 4 
 
 32 
 
 
 " " " 
 
 .12 cc. 
 
 4 
 
 33 
 
 6 
 
 
 
 
 c 
 
 .30 CC. per minute. 
 
 4 
 
 35 
 
 
 4 
 
 32 
 
 
 Spontaneously secreting. 
 
 .08 cc. per minute. 
 
 4 
 
 37 
 
 
 - 4 
 
 38 
 
 
 c I mm. 
 
 .7 cc. 
 
 4 
 
 38 
 
 
 - 4 
 
 45 
 
 
 Spontaneously. 
 
 ■5 cc. 
 
 4 
 
 45 
 
 
 
 
 
 c 
 
 .9 cc. per minute. 
 
 4 
 
 45 
 
 30 
 
 
 
 
 Clamped artery again. 
 
 
 4 
 
 46 
 
 30 
 
 - 4 
 
 47 
 
 30 
 
 c (coil 12) 
 
 •5 
 
 
 
 
 
 
 
 Gland still slowly secreting spontaneously. 
 
 4 
 
 48 
 
 30 
 
 - 4 
 
 49 
 
 3" 
 
 c 
 
 .1 
 
 4 
 
 50 
 
 
 - 4 
 
 51 
 
 
 c 
 
 •03 
 
 4 
 
 51 
 
 30 
 
 - 4 
 
 52 
 
 30 
 
 c 
 
 .005 in first thirty seconds, 
 then no more. 
 
 4 
 
 53 
 
 
 - 4 
 
 54 
 
 
 S 
 
 .03 
 
 4 
 
 54 
 
 
 - 4 
 
 55 
 
 
 c 
 
 .00 
 
 4 
 
 ■55 
 
 30 
 
 - 4 
 
 56 
 
 30 
 
 c coil 10 
 
 .00 
 
 4 
 
 57 
 
 
 - 4 
 
 58 
 
 
 s coil 10 
 
 .015 
 
 5 
 
 02 
 
 
 
 
 
 Undamped artery. 
 
 
 5 
 
 02 
 
 30 
 
 
 
 
 c 
 
 Readily secretes. 
 
 Blood rushes continuously out of vein a bright red on 
 unclamping the artery. 
 5 03 — 5 09 Gland secretes spontaneously ._5 cc. 
 
 5 09 - 5 10 s .05 
 
SECRETION PHYSIOLOGY. 345 
 
 5 13 30 Clamped artery. 
 
 5 13 40 - 5 14 40 c .5 
 
 5 14 40-5 17 30 No stimulation. 
 
 5 17 30 - 5 18 30 c .03 
 
 5 19 - 5 20 c .02 
 
 5 20 c .00 
 
 5 22 - 5 23 s .01 
 
 5 24 c .CO 
 
 5 25 - 5 26 s .01 
 
 5 35 c .00 
 
 5 35 30 . s .00 
 
 5 36 Undamped artery. Red blood rushes from tbe vein. 
 
 5 40 Chorda. Rapid secretion. 
 
 Gland secretes spontaneously. 
 
 5 45 - 5 46 Right Sympathetic. .1 cc. 
 
 5 47 - 5 48 Left Sympathetic. .04 cc. 
 
 Cut off head as rapidly as possible. Was unable to saw 
 
 5 49 30 through the vertebral column. All the muscles and skin 
 severed. 
 
 Right gland. 
 5 50 30-5 55 Intermittent stimulation of right chorda. .530 
 
 5 55 Chorda (coil 5) muscular contractions. No secretion. 
 
 5 57 Right S}Tnpathetic. .22 CC. 
 
 6 10 Right sympathetic. .04 CC. 
 
 ^ Left gland ; no secretion either from chorda or sympa- 
 
 5 50 tjjgti^_ 
 
 Experiment LIV. 
 
 Right submaxillary. Chorda and sympathetic cut. Dog 
 under morphine and ether. Tracheotomy. The dog's respira- 
 tions become very slow, and finally cease without any struggles, 
 and without ether. There was considerable fluid in the trachea. 
 
 4.46. Stimulate the chorda while dying, chorda effective 
 until 4.50. The secretion becomes less and less and finally 
 ceases. 
 
 I then stimulated the sympathetic and obtained a very copious 
 secretion of . 2 cc. No more secretion from either nerve. 
 
 Experiment LXIV. 
 
 Before cutting. 10 seconds stim. Coil 24. Secretes .79 cc. 
 Begin to cut at 4.50. i minute to sever head completely. 
 No secretion during operation. 
 
846 MATHEWS. 
 
 h 
 
 m 
 
 s 
 
 h m 
 
 
 4 
 
 57 
 
 - 
 
 4 58 
 
 Stimulates 3 times, 10 seconds at a time. 
 
 4 
 
 59 
 
 
 
 " 10 seconds 
 
 4 
 
 59 
 
 30 
 
 
 " 10 " 
 No more secretion. 
 
 Amount. 
 
 .515 cc 
 .150 cc. 
 .021 cc. 
 
 Total time of stimulation 50 seconds. Total amount. .686 cc. 
 
 From beginning to cut to end of chorda effect, 3 m. 30 s. 
 
 Experiment XXI. 
 
 Before cutting. Coil 20. 10 s. stimulation secretes .55 cc. 
 Begin to cut at 4.05. i minute to sever head completely. 
 No secretion during operation. 
 
 h 
 
 m 
 
 s 
 
 h 
 
 m 
 
 
 
 Amount 
 
 4 
 
 06 
 
 
 - 4 
 
 07 
 
 Stimulate 3 times, 10 seconds at a time. 
 Dog swallows. 
 
 I. 
 
 2. 
 3- 
 
 .235 
 .040 
 .090 
 
 4 
 
 07 
 
 
 - 4 
 
 08 
 
 " 3 times, 10 seconds at a time. 
 Swallows. 
 
 I. 
 
 2. 
 3- 
 
 .070 
 .040 
 .060 
 
 4 
 
 08 
 
 15 
 
 
 
 Coil to 10, muscular contractions, 10 sec. 
 
 
 .100 
 
 4 
 
 09 
 
 
 
 
 30 seconds stim. off and on (muscle). 
 
 
 .030 
 
 4 
 
 09 
 
 15 
 
 
 
 No more secretion. 
 
 
 
 4 
 
 10 
 
 
 
 
 Coil 4. Heavy contractions (escape of current). 
 
 000 
 
 
 Total time of stimulation, 85 seconds. Total amount, 
 
 
 .665 cc 
 
 Time from beginning to cut until end of chorda effect, 
 4 m. 15 s. 
 
 Experiment XVIII. 
 
 Before cutting. Coil 11. Stimulate 10 seconds. 
 Right gland secretes .64 cc. Left gland, .61 cc. 
 5.24.30 begin to cut head. Head severed in 30 s. 
 
 h m s h m RIGHT gland. Amount. 
 
 •125 
 .100 
 .080 
 .070 
 .050 
 
 3. .020 
 
 4. .010 
 
 5 27 - 5 28 "_ 40 seconds. .040 cc 
 
 5 28 30 " 10 " .000 
 
 5 25 - 5 26 Stimulate 3 times, lo seconds at a time. 
 
 5 26 - 5 27 " 4 " 
 
SECRETION PHYSIOLOGY. 347 
 
 Left Gland. Amount. 
 
 5 30 Stimulate left chorda 10 seconds. .070 
 
 next 10 " .010 
 5 30 30 << << chorda (strong muscular contrac- 
 tions). .070 
 
 5 31 Left chorda. No more effect except on mus- 
 
 cular contraction. 
 
 Summary. 
 Right gland. 
 
 Total time of stimulation, 120 seconds. Total secretion, .495 cc. From be- 
 ginning of cut to end of chorda effect, 4 minutes. 
 
 Left gland. 
 Total time of stimulation, 20 seconds. Total amount, .080 cc. Time from be- 
 ginning to cut to end of chorda effect (2)5 minutes, 30 seconds. 
 
 Experiment LXIV. 
 
 Before cutting. Coil 18. 30 sec. stimulation. Secretes 2.1 
 cc. Cut head at 4.30, 1 1^ minutes to sever completely. 
 
 h m s h m 
 
 4 31 40-4 36 Intermittent stimulation. Secretes .250 cc. 
 
 No more secretion after 4.35. 
 4 38 Stimulate sympathetic for two minutes, secretes .065 cc. 
 
 Time from beginning of cut to end of chorda effect 5 minutes. 
 
 Experiment XXII. 
 
 Before cutting. Coil 18. 10 sec. stim. Secretes .2 cc. 
 Cut at 6.07. 30 seconds to sever head completely. 
 
 h m s h m s 
 
 4 07 30 - 6 9. Stimulation, 1st 10 seconds .225 cc. 
 
 40 seconds stim. .060 cc. 
 
 6 09 20-6 19 10 Stimulate coil 18. 30 sec. stim. .150 cc. 
 6 10 30 Chorda no mre effect 
 
 6 12 Coil to 14. Muscular contractions .050 cc. 
 
 Total secretion .375 cc. 
 
 Time from cutting till chorda ineffective, 3 m. 30 s. 
 
 Experiment LXIII. 
 
 Small dog, Irish terrier, under ether. Canula in left Whar- 
 ton's duct. Tracheotomy. Chorda-lingual nerve cut. Pro- 
 
348 • MATHEWS. 
 
 tected electrodes on chorda. Vago-sympathetic not cut. Can- 
 ula connected with air reservoir in the head end of the left 
 carotid artery. 
 
 Before cutting, stimulation of the chorda, with secondary coil 
 at 200, causes a secretion of 0.15 cc. in 10 seconds. 
 
 Head rapidly severed at 4.17 P. M. As soon as it was 
 severed I opened the cock, letting air into the carotid. I then 
 stimulated the chorda tympani at 4.18. Stimulation of the 
 chorda causes a secretion of .02 cc. Secretion then stops 
 and no more can be obtained by any strength of stimulus. 
 
 Experiment LXVI. 
 
 Conditions of the experiment as in Experiment LXIII. Be- 
 fore cutting off the head stimulation of the chorda for 10 seconds 
 with secondary coil at 180 causes a secretion of .17 cc. 
 
 Head rapidly severed from body at 3.03. Chorda stim- 
 ulated at 3.03.45 for 20 seconds. Gland secretes .20 cc. 
 Air then forced into the carotid artery. 
 
 3.04.30—3.05.30 stimulation of the chorda with secondary 
 coil at 130 causes .07 cc. Thereafter no secretion with a stim- 
 ulation of any strength. 
 
 Experiment LXVII. 
 
 Conditions of experiment the same as in Experiment LXHI. 
 Before decapitation stimulation of the chorda for 10 seconds 
 with secondary coil at 230 yields a secretion of 0.2 cc. 
 
 Dog decapitated at 10.49. ^i^ forced into carotid' as soon 
 as cutting began. Head severed in 30 seconds. 
 
 h. 
 
 m. 
 
 s. 
 
 
 
 
 
 10 
 
 49 
 
 45 
 
 Chorda 
 
 10 seconds. 
 
 Coil 230 
 
 0.1 cc. 
 
 10 
 
 50 
 
 30 
 
 " 
 
 " " 
 
 " 200 
 
 0.05 
 
 10 
 
 52 
 
 
 " 
 
 20 " 
 
 " 180 
 
 0.05 
 
 Thereafter no more secretion. 
 
 Post-mortem examination shows the gland veins to be filled 
 with blood. The air does not seem to have penetrated the 
 gland. 
 
SECRETION PHYSIOLOGY. 349 
 
 c. The Nature of the Action of Atropine and 
 Pilocarpine. 
 
 Atropine permits vaso-dilation, on stimulation of the chorda, 
 but prevents secretion. The drug has been supposed to act, 
 not on the gland cell, but on the ends of the secretory nerve 
 fibers. The reasoning for this is as follows : In the dog's sub- 
 maxillary, atropine paralyzes the chorda secretion, but not the 
 sympathetic. If the sympathetic innervate the gland cell and 
 cause its secretion by action on the latter, the gland cells con- 
 nected with this nerve have evidently not been paralyzed. As 
 there is no reason to suppose these cells different from those 
 connected with the chorda, it is probable that the cells con- 
 nected with the chorda have not been paralyzed. But if the 
 gland cells have not been paralyzed, and the dilator action of 
 the nerve remains unaffected, we must assume that there is some 
 third element connected with the nerve which has been para- 
 lyzed. This must be the element causing secretion, /. t'., the 
 secretory nerve fiber. The latter must be paralyzed at the nerve 
 termination, since, as far as known, atropine does not act on the 
 nerve fibre. This argument is true only for the dog and not 
 for the cat^' since, in the cat, atropine paralyzes the sympathetic 
 as well as the chorda. The argument, as will be seen, depends 
 on the assumption that the sympathetic causes secretion by 
 action on the gland cells. This, as pointed out, is probably in- 
 correct. The sympathetic produces its secretion by action on 
 contractile tissue. There ' is, hence, no longer any reason to 
 suppose that the gland cells have not been paralyzed by the 
 drug. How it acts upon the cell is unknown, but the effect of 
 that action is to prevent or diminish the passage of fluid through 
 the cells. The variation in the susceptibility to its action of dif- 
 ferent glands in the same animal (compare the pancreas, salivary 
 glands and kidneys of dog), or of the same gland in different 
 animals (compare the pancreas of the dog and rabbit) points, I 
 believe, toward an action on the gland cell itself, the variations 
 in its action being due to variation in the chemical composition 
 of the cells. 
 
 THEOLOGICAL SEMINARY. 
 
 
350 MATHEWS. 
 
 That atropine does act on the gland cell is, perhaps, indicated 
 also by the action of its great antagonist pilocarpine. Pilocar- 
 pine, namely, produces a secretion of sweat two to three weeks 
 after cutting the sciatic of the cat, when the nerve is totally in- 
 active.''^ ^^ ^^ Luchsinger,*'' in commenting on this, says that 
 this secretion must be due either (i) to action on the secretory 
 cells themselves, or (2) to the non-degeneration of the nerve 
 ends. The second possibility is impossible since these nerve 
 ends are not provided with nuclei. A similar secretion may be 
 obtained in the dog's salivary glands, fourteen days after cut- 
 ting both chorda and sympathetic. The evidence is here not so 
 conclusive since the submaxillary ganglion does not degenerate. 
 In the sweat secretion, however, I believe the evidence is fairly 
 strong that pilocarpine does act directly on the gland cell. It 
 thus strengthens the evidence that atropine also acts on the cell. 
 
 There is also reason for believing that atropine acts in some 
 manner on the capillary wall, thus reducing, or preventing the 
 transudation of lymph. It might, in this way effect secretion 
 from glands. This possibility has not received the attention it 
 deserves.* 
 
 The evidence that atropine checks lymph transudation is as 
 follows : 
 
 If atropine permitted the transudation of lymph normally en- 
 suing on vaso-dilation, it would be expected that, after its injec- 
 tion, stimulation of the chorda would render the submaxillary 
 gland oedematous, since fluid no longer passes into the secre- 
 tion. Quite the contrary is the fact. I have repeatedly stimu- 
 lated the gland all day, after the injection of atropine, without 
 producing a trace of oedema. Heidenhain ^■'' himself says : 
 " After atropine on stimulation of the chorda tympani no in- 
 
 * Heidenhain's reasons for rejecting the possibility that atropine checks lymph 
 transudation and thus secretion will be found in Hermann's Handbuch. A strik- 
 ing instance of failure to consider this possibility is the following quotation from 
 Larigley : 
 
 "Atropine prevents the stimulation of the hilum from producing a secretion. 
 Nicotine does not do this, therefore, atropine acts upon structures more peripheral 
 than those acted upon by the nicotine. Since nicotine acts on nerve cells, and 
 atropine does not act on gland cells, atropine must produce its paralyzing result by 
 action on the secretory nerve endings." 
 
SECRETION PHYSIOLOGY. 351 
 
 crease in lymph flow occurs, even when during stimulation of 
 the chorda the medulla is stimulated and the blood pressure 
 greatly increased." Brunton in commenting on this says : 
 " It appears to me that this circumstance can hardly be explained 
 otherwise than by supposing that atropin not only paralyses the 
 secretary fibres of the chorda, but acts upon the blood vessels 
 in such a manner as to greatly diminish or prevent the exuda- 
 tion which would usually take place from them into the lymph 
 spaces." 
 
 Heidenhain^^ supposed that lymph normally left the blood 
 vessels on account of the secretory pull exerted by the gland 
 cell. Atropine prevented lymph transudation by paralysis of 
 the secretory chorda nerve ends. He was led to this conclusion 
 chiefly by the following facts : (i) No more lymph normally 
 leaves the blood vessels than passes into the secretion, and (2) 
 if one inject 4.9% solution of sodium carbonate, 0.5% hydro- 
 chloric acid or quinine sulphate into Wharton's duct the chorda's 
 secretory power is annihilated, but on stimulation the gland 
 becomes highly oedematous. If, however, atropine be injected 
 into the blood before the chorda is stimulated and after the in- 
 jection of quinine into the duct no oedema ensues, however long 
 the nerve be stimulated. I have fully confirmed these observa- 
 tions. The most probable interpretation of these facts, it seems 
 to me, is that quinine prevents the passage of fluid through the 
 glands by action on the gland cells, but does not prevent lymph 
 transudation. That atropine, however, acts directly on the 
 capillary wall, as well as upon the gland cell, in such fashion as 
 to prevent lymph transudation and secretion. 
 
 A further indication that atropine checks lymph transudation is 
 the diminution in thoracic lymph flow after its injection. Tschir- 
 winsky^® found that in morphinized animals thoracic lymph 
 flow fell from 3,75 cc. to 1.5 cc. and from 10 cc. to 4.2 cc. in a 
 given time. Atropine neutralized, also, the increased flow due 
 to curare. In the latter case it fell from 9 and 10 cc. to 2.5 
 and 5.3 cc. in a given time. As there is reason to believe 
 (Adami) that curare increases lymph transudation by direct 
 action on the capillary wall, the inhibiting action of atropine may 
 
352 
 
 MATHEWS. 
 
 be referred to an opposite action on the same structure. Not 
 knowing of Tschirwinsky's work, I had already performed simi- 
 lar experiments on the lymph flow, comparing it with pancreatic 
 flow on vagus stimulation and after pilocarpine. I found (Ex- 
 periment V that atropine temporarily neutralizes the large 
 increase in lymph flow which occurs concomitant with increased 
 panceas secretion during rythmic stimulation of the vago-sym- 
 pathetic after division of the cervical cord, and also neutralizes 
 the increased lymph flow due to pilocarpine. 
 
 Experiment Vb. 
 
 Medium-sized dog. Ether. Temporary pancreatic fistula. 
 Tracheotomy. Cervical cord cut. Artificial respiration. Tho- 
 racic duct prepared. Lymphatics of head and neck ligatured. 
 Readings every minute in cubic centimeters : 
 
 Thoracic Duct 
 
 Pancreas. 
 
 Thoracic. 
 
 
 
 Pancreas. 
 
 Th 
 
 oracic. 
 
 Pancreas. 
 
 Vagi 
 
 uncut. 
 
 .050 
 
 
 
 .009 
 
 
 .197 
 
 
 .009 
 
 .220 
 
 .02 
 
 .110 
 
 
 
 .013 
 
 
 .180 
 
 
 .004 
 
 .220 
 
 .02 
 
 .115 
 
 
 
 .012 
 
 
 .180 
 
 
 .008 
 
 .200 
 
 .02 
 
 y^ hour interval. 
 
 
 .150 
 
 
 — 
 
 .200 
 
 .015 
 
 .120 
 
 
 
 .013 
 
 
 .190 
 
 
 .006 
 
 .180 
 
 •015 
 
 .119 
 
 
 
 .012 
 
 Rt. 
 
 Vagus. 
 
 Ryth. 
 
 Coil 9. 
 
 .180 
 
 .010 
 
 .090 
 
 
 
 .009 
 
 
 .200 
 
 
 .000 
 
 .190 
 
 •015 
 
 .130 
 
 
 
 .Oil 
 
 
 .180 
 
 
 .000 
 
 .160 
 
 .013 
 
 .120 
 
 
 
 .010 
 
 
 .100 
 
 
 .002 
 
 .180 
 
 .017 
 
 .110 
 
 
 
 .010 
 
 
 .300 
 
 
 .068 
 
 •155 
 
 .017 
 
 .100 
 
 
 
 .008 
 
 
 — 
 
 
 •025 
 
 ■155 
 
 .018 
 
 .100 
 
 
 
 .009 
 
 
 — 
 
 
 .015 
 
 .170 
 
 .015 
 
 .102 
 
 
 
 .004 
 
 
 — 
 
 
 ■015 
 
 Cut vagi 
 
 in neck. 
 
 .100 
 
 
 
 — 
 
 
 
 Off. 
 
 
 .280 
 
 .015 
 
 Clot. 
 
 
 
 
 
 .160 
 
 
 .020 
 
 .220 
 
 .010 
 
 I shock 
 
 per 
 
 second. 
 
 
 .140 
 
 
 .010 
 
 .160 
 
 .005 
 
 Rt. Vagus. 
 
 Ryth. 
 
 Coil 10. 
 
 
 ■150 
 
 
 .005 
 
 .100 
 
 .003 
 
 .150 
 
 
 
 .009 
 
 
 — 
 
 
 .010 
 
 .120 
 
 .007 
 
 .220 
 
 
 
 .006 
 
 Rt. 
 
 Vagus. 
 
 Ryth. 
 
 Coil 9. 
 
 .100 
 
 .009 
 
 .250 
 
 
 
 .010 
 
 
 .360 
 
 
 .006 
 
 .100 
 
 .010 
 
 .170 
 
 
 
 .005 
 
 
 .200 
 
 
 .009 
 
 .060 
 
 .015 
 
 .230 
 
 
 
 .010 
 
 
 .240 
 
 
 .005 
 
 .050 
 
 .020 
 
 Current ofi 
 
 
 
 Coil to 4. 
 
 
 .090 
 
 .010 
 
 .200 
 
 
 
 .005 
 
 
 .200 
 
 
 .005 
 
 .120 
 
 .015 
 
 .190 
 
 
 
 .007 
 
 
 — 
 
 
 .008 
 
 .120 
 
 .003 
 
 .220 
 
 
 
 .007 
 
 
 
 Off. 
 
 
 .120 
 
 .007 
 
 220 
 
 
 
 .005 
 
 
 Clot. 
 
 
 0.15 
 
 .065 
 
 .010 
 
 .310 
 
 
 
 .002 
 
 
 '•■ 
 
 
 .011 
 
 •125 
 
 .010 
 
 .210 
 
 
 
 .001 
 
 Left Vagus 
 
 Ryth. 
 
 Coil 9. 
 
 .100 
 
 .oil 
 
 .180 
 
 
 
 — 
 
 
 •550 
 
 
 .030 
 
SECRETION PHYSIOLOGY. 
 
 353 
 
 Thoracic. 
 
 Pancreas. 
 
 Thoracic. ' 
 
 Pancreas. 
 
 Thoracic. 
 
 Pancreas. 
 
 .290 
 
 
 .005 
 
 .120 
 
 
 .005 
 
 .270 
 
 
 .015 
 
 — 
 
 
 .005 
 
 — 
 
 
 .010 
 
 — 
 
 
 .005 
 
 — 
 
 
 .005 
 
 •175 
 
 
 .025 
 
 .230 
 
 
 .005 
 
 — 
 
 
 .010 
 
 •225 
 
 
 ■ 045 
 
 .240 
 
 
 .000 
 
 Coil to 
 
 6 
 
 .015 
 
 .250 
 
 
 .055 
 
 — suddenly 
 
 .120 
 
 
 
 
 .060 
 
 .220 
 
 
 .110 
 
 .250 
 
 
 .080 
 
 — 
 
 
 .090 
 
 — 
 
 
 .120 
 
 Inject .5 
 
 cc. atrop 
 
 n into 
 
 .240 
 
 
 .100 
 
 .320 
 
 
 .140 
 
 supra-scap. vein 
 
 
 
 
 
 .090 
 
 .300 
 
 
 .130 
 
 Stimulation continued. 
 
 Off. Then 
 
 on by accident. 
 
 
 Off. 
 
 
 .250 
 
 
 .050 
 
 .230 
 
 
 .060 
 
 — 
 
 
 •115 
 
 .200 
 
 
 .070 
 
 
 Off. 
 
 
 .170 
 
 
 .065 
 
 — 
 
 
 .030 
 
 .140 
 
 
 •035 
 
 .200 
 
 
 .030 
 
 .180 
 
 
 :o20 
 
 .170 
 
 
 .030 
 
 .150 
 
 
 — 
 
 .140 
 
 
 .015 
 
 .160 
 
 
 .030 
 
 .200 
 
 
 .030 
 
 •145 
 
 
 .015 
 
 .120 
 
 
 .015 
 
 — 
 
 
 .020 
 
 •155 
 
 
 .015 
 
 .170 
 
 
 .015 
 
 — 
 
 
 .015 
 
 .110 
 
 
 .OJO 
 
 .130 
 
 
 — 
 
 .140 
 
 
 .015 
 
 .120 
 
 
 .015 
 
 .170 
 Left Vagus. 
 
 Ryth. 
 
 .010 
 Coil 6. 
 
 .140 
 •145 
 
 
 .010 
 .016 
 
 .120 
 
 Off. 
 
 .007 
 .008 
 
 .140 
 
 
 .007 
 
 •135 
 
 
 .009 
 
 .130 
 ■5 c 
 .080 
 .040 
 
 
 .010 
 
 .130 
 .060 
 .140 
 
 
 .002 
 .009 
 .090 
 
 .130 .005 
 .160 .017 
 Left Vagus. Rythmical. 
 
 :. atropin 
 
 .010 
 .010 
 
 .200 
 .290 
 
 
 .120 
 .130 
 .140 
 
 .160 
 .250 
 .210 
 
 
 .002 
 .008 
 .000 
 
 .090 
 .100 
 .100 
 
 
 .005 
 .007 
 .008 
 
 .235 
 
 
 .110 
 
 .240 
 
 
 .001 
 
 .120 
 
 .100 
 .100 
 
 
 .012 
 
 .280 
 .235 
 
 
 .130 
 .130 
 
 300 
 
 
 .015 
 .075 
 
 
 .007 
 .005 
 
 .250 
 
 
 .080 
 
 .300 
 
 
 .035 
 
 
 
 Off 
 
 
 .210 
 
 
 .045 
 
 . no 
 
 
 .005 
 .006 
 
 .340 
 
 .210 
 .190 
 
 
 .154 
 
 .116 
 
 .052 
 
 .350 
 .300 
 
 
 .too 
 
 .100 
 
 ito 
 
 Stim. Left Vagus. 
 .070 
 .100 
 
 Coil 6. 
 
 .010 
 
 .C08 
 
 .160 
 .190 
 .190 
 .170 
 
 
 ■043 
 .020 
 .020 
 .025 
 .011 
 
 Left Vagus 
 .220 
 .280 
 
 Ryth. 
 Off 
 
 .140 
 
 Coil 6. 
 110 
 .070 
 
 .120 
 .160 
 .160 
 .230 
 .200 
 
 
 .002 
 .000 
 .000 
 .000 
 
 — 
 
 
 .014 
 
 .200 
 
 
 .070 
 
 .260 
 
 
 .000 
 
 •15s 
 
 
 .oi5 
 
 .200 
 
 
 .050 
 
 .170 
 .200 
 
 
 .000 
 
 .150 
 
 
 .010 
 .010 
 
 .230 
 .180 
 
 
 .020 
 .025 
 
 Off. 
 
 .000 
 
 .150 
 
 
 — 
 
 .180 
 
 
 .030 
 
 .250 
 
 
 .000 
 
 Left Vagus 
 
 Ryth. 
 
 Coil 6. 
 
 •155' 
 
 
 .005 
 
 .170 
 
 
 .000 
 
 .210 
 
 
 .010 
 
 •145 
 
 
 .005 
 
 .190 
 
 
 .000 
 
 .190 
 
 
 .000 
 
 Left Vagus 
 
 . Ryth 
 
 Coil 6. 
 
 
 
 
 This experiment is of interest, not only as a clear confirmation of Pawlow and 
 Mett, but because of the invariable increase in thoracic lymph flow occun-ing on 
 stimulation of the vagus. I have repeatedly sought to obtain other experiments like 
 it, but never with such success. The operation is long and .apt to miscaiTy at some 
 point. 
 
354 
 
 MA THEWS. 
 
 Experiment XI. 
 
 Dog, etherized. Canula in thoracic duct. Readings in cc. 
 every minute. 
 
 Thoracic duct. 
 
 .150, .220, .200, .180, .300, .230, .250. 
 
 I cc. I % pilocarpine into left femoral vein. Dog perfectly quiet. 
 
 .250, .300, .500, .600, .400, .460, .400. 
 
 1 cc. pilocarpine. 
 
 .490, .410. 
 
 I cc. \fo atropine l%. 
 
 .240, .090, .060, .070, .170, .110, .120, .090, .090. 
 
 Moved head. 
 
 .220. 
 
 I cc. atropin. 
 
 .130, .100, .070, .060, .040, .120. 
 
 2 cc. pilocarpine. 
 .100, .080, .120, .130. 
 
 I hour interval. 
 .160. 
 
 It is not without interest in this connection that pilocarpine, 
 contrary to atropine, increases lymph flow. This was first ob- 
 served by Tschirwinsky.''^ My own experiments have yielded 
 a positive result generally, but not invariably. In all cases the 
 dogs had divided cervical cords, and generally divided vagi. 
 They were all under artifical respiration. The lymph was 
 measured in cc. for equal intervals of time. 
 
 Experiment. 
 
 Before pilocarpine 
 injection. 
 
 After the injection 
 of 1-2 cgs. of pilo- 
 carpine. 
 
 Remarks. 
 
 II 
 
 29 
 
 14 
 
 4 
 
 62 
 
 1.53 
 2.44 
 0.50 
 
 1-55 
 1. 41 
 
 3.00 
 6.09 
 1.72 
 10.40 
 1.69 
 
 7 minutes. Dog motionless. 
 
 Some movements of 
 abdomen. 
 
 Motionless. 9 minutes. 
 
 Movements. 
 
 No movements. Pancreas 
 did not secrete either. 
 
 In experiments 1 1 and 1 4 there were no visible movements. 
 The flow of the seven minutes after injection in No. 1 1 was 
 
SECRETION PHYSIOLOGY. 355 
 
 double that of seven minutes before, and in experiment 14 was 
 three times as great. In experiment 62, however, there was 
 scarcely any difference. 
 
 The evidence presented in the foregoing pages, if not conclu- 
 sive, certainly indicates that atropine restricts and pilocarpine 
 increases lymph transudation. They may in this manner affect 
 secretions. In any case, if the sympathetic causes its secretion 
 by action on contractile tissue in the gland, there is no longer 
 any reason against assuming that atropin acts directly on the 
 gland cell, in such manner as to check the passage of fluid 
 through it, and thus to prevent secretion. 
 
 d. The Action of Quinine and Nicotine. 
 
 We have considered the three main objections which have 
 been raised against the chorda salivary secretion being an osmosis. 
 There are, also, certain other phenomena which have been 
 thought indicative of the independence of the secretory and di- 
 lator action of this nerve, and, hence, are worthy of a short 
 criticism. 
 
 The first is the action of quinine, which when injected into 
 the gland duct causes a temporary vaso-dilation, but no secre- 
 tion. If, however, the chorda be stimulated, still greater dila- 
 tion ensues and secretion takes place. This secretion is less than 
 normal. Heidenhain^^ interprets this to mean that vaso-dila- 
 tion cannot of itself produce a secretion, but that the secretory 
 fibres must be aroused. (See literature reference No. 21, p. 85. 
 Also reference No. 23, p. 45,) 
 
 The facts may, however, be otherwise understood. Quinine 
 prevents the passage of liquid through the gland cell. This is 
 shown by the fact that ultimately it prevents chorda secretion, 
 even though the gland become oedematous. If the permea- 
 bility of the gland membrane be thus diminished, the slight 
 vaso-dilation caused by the drug may be insufficient to cause a 
 secretion, whereas a larger vaso-dilation on stimulating the 
 chorda might overcome this resistance. Another possibility is 
 that the quinine reaches a portion only of the alveoli, poisons 
 these, and throws their capillaries and arterioles' into dilation. 
 
356 MATHEWS. 
 
 On stimulating the chorda the secretion may be derived from 
 unpoisoned alveoli of which the blood vessels have not hitherto 
 been in dilation. 
 
 The value of Langley's and Heidenhain's observation, that 
 the secretory fibres of the chorda tympani recover, after nico- 
 tine poisoning, before the dilator fibres,- is seriously impaired by 
 a defective method of determining whether vaso-dilation did, or 
 "did not, occur. If we admit that the rate of flow of blood from 
 the gland's vein is a criterion by which we can determine 
 whether vaso-dilation has or has not occurred their conclusion 
 is justified. But reflection shows that if vaso-dilation be slight 
 the amount of water passing out into the secretion might so re- 
 duce the bulk of blood flowing through the gland as to mask 
 entirely all effects of the increased flow due to vaso-dilation. 
 In fact, the flow of blood from the vein would be a safe cri- 
 terion of dilation, only if there were no escape of liquid through 
 the capillary wall, a condition which manifestly does not here 
 exist. Langley's and Heidenhain's conclusion that the secre- 
 tory function recovers before the dilator is, hence, unjustified. 
 The same criticism applies, also, to Heidenhain's observation 
 that after the chorda tympani has been cut and allowed to de- 
 generate for three or four days stimulation still causes an in- 
 crease in the paralytic secretion, but no increase in blood-flow 
 from the vein. 
 
 e. Evidence of the Osmotic Character of the Salivary 
 Secretions which are Accompanied by Vaso-dilation. 
 
 wish now to summarize briefly those features of secretions, 
 accompanied by vaso-dilation, which indicate that they are of an 
 osmotic character. 
 
 (i) In structure the salivary glands have all the require- 
 ments of an elaborate osmotic mechanism They are, essentially, 
 extraordinarily thin -walled bags, possessing an enormous sur- 
 face, containing a mass of hydroscopic indiffusible substances. 
 The outer surface of this bag is in intimate association with a 
 mesh work of capillaries so coordinated by the nervous system 
 as to permit an almost instantaneous flooding of the gland mem- 
 
SECRETION PHYSIOLOGY. 357 
 
 brane. Plainly here are all the requisites of a delicate osmotic 
 mechanism adapted to the most rapid osmosis. 
 
 (2) Chorda secretion is closely dependent on blood supply. 
 (Compare p. 342.) Heidenhain has shown that partial occlusion 
 of the artery diminishes the rate of secretion (p. 88, Breslau 
 Studien IV.) 
 
 (3) If the osmotic equivalent of the blood be increased by the 
 injection of strong salt solutions the secretion is diminished or " 
 altogether inhibited. ^^ ^'^ 
 
 (4) If the osmotic equivalent of the blood be decreased by the 
 injection of water the rate of secretion is increased. '^'^ 
 
 (5) The rate of secretion is increased, other things equal, by 
 an increase in the rate of blood flow through the gland.'^*^ ^'^ 
 
 (6) The rate of secretion diminishes when the hj/logens are 
 washed out of the gland. (Paralytic secretions, secretion after 
 long stimulation.)"^ 
 
 (7) Substances may be absorbed with extraordinary rapidity 
 when injected into the duct (nicotine, atropine). 
 
 (8) If the percentage of salts in the blood be increased the per- 
 centage of salts in the saliva increases also. If the percentage of 
 salts in the blood be decreased, the percentage of salts in the 
 saliva decreases also.''"^ ^^ ^^ 
 
 (9) If the artery of the gland be clamped for 20-30 minutes, 
 and the blood thus completely cut off from the gland, on read- 
 mitting theblooda vaso-dilation ensues, so that the blood rushes 
 red from the gland veins, and this vaso-dilation is accompanied 
 by a spontaneous secretion. Stimulation of the chorda in no 
 way alters this secretion during the first minute, nor until the 
 dilation has somewhat diminished. This spontaneous secretion 
 is a close duplicate of that observed by Levy in the secretion of 
 sweat. [Experiment V (a).] 
 
 Although this spontaneous secretion might, perhaps, be ex- 
 plained by supposing that a direct stimulation of nerve-end or 
 cell by the oxygen has taken place, it seems more probable to 
 me to class it with the spontaneous secretion of sweat in the 
 horse, following section of the cervical sympathetic, and to refer 
 it to the direct effect of vaso-dilation. 
 
 Anxals N. Y. Acad. Sci., XI, September 13, 1898 — 24. 
 
358 MA THEWS. 
 
 f. Conclusion. The Physiology of Salivary Secretion. 
 
 If the sympathetic saHvary secretion shall be found to be due 
 to the action of contractile tissue, and if the criticisms of the ob- 
 jections to considering the salivary secretion, coincident with 
 vascular dilation, an osmosis, be sustained by subsequent work, 
 the following conclusions concerning the physiology of this 
 secretion may be drawn. 
 
 The salivary glands may be caused to secrete, either by the 
 action of contractile tissue under control of the sympathetic 
 nerve or by osmosis under control of the vaso-dilator nerve. 
 Probably in normal secretion both of these nerves come into 
 play, but of this evidence is as yet lacking. 
 
 Drugs, or other reagents, may arouse secretion by action on 
 either or both of these mechanisms. I would suggest that 
 secretion following strychnine injection, camphor, pikrotoxin, 
 physostigmin (after division of the chorda) are due to the con- 
 tractions of the contractile tissue. All of these drugs stimulate 
 the nerve centers and cause a pronounced vaso-constriction. 
 On the other hand, pilocarpine, nicotine, muscarine, curare and 
 chloral hydrate, or other drugs with a similar action on the 
 vascular system, probably cause secretion partly by vaso-dila- 
 tion and partly by increasing the permeability of the gland mem- 
 branes. Such drugs work through an osmotic mechanism. A 
 third class of drugs, such as quinine, atropine, hydrochloric acid 
 or sodium carbonate may produce vaso-dilation, but probably act, 
 also, on the gland cells in such manner as to diminish their per- 
 meability- Most of the work which has hitherto been done 
 upon the action of drugs on salivary secretion needs to be re- 
 peated with the possibility in mind that the chorda and sym- 
 pathetic induce secretion in these different ways. 
 
 The osmotic mechanism of secretion in the salivary glands is 
 probably dependent on the condition of the gland and capillary 
 membranes, upon the composition of the blood, upon the rate 
 of flow of the blood and the character and amount of hylogens 
 present within the gland. The evidence that the course of os- 
 mosis is controlled by the action of nerves directly on the gland 
 
SECRETION PHYSIOLOGY. 359 
 
 cells is open to serious criticism. That chorda salivary secre- 
 tion can ensue without vaso-dilation may be seriously doubted, 
 not only for the reasons already stated, but because in the 
 pancreas there is good reason to believe that secretion can not 
 take place without vaso-dilation. (See p. 361.) 
 
 V. SOME OTHER SECRETIONS. 
 
 The submaxillary gland, considered in the foregoing pages, 
 may be taken as a type of all the salivary glands, as each pos- 
 sesses a dilator secretory nerve, and a constrictor, sympathetic 
 secretory nerve. I wish now to consider some other secretion in 
 the light of the conclusions derived from the physiology of the 
 submaxillar}^ 
 
 a. The Physiology of Sweat Secretion. 
 
 There is reason to believe that the mammalian sweat glands 
 also have a double mechanism of secretion, a muscular and an 
 osmotic. These glands are surrounded by a sheath of muscle 
 fibres lying, like those of the skin glands of amphibia, be- 
 tween the cells and the basement membrane. From the obser- 
 vations of Ranvier, Joseph and others, who have shown that 
 upon stimulation of the sciatic this muscle contracts, there can 
 be little doubt that a secretion may thus be formed. Probably 
 sweat secretions ensuing coincident with vaso-constriction, upon 
 the injection of strychnine, upon stimulation of the sciatic in the 
 amputated limb or after compression of the blood vessels is due 
 to this mechanism. 
 
 On the other hand, certain secretions of sweat are too copi- 
 ous to be due to muscular constriction of the gland. That 
 those secretions probably fall under the second, or osmotic, 
 mechanism is shown by the following facts : 
 
 (i) The coincidence of vaso-dilation and sweat secretion. 
 Most sweat secretions are normally accompanied by vaso-dila- 
 tion. If the cervical sympathetic of the horse be severed, 
 strong hyperaemia and sweating occurs on the side of the neck 
 the nerve governs. This sweating ensuing after nerve division 
 
360 MATHEWS. 
 
 can hardly be explained, I think, on the basis of secretory cell 
 activity. 
 
 (2) Pilocarpine, which does not cause contraction of the mus- 
 cular sheath, causes a profuse secretion. 
 
 (3) The vaso-motor and secretory fibres in the cat follow the 
 same paths. 
 
 (4) Pilocarpine causes sweat secretions fourteen days after 
 nerve degeneration. 
 
 (5) If the blood supply be cut off, on readmitting the blood 
 after 30 minutes, a spontaneous secretion occurrs."*^ The sim- 
 ilar secretion in the submaxillary is invariably accompanied by 
 vaso-dilation. 
 
 (6) Increasing the capillary blood pressure or drinking large 
 quantities of water increases secretion. 
 
 The facts, as far as they go, are the same as those observed 
 in the cerebral salivary secretions and pancreatic secretion. 
 They justify us, I believe, in classing all three secretions in the 
 same category. That these sweat secretions are of an osmotic 
 character would thus be indicated. That other sweat secretions 
 are due to muscle there can be little doubt. 
 
 b. The Secretion of the Pancreas. 
 
 Secretion of the pancreas is normally accompanied by vaso- 
 dilation. In its relation to atropine, its increased content of or- 
 ganic bodies coincident with an increased rate of flow, and in 
 taking place after compression of the aorta, pancreatic secretion 
 resembles the submaxillary secretion on stimulation of the 
 chorda tympani. There is reason to believe, however, that the 
 pancreas cannot secrete unless the blood vessels dilate. Thus 
 the means employed by Pawlow,'''' Mett^^ and Kudrewetsky^- to 
 give the vagi a secretory function are just the means used by 
 Bowditch, Luchsinger and others^*^ to give the sciatic and other 
 mixed dilator and constrictor nerves a dilator action. These 
 authors either cut the vagi and splanchnics, and allowed them to 
 degenerate three or four days, or else they stimulated them 
 with rythmic induction shocks, at the rate of one per second 
 after division of the cervical cord. There are two possible ex- 
 
SECRETION PHYSIOLOGY. 361 
 
 planations of the fact that stimulation of the normal nerve with 
 the cord undivided causes no secretion. Either the nerve carries 
 inhibitory secretory as well as secretoiy fibres, or stimulation 
 of the nerve is unable to cause a secretion without vaso-dilation. 
 The first alternative Heidenhain has particularly combatted in 
 the case of the submaxillary, and it appears to me lacking all 
 proper experimental basis. The second alternative is probably 
 the true explanation, for the reason that stimulation of the nor- 
 mal nerve below the cardiac branches causes no alteration in 
 blood pressure, and for the reason that the treatment to which 
 the nerve is subjected is calculated to give it a dilator action. 
 If this be true the pancreas would appear fundamentally differ- 
 ent from the salivary glands, unless, as I have endeavored to 
 show, the latter are, also, in reality, unable to secrete on stim- 
 ulation of the chorda or other cerebral nerve, unless vaso-dila- 
 tion ensues. 
 
 Further evidence of the dependence of pancreatic secretion 
 on vaso-dilation is furnished by the action of pilocarpine, chloral 
 hydrate^^ and curare, drugs which cause vaso-dilation and secre- 
 tion, and by str}'Xhnine,^^ or digitalis, drugs which cause vaso- 
 constriction and inhibit secretion. Heidenhain, ^^ also, has 
 observed a close correspondence between vaso-dilation and 
 secretion, and between vaso-constriction and the cessation of 
 secretion. This parallelism between vaso-dilation and secretion 
 can not be accidental. It indicates, I beheve, that the dilation 
 is the cause of the secretion, other things being normal. 
 
 VI. GENERAL CONCLUSION. 
 
 We have now considered the evidences of the existence of 
 secretory nerves, and the reasons for believing that secretion is 
 a function of the gland cells. While readily admitting the pos- 
 sibilities that secretion may in certain instances be a function of 
 the gland cell, controlled by the action on it of secretory nerve 
 fibres, we have seen reason to believe that certainly many so- 
 called secretions are due not to the gland cell, but to the action 
 of contractile tissue either within or about the p;land. Among 
 
362 MATHEWS. 
 
 such secretions are the sahvary secretions following stimulation 
 of the sympathetic, certain secretions of sweat, the secretion of 
 the cephalopod salivary glands and of the skin glands of am- 
 phibia. 
 
 Whether those secretions which are normally accompanied by 
 vaso-dilation, such, for instance, as the salivary secretions follow- 
 ing stimulation of the cerebral nerves and the secretions of the 
 alimentary tract and its appendages, are governed by nerves act- 
 ing directly on the gland cells, or indirectly through the vascu- 
 lar system, cannot with certainty be said. But I believe it has 
 been shown in the present paper that the evidence which has 
 hitherto been offered that such secretions are controlled by 
 nerve action on the gland cell is open to serious criticism. The 
 remarkable parallelism between the hypothetical secretory and 
 vaso-dilator fibres, the close dependence of such secretions on 
 the vascular system, the general features of such secretions and 
 the structure of glands, all indicate, I believe, that osmosis is 
 the essential cause of these secretions, and that they are con- 
 trolled by the action of nerves on the vascular system. No 
 one would deny that the course of these secretions is modified 
 by the condition of the gland or capillary wall, and that that 
 condition is easily affected by drugs, but that nerve action di- 
 rectly affects that condition, I do not believe the evidence 
 entitles us to say. 
 
 Probably the study of these secretions from the standpoint of 
 osmosis will bring to light facts difficult to reconcile with our 
 present knowledge of osmosis. But while our knowledge of the 
 latter process through membranes undergoing chemical change, 
 such as gland membranes, remains in its present fragmentary 
 state, I do not believe that we are justified in assuming a special 
 sort of secretory activity on the part of the gland, or capillary 
 cell, unless the facts are certainly irreconcilable with any other 
 hypothesis. 
 
 In short, while fully admitting the possibility that nerves may 
 act on gland cells, in some way affecting osmosis through them, 
 it appears to me that, in the present state of our knowledge of 
 secretion, the assumption of a particular secretory function of 
 
SECRETION PHYSIOLOGY. 363 
 
 cells, and of special secretory nerves, is unwarranted, unneces- 
 sary, and, in certain particular cases, opposed to the phenomena 
 of the secretion itself. 
 
 SUMMARY OF RESULTS. 
 
 (i) The sympathetic nerve induces salivary secretion by 
 acting on contractile tissue in the glands and thus causing a 
 compression of ducts and alveoli. 
 
 (2) The chorda tympani, or other dilator salivary, secretory 
 nerve probably causes secretion by its dilator action on the blood 
 vessels, thus increasing osmosis. 
 
 (3) The evidence that the chorda tympani acts on the gland 
 cells is open to serious objections, as follows : 
 
 {a) Atropine probably acts directly on the gland cells and 
 capillary endothelium, diminishing their permeability. 
 
 ip) The post-mortem chorda salivary secretion is possibly 
 due to a back flow of blood from the veins owing to a dilation 
 of the capillaries. 
 
 (it) The increased content of organic matter in a secretion 
 coincident with an increased rate of secretion is of little value 
 as evidence of secretory nerves, because (i) saliva is generally 
 not a true solution, and (2) a weak stimulus probably arouses 
 but a portion of the gland. 
 
 id) The evidence derived from the action of nicotine and 
 the degenerated chorda tympani that secretion may ensue on 
 stimulation of the chorda without vaso-dilation is of doubtful 
 value, because of an erroneous method of determining that 
 vaso-dilation had not occurred. 
 
 (4) The sweat glands and the amphibian skin glands, like the 
 salivary glands, receive a double nerve supply and probably pos- 
 sess a double mechanism of secretion, /. e., a muscular and an 
 osmotic. 
 
 (5) Whether secretory nerves exist or whether secretion is 
 ever a function of the gland cell must be considered at present 
 an open question. 
 
 (6) The thoracic lymph flow in dogs reacts to nerve stimula- 
 
364 MATHEWS. 
 
 tion and drugs very similar to pancreatic secretion. It is in- 
 creased by rhythmical stimulation of the vagi after division of 
 the cervical cord and by pilocarpine and chloral hydrate, and 
 decreased by atropine. 
 
 Columbia University, April, 1898. 
 
 VII. LITERATURE. 
 
 1. Adami. On the Nature of Glomerular Activity in the Kidney: 
 Journal of Physiology, VI . 1885. 
 
 2. Adamkiewicz. Ueber Schweisssecretion : Archiv. f. Anat. u. 
 Physiol. {Physiol. Abth.). 1887 and 1880. 
 
 3. Afanassiew and Pawlow. Beitrage zur Physiologic des Pan- 
 creas : Archiv f. d. gesani. Physiologic, XVI. 1878. 
 
 4. Barfurth. Ueber den Bauund die Thatigkeit der Gasteropoden 
 Leber: Archiv f. Mik. Anatomie, XXII. 1883. 
 
 5. Bayliss and HiU. The Formation of Heat in the Salivary 
 Glznds: Jourfial of Physiology, XVI. 1894. 
 
 6. Berkely, John. The Intrinsic Nerves of the Submaxillary Gland 
 of Mus Musculus : The Johns Hopkins Hospital Reports, Vol. IV. 
 Cajal. Barcelona, 1893. 
 
 Huber. Observations on the Innervation of the Sublingual and 
 Submaxillary Glands : Journal of Exp. Medicine. No\. I. 1896. 
 
 Marinescu. Ueber die Innervation der Driiseu der Zungen- 
 basis : Archiv f. Anat. u. Phys. {Phys. Abth.^. 1891. 
 
 DogieL Die Nervendigungen in der Thranendriise der Sauge- 
 thiere : Arch. f. Mik. Anatomie, XLII. 1893. 
 
 Retzius. Biologische Untersuchimgen, III, 1892 ; IV. 1893. 
 
 7. Claude Bernard. Legons sur les proprietes physiol. et les al- 
 terations pathologiques des liquides de I'organisme, II, 1859. 
 
 8. Blochmann. Ueber die Driisen des Mantelrandes bei Aplysia 
 jnd verwandten Formen : Zeitschft. f. Wissensch. Zoologie, 
 XXXVIII. p. 414. 1883. 
 
 r 
 
 g. Brown-Sequard. Journal de Physiol, W. 1859. 
 
 11. Drasch. Beobachtungen an lebenden Driisen mit und ohne 
 Reizung der Nerven derselben : Archv. f. Anat, u. Phys., 1889. 
 (Phys. Abth.). 
 
 12. Dupuy. Journal de Med., XXXVII. 1816. See also Golz. 
 Arch.f d. Ges. Phys., 1875. 
 
SECRETION PHYSIOLOGY. 365 
 
 13. Eckhard. Der Sympathicus in seiner Stellung zum Secretion 
 in der Parotis des Schafes : Eckhara' s Beitrdge, IV. Also Zeit. 
 /. rat. Mcdicin., XXIX. 
 
 14. Ellenberger und Hoffmeister. Beitrag zur Lehre von der 
 Speichelsecretion : Archv.f. Anat. u. Fhys, iSSy. (Si(/>J>.) 
 
 15. Ellis. Plethysmographic and vaso-motor experiments upon 
 Frogs: Jourjial of Physiology, VI. 1885. 
 
 16. Engelmann. Ueber das Vorkommen und die Innervation von 
 contractilen Driisenzellen in der Froschhaut : Arch. f. d. ges. Phys. 
 IV. 1871. 
 
 17. von Frey, Max. Ueber die Wirkungsweise der erschlaffenden 
 Gefassnerven : Ludwig's Arbeiten aits der. Physiol. Anstalt zu 
 Leipzig, XI. 1876. 
 
 18. Giannuzzi. Von der Folge des beschleunigten Blutstromes 
 fiir die Absonderung des Speichels : Ber. d. Konig. Sachs, Gesell- 
 schaft d. Wissench, 1865, p. 68. 
 
 ig. Gottleib. Beitrage zur Physiologie und Pharmakologie der 
 Pankreas Absonderung : Archiv f. exper. Path. u. Phar?nakol. 
 XXXIII. 1893-94. 
 
 20. Grunhagen. Iris u. Speicheldriise : Zeits. f. rat. Medicin^ 
 XXXIII. 1868. 
 
 21. Heidenhain. Beitrage zu der Lehre von der Speichel Abson- 
 derung : Studien des physiol. Lnstituts zu Breslau, IV. 1868. 
 
 22. Heidenhain. Ueber Secretorische u. Trophische Driisen- 
 nerven : Archv. f. d. gesain. Physiology, XVII. 1878. 
 
 23. Heidenhain. Hermann's Handbuch der Physiologie, Bd. V, 
 Leipzig, 1880. 
 
 24. Heidenhain. Ueber die Wirkung einiger Gifte auf die Nerven 
 des Glandula Submaxillaris : Archv. f. d. gesam. Physiologie, V. 
 1872. 
 
 25. Heidenhain.- Einige Versuche an den Speicheldriisen : Archv. 
 f. d. gesam. Physiologie, IX. 1874. 
 
 26. Heidenhain. Versuche und Fragen zur Lehre von der Lymph- 
 bildung : Archv. f. d. gesam. Physiol., XLIV. 1891. 
 
 28. Hering. Ueber die Ursache des hohen Absonderungsdriick in 
 der gld. Submaxillaris : Ber. d. Wiener A had. Bd. LXVI, 
 Abth. III. 1872. 
 
 29. Joseph, M. Ueber Schweiss und Talgdriisensecretion : Archv. 
 f. Anat. u._ Phys. 1891. 
 
 30. Kendall and L#uchsinger. Zur Innervation der Gefasse : 
 Archv.f. d. ges. Physiol., XIII. 1876. 
 
366 MATHEWS. 
 
 31. Krause. Die Speicheldrusen der Cephalopoden : Cenh-albl. f. 
 Physiologic, IX. 1895. 
 
 32. Kudrewetsky. Chemismus der Pankreasabsonderung unter 
 dem Einflusse der Nervenreizung : Archv.f. Anat. u. Phys. 1894. 
 (Phys. At)th.) 
 
 33. Kuhne u. L»ea. Beobachtungen iiber das Secretion des Pan- 
 creas : Stiidien aiis dem Physiol. Instituts zu Heidelberg, II. 1882. 
 
 34. Lang. Text-Book of Comparative Anatomy, II, English 
 Translation. Macmillan, 1896, p. 399. 
 
 35. Langley, J. N. Zur Physiologie der Speichelabsondermig I. 
 Vom Eniflusse der chorda tympani und des n. Sympathicus auf die 
 Absonderung der Unterkieferdriise der Katze : Untersuch. aus dem 
 Physiol. Institut. zu Heidelberg, I. 1878. 
 
 36. Langley, J. N. The Antagonism of Poisons: Journal of 
 Physiology, III. 1881. 
 
 37. Langley. On the Physiology of Salivary Secretion. IV. The 
 effect of atropin upon the supposed varieties of Secretory Nerve 
 Fibres: Journal of Physiology, YS^. 1888. 
 
 38. Langley and Fletcher. On the Secretion of Saliva. In 
 particular the Secretion of Salts in it : Philosoph. Transact, of the 
 Royal Soc, CLXXX, B. 1889. 
 
 39. Langley. On the Physiology of Salivary Secretion. V. The 
 effect of stimulating cerebral secretory nerves upon the amount of 
 saliva obtained by stimulating the sympathetic nerve : Journal 
 of Physiology, X. 1889. 
 
 40. Langley. On the Changes in Serous Glands during Secretion: 
 Journal of Physiology, II. 1879—80. 
 
 41. Langley. On the Physiology of Salivary Secretion, VI: 
 Journal of Physiology,!^. 1890. 
 
 42. Langley. On the Physiology of Salivary Secretion, III. 
 Paralytic Secretion : Journal of Physiology, N\. 1885. 
 
 44. Levy, Max. Ueber Blutfulle der Haut u. -Schwitzen: Deutsche 
 Med. Wochenschr., p. 615, 1892. 
 
 45. Leydig. Die Circumanaldriisen des Menschen : Sitzungsber. 
 d. Wiener Akad., II, p. 329, 187 1. 
 
 46. Luchsinger. Die Wirkung von Pilocarpin u. Atropin auf 
 die Schweissdriisen der Katze : Archv. f. d. ges. Physiol. , XV. 
 1877. 
 
 47. Luchsinger, Hermann^ s Handbuch der Physiologie, Bd. V, p. 
 426. 1880. 
 
SECRETION PHYSIOLOGY. 367 
 
 48. Ludwig. Neue Versuche iiber die Beihilfe der Nerven zu der 
 Speichelsecretion : Zilrischer Natiirf. Gesellschft. Mittheil. 50. 
 Also Zeit. f. rat. Medicin, N. F. I. 185 1. 
 
 49. LfUdwig, C. und Spiess. Vergleichung der Warme des Un- 
 erkieferdriisen Speichels und das gleichzeitigen Carotidenblutes : 
 
 Zeitsch. f. rat. Medicin, N. F. XL 1857. 
 
 51. Macleod. Notice sur I'appareil venimeux des Araneides : 
 Archive d. Biologic, I. 1880. 
 
 52. Marme. Nachber. d. kbn. Gesellsch d. Wiss. zu Gottinge^i, 
 p. 102. 1878. 
 
 53. Mett. Beitrage zur Physiologic der Absonderung, Zweiter 
 Mittheil. Zur Innervation der Bauchspeicheldriise : Archv. f. 
 Anat. u. Phys. {Fhys. Abth.). 1894. 
 
 54. Novi. Ueber die Scheidekraft der Unterkieferdriise : Arch. f. 
 Anat. u. Phys. {Physiol. Abth.'). 1888. 
 
 55. Nussbaum. Ueber den Bau und die Thatigkeit der Drlisen : 
 Archiv f. 7711k. A7iat., 1882. 
 
 56. OstroumofF. Versuche iiber die Hemmungsnerven der Hautge- 
 fasse : Archiv f. d. Physiologic, XII. 1876. 
 
 57. Pawlow. Beitrage zur Physiologic der Absonderungen. Inner- 
 vation der Bauchspeicheldriise : Arch. f. A7iat. u. Phys. {Phys. 
 Abth.). 1893. 
 
 58. Pawlow and S. — Simanowskaja. Die Innervation der 
 Magendriisen beim Hunde : He7-77iann nnd Schwalbe Jahresbe7'icht, 
 1889. 
 
 59. Puelma u. Luchsinger. Zum Verlauf der Gefassnerven im 
 Ischiadicus der Katze : Arch. f. d. ges. Physiol., XVIII. 1878. 
 
 60. Pfluger. Die Speicheldriise, Strieker's Handbuch der Lehre 
 V. d. Gewebe, Leipzig, 187 1. 
 
 61. Rahn, C. Untersuchungen iiber Wurzeln und Bahnen der Ab- 
 sonderungsnerven des Glandula parotis beim Kaninchen : Mit- 
 theil. der Zilricher Natm'for. Gesellschft, d-r^. 
 
 62. Ranvier. Le Mechanisme de la Secretion : Jour7ial de Mi- 
 C7^og7-aphie , p. 150, 1880. 
 
 63. Rawitz. Ueber den feineren Bau der hinteren Speicheldriisen 
 der Cephalopoden : Archv. f. Mik. A7iatoi7iie, XXXIX. 1892. 
 
 64. Ribbert. Die normale u. pathol. Physiologic u. Anatomic der 
 Nierc : Bibliotheca Med. Abth. C. Casscl, 1896. 
 
 65. Schiff. Lemons sur la Physiologic de la Digestion, II, p. 304. 
 1867. 
 
368 MATHEWS. 
 
 66. Schluter. Disquisitiones microscopicae et physiolog. de Glan- 
 dulis Salivalibus : Inaug-Diss. Breslau, 1865. (Quoted from 
 Pfliiger). 
 
 67. von Sobieranski. Ueber die Nierenfunction und die Wirk- 
 ingsweise der Diuretica : Archv. f. exp. Path. u. Pharni., Bd. 
 XXXV. 1895. 
 
 68. Starling, E. H. On the -Mode of Action of Lymphagogues : 
 Journal of Physiology, y^Yl. 1894-5. 
 
 68a. Tiegerstedt. Lehrbuch der Physiologie des Menschen, I, 
 
 Leipzig, 1897. 
 6g. Tschirwinsky. Beobachtungen iiber die Wirkung einige 
 
 phamakolog. , Mittel auf die Lymph Ausscheidung : Arch. f. Exp. 
 
 Path. u. Pharm., Bd. XXXIIL 1893-4. 
 
 70. Unna, P. G. Zur Theorie der Drusensecretion, insbersondere 
 des Speichels : Cent./, d. Med. Wissensch, 1881, p. 258. 
 
 71. Vierheller. Beitrage zur Anatomic u. Physiologie der Parotis 
 desSchafes: Zeit. f. rat. Med. XXXI. 1868. 
 
 72. Vulpian. Comptes rendtis, LXXXVII. 1878. 
 
 73. Van Gehuchten. Recherches histologiques sur I'appareil di- 
 gestif de la larvc dc la Ptychoptera contaminata : La Cellule, VI. 
 1890. 
 
 74. W^eber. Ueber den Bau und die Thatigkeit der sog. leber der 
 Crustaceen : Archv. f. mik. Anatoniie, XVII. 1880. 
 
 75. 'Werther. Einige Beobachtungen iiber die Absonderung der 
 Salze im Speichel : Archv. z. d. gesam. Physiol. XXXVIII. 1886. 
 
 76. Wiedersheim. Text-book of Comparative Anatomy. 
 
 77. von Wittich. Sympathicus und Parotis : Virchow' s Archv. 
 /. Path. Andt. u. Phys., XXXIX. 
 
 78. von Wittich. Virchow^ s Arch. f. Path. Anat. u. Phys., 
 XXXVII. 
 

 Date Due 
 
 
 i»Uk \ Q 
 
 '4a 
 
 
 
 ^6 J« 
 
 > lAin 
 
 
 
 Aiifi •? 
 
 
 
 
 MU»3 f 
 
 1944 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (|) 
 
 
 
 
QP190 
 
 Mathev/s 
 
 M42 
 
■ •