|Ei|[inJ puT][rOgrrirg(pii3fiTinlf ruiirf^ THE LIBRARIES COLUMBIA UNIVERSITY Medical Library lEi rinJ[ruiinirnfruilfr5ilfnnlfriJTJ[]uil^ Digitized by the Internet Archive in 2010 with funding from Open Knowledge Commons http://www.archive.org/details/electromotiveproOOreid [ 335 ] IV. The Electromotive Properties of the Skin of the Common Eel. By E. Waymouth Eetd, Professor of Physiology in University College, Dundee, St. Andrew's University, N.B. Communicated by Professor M. Foster, Sec.R.S. Received November 19, — Read December 15, 1892. Heemann (1) has investigated the electromotive properties of the skins of Fish. His object in so doing was to attempt to determine, by the employment for experiment of a skin usually credited with being bereft of glands, whether the marked " current of rest " exhibited by the skins of Amphibia is with greater proba- bility to be ascribed to glandular processes in accordance with the opinion of T>v Bois Reymond (2), or whether the phenomena are not explained with greater simplicity upon the principles of his own ' Alterations-Theorie.' Du Bois failed to obtain evidence of a " current of rest " in the four kinds of Fish with which he worked (Eel, Tench, Pike, and Perch), a fact, which taken in conjunction with the absence of " glands," satisfied him that in the richly glandular Amphibian skin the source of the E.M.F. must lie in the secreting structures. Hermann himself, previous to his examination of the skins of Fish, shared to some extent the opinion of du Bois, for in a paper published in 1878 (3), he incHnes to the idea, that preparatory processes of glandular origin are the cause of the E.M.F. of the current of rest in the skin of the Frog, but also advances the supposition of a possible contribution from epithelial action at the surface. Finally, as is well known, Hermann demonstrated the presence of an ingoing current of rest in the skins of some ten genera of Fish, but found that its E.M.F. was far less than that exhibited by the Amphibian skin. After noting that substances, the application of which destroys the current of rest in Amphibian skin, cannot be traced microscopically beyond the upper layers of epi- dermic cells, and recalling the fact that an electromotive excitatory change was demonstrated by Bach and Oehler (4), in the skin of the Frog, after complete removal of the current of rest, by the action of corrosive sublimate applied to the outer surface, Hermann concludes that the E.M.F. of the current of rest, and that of the current of action are of different origin. In speaking of the origin of the current of rest, he makes the following statement, upon the strength of his demonstration of such a current in the " non-glandular" skins of Fish, "dass nicht, oder nicht in erster Linie, die Drusen, sondern die Epithelschicht, der Sitz der elektromotorischen Haut- MDCCCXCIII. — B. 31.5.93. 336 PEOFESSOR E. W. REID ON THE ELECTROMOTIVE wirkung ist." Finally he bases his explanation of the source of the E.M.F. of the current of rest of the skins of both Amphibians and Fish upon the axioms of the " Alterations-Theorie." According to this hypothesis, the processes of dying or excitation in the continuity of protoplasm cause the more altered parts to be negative electrically to the less altered, so Hermann says " Nun haben wir aber zunjichst in alien verhornenden Epithelgebilden eine dem Absterben vollig analoge Alteration, welche von aussen nach innen fortschreitet (und durch den Nachwuchs compensirt wird), namlich die Verhornung." Thus in the case of the Amphibian the keratinized superficial cells of the epidei-mis are supposed to form a demarcation surface whose electrical sign is negative to that of the deeper less altered portions of the skin tissue. Analogous to the " keratin-metamorphose " of protoplasm stands in this connection a " mucin-metamorphose," and Hermann remarks " Am Aal und an der Triische kann man direct sehen, wie die ausseren Zellenden unter Mucinbildung zu Grunde gehen." A protoplasmic alteration, then, on the down-grade side of cell life is the explana- tion offered of the source of the E. M. F. of the current of rest of the skin of Amphibian and Fish, and the case of the latter is taken to throw light upon the former, since the complication due to the presence of glands is absent. The explanation for the case of the Fish (mucin-metamorphose) is further supported by a reference to Rosenthal's (5) discovery of an ingoing current in the mucous surface of the stomach and gut of the Frog and Rabbit, and the statement that Oehler has demonstrated a current with similar direction m the integument of the Snail. Is it correct to consider the skins of Fish as being non-glandular ? If the term "gland" be restricted to collections of secretory cells, in the form of definite organs, such as exist in Amphibian skins, then it is certain that, with the exception of the special case of the dermal glands of Myxinoids, " glands" are absent from the skins of Fish. But if one include under the designation of "gland" any cells capable of forming a secretion — calling such cells, as is usual, "unicellular glands" — then, on the other hand, it must be admitted that the skins of Fish are richly glandular. One has but to turn to the histological works of Leydig (6 and 7), F. E. Schulze (8), FcETTiNGER (9), LiST (10), and many others, in order to be convinced that unicellular secretory organs not only exist in the skins of Fish, but are in many cases of a high degree of complexity. In the case of the common Eel, ordinary goblet cells, and also the special "kolben" first described by Kollikeb (11) for the skins of Myxine and Petromyzon, compose more than half the structures of the epidermis ; and, by comparison of the histology of resting and stimulated skins in this Fish, I have convinced myself that both these structures are engaged in the act of secretion of slime, the goblet cells supplying a mucinous matter, and the " kolben" passing into the fine threads that occur in the slime. The details of this process of secretion must be reserved, however, for another communication. The demonstration, therefore, of a PROPERTIES OP THE SKIN OF THE COMMON EEL. 337 current of rest ia the skins of FishesJ is not of necessity a proof of Hermann's explanation, seeing that active secretory processes are at work in such structures. Moreover, Hermann's statement that one " kann direct sehen wie die ausseren Zellenden unter Mucinbildung zu Grunde gehen" is not substantiated by microscopic examination of the skin of the Eel. In the metachromatic staining reaction of thionin, as worked out by Hoyer (12) one has, I am convinced, a reliable microscopic test for the presence of mucin. If sections of Eel-skin, treated with corrosive sublimate, be stained with this substance, it is the goblet cells alone — and these occur in the deep layers of the epidermis as well as near the surface — that give the charac- teristic reddish- violet colour, the surface epidermic cells staining bright blue. I propose in the following pages to attempt to demonstrate that the ciirrent of rest in the case of the Eel is more probably associated with active preparatory processes in its glandular elements, as Hermann originally thought to be the case for the Frog, than with any keratinous or mucinous demarcation surface in accordance with his later writing ; at the same time it will be shown that both negative and positive variations of the current of rest can be elicited by suitable experimental treatment of the removed skin of the Eel. The Eel has been chosen for experiment principally on account of the fact that it can be flayed without the removal of any of the subjacent muscular tissue, but also because it is easily obtained with little injury, and lives well in captivity. Method. Eels, caught in the River Tay without the use of a hook, were transferred as quickly as possible to the laboratory, and kept in running water. Death was effected by transfixion of the medulla, the utmost care being taken not to injure the surface of the skin. Pieces of skin of an area of from 9 to 16 square centims. were removed from the back and sides, and were pinned by means of hedgehog bristles upon a suitable cork frame. The supported piece of skin was fixed in the special moist chamber devised for the electrical examination of membranes by Engelmann (13), and led oflfby the ordinary non-polarisable electrodes, provided with crochet cotton contacts soaked in physio- logical salt solution. The galvanometer (Elliott, 20,000 ohms R.) was arranged with the usual compensating circuit supplied with a Daniell cell of very low resistance. The general plan of circuit is evident from the wood-cut, fig. 1. By means of a paraffin switch, C, it was possible to stimulate the skm by the current of the secondary coil, S, through the electrodes, with the galvanometer circuit broken, and immediately, by turning over the key, to turn the skin current into the compensating circuit. The reverser, R', was introduced for purposes of testing for possible polarisation of electrodes, but with the strengths of stimulating current employed this was not found to occur. The moist chamber was furthermore MDCCCXCIIL — B. 2 X 338 PROFESSOR E. W. REID ON THE ELECTROMOTIVE provided with an inlet and outlet tube, so that gases or vapours could easily be introduced and subsequently removed without any disturbance of the electrical contacts. Fig. 1. D The Electroinodve Force and Direction of the " Current of Rest." [Hermann quotes '003 — "0070 for the value of the E.M.F. of the current of rest of the Eel's skin, while Bayliss and Bradford (14) give "00569 volt for one case. In the course of these experiments I have observed variations from "00072 — 00936D. As regards the direction of the current, both Hermann and Bayliss and Bradford found it to be ingoing. Though this is the rule, and is always the final result, yet it is iiot uncommon to find an outgoing current indicative of inner surface negative to outer immediately after setting up the preparation. With time the direction of this abnormal current reverses to the normal. Even if the current is in the normal direction at the first, its E.M.F. generally continues to rise for some time. The conditions that favour the presence of an abnormally directed current of rest are ; (i.) the examination of pieces of exsected skin immediately after a process of capture in which the skin secretory mechanism has been greatly tried ; (ii.) absence of special care in removal. As regards the source of the E.M.F. of the normally directed current of rest, it is not in difference of reaction between the two surfaces, for the inner surface is normally more alkaline to litmus than the outer, a condition that would, per se, lead to an outgoing current. Moreover, I cannot see that mucin-metamorphosis can be adduced, seeing that microscopic observation shows no evidence of this in the surface scales of the epidermis. It is more probable, as will be seen in the sequel, that the source of the E.M.F, of the current of rest lies in glandular preparatory processes occurring in the secreting elements of the epidermis. The amount of the E.M.F. of the normal current is associated with the vigour of the animal — a fact noted for the Frog by du Bois, Budge (15), and Engelmann. As regards the explanation of the abnormally directed current of rest observed PROPERTIES OF THE SKIN OF TEE COMMON EEL. 339 often immediately after putting up the preparation, and especially in Eels fatigued in tlie process of capture, it is possible that either or both of two causes may contribute to its production : (i.) injury to the inner surface during removal tending to produce an outgoing current ; (ii.) depression of epidermic activity from fatigue, or "shock" of removal, causing a temporary diminution in the E.M.F. of the normal ingoing current. Two contacts upon the outer surface will often give evidence of difference of potential, but I have never read more than -00120, about one-fifth of the average E.M.F. between inner and outer surface. This difference between two points on the outer surface is to be attributed to an unequal distribution of secreting elements (especially the " kolben ") often observed in microscopic sections. Mechanical stimulation in the region of one of the externally placed electrodes causes an increase of the negativity of that electrode (see p. 359), a fact much against the mucin-metamorphosis theory of origin of E.M.F. The fact of proximity of an electrode to a gland mass favouring its negativity is demonstrable on the outer surface of the skin of the Toad, where with two thread electrodes it is found that there is greater difference of potential when one is on a " gland wart " and the other between two, than when both are over glands. Here, however, mechanical stimulation leads to diminution of negativity. Two points on the inner surface of the skin of the Eel show no appreciable difference of potential. I am of opinion, therefore, that it is probable that the source of the E.M.F. of the current of rest of the skin of the Eel is in the secretory activity of its unicellular glands.— March 2, 1893.] Action of Carbonic Acid Gas. If, as would appear to be probable, the E.M.F. of the current of rest of the skin of the Eel, is the result of secretory processes occurring in its glandular elements instead of a " mucin-metamorphose " of the ordinary epidermic cells, it should be possible by experimental treatment to obtain increments and decrements of the E.M.F. according as the active processes are stimulated or depressed. The first experiments in this direction were made with carbonic acid gas. Engelmann (13) obtained enormous falls of the E.M.F. of the skin of the Frog by the use of this reagent, and also saw that upon replacing the COg by air the original level of potential was rapidly regained. He associated this fall of E.M.F. with the contraction of the skin glands, which he had previously noticed to occur when the skin or membrana nictitans was subjected to the action of this reagent. In the skin of the Eel there are no contractile elements similar to the muscle cells which Engelmann (17) describes as sheathing both varieties of skin glands in the Frog, so that it becomes a rather simpler matter to determine upon what elements of the skin a reagent affecting E.M.F. exerts its action. 2x2 340 PROFESSOR E. W. REID ON THE ELECTROMOTIVE The CO3, prepared from marble by the action of HCl, was washed with water and saturated solution of sodic carbonate, and then passed as required into the gas chamber containing the skin in contact with its electrodes. As has been ali-eady mentioned the outer surface of the skin of the Eel immediately after removal may be either negative or positive to the inner surface, though, if time be given, it is the rule that the condition of outer surface negative to inner is that which finally obtains. It is therefoi'e necessary to observe the action of CO3 upon the E.M.F. of the skin current in both of these conditions. The following two experiments are illustrative. In Experiment A the gas was applied to a piece of skin in the normal condition, i.e., with outer surface negative to inner, wliile in Experiment B a case was selected where, at the commencement, the outer surface was positive, though, as is the rule, it was gradually passing towards the normal state. N.B. — In these and in all subsequent experiments — A north galvanometer deflection (N.) signifies outer surface negative to inner ; A south galvanometer deflection (S.) signifies inner surface negative to outer. Experiment A. Freshly removed Eel's skin led off from inner and outer surfaces. Outer surface negative to inner. Deflection of galvanometer N. The numbers in the right hand column indicate compensator degrees of which 1 = -000008 D. 11.25 A.M. 250 N. 11.28 200 N. 11.30 180 N. CO3 introduced into gas chamber. 11.301 A.M. 11.31^ 120 s. Eeversal of current. Air circulated and CO3 expelled. 11.34 A.M. 11.39 90 N. Return to normal dii-ection of current. 11.44 130 N. 11.49 100 N. 11.58 70 N. COa again introduced. 12.0 noon. 70 S. Reversal of current. Air circulated aud CO3 expelled. 12. 3 p.m. 90 N. Return to normal direction of current. 12. 8 40 N. 12.80 PROPEETIES OF THE SKIN OF THE COMMON" EEL. 341 Experiment B. Freshly removed Eel's skin led off" from inner and outer surfaces. Outer surface positive to inner. Deflection of galvanometer S. The numbers in the right-hand column have the same unit as in Experiment A. 2.47 P.M. 440 S. 2.51 ■ 350 S. 2.55 320 S. 2.5(i CO3 introduced into the gas chamber. 2.57 890 S. 2.581 260 S. 3. 4 250 S. Air circulated and CO2 expelled. 3.10 P.M. 120 S. 3.11 COo again introduced. 3.13 150 S. 3.15 190 S. Air circulated and CO.3 expelled. 3.17 p.m. 120 S. 3.18 50 S. 3.20 25 N. Reversal. 3.25 CO3 again introduced. Second reversal occurred. Deflection now again S. 3.26 P.M. 200 S. Air circulated and COj expelled. 3.28 P.M. 150 S. 3.30 100 S. 3.32 SOS. 3.35 10 S. 8.40 100 S. 8.43 150 S. 8.46 170 S. 3.50 200 s. With reference to Experiment A it will be noticed that the exposure of the skin with normal direction of current to the action of CO^ leads to a diminution of the E.M.F., and finally to a temporary reversal of the direction of the current. A recovery occurs upon the admission of air, but it is not in this case complete, for, as will be seen by reference to the earlier stages of the experiment, a general fall of E.M.F. is taking place, hastened later on no doubt by the action of the gas. In 342 PROFESSOR E. W. REID ON THE ELECTROMOTIVE fact the }3recipitations of E.M.F. induced by the CO2 are seen to be imposed upon a gentle curve of descent. If, on the other hand, Experiment B be examined, a case where at the com- mencement the outer surface is positive to the inner, the following points will be evident. In the first place the negativity of the inner surface is diminishing, a fact that has already been alluded to in a previous section of this paper. In the second place the exposure of the skin to the action of the CO^ leads, in contradistinction to the former normal case of Experiment A, to an augmentation of the E.M.F. of the current in contra-normal direction, but, as before, upon the admission of air the effect is cut out, and the ordinary diminution of negativity of the inner surface continues. In the thii'd place it is seen that though the skin lay for 8 minutes in CO3, at the first application, recovery from the effect was noticed within 2| minutes, ivMle the skin was still subject to the action of the gas, and the effect was relatively a small one, while in the third application, though the action was only allowed to continue for 1 minute a far greater effect was produced, suggestive of a diminution of vital resistance, though the rapid admission of air excluded any observation as to whether recovery could have still occurred while the gas was still present. Finally it is evident that the repeated exposure to CO3 has put an end to the general tendency to replacement of negativity of inner surface by negativity of outer, for the state of matters at the end of the experiment is the reverse of that at the beginning, viz. , the negativity of the inner surface is now apparently increasing. It has been suggested that in a piece of freshly removed skin one has two opposing sources of E.M.F. at the two surfaces, the one, due to injury of the inner surface, and tending to diminish with time, the other due to the activity of secretory protoplasm situated nearer the outer surface, finally also tending to diminish, yet capable of increase, either as a result of recovery from shock of removal or on account of excitation, as will be shown later. Upon which of these sources does COg exert its action, and of what nature is that action ? It is not probable that CO3 has any action upon the inner surface, for as far as is known CO2 is without effect upon the electrical condition of connective tissue fibre. Du Bois (18) has shown that short exposure of muscle to CO3 has no appreciable effect upon the demarcation current, an experiment that I have often repeated, leading off from the tendon and belly of the gastrocnemius of the Frog, with con- firmatory result. In this case, if the current is not affected, there is action upon neither muscle nor tendon, the absence of action upon the former being probably associated with the fact that the muscle tissue is, so to speak, habituated to the presence of this gas. This is not, at any rate to like extent, the case with the protoplasmic structures at the outer surface of the skin, and it is reasonable to suppose that the action of CO3 is almost exclusively upon this surface. With the assumption that CO2 acts principally, if not exclusively, upon the PROPERTIES OF THE SKIN OF THE COMMON EEL. 343 structures of the epidermis, and that its action is, by lessening vital action, to cause a diminution of the E.M.F. dependent upon such action, the facts arising out of experiment became easily explicable. In the normal case (Experunent A) the application of COg causes, first a reduction of the E.M.F. of the outer surface to such a degree that it is balanced by that of the inner, so that there is no current, and by a continuance of its depressant action, allows the E.M.F. generated at the inner surface (probably as the result of injury during removal) to gain the upper hand, with the result that a current sets in in the reverse direction to normal. Aeration brings back activity to the protoplasm of the cells of the outer surface, and with renewed action the negativity of injury of the inner surface is again over-compensated, leading to a re-reversal of current back to normal. Tiu-ning to the second case (Experiment B), where, whether from interference with the activity of the cells at the outer surface, or greater injury in removal to the inner, the inner surface is negative to the outer, the same explanation will hold good. The balance is here in favour of the E.M.F. of the inner surface, hence any reagent, such as CO3, diminishing the activity of the protoplasm of cells at the outer surface that generate an opposing E.M.F., will lead to an increase in the original outgoing current, followed by a decrease when air is readmitted. In this case too a more permanent depression seems to have been produced by the action of the gas, for after the third admission of air, the general recovery that had been progressing throughout the experiment is no longer visible, but, on the other hand, has given place to a condition in which the E.M.F. of the inner surface is apparently gaining the upper hand. It is interesting to note that in the early part of the experiment the cells of the outer surface are far more resistant to the action of the CO3 than later, for recovery occurs while the skin is actually in the gas. The explanation offered then of the action of CO3 upon the current of the skin of the Eel, is that by depressing the activity of protoplasmic structures in the epidermis (presumably the secretory cells) it lessens the E.M.F. constantly being generated therein by vital processes. Action of the Vapour of Chloroform. The effect of chloroform vapour upon the E.M.F. of the skin of the Frog was observed by Engelmann (13) to be very similar to that of CO3. I have repeated Engelm Ann's experiments, and am able to confirm the results he obtained. In the experiments with the skin of the Eel the vapour was introduced into the gas chamber containing the skin by means of gentle pressure upon a rubber ball connected with a large vessel containing a sponge soaked in the drug. The effect of chloroform vapour upon the E.M.F. of the skin of the Eel is of the 3i4 PROFESSOR E. W. REID ON THE ELECTROMOTIVE same nature as that exerted by COo, but, as might be expected, the action of chloroform is the more powerful of the two. If the outer surface of the skin is negative to the inner (normal condition), chloroform vapour reduces the E.M.F. of the ingoing current till the two surfaces are equipotential, and, if the action be allowed to continue, the ingoing is replaced by an outgoing current. The replacement of the chloroform vapour by air causes a diminution of the outgoing current, and, if the action of the vapour has been short, the normal con- dition of an ingoing current is soon again attained. On the other hand, should the inner surface of the skin be negative to the outer at the commencement of the experiment, then, as with COo, the immersion in chloroform vapour leads to an increase in the E.M.F. of the outgoing current. The following experiment is illustrative of the action of chloroform upon the E.M.F. of a piece of skin with normal direction of current : — Experiment C Freshly removed Eel's skin led off from inner and outer surfaces. Outer surface negative to inner. — Deflection of galvanometer N. The numbers in the right-hand column indicate compensator degrees, of which 1 = -00008 D. 12.32 p.^i [. 100 N. 12.35 140 N. 12.38 150 N. 12.41 180 N. 12.44 200 N. 12.47 220 N. 12.49 Chloroform vapour passed into gas chamber. 12.50 50 N. 12.50 L Reversed current. Deflection S. 12.511 170 S. 12.52 Air admitted. 12.53^ Skin surfaces equipotential. 12.54 Ee-reversal. Deflection N. 12.561 90 N. 12.58 110 N. 1.0 120 N. It is to be noticed that, though the E.M.F. of the ingoing current is rising rapidly at the commencement of the experiment, the immersion of the skin in chloroform vapour causes a precipitate fall, leading finally to a complete reversal of current. PROPERTIES OF THE SKIN" OF THE COMMON EEL. 345 Recovery occurs upon the admission of air, but the E.M.F. of the outer surface never attains its original maximum, though the skin was only in the vapour for 3 minutes, and was perfectly fresh. The rapidity of recovery upon removal of the vapour is as striking in this case as it is with COj. ■ By prolonging the action of the vapour, however, the recovery is found to be slower, and may be delayed for some time after the admission of air. On the other hand, if the action of the vapour is pfolonged even further, recovery may actually occur while the skin is still subjected to it. In the following experiment the skin was left in chloroform vapour for successive periods of 5, 10, and 30 minutes, and it will be seen that, though recovery occurs immediately upon admission of air in the first case, in the second case a period of about 3 minutes elapses, while, in the third case, after the maximum eflfect has been produced, a period, daring which there is no change of potential, occurs, and, finally, recovery commences after some 12 minutes' immersion, while the skin is still subject to the action of the vapour. Experiment D. Freshly removed Eel's skin led off from inner and outer surfaces Outer surface positive to inner. Deflection of galvanometer S. The numbers in the right hand column indicate compensator degrees of which 1 = -000008 D. 3.42 P.]M [. 500 S. 3.45 150 S. 3.4(5 80 S. 3.49 200 N. Reversal. 3.51 410 N. 3.521 450 N. 3.54 420 N. 3.55 Chloroform vapour applied. 3.57 200 N. 4.0 50 N. Air admitted. 4.3 P.M. 140 N. 4.6 130 N. 4.9 80 N. - 4.10 Chloroform vapour applied, 4.11 Faint N. 4.14 50 S. Reversal. 4.17 120 S. 4.20 190 S. Air admitted. MDCCCXCIII. — B. 2 Y 346 PROFESSOR E. W. REID ON THK ELECTROMOTIVE 4.23 P.M 220 S. 4.26 ISO S. 4.29 150 S. 4-32 145 S. 4.33 Chloroform vapour applied. 4.35 150 S. 4.38 180 S. 4.41 230 S. 4.44 230 S. 4.47 220 S. 5.3 160 S. If chloroform vapour be applied to a skin whose cui'rent is in the contra-normal direction, i.e., outgoing, the effect is similar to that already seen in the case of COg. Under these conditions an increase in the outgoing current takes place attributable to a diminution in the E.M.F. of the opposing ingoing current. A case of this is seen in the commencement of Experiment O, p. 362. If the vapour of chloroform be applied exclusively to the outer or inner surface of the skin, it then becomes evident that it is upon the former almost exclusively that its action is exerted. In the following experiment a stream of chloroform vapour was gently directed upon the surface of the skin in the region of the electrode. A blank experunent with electrodes only showed that there was no appreciable galvanometer deflection from cooling of electrodes. Experiment E. Freshly removed Eel's skin led off from inner and outer surfaces. Outer surface negative to inner. Deflection of galvanometer N. The numbers in the right hand column indicate compensator degrees, of which 1 = 000008 D. 2.18 P.M. 255 N. 2.20 265 N. Chloroform vapour applied to Inner surface. 2.21 P.M. 270 N. 2.22 270 N. Chloroform vapour applied to Outer surface. 22.3 P.M. 250 N. 22.4 270 N. 22.5 268 N. Chloroform vapour applied to Inner surface. 2.26 P.M. 270 N. 2.27 268 N. PROPERTIES OF THE SKIN OF THE COMMON EEL. 347 Chloroform vapour applied to Outer surface. 2.28 P.M. 225 N. 2.29 270 N. The results then of the action of chloroform vapour upon the electromotive pheno- mena of the current of rest of Eel's skin, seem to be capable of the same explanation as that offered in the instance of the action of CO^, viz., that the reagent leads to a diminution of the activity of the processes going on in the seci-etory structures of the epidermis, with a concomitant fall in E.M.F. It must here be mentioned that chloroform vapour may act in quite an opposite manner upon the skin of an intact Eel to that in wHich it works upon a piece of removed shin. ' A reflex copious secretory action occurs when a pithed Eel is sub- jected to the vapour, and the histological features of the skin of such an animal give clear indications of the process ; these features are absent when the chloroform is applied to exsected skin. It seemed of interest in connection with the action of chloroform vapour upon the skin to test the effect produced upon the demarcation current of muscle, for here, unlike the case of COo, which is hardly a foreign substance to muscle, one would expect to get a marked result. The results lend some support perhaps to the theory of the action of chloroform upon the skin of the Eel just stated, for it is found that an increase in the positivity or decrease of the negativity of a contact upon the surface of a muscle occurs as the result of exposure to chloroform vapour. Three cases are quoted below in illustration. In the first, a muscular surface made negative by section became positive when the muscle was surrounded with chloroform vapour. In the second, the contact on the belly of a removed muscle which was positive to that on the tendon, became more so from treatment with chloroform vapour, but recovered completely later. In the third case, an instance is taken where the belly of the muscle was weakly negative to the tendon, and in this case a reversal and development of a strong opposing current with belly positive to tendon resulted from the action of the vapour. These results were obtained by the application of a vapour weak in chloroform. If, on the other hand strong vapour is pumped directly on to the muscle, the results are not constant, for injury currents from coagulation of the muscle substance are then originated. (i.) Gastrocnemius of Frog led off from artificial section and tendinous end. 10.38 A.M. E.M.F. of Demarcation current '03032 D. Cut end negative. Deflection N. Chloroform vapour passed into the gas chamber. 10.39 a.m. Current reversed. Cut end positive. Deflection S. 2 Y 2 348 PKOFESSOR E. W. RETT) ON THE ELECTROMOTIVE Air circulated, 10.42 A.M. E.M.F. of reversed current -0356 D. 10.55 „ „ -0312 D. (ii.) Gastrocnemius of Frog led off from longitudinal surface and tendinous end. 11.5 a.m. E.M.F. of Demarcation current -0184 0. Muscle surface positive. Deflection S. 11.6 E.M.F. of Demarcation current -0168 D. Cbloi'oform vapour passed into the gas chamber, 11.7 A.M. E.M.F. of Demarcation current '0280 D. Deflection S. 11.8 „ „ -0192 0. 11.9 „ „ -0168 D. (iii.) Gastrocnemius of Frog led off from longitudinal surface and tendinous end. 11.20 A.M. E.M.F. of Demarcation current -004 D. Muscle surface negative. Deflection N. Chloroform vapour passed into gas chamber, 11.21 A.M. Current reversed. Muscle hecomes positive. Deflection S. E.M.F. -0368 D. 11.211 „ -0312 0. Deflections. 11.22 „ -0296 D. 11.27 „ -0232 0. It is evident that in all the above cases the action of the chloroform vapour was to cause diminution of negativity or increase of positivity of the muscular substance, i.e., the same efiect as that produced upon the outer surface of the Eel's skin. Excitation. In the above pages, the experiments quoted have dealt solely with depressions of the value of the E.M.F., generated at the outer surface of the skin. It is now neces- sary to turn to the question whether it may not be possible to obtain increments of E.M.F. as a result of excitation. Excitatory electrical variations in glandular structures have long been known. Valentin (19), in 1861, and Roeber (20), in 1869, demonstrated the fact for the skin of the Frog. Hermann and Luchsinger (21), in 1878, did the same in the case of the tongue glands of the Frog and the sweat glands of the paw of the Cat, Luch- singer (22), in 1880, in the glands of the snout of the Pig, Goat, and Cat, and Bayliss and Beadforb (16), in 1887, in the salivary glands of the Cat and Dog. The case of the skin of the Frog has been more frequently examined than the others, but the results have proved most conflicting at the hands of the several observers PROPERTIES OP THE SKIN OP THE COMMON EEL. 349 The conditions here which appear to affect the nature of the electrical variation upon excitation of the cutaneous nerves are numerous. They are given as follows by the different authors. The magnitude of the original current of rest (RoebeRj Hermann, Bach, and Oehler). The state of humidity of the surface of the skin (Roeber and Engelmann). The strength of the stimulus (Hermann, Engelmann, Bach, and Oehler, and Bayliss and Bradford). The temperature of the skin (Bach and Oehler). The time that has elapsed since the commencement of the experiment (Bach and Oehler). Finally, the condition of the animal, with special reference to that obtaining during the breeding season (Bayliss and Bradford). On the whole, perhaps, the mass of the evidence seems to point in favour of a diphasic variation as a result of excitation of the skin of the Frog, the first phase being negative, and the second positive. Of these two phases, the second, or positive, appears to be the more marked, the first being often spoken of as a "Vorschlag," though under various conditions of experiment the reverse is the case, so that Bayliss and Bradford (14, p. 223) observe, "It is scarcely possible to speak of a normal excitatory variation." My own experiments upon the excitatory variation of the current of the Frog's skin are as yet too incomplete to justify any entrance upon a discussion of the nature of the vaiiation in this paper ; it appears, however, that the chief difficulty in predicting the variation which will follow a stimulus is associated with the great variability in the states of conti'action and expansion of the skin glands, as demonstrated by Engelmann (17). In the lingual glands of the Frog, Hermann and Luchsinger noted, upon stimu- lation of both the glosso-pharyngeal and hypoglossal nerves, a strong positive electrical vai'lation of the rest current, interrupted by a negative phase so short in duration that the positive phase outlived it, and was still able to make itself subsequently evident. In the Cat's paw, the same observers saw that a current directed from the outer to the inner surface of the skin (corresponding to a "positive variation") was developed, as a result of excitation of the sciatic nerve. In the glands of the snout of the Pig, Goat, Dog, and Cat, Luchsinger also observed a positive variation of the current of rest. The observations of Bayliss and Bradford upon the variations of the current of rest of the submaxillary glands of the Dog and Cat led them to the conclusion that " the sign of the electrical disturbance varies with the nature of the secretion, as measured by its amount and its viscidity," so that the excitation of a nerve trunk in which "secretory" fibres predominate leads to a positive variation of the rest current, whereas the stimulation of one containing an excess of "trophic" fibres leads to a negative variation. In the glands, then, of the paw of the Cat and the snout of the Pig, Goat, Dog, and Cat, the excitatory variation observed has been purely positive, while in the Frog's tongue it has been mainly positive, but interrupted by an interpolated negative phase. In the case of the skin of the Frog the evidence is not perfectly clear, but of the 350 PROFESSOR E. W. REID ON THE ELECTROMOTIVE two phases of the variation the positive appears to be tlie more marked. Finally, in the Mammalian salivary gland, there VFOuld appear to be a definite connection between the nature of the nerve fibres stimulated and that of the electrical change. In experiments upon excitation of the skin of the Eel it is unfortunately impossible to make use of indirect stimulation by cutaneous nerves. This fact is occasioned b}^ the absence of a large subcutaneous lymph space, such as exists in the case of the Frog and Toad, so that a "nerve-skin preparation" is not feasible. It is, therefore, necessary to resort to direct stimulation, which may be electrical, thermic, or mechanical. In the circuit represented in fig. 1, p. 338, it is evident that, by means of the switch C, induction shocks can be applied to the skin through the leading-ofi" electrodes, and the current of the skin subsequently passed into the previously compensated galvano- meter circuit. With carefully made " non-polarisable " electrodes, and by reversals of the direction of the stimulating current, it is possible to assure oneself of freedom from error due to polarisation. Whatever method of stimulation is employed, it is found, firstly, that the skin of the Eel is excitable ; and, secondly, that the electrical variation is one indicative of increase in the negativity of the outer surface, i.e., a positive variation, if the resting current is in the normal direction. Stimulation with Single Induction Shocks. When a single induction shock is passed through the skin and the electrodes sub- sequently connected with the galvanometer, the following points are observed. Firstly, an electrical variation, indicative of increased negativity of the outer surface, occurs ; secondly, the excitatory state rapidly reaches its maximum, falls quickly at first, and then very slowly. Thirdly, the amount of the primary variation is dependent upon the strength of the stimulus. These points are illustrated by the following example : — Experiment F. Freshly removed Eel's skin, led off from outer and inner surfaces. Outer .surface 7iegatioe. — 'North deflection of galvanometer. Stimulation with break induction shock. Coil 15 centims. Positive excitatory variation. N. deflection. Reversal of direction of stimulating current did not affect variation. Make shock at 15 centims. coil : no effect. Stimulation tvith coil ai 10 centims. — Break shock : larger deflection than before. Make shock : no effect. Coil at 5 centims. — Make shock : effective. PROPERTIES OF THE SKIN OF THE COMMON EEL. Stimulation with Coil at 5 centims. 351 Make shock. Break shock. Seconds after G-alvanometer Seconds after Galvanometer stimulus. reading. stimulus. reading. 6 U 5 45 5 230 10 30 20 120 • 20 20 30 60 30 15 40 50 40 12 50 i 40 50 10 60 1 35 60 8 70 1 30 70 8 90 22 90 7 100 1 20 110 6 110 1 17 130 3 130 j 15 150 2 150 11 200 180 8 230 5 260 3 300 The rapid arrival of the E.M.F. of the excitatory variation at its maximum is remarkable, and, as far as my own experience goes, considerably more marked than in the case of the Frog's skin directly excited. The slowness, however, of the subse- quent decline agrees closely with the phenomena of the variation of the Frog's skin, as already noted by Hermann and Bayliss and Bradford. In the case of the sub- maxillary gland, Bayliss and Bradford mention the occurrence of a quick maximum and slow decline of potential as a result of chorda stimulation. As regards the length of the latent period, I have as yet been unable to make any exact measurements ; all that can be here stated is that it is shortened by increasitig the strength of the stimulus — a fact already noticed by Engblmann in the case of the skin of the Frog. The relation, too, between the magnitude of the variation and the strength of the stimulus evident in Experiment F. (where the break shock gave a deflection of 230° and the make only 45° with coil at 5 centims., while at 10 the make shock was ineffectual), has been observed by both Boeber and Engelmann in the case of the Frog's skin, the former using indirect stimulation by means of the cutaneous nei'ves, the latter stimulating directly through the leading-ofi" electrodes. 352 PROFESSOR E. W. REID ON THE ELECTROMOTIVE Stimulation by Faradisation. The excitatory effect following faradisation of the skin is, as would be expected, far more marked than that produced by the application of a single induction shock. The increase of E.M.F. upon stimulation, as measured by the method I have used, is, of course, no criterion of that actually developed in the skin, on account of the short circuiting that occurs in the skin structures themselves, but the percentage value in terms of that of the current of rest is of interest. Some observations of the increase of the E.M.F. upon faradic stimulation of the skin of the Eel will be found in Table I. Table I.- -Increase of E.M.F. of the Skin of the Eel as a Result of Faradic Excitation. I. II. III. IV. V. VI. Number. E.M.F. of cuvrent of rest. Increase of E.M.F. following excitation. Percentage increase of E.M.F. in terms of tliat of current of rest. Stimulus. Remarks. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 •00128 D •00320 D 00212 D •00248 D •00276 D ■00264 D •00280 1) •00288 D •00292 D ■00116 D •00112 D •00076 D •00120 D •00116 D •00032 D •00032 D •00040 I) •00044 D •00036 D •00040 D •00048 D •00068 D •00056 D •00112 D •00060 D •00144 D ■00036 D ■00120 D 25^0 10^0 18^8 17^7 13^0 151 17^1 23-6 191 96-5 53^5 189^4 30^0 1034 Coil 15 for 5" Coil 15 for 15" Coil 20 for 5" Coil 15 for 5" Coil 20 for 6" j Same Eel. ► Same Eel. J >Same Eel. > Same Eel. The method employed in obtaining these measurements, viz., compensation as rapidly as possible after stimulation, must give a value considerably below the maximum, for, as has been already noted, the rise of E.M.F. on stimulation is sudden and at first declines rather rapidly ; but as will be seen in Case 12 the E.M.F. of the action current may almost double that of the current of rest ; this is, however, very exceptional. Engelmann observed a variation of 25 per cent, to 30 per cent, in the E.M.F. of the Frog's skin upon stimulation with a single induction shock and a variation of 40 per cent, to 50 per cent, with rep)eated stimuli ; these variations were, however, of the nature of falls of the E.M.F. of the current of rest. PROPERTIES OF THE SKIN OF THE COMMON EEL. 353 Bayliss and Bradford state that the E.M.F. of the positive phase of the vai'Iation of the Frog's skin current may exceed '022 volt, but since they give no relative values for the E.M.F. of the current of rest this number is of little value, though taking "1 volt as an average E.M.F. for the current of rest, an increase of 22 per cent, is a possibility. In the case of injured muscle, Sanderson and Gotch (23) have demonstrated that the E.M.F. of the current of action may exceed considerably that of the demarcation current. Tn the gastrocnemius of vigorous Frogs they observed that the E.M.F. of the action current might be more than double that of the current of rest ('084 D as against "04 D) ; in the sartorius, however, though the E.M.F. of the action current undoubtedly exceeds that of the demarcation current, the ratio is not so large as in the case of the gastrocnemius. These results, however, are not comparable with those stated in Table I., for not only were they gained with a single induction shock as stimulus, but the method of measurement was far more accurate than that which I have employed. If stimulation be applied to a piece of Eel's skin, which is in the condition of outer surface positive to inner, the variation of the current of rest is negative instead of positive, for, as has been stated above, excitation always results in an increase in the negativity of the outer sui-face. An instance of this is given in the following experiment. Experiment G. Freshly removed Eel's skin, led off from inner and outer surfaces. Outer surface jwsitive to inner. Deflection of galvanometer S. The numbers in the right-hand colimm indicate compensator degrees, of which 1 = -000008 D. 8.50 P.M. 8.51 8.52 8.53 8.57 Stimulation, coil 15 for 5". 8.58 P.M. 9.0 Stimulation, coil 15 for 5''. 9.iip.M. 9.2 9.3 9.4i 9.5 MDCCCXCIII. — B. 350 S. 305 S. 280 S. 255 S. 195 S. N. variation, i.e., negative. 165 S. 135 S. N. variation, i.e., negative. 95 S. 106 S. 95 S. 85 S. 78 S. 2 z 354 PROFESSOR E. W. REID ON THE ELECTROMOTIVE Stimulation, coil 15 for 5".* N. variation, i.e., negative. 9.6 P.M. 49 S. 9.7 55 S. 9.8 -49 S. 9.9 35 S. The reductions of E.M.F. which folloM^ excitation in this experiment are obviously the result of the excitatory increase of the E.M.F. of the ingoing current, v^^hich at the time the experiment was commenced was below that of the outgoing current, though evidently gradually gaining the upper hand, as is the general rule. When, on the other hand, the skin is in the normal condition, i.e., with outer surface negative to inner, the variation upon faradisation is invariably, as far as I have seen, of a positive nature. The following experiment is typical of the normal variation upon faradisation. Experiment H. Freshly removed Eel's skin led off from inner and outer surflices. Outer surface negative to inner. Deflection of galvanometer N. The numbers in the right-hand column indicate compensator degrees, of which 1 =: -000008 D. 3.0 P.M. Stimulation, coil 20 for 15". 3.1 P.M. 3.4 3.7 3.10 3.13 3.16 Stimulation, coil 20 for 5". 3.17 P.M. 3 17J^ 3.201 . ■ . . ^-^^^ Stimulation, coil 15 for 5".* 3.24| P.M. 3.25 3.28 3.31 145 N. N. variation, i.e., positive. 285 N. 195 N. 195 N. 185 N. 170 N. 140 N. N. variation, i.e., positive. 215 N. 185 N. 155 N. 95 N. N. variation, i.e., positive. 275 N. • 225 N. 210 N. 185 N. Indicates reversal of direction of stimulating current. PROPERTIES OF THE SKIN OP THE COMMON EEL. 355 The quick rise of E.M.F. upon excitation and the slow return to the previous value are clearly seen in this experiment, and also the dependence of the magnitude of increase of the E.M.F. upon the strength of the stimulus. Since RoEBER had noticed the occurrence of fatigue phenomena in the case of the excitatory variation of the skin of the Frog, I deemed it of interest to search for evidence of the fact in the case of the Eel. Pieces of skin, the E.M.F. of w^hose current of rest was increasing with time, were chosen for experiment, since, if any evidence of fatigue were forthcoming as a result of repeated excitation, it would obviously be more convincing in such instances. The following experiment is an instance of an attempt to produce a condition of fatigue. Experiment K. Freshly removed Eel's skin led off from inner and outer surfaces. Outer surface negative to inner. Deflection of galvanometer N. The numbers in the right-hand column indicate compensator degrees of which 1 = -000008 0. 1.35 P.M. 125 N. 1.37 135 N. 1.39 155 N. 1.41 180 N. 1.43 190 N. 1.45 200 N. 1.47 215 N. 1.49 225 N. 1.51 245 N. 1.53 255 N. 1.55 265 N. Stimulation, coil 15 for 5". N variation, i.e.. , positive. 1.55^ P.M. 315 N. 1.57 295 N. 1.58 300 N. 1.59 295 K 2.1 310 N. Stimulation, coil 15 for 5". 2.1^ P.M. 2.3 2.5 2.7 N variation, i.e.. 365 N. 350 N. 345 N. 345 N. , positive. Stimulation, coil 15 for 5". 2.7i P.M. N variation, i.e., 390 N, , positive. 2 z 2 356 PROFESSOR E. W. REID ON THE ELECTROMOTIVE 2.9 P.M. 375 N. 2.11 370 N. Electrodes moistened. 2.13 P.M. 330 N. Stimulation, coil 15 for 5". N variation, i.e., positive. 2.131 P.M. 380 N. 2.14 360 N. 2.15 350 N. 2.17 350 N. Stimulation, coil 15 for 5". N variation, i.e., positive. 2.l7iP.M. 410 N. 2.19 375 N. 2.21 360 N. 2.23 360 N 2.25 360 N. Stimulation, coil 15 for 5". N variation, i.e., positive. 2.25^ P.M. 445 N. 2.26 405 N. 2.27 375 N. 2.29 370 N. 2.31 370 N. 2.32 365 N. 2.35 365 N. 2.37 365 N. Stimulation, coil 15 for 15", N variation, i.e !., positive. 2.371 P.M. 435 N. 2.39 410 N. 2.41 400 N. Stimulation, coil 15 for 5". N variation, i.e., positive. 2.41^ P.M. 435 N. 2.43 410 N. 2.45 405 N, Stimulation, coil 15 for 5". N variation, i.e.. , positive. 2.45|- P.M. 445 N. 2.47 425 N. 2.49 415 N. Stimulation, , coil 15 for 60" . N variation, i.e., positive. 2.51 P.M. 470 N. 2.53 440 N. Stimulation. , coil 15 for 60" . N variation, i.i ;., positive. 2.54i P.M. 490 N. PEOPERTIES OF THE SKIIST OF THE COMMON EEL. 357 2.55 P.M. 475 N. Coil now put to 0, and stimulation continued for 60". E.M.F. of current of rest SOON. Stimulation, coil for 10". N variation, i.e., positive. ^ minute later 510 N. It is noticeable in this experiment that in spite of the repeated stimulation, the E.M.F. of the current of rest continues to rise, and during the early period of the experiment, at any rate, there is no evidence of any diminution in the amount of the positive response to excitation. The amount of the variation is a little reduced in the case of the eighth and ninth excitations, in spite of the longer period of stimula- tion ; but it is not til] the skin has been subjected to stimulation for a period of sixty seconds, with the coil at zero, that any marked diminution in the excitatory variation becomes evident, though even then no faU in the E.M.r. of the current of rest occurs. It would appear then, that the skin of the Eel is remarkably resistant as regards fatigue. Whether the continuation in the rise of E.M.F. throughout this experiment is in part due to some beneficial effect of the stimulus, somewhat of the nature of the rise in the E.M.F. of the rest current of the submaxillary gland noted by Bayliss and Bradford after stimulation of the sympathetic, it is difficult to decide, for in this case the E.M.F. of the current of rest was rising before the stimulation was com- menced. I have attempted to convert a case of falling E.M.F. of current of rest into one of rising E.M.F., by repeated excitation, but so far without success. Any attempt to explain the continuation of increase of the negativity of the outer surface upon a hypothesis of increase in alkalinity as a result of excitation is negatived by the fact that long faradisation of pieces of skin leads to the opposite effect, i.e., diminution of alkalinity of the outer surface. Thermic Stimulation. Engelmann obtained a negative electrical variation upon the application of heat to the skin of the Frog, using a platinum wire traversed by the current from two Grove's cells. This is, as far as I am aware, the only case recorded of the effect of heat upon the electrical phenomena of a glandular structure. In the case of the skin of the Eel, I applied heat by bringing a small heated copper spatula into the neighbourhood of the skin. Such a position of the electrodes was always selected that the thermo-electric variation due to heating of the electrode nearer the spatula was against any electrical variation arising in the skin itself I found, as in the case of electrical excitation, a pure positive variation indicative of increase in the negativity of the outer surface, in all cases where the outer surface of tlie skin was heated above the temperature of the inner. The following experiment is given in illustration. 358 PROFESSOR E. W. REID ON THE ELECTROMOTIVE Experiment L. Electrodes placed in contact with one another lie east and west. Heating ivest electrode. S deflection of galvanometer. ,, east „ N „ „ Freshly removed Eel's skin now led off by the electrodes. West electrode in contact with outer sui-face. West electrode negative. Deflection N, E.M.F. -00128 D. Heated spatula held 3 to 4 centims. from outer surface. N deflection, 20° of scale. Return to zero in 2 minutes. Heated spatula held near inner surface. Faint N deflection (thermo-electric current from east electrode ?). Repeated many times with similar result. The positive excitatory variation (N deflection) in the case of heating of the outer surface, was evidently sufiiciently strong to swamp any thermo-electric current (S deflection) due to heating of the electrode in contact with the skin. Mechanical Stimulation. Engelmann has noticed that the current of rest of the skin of the Frog is subject to a negative variation upon mechanical stimulation, and I have already mentioned that the same holds good for the skin of the Toad. Indeed, the skin of the Toad is even more sensitive to this kind of stimulus than that of the Frog, the slightest pressure with a glass pointex-, and without any possibility of shifting of the thread electrodes, leading to a diminution of the negativity of the electrode in the neighbour- hood of the stimulated spot. Two electrodes on the outer surface of the skin of a Toad spread on glass. First reading ... Electrode A. Electrode B. + + + + -1. 1' later, spontaneous change to Fall of potential, marked by irregularities, to . 4" later, reversal to Pressure near B change to „ ,, A ,, back to The above is an extreme case, and is not only noticeable for the complete reversal PROPERTIES OP THE SKIN OF THE COMMON EEL. 359 of sign upon irritation, but also for the spontaneous reversal and re-reversal at the commencement of the experiment. With the glands of the " ear swellings " I have obtained similar results. With the skin of the Eel, I have exclusively obtained a variation indicative of increased negativity of the spot excited upon mechanical stimulation. It is, of course, impossible to apply mechanical stimulation with safety, when the skin is on the cork frame and led off from the inner and outer surfaces, on account of th*e danger of shifting the electrodes. It is necessary to spread the skin upon a glass plate with the outer surface uppermost, and then stimulate by pressure with a fine glass pointer with knobbed extremity in proximity to one electrode. The following experiment is an instance of mechanical excitation : — Experiment M. Freshly removed Eel's skin led from two points on the outer surface. E.M.F. of current of rest "000768 D. Deflection S. Electrode A. Electrode B. + Slight pressure near B. S deflection of 23°, i.e., increased negativity of B. Return to zex'o not complete. Rest current recompensated. Slight pressure near A. N deflection 20°, i.e., increased negativity of A. The variation, therefore, upon mechanical stimulation corresponds to that obtained when excitation is electric or thermic. Any discussion as to the reason of the opposite nature of the variations in the Batrachian skin on the one hand, and in the Fish skin on the other, in the cases of both direct electrical and mechanical excitation, is out of place in this paper ; it must, however, be here noted, that though the excitatory variation in the skin of the Eel is invariably such as to indicate increased negativity of the outer surface, pure and simple, i.e., a positive variation with normal direction of the current of rest, the variation upon direct excitation of the skin of the Frog and Toad is often not purely negative, but presents a second positive phase, preceded by a negative " Vorschlag " similar to that obtained upon nerve excitation, and already noted by Hermann, Bach and Oehlek, and Bayliss and Bradford. Action of Atropine. The Eel appears to be extraordinarily resistant to intoxication with a,tropine. It is, of course, impossible to employ subcutaneous injection, for the animal cannot be handled for the purpose if regard is to be had for the integrity of the epidermis. 360 PROFESSOR E. W. REID ON THE ELECTROMOTIVE Atropiuisation has therefore to be effected by adding the drug to the water in which the animal is placed. Tiegel (24) has used this method with success for the purpose of intoxication of Eels with strychnia, obtaining evidence of absorption of the drug in one and a-half hours when six milligrammes of strychnia were added to the litre and a-half of water in which the Fish was swimming. I have kept Eels for two days in "IS per cent, solution of atropine sulphate aerated by a Bunsen pump, with apparently little effect as regards general vital phenomena. The animals were lively at the end of this period, and though the pupil was slightly dilated it reacted to light, yet chemical tests applied to proteid free watery extracts of the livers showed that atropine had been absorbed. The removed skins of such atropinised Eels are found to give either a very feeble ingoing current of rest or outgoing currents which do not reverse with time. This fact is perhaps best explained upon the hypothesis that atropine diminishes the activity of the glandular preparatory processes, and is of interest in connection with the statement by Bayliss and Bradford that the E.M.F. of the current of rest of the skin of the Frog is reduced by treatment with atropine. When the skin of an atropinised Eel is subjected to stimulation by means of the break induction shock, it is found that the normal excitatory variation can no longer be obtained, provided the atropinisation is sufficient. The following experiment is illusti^ative of this fact. Experimeyit N. Eel allowed to live in '15 per cent, solution of atropine sulphate for 48 hours. Pupil slightly dilated but reacted to light. Skin removed immediately after pithing and led off from outer and inner surfaces. Outer Surface negative to inner. Deflec- tion N. The figures in the right hand column indicate compensator degrees of which 1 = -000008 D. 5.10 130 N. 5.13 113 N. 5.15 85 N. ■ 5.16 Stimulation, break induction shock, coil 15. No effect. 5.17 75 N. 5.18 60 N. 5.19 Stimulation, break induction shock, coil 10. No effect. 5.20 60 N. 5.21 Stimulation, break induction shock, coil 5. No effect. 5.24 35 N. 5.25 20 N. 5.26 Stimulation, break induction shock, coil 0. No effect. 5.42 Current reversed. Outer surface +• S. deflection. 55. Stimulation, break induction shock, coil 0. No efiect. PEOPERTIES OF THE SKIN OF THE COMMON EEL. 361 Liver and skin extracted with water for 1 7 hours at 40° C. Extract freed from proteid by saturation with ammonium sulphate, evaporated with HNO3 to dryness and alcohohc soda added, reddish-violet reaction, presence of atropine. The abolition of the excitatory variation by atropine lends considerable support to the glandular hypothesis of origin of its E.M.F. Exclusion of the Excitatory Variation hy Narcotisation. It has been seen that by means of CO;, and also by the vapour of chloroform, the negativity of the outer surface of the skin of the Eel may be diminished ; while, by electrical, thermic, and mechanical stimulation it may be increased, so that, according- to the sign of the outer surface of the skin at the moment that it is the subject of experiment, a variation is induced which, in the normal state of outer surface negative to inner, is negative with the depressants and positive with excitation ; while, should the inner surface be negative to the outer, the reverse is the case. Thus :— (a.) Outer surface of skin negative to inner surface (normal) . . . (6.) Outer surface of skin positive to inner surface Sign of variation. CO3. Chloroform. Excitation. + + + The fact that in skin with normal direction of rest current, the E.M.F. may be brought down to zero by chloroform, and by continued action a reversal of current occasioned, so that the inner surface is negative to the outer, but that recovery to the normal is possible upon readmission of air [vide Experiment C, and Experiment 0) is very strongly in favour of the source of the E.M.F. of the normal current of rest being in some vital processes going on near the outer surface whose activity can be temporarily reduced by narcotisation. That it is one and the same set of vital processes that is acted upon, in the direction of augmentation by excitation, and diminution by the presence of a narcotic, is shown by the interesting fact that it is possible to narcotise the skin with chloroform vapour, with accompanying diminution in the negativity of the outer surface, to such a degree that it absolutely fails to give a positive response to stimulation by elec- tricity during the continuance of the narcotisation, though removal of the vajDour allows of complete recovery. The following experiment is illustrative of this point in particular : — MDCCCXCIII. — B. 3 A 362 PROFESSOR E. W. REID ON THE ELECTROMOTIVE Experiment 0. Freshly removed Eel's skin led off from inner and outer surfaces. Outer surface 2}0sitive to inner. — Deflection of galvanometer S. The figures in the right-hand column indicate compensator degrees, of which 1 = -000008 D. 3.47 P.M. 415 S. 3.49 295 S. 3.51 195 S. 3.52 175 S. Stimulation, coil 15 for 5" N deflection, i.e., negative variation. 3.53 P.M. 125 S. 3.55 85 S. 3.56 75 S. Chloroform vapour passed into gas chamber. 3.57 P.M. 205 S. Air admitted. 3.58 P.M. 150 N, i.e., reversal to outer surface negative. 4.1 130 N. Chloroform vapour passed into gas chamber. Reversal so that outer surface now again positive. 4.3 P.M. 75 S. Air admitted. Reversal again to condition of outer surface negative. 4.5 P.M. 195 N. 4.7 170 N. Stimulation, coil 15 for 5". N deflection, i.e., positive variation. 4.7i P.M. 215 N. 4.9 185 N. Chloroform vapour passed into gas chamber. 4.11p.m. 15 N. Stimulation, coil 15 for 5" while in chloroform vajooiir. No effect. Air admitted. 4.13 p.m. 35 N. 4.17 95 N. 4.19 225 N. 4.24 255 N. CO2 passed into gas chamber. Reversal to outer surface positive. 4.26 P.M. 40 S. Stimulation, coil 15 for 5" in atmosphere of COg. N deflection. PROPERTIES OF THE SKIN OF THE COMMON" EEL. 363 4.26^ P.M. 25 S. Air admitted. Reversal to outer sui-face negative. 4.30 P.M. 65 N. 4.32 155 N. 4.35 185 N. The early part of this experiment illustrates points already referred to, viz., the action of electrical stimulation and of chloroform vapour upon the skin in the condition of outer surface positive to inner. The main point of interest is seen between the times of 4.7 and 4.24. It will be noted that the same stimulus which at 4.7 liberates a positive variation of the rest current while the skin is in the normal condition, at 4.11, when the skin is under the influence of chloroform vapour, is unable to produce any effect. This is not due to any permanent depression of vitality, for it will be noticed that by 4.24 the E.M.F. of the current of rest is considerably above the level at which it stood at 4.7. It is also to be noted that with COo the same degree of narcotisation is not produced, for, in this case (4.26), the skin is still able to respond to the stimulus, though not to such good effect as before. Conclusion. In the foregoing pages it has been demonstrated that an electromotive force is developed as a result of excitation of the skin of the Eel, which manifests itself as an increase of the negativity of the normally negative outer surface. . Is this E.M.F. to be rightly considered as due to an increase in the activity of the glandular pre- paratory processes, to wliose charge that of the normal rest current has been laid, or are the E.M.F. of the current of rest and that of the current of action of different origin ? Hermann, in his explanation of the negative " Vorschlag " and positive after variation of the Frog skin current as a result of indirect excitation, has assumed, mainly upon the ground of his results with the Fish skin, that the E.M.F. of the current of rest is mainly epidermic in origin, while that of the action current is glandular. It has been seen that an epidermic origin for the E.M.F. of the rest current of the Eel's skin upon the theory of mucin-metamorphosis will not hold good, and that it is far more probable that the explanation that the source of the E.M.F. of the current of rest is in the preparatory processes going on in glandular structures, is the correct one. If such be the case, the most simple explanation of the origin of the E.M.F. of the current of action in the Eel's skin is that it is simply due to an increase in the activity of processes whose molecular changes give rise to that of the current of rest. Most striking support is given to this explanation by the fact, evident in Experi- 3 A 2 364 PROFESSOR E. W. REID ON THE ELECTROMOTIVE ment O, that, when the E.M.F. of the vest current is jeduced nearly to zero by the action of chloroform vapour, the normal excitatory developinent of E.M.F. is absent. It remains now to recapitulate a few of the more important points with reference to the electromotive properties of the skin of the Eel dealt with in this paper : 1. The assumption that the E.M.F. of the current of rest of the skin of the Fish is entirely due to ejDidermic mucin-metamorphosis, and that it is not possible to attribute it to the presence of glandular elements, is negatived, in the case of the Eel, by the absence of any such mucinous change in the superficial epidermic cells, and by the presence of abundance of secretory cells thi-oughout the structure. 2. The existence of considerable differences of potential between two contacts upon the outer surface of the skin, and the fact that such electromotive force is capable of excitatory increase upon mechanical stimulation, coincides with the assumption that the E.M.F. of the current of rest is the outcome of glandular processes of vai'iable activity, and is not compatible with the theory of origin of the E.M.F. in mucin-meta- morphosis. 3. The reductions in the E.M.F. of the normal rest current following exposure of the skin to carbonic acid gas and to the vapour of chloroform, and the subsequent recovery upon admission of air, are strong evidence that the origin of the E.M.F. is in some active vital processes taking place in the skin, and it is reasonable to assume that these occur in its glandular elements. 4. The demonstration that the E.M.F. of the skin of the Eel undergoes an excita- tory variation as a result of electrical, thermic, and mechanical stimulation, is in accordance with what is known to occur in other glandular structures, and the fact that such excitatory change manifests itself as a positive variation of the current of rest agrees in the main with the phenomena observed in other cases. 5. The fact that chloroform narcosis excludes the possibility of the excitatory variation upon stimulation at the same time as it reduces the E.M.F. of the normal current of rest to zero, supports the assumption that the E.M.F. of the current of rest and that of the current of action originate in one and the same source. 6. Finally, the reduction of the E.M.F. of the normally directed current of rest by atropinisation, and the complete absence of any excitatory variation under such conditions, are facts strongly in favour of the hypothesis that both the E.M.F. of the current of rest and that of the current of action are from a glandular source. BiBLIOGEAPHY. 1. Hermann. Neue Untersuchungen liber Hautstrome. PrLiiGER's Arcliiv, vol. 27, 1882, p. 280. 2. Du Bois-Reymond. Untersuchungen ilber thierische Elektricitiit, vol. 2, Abth. 2, p. 9-20. 3. Hermann. Uber die Secretionsstrome und die Secretreaction der Haut bei Froschen. Pfluger's Archiv, vol. 17, 1878, p. 291. PROPERTIES OF THE SKIN OP THE COMMON EEL. 365 4. Bach -mid Oehler. Beit]-age zur Lehre von den Hantstromeu. Pfluger's Archiv, vol. 22, 1880, p. 30. 5. Rosenthal, tjber das elektromotoi-ische Verhalten der Fi'osclihaut. Reichert und Du Bois-Reymond's Archiv, 1865, p. 301. 6. Leydig. tJber die Haut einiger Susswasserfische. Zeitsch. f. wissenscli. Zoologie, vol. 3, 1851, p. 1. 7. Leydig. Hautdecke und Hautsinnesorgane der Fische. Halle, 1879. 8. F. E. ScHULZE. Epithel- und Drilsen-Zellen. Arch. f. mikr. Anat., vol. 3, 1867, p. 137. 9. FoETTiNGER. Recherches sur la structiu-e de lepidenne des Cyclostomes, etc. Bulletin de TAcadfoiie royale de Belgique, 2™"^ serie, vol. 41, J 876, p. 599. 10. List. Zoolog. Anzeiger, Jahrg. 8, 1885, p. 388. Biologisches Centralblatt, vol. 5, 1885-6, p. 698. Archiv f. mikr. Anat., vol. 27, 1886, p. J 81. 11. KoLLlKER. Wurzburger naturwissensch. Zeitschrift, Hft. 1, 1860. 12. HoYER. Uber den Nachweis des Mucins in Geweben mitfcelst der Farbemethode. Archiv f. mikr. Auat., vol. 36, 1890, p. 310. 13. Engelmann. Die Hautdrilsen des Frosches. Pfluger's Archiv, vol. 6, 1872, p. 97. 14. Bayliss and Bradford. On Electrical Phenomena accompanying Secretion in the Skin of the Frog. Journal of Physiology, vol 7, p. 217. 15. Budge. tJber den galvanischen Strom, welcher sich in der Haut des Frosches zu erkennen giebt. Poggendorff's Annalen, 1860, p. 537. 16. Bayliss and Bradford. The Electrical Phenomena accompanying Secretion. Internat. Monatsschrift fllr Anat. u. Physiol., vol. 4. 17. Exgelmann. Die Hautdrlisen des Frosches. Pfluger's Archiv, vol. 5, 1872, p. 498. 18. Du Bois-Reymond. Untersuchungen, vol. 2, Abth. 1, p. 187. 19. Valentin. Zeitsch. f. rat. Med., vol. 15, 1861, p. 208. 20. Roeber. Uber das elektromotorische Verhalten der Froschhaut bei Reizung ihrer N erven. Reichert und Du Bois-Reymond's Archiv, 1869, p. 633. 21. Hermann und Luchsinger. Uber die Secretionsstrome der Haut bei der Katze. Pfluger's Archiv, vol. 17, 1878, p. 310. Hermann und Luchsinger. Uber Secretionsstrome an der Zunge des Frosches. Pfluger's Archiv, vol. 18, p. 460. 22. Luchsinger. Neue Beitriige zur Physiologie der Schweiss-secretion. Pfluger's Archiv, vol. 22, 1880, p. 152. 23. BuRDON Sanderson and Gotch. Journal of Physiology, vol. 12, No. 4. Proceedings of Physiological Society. Oxford, June 20, 1891. 24. TiEGEL. Vom Rlickenmark der Schlangen und der Aale. Pfluger's Archiv, vol. 17, 1878, p. 594. [ 367 ] V. The Cerebriim of Ornithorhynchus paradoxus. By Alex Hill, M.D., Master of Downing College. Communicated by Alexander Macalister, F.R.S. Received and read June 16, 1892. [Plates 20-22.] Sum,m,ary. The brain of Ornithorhynchus is completely devoid of convolution, simple in structure, and similar in many respects to the brains of the embryos of higher Mammals. It is, however, strictly Mammalian and not Avian in type. It differs from most, if not from all, other Mammahan brains in many respects, of which the following are the most notable : — It is destitute of corpus callosum ; the structure which has been described under this name is limited exclusively to the hippocampus, and belongs therefore to the fornix. It forms a decussation or commissure above the anterior commissure. Fibres reach this decussation from all parts of the hippocampus, both in front and behind. The fibres from the decussation turn downwards in three sets (1) in front of the anterior commissure ; (2) between the anterior commissure and the soft commissure ; (3) above the soft commissure. The hippocampus extends to the extreme anterior end of the hemisphere above the ventricle. The rhinencephalon (pyriform lobe, natiform protuberance, &c ) extends at first along the base and afterwards along the mesial aspect of the liemisphere to the posterior extremity of the ventricle. In the absence of the corpus callosum and septum pellucidum the cortex of the mesial wall of the hemisphere is continuous through the anterior perforated substance with the corpus striatum. In a similar manner the putamen of the nucleus lenticularis is continuous, behind the cerebral crus, with the cortex of the rhinencephalon. Appearances therefore suggest that nucleus caudatus and putamen belong to the mesial wall of the hemi- sphere which has been involuted by the crus, the optic thalamus, and perhaps by the remainder of the nucleus lenticularis. Historical. The brain of the Ornithorhynchus, like all other parts of the animal, is very elabo- MDCCCXCni. — B. 4.9.93 368 DR. A. HILL ON THE CEREBRUM OF ORNITHORHTNCHUS PARADOXUS. rately aud beautifully figured in Meckel's monograph.* He represents correctly the pyiiform shape of the hemispheres ; their surface perfectly smooth save for the deep grooves cut by the blood-vessels which, emerging from beneath the outer sides of the olfactory bulbs, radiate backwards. Meckel does not appear, however, to have noticed the middle cerebral artery. He figures the blood-vessels which groove the internal surface of the hemisphere, as well as the deep groove which is continued forwards from above the hippocampus to the anterior extremity of the brain. The figure of a sagittal section in the median plane (Plate 7, fig. 7) is drawn with great accuracy, but the references to the figure are singularly incorrect, so incorrect, in fact, that we should infer that the pointer-lines were wrongly placed by the engraver if their disposition did not agree entirely with the description in the text. As the mistakes are made with regard to just- those features which are, owing to the position of this animal amongst vertebrates, of highest importance, it is necessary to quote the words of the text in connection with the lettering of the figures. In the figure the hippocampus or ridge produced by the projection of the fascia dentata is called the " corpus callosum." The transverse commissure, which seems to take the place of the corpus callosum, is lettered as the " septum pellucidum." That there is no mistake in the lettering of the figures is evident from the following description of the former structuret : — •" Corpus callosum adest quidem, sed breve, quum baud quatuor lineas longitudine aequet. Memorabilius etiam videtur, in dimidia duo lateralis, linea mediana haud confluentia, esse disjunctum. Equidem saltem in faciebus sese spec- tantibus internis nullum dilacerationis vestigium in venire potui." As Meckel makes no reference in the text to the structure which he figures as the septum pellucidum, it is impossible to tell how it came about that a lai-ge transverse commissure was thus named. Owen's description^ is even less easy to understand than Meckel's, although he corrects Meckel's mistake with regard to the corpus callosum. " My doubts as to the great development of the corpus callosum of the Ornitho- rhynchus were further justified by the indication of its nearer approach to the Oviparous type afforded by the simple bipartite condition of the tubercles called ' quadrigemina.' Well preserved specimens of Ornithorhynclms presented to me by Mr. Thomas Bell, Surgeon, R.N., in 1838, have enabled me to determine this question. There is neither corpus callosum nor septum lucidum in the Ornithorhynchus. " The part described by Meckel as the corpus callosum corresponds with the fornix and hippocampal commissure, as it exists in the Marsupialia, excepting that the essential function of the fornix, as a longitudinal commissure, uniting the hippocampus major with the olfactory lobe of the same hemisphere, is more exclusively maintained in the Ornithorhynchus, in consequence of the smaller size of the transverse band of * Meckel, ' Ornithorhynclii paradoxi, Descriptio Anatomica,' Leipsic, 1826. t Loc. clt. p. 33. J Todd's ' Encyclopsedia of Anatomy and Physiology,' pp. 382, 383. DR. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. 369 fibres uniting the opposite hippocampi, and representing the first rudiment of the corpus callosum, as it appears in the development of the placental embryo. The thin internal and superior parietes of one lateral A'entricle are wholly unconnected with those of the opposite ventricle." " The part described by Meckel as the corpus callosum corresponds with the fornix and hippocampal commissure." This is not the case ; it corresponds with the projection of the fascia dentata. " The essential function of the fornix, as a longitudinal commis- sure, uniting the hippocampus major with the olfactory lobe of the same hemisphere, is more exclusively maintained in the Ornithorhynchus, in consequence of the smaller size of the transverse band of fibres uniting the opposite liippocampi and representing the first rudiment of the corjDus callosum, as it appeal's in the development of the placental embryo." It would appear from this to be clear that what Owen regarded as the " longitudinal commissure," is the ridge on the mesial surface made by the fascia dentata, and not a commissure at all. It cannot well be the commissure which I have lettered as D.F., since this runs transversely. Again, which is the transverse band of small size uniting the opposite hippocampi, and representing the first rudiment of the corpus callosum ? Owen has said above that there is no corpus callosum. The transverse commissure (Meckel's septum pellucidum) is not a small band by any means, although smaller than the anterior commissure. Owen could not so have described the structure which was figured in Meckel's plate. Either description is to me an inscrutable mystery. I cannot understand how Meckel could have mistaken this thick round commissure for the septum pellucidum, or how Owen could either have overlooked it altogether or else have described it as " a small tra.nsverse band." On what ground, too, does Owen maintain that " the essential function of the fornix, as a longitudinal commissure ... is maintained " when, as will be shown presently, all such fornix as exists in OrnithorhyncJms decussates in the middle line, is not united with the olfactory bulb, and may be, for all one can tell to the contrary, not a longitudinal commissure at all, but a series of tracts uniting together corresponding parts on the two sides ? The chief interest of the brain of this animal clearly centres about the question as to whether it does or does not contain a corpus callosum, and, therefore it appears to me desirable to sum up the literature of the subject in connection with this question. Corpus callosum. — In the ' Philosophical Transactions of the Royal Society ' for 1837 (p. 80), Owen asserted that no corpus callosum or septum pellucidum are to be found in either Monotremes or Marsupials. He figures the brains of a large number of species {Ornithorhynchus is not amongst them) as showing a "fornix commissure" or transverse commissure superior to the anterior commissure, which represents, however, he says in one place " besides the fornix, the rudimental commencement of the corpus callosum." The same expression is used in his article " Marsupialia," in ' Todd's Encyclopsedia,' 1847, for Owen was evidently not convinced that the corpus callosum is really absent MDCCCXCIIL — B. 3 B 370 DR. A. HILL ON" THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. ill these auimals, although his whole description implies tliat this is the crucial dis- tinction between non-placental and placental Mammals; a difference which he accounts for by the cui-tailment of intra-uterine life :• — " I have since derived the most satis- factory confirmation of this coincidence from the repeated dissection of the brains of Marsupials belonging to different genera ; and although unable to explain how a brief intra-uterine existence, and the absence of a placental connection between the mother and foetus can operate (if it be really effective and anything more than a relation of simple co-existence) in arrestmg the develojament of the brain, yet it is a coincidence which has not been suspected, and is, in various points of view, perhaps the most interesting of the anatomical peculiarities of the quadrupeds here treated of." In liis article " Monotremata," Owen refers to the transverse commissure which he figures in the brain of Echidna as " the hippocampal commissure," biit again he says later on that " the short transverse commissure above mentioned is the sole rejDre- sentative of the corpus callosum and fornix." Whether or not OwEN asserted that the corpus callosum is totally absent in non- placental Mammals (and I cannot find such a definite statement made about any animal except Ornithorliynclms), he was understood to mean this, and he himself quotes with approval a table given by MM. Eydotjx and Laurent (' Voyage de la Favorite,' p. 166) in which it is thus stated, " Corps calleux ; Monodelphes (existe) ; Didelphes (manque) ; Ornithodelphes (manque) ; Oiseaux (manque)." These writers add, " Meckel a cependant admis dans les figures I'elatives a I'encephale de VOrnitho- rhynque I'existence du corps calleux ; mais, en etudiant avec soin I'encephale de notre JEchidne, nous nous avons reoonnu que les descriptions de M. E. Ov^en sont plus exactes que celles de Meckel, et que les determinations de I'anatomiste Anglais doivent etre adoptees." Flower published in the 'Philosophical Transactions' for 1865 (p. 633, et seq.) a series of observations upon the brains of Echidna and various Marsupials, with the object of disproving Owen's teaching with regard to the corpus callosum. He did not study the brain of Ornithortiynchus, but in the brain of Echidna he says, " is seen the superior transverse commissure, very much reduced in extent, and in which the two portions, upper and lower, observed in the Kangaroo are no longer distinguishable," but he adds, " whatever parts of the placental Mammalian brain are represented by this commis- sure in the Kangaroo are also represented by it, though in a reduced degree, in Echidna." Owen recognized that the commissures of the non-placental Mammals are not disposed in the same way as, perhaps are not honiologous with, the commissures of placental Mammals. He saw the diflS.culty but not the way out of it ; he relied so far as one can judge from his description upon naked-eye observations and very simple dissections, and did not trust his own conclusions. At one time he quotes with approval, as a summary of his own views, the bald statement that neither Marsupials nor Monotremes possess a corpus callosum ; at another time he talks about their having DR. A. HILL ON THE CEREBRUM OF ORNITHORHYISrCHUS PARADOXUS. 371 the " rudimental commencement of a corpus callosum," and again at another time he gets over the difficulty by naming the commissure the " commissure of the hippo- campus." Flower* gives a most admirable and philosophical summary of the position of the question as it rests upon his observations, but in his conclusion he is, as it appears to me, unfair to Owen. He may well resent the uncertain language which his pi'e- decessor used, but he should not deny him the merit of recognising an essential difference between the brains of placental and non-placental Mammals ; and had his own observations been carried out, as they would be if made at the present time, by means of a series of sections, he would, I think, have maintained the justice of Owen's contention that the upper transverse commissure in non-placental Mammals is not sufficiently described as " corpus callosum " ; indeed he would, I think, have allowed that the Monotremes, at any rate, have no corpus callosum at all. Flower's summary is too long to quote in extenso, but certain sentences may be extracted without giving a false impression. " The commissure radiates over the whole of the inner wall." " They are part of the great system of transverse fibres which bring the two hemispheres into connection with one another." " They cannot in any sense be confounded with the posterior crura of the fornix." " In all Marsupial and Monotreme animals it lies above the septuai ventriculorum and especially above the precommissural fibres of the fornix." " Moreover, passing outwards into the hemi- spheres, it overarches or forms the roof of the lateral ventricles of the cerebrum." Tlie test of a corpus callosum would appear to be summed up in this last sentence. The commissure which we know as corpus callosum consists of fibres uniting together homologous (and also, as Sherrington has shown, heterologous) parts of the convex surfaces of the hemispheres. In Oniithorhynchus the transverse commissure in question does not, as my sections show, surpass the limits of the hippocampus. With regard to other non-placental Mammals I cannot speak from observation, but Flower's picture of the brain of Echidna represents a condition almost identical with that of Orniihorhynchus and Herrick's figures of sections of the brain of Didelpliys virginiccff show that in this Marsupial animal the disposition of the commissure is the same as in Ornithorliyndius. In certain Monotremes and Marsupials, therefore, the commissure does not fulfil Professor Flower's test, and it is open to question whether any animal in which the hippocampus extends the whole length of the brain, above the ventricles, can have a corpus callosum. The hippocampus is the doubly folded margin of the cortex supported by a rod of white fibres, the fimbria or posterior crus of the fornix. If the hippocampus lies above the corpus callosum the fimbria (or fornix) must also have this situation, a curious reversal of the relation which these parts bear to one another in animals in which the corpus caUosum is well developed ! This is the only arrangement possible * Tjoo. cit., p. 648. t ' Jl. of Comp. Neurology,' vol. 2, p. 1. February, 1892, 3 B 2 372 DR. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. unless, indeed, the fibres from the alveus to the fornix interlace with those of the corpus callosum. In Oniithorhynchus, however, appearances are altogether against the existence of callosal fibres ; the fibres of the anterior commissure which sweep beneath the corpora striata supply the whole of the convex surface, as well as a part of the mesial surface of the pallium. Professor Flowee's pictures of the inner surface of the hemisphere suggest that a more or less horizontal plate of commissural fibres, which constitutes the great corpus callosum of Man and most Mammals, is present in a rudimentary form in Macropus, Phascoloimjs, and Thylacinus (with which he says that Phalangista and Didelphys agree), but totally absent in Echidna. I find it quite impossible to gather Professor Herrick's views with regard to the dorsal commissure in Didelphys. Under the heading " Callosum and hippocampal commissure," he says ■"' " it is not necessary to recount the various opinions and discussions of the callosum in the Marsupials. Until Osborn, most authors had agreed that the callosum is absent, and functionally replaced by the anterior com- missure." This is curiously oblivious of Flower, who especially opposed Owen on this question, not to mention Leuret, Foville, Mayer (who described Didelphys virginiana in 1842), and Pappenheim, who, speaking about the dorsal transverse commissure in the same animal, says,t "le corps en question est bien un corps calleux . . . Cette commissure n'est done ni un fornix ni un melange du fornix avec le corps calleux." Herrick continues, " Professor Osborn has done much to place this whole subject in its proper light, and Ave agree with him in respect to the essential homologies of the dorsal commissural system . . . The motor cortex as such is thrown well cephalad, and the fornicate gyrus is carried forward along the mesial surface, as may be seen from an inspection of the transverse sections of Plate A. Thus it happens that the caudal portion of the dorsal commissural system is much more highly developed than the cephalic or callosal portion. The latter consists of a few fibres which sjDring from the region about the anterior prolongation of the splenial fissure, if this term may be applied to the fissure which bounds the cephalad contiiiuation of the fornicate gyrus." No description of the tract in question is contained in the paper. ZuckerkandlJ also describes the brain of Ornithorhynchus, although he is careful to state that the brain was badly preserved. He also simply names the decussation of the fornix " der rudimentiire Balken," and, although the "Randbogen" is the structure under discussion, he overlooks the fact that it passes right above his sup- posed corpus callosum to the front of the hemisphere. If, as appears to me to be the case, the corpus callosum is totally absent in * Loc. cit., p 7. t " Notice preliminaire sur ranatomie du Sarigue femelle (Didelphys virginiana) " ' Comptes Rendus,' vol. 24, p. 186 (1847). X ' Ueber das Rieclicentrum.' Stuttgart, 1887. DR. A. HILL ON THE CEREBRUM OF ORNITHORHTNCHUS PARADOXUS. 373 Ornithorhynclms, it will be necessary to reconsider the whole question of the homology of the commissures in Vertebrata. Sir William Turner does not describe the interior of the brain in his paper on Ornithorhynchus,* but he mentions the appearance of what he considers (adopting Flower's view) to be the corpus callosum. " The anterior commissure and the rudimentary corpus callosum were seen at the surface of the section, and behind these was the hippocampus. A shallow antero-posterior fissure 6 millims. long was observed on the mesial surface of the pallium in front of and above the divided corpus callosum. It might possibly be the splenial fissure, and the slender band of the palliiun between it and the corpus callosum would thus represent a rudimentary callosal convolution." The fissure in question, as seen in sections 5 and 6, is the extremely deep dentary fissure (or fossa, as I prefer to call it). The band of pallium, the fascia dentata. Unaware of the continuation forwards of the hippocampus, Turner saw no reason for considering the dorsal commissure as anything but corpus callosum. Method. The whole animal was hardened in spirit. The skull had been opened, and there- fore the spirit had at once reached the brain, which was in fair condition. The left hemisphere was stained in carmine en hloc and then cut into an irregular series of sections. Finding, however, that it presented certain points of extreme interest, I determined to treat the right hemisphere with much greater care. It was therefore placed for a fortnight in a two per cent, solution of bichromate of ammonia, for even a brain which, like this one, has been for years in spirit will yield sections which can be stained by Weigert's method if It is placed in a chrome-salt for a time. The hemisphere was next placed in a solution of carmine-alum for a week, washed in water, and after dehydration by alcohols of increasing strength, embedded in celloidin. When the celloidin ha^d been firmly set by chloroform the brain was very carefully divided into fourteen blocks along the lines marked in the photo- lithograph (Plate 20, fig. 1). Each block was then cut into a series of sections of which a number, varying according to the apparent interest of the region, were mounted and labelled A to D or M or N as the case might be; about 150 sections in all (not including some 120 already prepared from the left hemisphere). A certain number of sections from each block were stained by Weigert's method in order that the arrange- ment of the fibre- tracts might be determined with certainty. Figs. 6-10 are diagrammatic, inasmuch as the cortex is not coloured at all, and the fibres are shown as they appear in the nearest section of the series which was stained with hsematoxylin. * ' Jl. of Anafc. and PhysioL,' vol. 26, p. 357, April, 1892. 374 DE. A. HILL ON THE CEREBRUM OP ORNITHORHTNCHUS PARADOXUS. Description of Several Parts of the Brain, The olfactory bulb is of moderate size, its maximal coronal dimensions being vertically 3 millims., ti'ansversely 4 millims. Its length is 3 millims. It is absolutely free from the cerebral hemisphere, the rounded neck of the crus, if such an expression is allowable, being crossed by the large anterior cerebral artery on its way to the outer surface. The bulb is cupped on the under side, as in Didelphys virginica described by Herrick ;* the cupping is not, however, visible from the under surface, for it is occupied by glomeruli which lie in loculi of connective tissue. I do not gather from Herrick's description or pictui-es that such an arrangement is found in Didelphys, nor am I acquainted with any animal in which it obtains. The ventricle of the bulb is obliterated. In minute anatomy the bulb presents no feature of particular interest. The olfactory crus remains for some distance distinct from the hemisphere. It becomes progressively thinner until just before its attachment it is a flat band 2 millims. wide by '5 millim. thick. The fibres of the olfactory tract lie on its ventral surface. It is situate remarkably near to the middle line at the spot where it adheres to the under surface of the hemisphere. External form. — The general form and external appearance of the brain have been described so recently by Sir William TaRNERt that it is unnecessary for me to x'epeat a description which is already iu the hands of all students. I would merely point out certain respects in which the brain is somewhat remarkable. Although the distinction between the rhinencephalon (using this tei'm in an inclusive sense) and the rest of the hemisphere is deeply marked, the brain is absolutely destitute of convolu- tion in the proper sense of the word. It is almost a pity that the fun-ow (the ectorhinal fissure) which sepai-ates the rhinencephalon from the rest of the hemisphere, is named a "fissure," for its origin is not due to the same mechanical causes as the origin of the fissures proper. Rather does it fall into the same class with the incisura pallii longitudinalis (longitudinal fissure) and incisura pallii transversa (transverse fissure) which are the gaps between the difierent organs which collectively constitute the brain. It is almost a crime to add another to the terms with which the nervous system is already smothered, but I am inclined to suggest that just as longitudinal and transverse fissures have given way to incisura pallii longitudinalis and incisura pallii transversa, so the ectorhinal (or, in human anatomy, collateral fissure) should give way to " incisura rhinalis." The so-called "choroidal fissure" is not, of course, a fissure in any sense of the word. I have in another paperj suggested that it should be termed " hiatus ventriculi." * ' Jl. of Comp. Neurology,' vol. 2, p. 1. February, 1892. t ' Jl. of Anat. and Physiol.,' vol. 26, p. 357. April, 1892. % " The Hippocampus," see p. 389, infra. DR. A. HILL ON THE CEREBRUM OP ORNITHORHrNCHUS PARADOXUS. 375 Is the dentary fissure a true fissure ? The fact that the same terminology is used in this as in other parts of the brain has led, as I think, to fruitless attempts to homologize the several structures which make up the "hippocampus" with other parts of the pallium. The hippocampus is altogether different in constitution from the rest of the pallium, and forms, as was recognized by Bkoca and by Zucker- KAis'DL, a part of the rhinencej^halon. Topographically it belongs to the rhinen- cephalon. It varies in size with the rest of the rhinencephalon. Its essential con- stituent, the fascia dentata, is, as I have shown in the paper referred to, absent when the olfactory bulb is absent. The cerebral cortex is divisible, therefore, into the part which belongs to the rhinencephalon and the part outside it. If we accept the view that the convolution of the brain is due to mechanical causes, that is to say, to the necessity for disposing of a layer of superficial tissue, the nutrient needs of which forbid its increase in thickness beyond a certain maximum, which is reached in very small brains, and which can only be disposed upon the surface of larger brains in a plicated manner ; since, while the contents of a sphere equals nr^, its superficies equals ttt' only ; it is obvious that the use of the word " fissure" in two distinct senses can only lead to confusion. The formation of the dentary fissure is clearly not due to the causes just mentioned ; indeed, the rhinencephalon is distinguished from the rest of the hemisphere by its slight tendency to convolution, and, thei'efore, in this case also, the use of the word " fissure " might be dropped with advantage and the term " fossa," i.e., dentary fossa, used instead. The hemispheres are absolutely destitute of convolution, although the projection upwards of the lateral lobe of the cerebellum causes the formation of a deep pit, which separates the occipital from the back of the temporo- sphenoidal region. The external surface is deeply channelled by blood-vessels which lie in rounded grooves. The greater number of these vessels are branches of the anterior cerebral artery, which reaches the external surface by passing just above the base of the olfactory bulb, separating its crus from the frontal portion of the hemisphere which lies above it. Immediately on reaching the outer surface it divides into two branches, from which are derived other vessels which, diverging from one another, groove the outer surface of the hemisphere as far as its posterior border. A small branch of the anterior cerebral passes to the outer surface beneath the olfactory bulb. The middle cerebral artery reaches the outer surface at a spot which may probably be regarded as the situation of the fissure of Sylvius, and breaks up into branches which do not supply so extensive an area as the anterior cerebral. The ventral surface presents, at the front, the olfactory bulb. To the outer side of the bulb lies the optic nerve, and to the outer side of this again the immense fifth nerve. The dura mater constitutes a firm sheath for these structures which could not in my specimen be removed without injuring them. 376 DR. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. The mesial surface preseiiits at its ujDper part the unconvoluted internal surface of the hemisphere, of uniform breadth [5 millims.). It is grooved by five small artei'ies, which pass to its upper border from the posterior cerebral artery. They incline backwards as they pass towards the upper edge of the hemisphere. Behind the corpus callosum, i.e., for more than half its length, the cerebrum forms a flat cap, which rests upon the optic thalamus, the mid-brain, and the cerebellum. The brain is remarkable amongst Mammals in the continuation forwards of the hippocampus to its extreme anterior end ; and, lastly, if my observations are correct, in the complete absence of corpus callosum. The Hippocampus. — The margin of the ventricular slit is formed by the rhinen- cephalon below, the hippocampus above. These two meet in block 11 at the back of the slit. The hippocampus lies, therefore, entirely above or dorsally to the velum inter- positura. The ventricle has no descending horn, but extends in a horizontal plane above the crus and thalamus. The cortex cerebri is strongly folded upon itself several times. First at the dorso- mesial edge of the hemisphere it descends as a broad flat surface which borders the incisura pallii (longitudinal fissure). This surface has a breadth of 5 millims. Then the cortex returns upon itself about the dentary fossa : its pyramids being at once arranged in the single sheet which is characteristic of the subiculum. The subiculum sweeps round into the cap of fascia dentata. At no part of the hippocampus is there any ridge of longitudinal white fibres such as is characteristic of the fimbria, but from the very commencement of the hippo- campus the convex surface of the cortex, which projects into the ventricle, is covered with a sheet of longitudinally running white fibres. It is best to call this indifferently the fimbria or fornix, although it is the homologue of the fimbria and alveus of higher Mammals, and is by no means a " corjDus fimbriatum." It represents both pars libra and pars fixa of Zuckerkandl. Nor is there from behind forwards a distinction in the tract into fimbria, posterior pillar and body of the fornix. The arrangement of this tract with regard to the fascia dentata is interesting. As seen in IOg it enters with the pyramidal cell-layer into the concavity of the cap of fascia dentata. The fascia dentata is quite devoid of fibres on its surface, but consists, as in other animals, of a layer of very small cells (granules) three or four deep, covered with molecular grey matter of rather greater density than the super- ficial layer of the cortex in other regions. In 9 the fornix (or alveus or fimbria) is thicker and slightly folded upon itself on the deep surface of the fascia dentata. In 8 it is still thicker and sw eeps over the whole of the convexity of the subiculum* to form a continuous white lining for the ventricle. * Using the term in a sense suggested in my paper on the hippocampus. "DR. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. 377 In 7g the fibres are beginning to collect on the mesial side of the hippocampus, iii the angle between the fascia dentata and the convex surface of the subiculum. At 7p the fibres of the two sides decussate in the so-called " rudimentary corpus callosum," or "hippocampal commissure." In front of the decussation certain columns of fibres are seen running longitudinally in the substance of the mesial wall of the hemisphere. Others, a thick sheet, sweep up the inner side of the ventricle. A third group, now detached from the rest, lies in the angle beneath the outer limb of the fascia dentata. Some of the fibres of the last groups are longitudinal, others sweep into the nucleus fasciae dentatse. The fascia dentata is at first (behind) uniformly curved, then folded at two right angles, then at one acute angle, which is sharpest at 6d, where the mesial limb is continued for some distance on the surface of the hemisphere. (See sections 10, 9, 8, 7, 6.) The fascia dentata in front of the decussation constitutes a narrow convolution which is continued to the front of block 5, as seen in the lithograph and in the sections. Its extreme anterior end is shown in the figure 5a. I have found it impossible to judge the nature of the decussation or commissure of the fornix from the sections. It is very clear that few, if any, of the fibres of this fornix system extend beyond the hippocampus at either end. It shows, at its occipital end, such a thin coat of white matter that it is impossible that any considerable number of fibres pass beyond the limits of the hippocampus into the medullary centre of the occipital lobe. Throughout its whole extent the sheet of white fibres which covers the subiculum on its convex (outer) surface, i.e., the alveus or white lining of the inner wall of the ventricle, grows thinner and thinner as the angle between the inner and outer walls is approached. The rest of the ventricle is, as will be presently shown, lined with white matter belonging to the system of the anterior commissure, and it is more likely that the white lining of the inner wall is reinforced by fibres from the com- missure than that the lining of the outer wall receives fibres from the alveus. The numbers of fibres crossing in the decussation is so large that we may infer that but few remain throughout their course in one hemisphere only ; but I cannot see in the sections any appearances which enable me to decide whether the fibres from the front of one hemisphere pass to the front or to the back of the other. Immediately in front of, or rather beneath, the decussation the main body of fibres (excluding the precommissural fibres already referred to) turn downwards and curve backwards as the anterior pillars of the fornix which will be described in connection with the corpus mammillare. The Anterior Commissure is a transverse tract of great size, its cross-section measuring in my specimen 2'5 X 3*25 millims. It is separated from the decussation of the fornix by the thin internal edge of the corpus striatum. It rests below upon the grey matter which forms the inner wall MDCCCXCIII. — B. , 3 c S78 DR. A. HILL ON THE CEREBRUM OF ORNITH.ORHYXC'HUS PARADOXUS. of the hemisphere, intervening between the rhinencephalon and its mesial surface. It is, therefore, surrounded by grey matter on all sides. Its fibres form a broad thick sheet which invest and support the corpus striatum. It is fix'st cut in the middle line in 6e, but its anterior edge being concave, the thick flattened column is met with, obliquely cut, in sections farther forward. Indeed it may with propriety be described as having a forceps anterior and forceps posterior, for the fibres into which it spreads out on either side pass to all jiarts of the hemisphere. In front of the decussation it is easy to follow the course of its fibres, for they are strictly limited to the region between the corpus striatum and the cortex ; they form a thick external capsule for the corpus striatum ; above the corpus striatum they constitute, as already said, the white lining of the outer wall of the ventricle, and supply, doubtless mingled with, peduncular fibres, all parts of the cortex. Behind the decussation the fibres are more difficult to follow, for they become involved with the peduncular fibres, but again, by block 9, they appear to have become distinct, for a mass of fibres lies beneath the nucleus lenticularis in the angle between it and the meso-ventral edge of the hemisphere, which seems to be continuous with the forceps posterior of the commissure, and but little affected by the peduncular fibres which lie dorsally to this part of the nucleus lenticularis. The Grey Matter of the Hemisphere may be classified in three divisions : — - 1. The grey matter of the mesial wall, with the nucleus lenticularis and nucleus caudatus. 2. The cortex of the rhinencephalon, with (or including) the hippocampus. 3. The cortex in general or pallium. As there is no septum pellucidum, and the lateral ventricle extends for some distance in front of the commissures, the inner wall of the hemisphere includes a triangular region, bounded by the hippocampus above, the ventro-mesial edge below and the commissures behind, which is very difficult to homologize with parts of the human brain. Pei-haps it is not possible to homologize it, for the smooth surface of the cortex which in the human brain lies in front of the rostrum, the " terrain desex't," belongs to the pallium, and constitutes the common starting ground of the gyrus fornicatus and the gyrus geniculatus (Zuckerkandl) which latter dies away into the nerve of Lancisi ; while the region which lies in front of the commissure in the brain of Ornithorhynchus is completely separated from the gyrus fornicatus by the hippocampus. It appears to be the part which, in higher Mammals, atrophies in becoming the septum pellucidum, as well as the anterior perforated substance. It is composed of grey matter 1 "3 to 1"5 millims. thick ; more compact on either surface than in its centre, and containing branched cells of the pyramidal type. The cells are rather larger than those which are scattered uniformly throughout the corpus striatum, about 11 /x broad by 15 ju, long, with apical, basal, and axis-cylinder processes ; but partly because the tissue, although in many places well preserved, DE. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. 379 is hardly fit for accurate histological work, and partly because the shape of the cells depends upon the plane in which they happen to be cut, I do not propose to draw far-reaching conclusions from apparent differences in minute structure. More interesting than its minute structure is the fact thai this tissue passes over at the ventro-mesial angle of the hemisphere into the corpus striatum, without the possibility of any line of demarcation being drawn between them. It would appear, therefore, that the grey matter which lies on the inner side of the hemisphere is a part of the same formation as the nucleus caudatus. This reduces the latter very distinctly to the level of a cortical formation, differing morphologically from other pai"ts of the cortex in the absence of a white coat. The grey matter of the corpus striatum lies on the axial side of the rhinencephalon, which in turn lies on the axial side of the pallium. This grey formation reaches almost as far forwards as the ventricle, but projects in front of the part common to it and the corpus striatum as a rounded boss. Con- sequently in 4 the curious appearance is presented of a round nucleus of grey matter near the ventral edge of the mesial wall, quite distinct from the oval anterior end of the nucleus caudatus. The anterior commissure passes, as already said, right through this grey matter, so that we find in 6j the appearance of a nucleus of grey matter between the anterior commissure and the commissure of the fornix. It is noticeable that as the corpus striatum increases in width so also does the grey matter beneath the anterior commissure ; the line which marks the outer boundary of the corpus striatum sweeps, therefore, in an even curve beneath the subcommissui'al grey matter. The foramen of Monro is seen in block 7. It is bounded above by the hippo- campus and fornix, beneath by a small portion of this grey matter which rests on the anterior commissure. It is not easy to define the posterior limits of this grey matter. The anterior commissure is last seen in 7 J. By this time, as may be seen in the lithograph, the section passes through the third ventricle, and its mesial wall is the epithelial lining of the ventricle. Still the large-celled grey matter, which is characteristic of and common to both the mesial wall of the hemisphere and the corpus striatum, is continued backwards into the angle between the cerebral crura. It appears, therefore, that grey matter of uniform structure constitutes the mesial wall of the hemisphere in front of the anterior commissure, the corpus striatum, and the wall of the third ventricle. Such at any rate is the conclusion to be drawn from my sections, but I should be very sorry to assert that it would not be possible in a really well-preserved brain, successfully stained, to recognize as many difierent formations as are described in human anatomy, although I am inclined to think that anatomical distinctions have been pushed too far, The minute structure of the brain should always be studied simultaneously with its configuration. 3 C 2 380 DR. A HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXCTS. Central Grey Masses. — In describing the grey matter of the interior of the cerebrum, I have purposely used the expression corpus striatum, because in its anterior position it is not distinguishable into nucleus caudatus and nucleus lenticularis. The corpus striatum is first seen towards the back of block 4 on the outer side of ihe ventricle, very near to, but not quite at, its anterior extremity. Its form in section, at this level, is a long oval placed obliquely. It is pierced by a few large bandies of peduncular fibres. It has a thick tract of longitudinal fibres at its upper end, between it and the epithelium of the ventricle, and it is supported on its outer side by ascending fibres of the anterior commissure. Its ground substance is dense and stains darkly. Its cells are round, fusiform, or triangular with rounded angles. They measure 12 ^u, to 14 /a in diameter. They are frequently grouped in twos, threes, or fours. In 5 the transverse section of the corpus striatum is much larger than in 4. The number of peduncular fibres by which it is traversed, especially at its upper jiart, is much greater. Although the cells in the grey matter on the mesial wall of the ventricle are somewhat larger than those in the corpus striatum, it appears, as already remarked, impossible to make a distinction between these two formations. The three regions, mesial wall of hemisphere, substantia perforata anterior, and corpus striatum are anatomically indistinguishable. In 6 the corpus striatum is limited below and on the outer side by the anterior commissure. The number of penduncular fibres has increased greatly and bundles are beginning to divide its upper part into separate nuclei. In 7 the number of fibres is still further increased so that the grey matter is honeycombed with them, and a portion of the grey matter which lies beneath the ventricular epithelium is continued across the middle line beneath the velum inter- positum. This is rather an unexpected arrangement, for in the last sections of block 6, where the anterior commissure is at its maximum development, the ventricular slit reaches right down to the anterior commissure save for a thin sheet of ependyma, and the grey matter of the mesial wall is therefore completely separated from the corpus striatum, but as the vertical height of the anterior commissure is diminished, grey matter again intervenes between the commissure and the ventricle and follows the commissure across the middle line, so that the two corpora striata are in continuity through the medium of a thin sheet of grey matter which invests the anterior com- missure. It is but a very thin sheet, for it is soon occupied by the descending fibres of the fornix, which it separates fi-om the back of the anterior commissure, and yet it seems to me of importance as showing that in the absence of a corpus callosvim, the grey matter which forms the lateral and anterior w^alls of the third ventricle passes into the grey matter of the mesial wall of the hemisphere {i.e., of the region which, in brains containing a corpus callosum, becomes the septum pellucidum) and also into the grey matter of the corpora striata. 7i shows the round anterior end of the optic thalamus with fibres on the outer side DR. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. 381 separating it from the nucleus caudatus ; the velum interpositum resting on its upper surface and descending fibres of the fornix thrusting it aside from the ventricular cavity. This will be described immediately. The nucleus caudatus is hardly as yet separated from the nucleus lenticularis, but a very large number of bundles of fibres, some large, some small, are cut transversely. They clearly enter the corpus striatum from behind, and are, for the most part, destined for its substance, but in all sections a certain number may be seen emerging from the outer surface of the nucleus and joining the system of fibres belonging to the anterior commissure. It is impossible to follovvr the course of such bundles of fibres in a series of transverse sections, but there cannot, I think, be much doubt that (1) the majority of the peduncular fibres which enter the corpus striatum from behind are lost in its substance ; (2) a number of bundles traverse the corpus striatum to reach the cortex ; (3) while it is impossible to say whether all the bundles of fibres which are seen partly within and partly without this body are passing directly from the peduncle to the cortex, few if any are entering from its cortical surface to lose themselves in its substance. A remarkable difference in structure between the ventricular and extra- ventricular portions of the corpus striatum has now become established. Towards its convex surface, its cells are, as in its anterior part, both large and numerous. The ventricular portion, on the other hand, shows a denser more deeply staining ground substance, but few cells, and these are not so large as in the outer portion. 7l. The peduncular fibres may now be said to form an internal capsule. The grey commissure is seen bridging across between the thalami. 8a. The nucleus lenticularis, as a thin capsule of grey matter, sweeps round the outside of the internal capsule and above the portion of the anterior commissure exposed in this section, into the cortex of the pyriform lobe. It is traversed, or partially traversed by many bundles of fibres ; those which separate it from the nucleus caudatus difiering from those which traverse it elsewhere merely in being an association of bundles larger than the rest. On its basal side the bundles are so numerous as to leave but a network of grey matter amongst them. The nucleus caudatus is contracted into an oval column. 9h. An inner nucleus fills the concavity of the shell of the nucleus lenticularis. It occupies, with regard to the latter, the relation of nucleus ruber to putamen. It seems, however, to be split off from the optic thalamus rather than from the corpus striatum, and, although its cells are not so large as those in the rest of the thalamus, they are nevertheless larger than those of the nucleus lenticularis. 10. The putamen is so thin as almost to resemble a claustrum. It is, however, quite distinct in structure from the large nucleus ruber. A few large cells lie on its concave surface. Its lower thicker border has withdrawn from the cortex of the pyriform lobe. llG, The nucleus ruber has almost disappeared, but the putamen is still distinct. 382 DR. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. The tail of the nucleus caudatus is thicker and inclines towards the temporal cortex. 12. The putamen and tail of the nucleus caudatus, reunited, form a capsule for the central white core of the occipital lobe. It is last seen in 12d. I am unwilling to draw wide conclusions from the study of a single brain, and shall hope to compare with the brain of Ornithorhynchus brains of Marsupials and Rodents, both adult and embryonic ; nevertheless, I can hardly leave the basal grey matter of the hemisphere without pointing out that the appeai-ances presented by a series of sections carried through this region strongly suggest its formation by reduplication in front of the peduncular fibres. Nucleus caudatus and putamen form a common structure which is continuous with the wall of the hemisphere both in front of and behind the cerebral crus, and appears to be a portion of the wall of the hemisphere which has been thrust outwards (or pitted in) by the crus and optic thalamus. In minute structure the nucleus ruber resembles the thalamus rather than the corpus striatum. Cortex. — The brain is, as already remarked, totally destitute of convolution. Save for the grooves made by blood-vessels, and for the bulging of the hemisphere into occipital and temporo-sphenoidal lobes, the cortex forms a uniformly curved case for the white matter and central grey masses. It is, relatively to the size of the brain, very thick, measuring in the spiiut-hardened brain at its thinnest part, i.e., on the upper surface of the hemisphere near its internal border, 1"3 mm., at its thickest part, i.e., near the meeting place of the external and basal surface, 3"5 ram., the maximum diameter of the hemisphere being 15 mm. In minute structure it is extremely uniform, presenting neither marked stratifi- cation of the elements of which it is composed at any given spot nor notable differences in structure in diffei-ent regions, and yet it is easy to see that, both in their relative number and in their arrangement, the elements vary in disposition in different regions. The "pyramids" are very irregular in form, being as often triangular, stellate, or fusiform as distinctly pyramidal in section. Very frequently the process which would be regarded as apical, were the orientation of the section not known, is directed centrally not peripherally. Large pyramids are undoubtedly most numerous in the anterior and dorsal part of the cortex, i.e., in blocks 5, 6, 7 , dorsal to the corpus striatum. Only in the cortex of the rhinencephalon, and in certain parts of tlie inferior and external surface of the hemisphere, are the small pyramids collected into so distinct a layer as in the brain of most Mammals. The superficial stratum moleculare has a uniform thickness of about 30 /a ; it is almost destitute of cells. The small pyramids commence rather abruptly beneath the stratum moleculare DR. A. HILL OS THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. 383 and give way with every transition in size into large pyramids, the maximum diameters of which are about 22 fi and 12 jx; in the occipital region some of the pyramids have a transverse diameter of as much as 1 7 //.. Granules are scattered uniformly throughout the layer of pyramidal cells, and nowhere collected in a separate stratum. They are more numerous in the basal and posterior region of the cortex than in the dorsal and anterior region. Fusiform cells are seen in the deepest stratum of the cortex, but it is only in certain places that they can be said to form a layer. Perhaps it is justifiable to say, although our standard is uncertain, that the cortex is simple in minute structure. Optic Thalamus. — This mass of grey matter, although a part of the primary fore- brain and not of the cerebral hemisphere, is conveniently included in a description of the latter. It appears first in 7i as a round darkly staining mass supported and enveloped by fibres which enter its substance in thin bundles. The head of the thalamus is in this section separated from the ventricular epithelium by descending fibres (anterior pillai-s) of the fornix. 7k. The descending fibres of the fornix have divided into two almost equal groups ; the one of which runs as a thick bundle along the upper surface of the thalamus, the other descends to the large corpus mammillare. The coi-pus mammillare depends from the floor of the ventricle as a thick round tubercle of grey matter. Between the two bundles, the thalamus reaches the middle line, in the grey commissure. 7l. An immense number of bundles enter the thalamus from its under and outer sides. Large when they first enter, they rapidly diminish in size. Beneath the thalamus they interlace with fibres of the internal capsule and curve both inwards and outwards, in a manner which renders it unwise for me to make assertions with regard to their destination. 8 a. The thalamus has increased greatly in size and its outer part, which is traversed by many bundles of fibres, differs from its inner portion in structure. The middle, soft, or grey commissure occupies the whole of block 8 and a large part of block 7. The bridge is not by any means, however, limited to grey matter, for a distinct and considerable tract of fibres crosses in its posterior and lower portion. They spread out in all directions after passing the middle line. The portion of the thalamus which enters into the formation of the wall of the ventricle is continuous with the corpus mammillare and projects between the crura cerebri. It causes the crura to be thrown far outwards so tha,t in 8c, for example, the fibres of the crus lie on the outer side of the thalamus rather than beneath it. The thalamus is still seen in the anterior sections from block 10 in which the aqueduct of Sylvius and the grey matter around it also appear, but its outer margin is ill defined. Comparatively simple although the structure of the brain of this lowest of Mammals is, it is nevertheless vastly too intricate to allow of the analysis of its fibre-tracts by 384 DR. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. means of series of coronal sections only. I have, therefore, confined myself in my description to its more strilting features. In the hope that I may soon receive from Australia some brains of Ornithorliynclius thoroughly well hardened in a chrome-salt by an expert collector, I have abstained from describing in this paper many things which my sections show in a fairly satis- factory and unmistakable manner. Corpus Mammilla7'e. — This body is large and well-defined, and great interest attaches to its structure in this animal on account of the great size of the fornix, for it would seem to offer an ojDportunity of settling the vexed question of the relation, to the corpus mammillare and to one another, of the descending pillar of the fornix and the bundle of Vicq d'Azyr. From the decussation of the fornix a certain number of fibres enter the mesial wall of the hemisphere in front of the anterior commissure, but by far the greater number curve backwards and downwards on either side in the wall of the ventricle as the anterior pillars of the fornix, which almost rest upon the back of the anterior com- missure. In 7i the whole column appears to be descending, but in 7 J it is seen that only part, about one-half, of the fibres descend, the rest begin to ascend again, resting on the upper surface of the optic thalamus. The descending fibres sweep down at some little distance from the ventricular epithelium towards the corpus mammillare, or rather towards the deep prolongation of the floor of the ventricle which occupies the situation of this structure between the crura cerebri. A small number of large cells, with their long axes placed vertically, lie amongst these fibres just above their entrance into the back and outer side of the corpus mammillare. The bundle which passes backwards beneath the foramen of Monro and then skirts along the dorso-mesial edge of the thalamus in the situation of the tsenia thaJami needs investigation. Had I stained all the sections on one side of the brain b}^ Weigert's method it would doubtless have been possible to follow the course of these several bundles of fibres ; as it is I can merely point out that the study of the brain of Ornithorhynchus will probably enable us to determine the course of several at present ambiguous tracts. Minute Structure of the Thalamus. — The appear-ance presented by this tissue in my sections depends upon its deficient preparation ; but while I am unwilling to enter into a detailed description of the microscopic appearance of a tissue which has not been prepared especially for this purpose, I am bound to say that this brain is sufiiciently well hardened to show all the more important features of its structure, and that the different parts of the brain may well be compared with one another. The anterior tubercle of the thalamus consists of a dense, darkly- staining matrix, containing large round cells. These cells differ, however, in a marked way from the cells of the corpus striatum and of the cortex, for instead of staining darkly and exhibiting well-defined angular bodies, they belong to the type of soft cells, the bodies DR. A. HILL ON THE CEREBRUM OF ORTSTITHORHYNCHUS PARADOXUS. 385 of which are Reldom seen, since they break down in tlie hardening reagents, leaving- behind nothing but clear round vesicular nuclei lying in round spaces. It is not uniform in structure, however. Its posterior portion which spreads far outwards into the internal capsule, by the bundles of which it is traversed, contains cells which are much larger and more definite in form than any to be found in its anterior portion or in the corpus striatum. The largest of these ai'e found on its under side where they form an almost distinct group, homologous probably with one of the subthalamic nuclei. They measure as much as 30 /a by 14 jjl. Their long axes are placed horizontally, and they appear to be in connection with the white fibres which cross the middle line on the under side of the commissura mollis. Athough the largest of the cells form an ill-defined group by themselves, others almost as large are found throughout this region, and extend across the middle line in the commissura mollis. [This paper contains little more than a series of detached observations, which are published with some reluctance. Although I had been familiar with the main facts in the cerebral organization of Ornithorhynchus for some years, I had postponed their publication in the hope of obtaining a supply of well-preserved specimens, from which it would be possible to consti'uct a complete monograph. The interest which attaches, however, to the description of the corpus callosum as a chapter in the history of science, and the peculiar features in the disposition of the rhinencephalon, which may serve to throw light upon the relation to one another of the two portions of the cerebral hemisphere, seemed to justify a preliminary communication. Since this paper was written, Dr. Johnson Symington has published observations which conhrm my conclusion that the corpus callosum is absent in Ornithorhynchus^ Dr. Symington, indeed, extends this statement so that it includes all non-placental Mammals: — "It appears to me, therefore, that the cerebral commissures in the Monotremata and Marsupialia present several characters by which they can be distinguished from those of the placental Mammals, viz., (1) the anterior commissure is as large, and generally much larger, than any other transverse commissure of the cerebrum, and it unites the whole of the cortex of the two hemispheres, except the gyri dentati and hippocampi majores ; (2) they have no true corpus callosum ; and (3) the superior transverse commissure is simply a commissure for the gyri dentati arid hippocampi majores."* — A. H., June, 1893.] * Johnson Symington. "The Cerebral Commissures in the Marsupialia and Monotremata." 'J. of Anat. and Physiol.,' vol. 27, p. 69, October, 1892. MDCCOXCIII. — B. 3 D 386 DR. A. HILL ON THE CEREBRUM OF ORNITHORHYNCHUS PARADOXUS. Description of Plates. PLATE 20. Pig. 1. — Photo-lithograpli of the right half of the brain of Ornhhorhynchus from the mesial aspect. Magnified 2 diameters. The lines indicate the divisions between the blocks into which the brain Avas cut. A series of sections was taken from each block. The dura mater was left on the basal aspect of the brain. Section 4e. — Carried through the extreme anterior end of the lateral ventricle, shows the junction of the flat olfactory crus with the under side of the hemisphere. Section 4e.i exhibits the commencement of the nucleus caudatus on the outer wall of the ventricle. A distinct nucleus of grey matter is also seen at the ventro- mesial angle of the section ; the fibres of the olfactory crus form a narrow area to the outer side of this nucleus. Section 5a. — The anterior end of the hippocampus. The fascia dentata lies almost at right angles to the plane of the lai-ge pyramids of the subiculum. Fibres from the front of the commissure of the fornix are descending in the ventricular wall. Section 6d. — Diagrammatic view of a section in front of the commissure of the fornix and the anterior commissure. The appearance of the fibres as stained with Weigert's hgematoxyliu, has been added to a drawing made from a carmine preparation. The cortex is curving over into its scabbard of fascia dentata. The fibres of the fornix system constitute a capsule for the hippocampus, to which they are limited. The lateral, superior, and part o"f the mesial cortex is supplied by fibres of the anterior commissure. Section 6j passes through the anterior edge of the commissure of the fornix, and cuts the anterior commissure. It shows the continuity of the grey matter of the mesial wall above and below the anterior commissure with the nucleus caudatus. On the ventral side of the section is seen the rhiuencephalon, bounded both on its mesial and lateral aspects by fissures, which leave but a narrow sheet of fibres as a connection between it and the rest of the hemisphere. DR. A. HILL ON THE CEREBRUM OP ORNITHORHYNCHUS PARADOXUS. 387 PLATE 21. Section oBj, through the extreme anterior end of the hippocampus, shows the relation of the fibres of the fornix-comraissure to the mesial wall, and the descent of its precommissural fibres. The tract of cortex to which the fibres of the olfactory cms are applied is beginning to be separated from the general cortex by mesial and lateral fissures. Section 7f cuts the commissure of the fornix. Section 7g is carried just behind this commissure. Sections 8, 9, IOg show the way in wjiich the fornix dies away upon the sm'face of the hippocampus behind its commissure. PLATE 22. Sections 7e, 7j, 7k„, Be, 9e, from preparations stained after Weigert's hsematoxylin- method. F. Fornix. D.F. Decussation or commissure of the fornix. D.P.F. TJescending pillars of the fornix. S.C.B. Supra-commissural bundle (tsenia thalami). A.C. Anterior commissure. W.C White fibres in the commissura mollis. O.T. Optic tract. CO. Optic chiasm. F.L. Pyriform lobe (rhinencephalon), C.S. Corpus sti'iatum. X.C. Nucleus caudatus. P. Putamen. 3 B 2 Phil. Trans. im3B.} 4e 4e^ 5a Phil. Trans. \m3B.Plate 21 Phil. Trans. IS'^^. B.Plate 22. 9e. 7 J. [ 389 ] VI. The Hippocampus. By Alex Hill, M.D., Master of Doxvning College. Communicated by Professor Alexander Macalister, M.D., F.R.S. Received May 4, 1892,— Read June 2, 1892. [Plates 23-25.] Nomenclature. . It is only within quite recent times that anatomists have realized that the compariscns with which they help out their descriptions of parts of the brain are likely to become fixed as the names of the parts. Even in the beginning of this century — indeed the custom has not altogether ceased even yet — they quoted the illustrations used by their predecessors, to replace them at once by new ones, if any, more apposite, occurred to them, without recogTiising the I'isk of confusion likely to arise from the multiplicity of names. There is a curious alternation between the use of the simile as an unalterable symbol and as an aid to description. In other words, we strike in quite recent anatomical strata upon stages in word-formation which are deep down in the history of other sciences. Galen converted illustrations into cognomens for most of the bones, muscles, and other parts of the body, but the anatomists of the I7tli and 18th centuries, when writing of the parts of the brain, considered it as immaterial whether or not they used the same symbols or illustrations as their predecessors ; they did not regard its terminology as in any degree fixed, and if a new illustration occurred to them which conveyed a better idea of tlie form of the part than any as yet current, by all means let posterity make use of it ! Thus we find the region about to be described compar-ed to and named as : — Hippo- campus, pes hippocampi, bombyx, appendix bombycinus, vermis bombycinus, baco, processus cerebri lateralis, protuberantia cylindroides, cornu arietis, cornu Aramonis ; with innumerable variations in the translation of the foregoing terms into French and German :— hippocampe, corne de belier, corne d'Ammon, Ammonshorn, Widderhorn, Reil'sche Kolbe, &c. Nor does the confusion end with the selection of a suitable name. The hmitation of the part is as uncertain as its nomenclature. Thus, we find the term " pes hippo- campi" applied to the swelling within the descending horn of the lateral ventricle ; MDCCCXCIII. — B. 4.9.93. 390 DR. A. HILL ON THE HIPPOCAMPUS. to the anterior part of this swelling only, with a loose use of the term "hippocampus" for the swelling as a whole ; to the swelling plus its fimbria, or plus its fimbria and the fascia dentata. Indeed, the use of the term as including all these parts appears to be the older : " para longior cum proxime cruribus fornicis interjecta, cerebri medulla unitur et cum hippocampo comparata in arcum flectitur."* If out of all the terms in use it were open to us to choose the most apposite, "cornu Ammonis" would be the one upon which our selection would fall. This is the term preferred by ScHWALBE,t although he seems to me to overlook its descriptive value when he limits it to the " halbmondformig gekriimmter Wulst von ungefahr 50 millims. Bogenlange, der vom Balken wulst an zunachst dem vorderen concaven Rande des Trigonum ventriculi la.teralis folgt." The peculiar applicability of the term is not appreciated until the brain is dissected away from above, as often happens, almost by accident, when the brain is semi-pixtvid, and the two gently curling horns exposed in their whole extent. Exclude the fimbriae, and the illustration is almost pointless ! I should have selected the expression " cornu Ammonis " for application to the whole region had it appeared justifiable to neglect the classical expression " hippo- campus or pes hippocampi " ; but it appears that this name was already fixed before its likeness to the horns of the Horned-Jupiter occurred to any anatomist. The part was first described and named by AeantiusJ " hippocampi vel bombycini vermis ventriculos appellare libuit." DiEMERBROEK, writing in 1632, attributes the name "pes hippocampi" to Hippo- crates, but I cannot find the reference, and the anatomists of the 17th and 18th centuries who were likely to be deeply read in Hippocrates are unanimous in assigning it to Arantius. Vteussens, in his ' Neurographia Universalis,' written in 1585, just before the date of Arantius' work, while giving a somewhat elaborate description of the part, uses no name : " crura posteriores corporis callosi seu veri fornicis columnas nuncupamus .... cruribus intermediam fimbriatam veri fornicis appendicem nominamus." Caspar Bauhin,§ in 1621, says: "... parte posteriore fulcra duo obtinet, quae posterias deorsum reflexa velut novas cavitates constituere videntur, quas ventriculos hippocampi, ab aniinalis forma, posteriores denominare solent. Pedes hippoca/mpi noininantur ab Arantio." E,iDLEY,|| in 1695 : " Hippocampi Arantii or bombyces." J. and C. WenzelIT explain : " Processus cerebri lateralis, vulgo hippocampus ab Arantio qui jorocessum istum primus descripsit." * Haller, ' Elementa physiologise,' lib. 10, sect. 1, 1762. t ScHWALBE, ' Lehrbuch der Neurologie,' p. 513. J ' Anatomicarum Observationes,' cap. 3, 1687. § ' Theatrum Anatomicum,' lib. -3, cap. 13. II ' Anatomy of the Brain.' ^ ' De penitiori strnctura cerebri hominis et brutorum.' Tiibingen, 1812. DR. A. HILL ON THE HIPPOCAMPUS. 391 It seems to be cleai-, therefore, that this region was first described by Arantius, and that his comparison to the " pes equuU marini" was adopted as its name. I have been unable to trace the origin of the term " cornu Ammonis." MoRAND'" terms the region " cornes de beher " ; but, he remarks, " qu'une partie de I'embarras qui se trouve a conciher les anatomistes, vient de ce qu'ils ont donne indifferemment le nom de cornes de belier aux ventricules et aux parties qui y sont contenues. Ce sont celles-ci qui sont vraiment les cornes de belier ou I'hippocampus suivant Akantius." The three cornua of the lateral ventricle (first called lateral ventricle by Moeand) were named some time later by YiCQ d'Azyr ; the confusion did not arise, therefore, from the application of the tei'm " corne de belier" to the descending horn. GuNzt uses the term " cornu arietis." " Qu£e cornu in modum flexse sunt, atque rotundulse, totaeque in illo pai'iete exstant, hippocampi, vel arietis cornua nomi- nantur " ; but the first use of the term " cornu Ammonis " which I have been able to trace is in a series of papers " On the Human Brain," laid before the French Academy by ViCQ d'Azyr, in 1781. In obs. ix.' (p. 518) he treats of "Les cornes d'Ammon ou hippocampes," and says that " une circonvolution recourbee en maniere de crochet designe leur situation et leur forme." From his description it would be difficult to gather his view as to their external boundary, but his figures make it quite clear that the cornu Ammonis, as he understood it, included the floor of the descending horn as well as the fascia dentata. The latter he terms "la portion crenelee ou bord dentele de la corne d'Ammon." In his ' Vocabulaire,' ViCQ d'Azvr defines : " Hippocampi, pedes hippocampi, cornua Ammonis ; les protuberances medullaires recourbees qui sont situees dans la partie posterieure et inferieure des ventricules lateraux." He adds, " je distingue deux especes d'hippocampes. Les grands . . . et les petits ou ergots dit de Morand." Perhaps a more thorough search through the literature of the period preceding the French revolution would reveal the origin of the term, or it may be that ViCQ d'Azyr transformed " corne de belier " into " corne d'Ammon " without thinking the change of name worth noticing. The quotation from Haller, which is given above, is important in this connection, for Haller, although extremely careful to give synonyms and references, says nothing about cornu Ammonis, Vulgo " hippocampus " appears to be the established term. The " pes " may be omitted, for it has no meaning when apphed to the hippocampus minor, and can hardly be retained without rendering the expression cumbrous when the subiculum (pedis) hippocampi, gyrus (pedis) hippocampi, &c. are referred to. * ' Histoire de I'Academie,' 1744, p. 6. t ' Prolusio observationes anatomicas de cerebro continens altera.' Leipsic, 1750. 392 DR. A. HILL ON THE HIPPOCAMPUS. Interpretation of the Hippocampus. By all anatomists with whose works I am acquainted the whole of the hippocampus is regarded as a part of the cortex. " On doit regarder la corne d'Amraon ou grand hippocampe comme une circonvolution cerebrale d'une forme particuliere," says ViCQ d'Azyr ; containing, that is to say, the same elements as are found in other parts of the cortex (although they are differently arranged) and no elements which are not found elsewhere. The hippocampus is always spoken of as a composite of convolu- tions, the only difference of opinion being as to the number of convolutions repre- sented. Since the time when Zuckerkandl* pointed out that the strise longitudinales lateralis et medialis (the stria obtecta and nervus Lancisii, as they are usually called) belong to the mesial surface of the brain, being developed from the dorsal portion of the outer arcuate convolution of Arnold, while the ventral portion of the same con- volution becomes the fascia dentata, the problem has altered somewhat in form, but it remains essentially the same as in the days of ViCQ d'Azyr. It is taken for granted that the arcuate convolutions of Arnold are portions of the mantle, homologous with those other parts which develop into the convolutions of the general surface of the adult brain. In a recent memoir, HoNEGGERt has most laboriously analyzed the literature of this subject, and it would therefore be unnecessary for me to go through it a second time, even if I thought that it would thi'ow any light upon the subject of this paper. The sum of this investigation is that Zuckerkandl traces the stria lateralis (sen obtecta) into the fascia dentata ; Honegger traces the stria medialis into the fascia dentata, while the lateral stria passes, he says, into the superficial layer of the gyrus hippocampi. GanserJ denies that the fascia dentata is continued in any way on to the dorsal surface of the corpus callosum. Giacomini's views will be discussed later. It is doubtless of great importance that these stages in the development of the margin of the porta should be traced with exactitude, but before the facts can be applied to the determination of the homology of the hippocampus certain antecedent questions nuist be answered, and still earlier stages in the development of Arnold's convolutions must be worked out in order that we may be placed in a position to judge whether they are in any respects comparable to the convolutions of the general surface of the great brain. In particular we need to know the relation which Arnold's convolutions bear to the olfactory crus and bulb. If the external stria of the olfactory tract passed into any part of the hippocampus we should have data from which to * Zuckerkandl, ' Das Riechcentrum.' Stuttgart, 1887. t Honegger, " Vergleiohend-aiiatomisclie Untersuchungen iiber den Fornix und die zn ihm in Bezieh- ung gebrachten Gebilde im Geliirn des Menschen und der Saugethiere." ' Recueil ZooL Suisse,' vol. 6, Nos. 2 and 3, pp. 201-435. December, 1890. I Ganser, " Gehirn des Maulwurfes." ' Morphol. Jahrb., vol. 7, 1882, pp. 691-725. DR. A. HILL ON THE HIPPOCAMPUS. 393 speculate ; but this is not the case. Although I have cut a great number of sections in all possible planes from the temporal pole of brains of various animals, I am unable to trace the ultimate connection of the external stria, but T can -with, confidence assert that it does not go, as Herrick* conjectures, into the lamina nuclearis and superficial bundle of the fascia dentata. Until, therefore, something more is known as to the morphogeny and histogeny of Arnold's convolutions, no light will be thrown upon the morphology of the hippocampus by the study of the relations of the several portions of these convolutions in the adult brain. Whatever its origin, all anatomists who have written on the subject up to the present time agree in regarding the fascia dentata as a convolution, or as a part of a convolution, and as this is not the view of the writer of this paper, little would be gained by giving an account of the attempts which have been made to homologize it with other convolutions. Sueli theories as haye been published may be classified in three groups : — (a) Fascia dentata 2J^«s nucleus fascias dentatse constitute one convo- lution, the small cells of the stratum granulosum being the small pyramids of the gyrus (subiculum) hippocampi collected together in a sepai-ate cluster (Henlb, Krause, Meynert, Hugtjenin, Schwalbe, Obersteiner, &c.) ; (b) the fascia dentata is an independent convolution, cells of all layers being represented in its granules (Ditval) ; (c) it is an independent convolution, but inverted (Golgi, Sala). The view last citedt would appear at the first moment to approach most nearly to my own. Golgi says that the reflected cortex (strato grigio circonvoluto), and the fascia dentata " non devono essere considerati come due zone di un medesimo strato, ma bensi come due distinti circonvoluzioni . . . che la fascia dentata sia una diritta continuazione anzi un' espanzione della lamina grigia circonvoluta." Such embryo- logical investigations as I have yet been able to make, lead me to believe that the inner margin of the \^'all of the great brain is folded into a pouch:, the mesial wall of which folds back to the fimbria, and is carried into the descending horn of the ventricle as the choroid plexus. As said elsewhere, I lay very little stress upon these observa- tions, which require to be checked by a full examination of the earliest stages in the development of this region, and I stould have prefei'red not to mention them at all except that they confirm Golgi and Sala's view that the layers present a " disposizione in versa." Certain appearances presented by the hippocampus at its extreme anterior end are highly suggestive of the folding over of the margin of the cortex into a pouch or trough; as is also the arrangement of the small cells in this region in animals in which the fascia dentata is rudimentary or not developed at all (see figs. 2 and 7). It does not necessarily follow, however, that the margin of the general cortex (subiculum hippocampi) rolls over into the trough in the manner imagined by Golgi. It is clear * Hereick, " Cerebrum of Opossum." ' Jl. of Comp. Neurology.' February, 1892, p. 12. t Golgi, ' Sulla fina anatomia degli organi central! del sistema nervosa,' pp. 80-112. MDCGCXGIII. — B. 3 E 394 DR. A. HILL ON THE HIPPOCAMPUS. that there is one other way in which the fascia dentata might be applietl to the subiculum ; the nucleus fasciae dentatse, or rod of large pyramids, might press against the outer side of the lateral wall of the trough. The fascia dentata would, in this case, be two-layered, and either its inner or its outer wall might be developed into nervous matter. Some appearances which I have seen, have led me to suspect that the fascia dentata is covered by a layer of undeveloped cerebral tissue (ependyma), but, on the other hand, in many cases an observation of the adult tissue favours the idea that the thickened margin of the general cortex has rolled over into the trough of fascia dentata. It is a question which can only be settled by ontological investigations. Sala* represents the fascia dentata as containing rounded cells, 10 to 20 ju. broad by 15 to 30 /A long, which give an abundantly branched process towards the periphery, after the manner of Purkinje's cells in the cerebellum. The nerve-network formed by the nervous process of the cells (which belong to Golgi's Types 1 and 2) is in apposi- tion with the nerve-network formed by the processes of the large pyramids of the nucleus fasciae dentatse. The small and large pyramids are therefore placed, as it were, back to back. As in many other cases in which portions of the brain are described and figured after the manner initiated by GoLGi, it is quite impossible to know how to harmonize or compare the descriptions with other observations. The beautiful pictures which accompany Sala's description show a single sheet of black cells, placed at considerable intervals apart, and each provided with an _ elaborately branched black peripheral process, and a single red central process which breaks up into an open network. The protoplasmic process unites with the processes of neurogleial cells. These pictures are, of course, drawn from mercury and silver impregnations. Prepared in any other way, the fascia dentata appears as a layer of " granules " close set in three or four rows, and mixed with a certain number of small fusiform or pyramidal cells (see fig. 1). I agree, therefore, with GoLOi and Sala in looking upon the fascia dentata as a peculiarly-folded portion of the wall of the fore-brain, independent of the nucleus fascige dentate in its origin, and I am not disinclined to believe from such imperfect observations as I have been able to make into its mode of origin, that the layers may be inverted. We are not at present in a position to state, however, that an embryo- logical inversion would permanently result in the projection of the processes of the neuroblasts in opposite directions. I am absolutely opposed, however, to the writers just named, and, indeed, to all anatomists with whose works I am acquainted, in their attempt to homologize the elements found in the fascia dentata with those found in other parts of the cortex ; or in other words to prove that the fascia dentata is a convolution comparable with * Sala, 'Zeitscbrift f. wiss. Zool.,' vol. 52, 1, pp. 18-46. DR. A. HILL ON THE HIPPOCAMPUS. 395 other convolutions of the general cortex. Firstly, because in structure it does not appear to me to resemble the general cortex, and, secondly, because it does not behave like a convolution. For example, it does not vary in its development with the convo- lutions of the general surface. In perfectly smooth brains, the fascia dentata is as large, or proportionally larger, than in convoluted brains. As I have pointed out in my paper on the brain of Ornithorhynchus* the brain may be absolutely destitute of convolution, and yet the rhinencephalon be cut off from the general surface with umxsual distinctness and the fascia dentata exceedingly broad. In the Ornitho- rhynchus, the fascia dentata lies wholly in the dorsal half of the cerebral hemisphere. It joins the pyriform lobe or rhinencephalon at the back of the porta. It is also interesting to notice that, since in this animal there is no corpus callosum, nothing has happened to destroy the external arcuate convolutions. If it could be proved that the stria longitudinalis lateralis passes anteriorly into true cortex as the gyrus geniculi, and posteriorly into the fascia dentata, a strong argument in favour of the latter being a convolution would be advanced, but the manner in which the fascia dentata terminates beneath the middle of the corpus caUosum in strongly osmatic brains, such as that of the Ox, have led me to doubt the validity of this description, and the views held by ViCQ d'Azyr (?), A. Ketzius, ZucKERKANDL, HoNEGGEK, GiACOMiNi, et ol. I fail to see any indication of the return of the fascia dentata to the under side of the spleniiim, in order that it may round the splenium and sweep forward in the nervus Lancisii in the manner required by the theory. To label the fascia dentata, gyrus uncinatus, as is sometimes done,t appears to me a gratuitous assumption which does not in the least facilitate the comprehension of the subject. Its development bears no relation to that of the gyrus uncinatus, as can be seen at a glance in the brain o£ Monodon monoceros, in which the uncus and gyrus uncinatus {seu hippocampi) are particularly well developed, while the fascia dentata is absent. A mystery surrounds the fascia dentata, which is not removed by attempts to homologize its several layers with those found in other parts of the cortex, and it would not serve my purpose to give in this paper a synopsis of the tables of homologies which have been drawn up by Schwalbe, Obersteinee, Golgi, and others. For purposes of comparison, it may be well, however, to give an account of the view generally held, as expressed, for example, by Schwalbe, before appending a summary of my own. Schwalbe speaks of the fascia dentata as the remains of the gyrus arcuatus Arnoldi : " . . . der rudimentare hintere untere Theil des Randbogens, der zwischen * Hill, ' PhU. Trans.,' 1893, p. 367. t Gf. Hekeick, " Brain of Opossum," he. cH., figs. 4, 10, &c. 3 E 2 396 DR. A. HILL ON THE HIPPOCAMPUS. Fimbria und Subiculum coiniu Ammonis gewissermassen eingefalzt ist und einen eigenthumlicla gekerbten granen Rand besitzt, als Fascia dentata Tarini." He treats it as a pai't of the cortex, and gives a table to sIioav the homology of the constituents of convolutions in general, the subiculum, the cornu Ammonis, and the fascia dentata. In this table the several constituents of the typical cortex are found in the fascia dentata by homologizing the fimbria and alveus with the medullary substance, the pyramids of the nucleus fascige dentatte with the large pyramids, and the stratum granulosum with the layer of small pyramids. It is easy to follow Schwalbe in his distinction between the fascia dentata and the '■' cornu Ammonis," but very difficult to see where he can draw a boundary-line between the latter and the subiculum. The ^^I'iter's view of the relation of the hippocampus to the general cortex is entirely dififerent, as he briefly pointed out in his translation of Obeesteiner's " Nervose Centralorgane."* The fascia dentata cannot, he thinks, be homologized with any layer which is to be found in the cortex elsewhere. Instead of the fascia dentata being part of the layer of small pyramids — (l) it consists, as already pointed out, of " granules," and a much smaller number of pyramidal or fusiform cells ; (2) it does not shade off into the cortex, but has a pronounced border on either side of the folded strip. Its upper border encloses in some situations a portion of the fimbria instead of its being everywhere (as it would be were it a portion of the general cortex) placed beneath the fimbria ; (3) the nucleus fasciae dentatee is not the homologue of the layer of large pyramids, but of the whole cortex. As the hippo- campus is approached through the subiculum, the many ranks of pyramidal cells arrange themselves in single file, the sheet thus formed enters the hilum fasciae dentatge, and, when within the fascia dentata, the pyramidal cells, which are not of exactly the same form or size as those in either of the layers of the cortex, form-up into an u-regular crowd. In certain sections of embryonic brains the cells of the nucleus fasciae dentatae are seen to belong to the series of the small pyramids, with which the large pyramids level-up from behind, as it were, and not with the large pyramids — so far as it is justifiable to describe these as forming a separate layer. In early embryonic brains the thin mesial wall of the lateral ventricle is seen to be doubly folded. The portion of the S-fold which is convex towai'ds the ventricle is the subiculum hippocampi, the deep concave part the fascia dentata ; but a description of the way in which these structures are formed from these folds, and of how it comes about that they interlock, must be reserved until the matter has been more fully investigated. The nucleus fasciae dentatae must, as I think, be regarded as the edge of the mantle * 'Tlie Anatomy of tlie Central Nervous Organs,' by Obersteinee and Hill, p. 362, et. seq. DR. A. HILL ON THE HIPPOCAMPUS 397 proper ; the fascia dentata as a sheath, into which this edge is received. In this sense the fascia dentata is a band of grey matter added to the cortex around a portion of its margin. It is useful to make a distinction between the cortex and the fascia dentata for descriptive purposes, albeit it is quite understood that both alike are developments from the wall of the fore-brain. I purposely use the expression " fore-brain," without the qualification " primary " or " secondary," since there are questions with regard to the development of this region which need, as it seems to me, further investigation. The Hippocampus of the Ox. I do not propose in this paper to give an elaboi-ate account of the minute anatomy of the well-developed hippocampus, nor could I do so with advantage without a large number of illustrations. Before, however, describing the hippocampus, or rather the want of hippocampus, in marine Mammalia, it is necessary to call attention to certain points in the structure of a well-developed hippocampus, and it is desirable that the animal selected should be as nearly related to the Cetacea as may be, and that its brain should be of large size. It seemed to me, therefore, that I could not do better than select for my standard of comparison, the brain of the Ox. In the Ox the outer margin of the slit through which the velum interpositum enters the lateral ventricle is folded over into the structure known as the hippo- campus, of which the fimbria and fascia dentata are best regarded as parts. Commencing on the inner side of the temporo-sphenoidal lobe it passes backwards and upwards to the under side of the corpus callosum, which it joins some little distance in front of the splenium. It lies in contact with the corpus callosum to about its centre, when the folding over ends rather abruptly, by a rapid diminution in the extent of the folding, that is to say, the fimbriee are then for a short distance separated from the corpus callosum by a small amount of grey matter only, after which, as the anterior pillars of the fornix, they sweep downwards along the back of the septum pellucidum. Except for a distance of about 3 millims. the fimbrise are separated from the corpus callosum by a considerable intei'val. There are no posterior pillars of the fornix in the sense of pillars which intervene between the body of the fornix in contact with the corpus callosum, and the fimbrise hippocampi. The hippocampus presents no digitations at its inferior end similar to those which mark the floor of the descending horn of the lateral ventricle in the Human brain, and from which the comparison with the foot of the Seahorse was derived. The hippocampus can only be described as a folding over of the cortex about the 398 DR. A. HILL ON THE HIPPOCAMPUS. dentaiy fissure ; its extreme edge ending in the fascia dentata Tarini, and its upper or convex surface bearing the fimbria, which gradually increases in thickness from its commencement at the apex of the ventricular slit. A typical section through the middle of the hippocampus in the Ox has the following characters : — As the cortex approaches the dentary fissure the ai'rangement of its cells is changed. If the section is really carried through the middle of the hippocampus, i.e., near the splenium corporis callosi, where the caudex of the rhinencephalon is narrow, the change in arrangement commences at the ectorhinal (collateral) fissure. If it happened that the section was carried through the pyriform lobe a special description of the arrangement of the pyramidal cells would be necessary. On the outer or under side of the ectorhinal fissure, the molecular layer has a uniform thickness of •4 millim., the pyramids are arranged, as^everywhere else in the cortex, according to size, the largest being the farthest from the surface, but in a section taken from behind the pyriform lobe, it is sufficient to state that the cortex has its ordinary appearance. In the gyrus hippocampi [seu uncinatus in the wider use of the term) the cells instead of forming a number of layers are pressed together into a sheet some two or three cells thick. The cortex falls rapidly outwards towards the lateral ventricle, from which it is separated by the thin sheet of white fibres known as the alveus. At the summit of the great curve upon which the fimbria rests, the cells are truly pressed into a single layer. Sweeping down for a short distance beyond the fimbria, the sheet of cells enters the hilum of the fascia dentata. As soon as this is entered the single layer gives place to a crowd of cells disposed with great irregularity, but still in most sections showing a tendency to concentration on its upper side (the concavity of the curve). Certain peculiarities in the form of the cells will be mentioned presently. Where the cortex begins to fall back towards the ventricle its molecular layer is greatly increased in thickness, from "4 millim. to 1"4 millim. It is not, however, increased in amount relatively to the superficial expansion of the layer of large cells, since it occupies the inside of a sharp curve. The molecular layer can hardly be said to enter the hilum fasciae dentatse. The number of white fibres on the surface of the molecular layer also undergoes a considerable increase, the fibres lying tangentially for the most part and marking the boundary between the molecular layer of the cortex and the molecular layer of the fascia dentata, where these would otherwise be in contact beneath the bottom of the dentary fissure. Blood-vessels which enter from the bottom of this fissure also limit the two regions in certain sections. The names which have been applied to these several layers do not, as it appears to me, aid the description, but rather introduce a source of confusion by suggesting differences in kind, where there is only a change in DR. A. HILL ON" THE HIPPOCAMPUS. 399 the arrangement of the elements which make up the cortex in the subiculum hippocampi and other parts. The only essential alteration consists in the disposition of the pyramidal cells in a thin sheet Between the bases of these cells and the ventricle lie the usual fusiform cells (stratum oriens) very few in number, the medullary substance (alveus), and ependyma. On the concave or mesial side of the curve lie the apical processes of the pyramids (stratum radiatum) ; the usual molecular layer, very thick, as ali'eady mentioned (stratum moleculare — I can see no object whatever in distinguishing the deeper part of it as stratum lacunosum) ; a sheet of white fibres such as are found else- where in the molecular layer, although not, perhaps, in such numbers ; and then, as far as the bottom of the dentary fissure, the pia mater or surface-ependyma, from the cells of which i-emarkably coarse processes radiate into the substance of the cortex. The fascia dentata I can only regard and describe as something superadded to the reflected cortex. It may, of coui'se, be objected that, if the fascia dentata is a portion of the wall of the fore-brain, it is, ipso facto, a portion of the cortex, but this is a very unworkable definition of the term " cortex." It is far better to limit the name " cortex " to all such parts of the grey matter on the surface of the cerebral hemi- spheres as have a certain typical structure. Portions of the wall of the fore-brain remain as " ependyma," other parts acquire the typical cortex structure, while still another part becomes the fascia dentata. The fascia dentata is a strip of tissue into which the thickened edge of the cortex is received. It is almost uniform in thickness and constitution. Its borders are well defined, and show no tendency to shade off into the superficial layers of the cortex. If it could be stripped ofi" from the border of the reflected cortex, it would be found to be a riband of uniform width and thickness ; in the Ox about 9 millims. wide by 50 millims. long. Proceeding from the nucleus fascise dentatse outwards it consists of the following layers : — •(!.) The stratum granulosum which appears in most prepara- tions as a sheet of nuclei some three or four deep, lying in clear spaces. The greater number of these nuclei belong to " granules." The sheet of nuclei comes quite to the surface on the mesial side. On the outer side it ends abruptly, and is in certain situations most clearly marked off" from the adjacent molecular layer of the cortex by the fibres which occur in the superficial layer of the latter. (2.) The molecular layer has a thickness of "4 millim. in its centre, but increases in thickness from its mesial to its outer bordei', and presents, so far as one can see, much the same constitution as in other parts of the cortex, unless it is, as I am inclined to think, somewhat more dense in texture. It is covered by (3.) the pia mater. Beneath the pia mater a few medullated fibres are to be seen running tangentially (stratum marginale). Fig. 1 was taken from a portion of the fascia dentata of the calf from near the anterior end. The brain was hardened in warm bichromate of ammonia (2|- per cent.), well washed out in 30 per cent., 50 per cent., 75 per cent, alcohol., stained in carmine- 400 DR. A. HILL ON THE HIPPOCAMPUS. alum en bloc (the excess of carmine being washed out in 30 per cent, alcohol), the tissue embedded in celloidin and cut frozen in gum. Sections were then stained with Weigert's hiematoxylin method. The previous staining in carmine-alum causes the cells and their processes to take a black or grey coloration. It is a modification of Weigert's method which I introduced six years ago, and have since used almost invariably in my own work. It may not be out of place perhaps to call the attention of histologists to the importance of stating how the tissue has been hardened before giving measurements of cells. Great confusion arises in the case of the central nervous system from want of information on this point, since we are often dependent upon accurate measure- ments of size for the only differentia between cells. Measurements are only com- parable when taken from specimens of brains not only hardened in precisely similar ways, but of exactly equal freshness at the commencement of the process. In sections stained according to the manner just described the blood corpuscles are dai'k and regular in outline, and may serve, therefore, as a test of the amount of shrinkage wliich the tissue has undergone. The blood-capillaries are numerous in the fascia dentata. They traverse it verti- cally or obliquely. The corpuscles in this specimen have an average diameter of 4 /x, i.e., about half then- size when living. In examining the fascia dentata we have to look to the nuclei for information concerning the nature of the cells of which it is made up. The section stained as above described shows, in addition to blood corpuscles, four forms of nucleus. 1. Nuclei exactly similar to those found throughout all parts of the brain, 3 "6 /a iir diameter, round, darkly-staining. These are the nuclei of neurogleia cells. They are often seen to occupy the centre of a clear space, which indicates that the ground substance has shrunk away from them. It is important to bear this in mind when examining the granules, as otherwise one might be teinpted to believe that the granules, which also occupy clear spaces, are surrounded during life by cell-bodies of exceedingly soft and deliquescent protoplasm which disappears in the hardening process. 2. Dark round or oval nuclei, 7 or 8 /a in diameter. 3. Clear dark -bordered nuclei 7, 8, or, in some cases (when oval) 12 or 13 fi diameter, with conspicuous nucleoli. The differences between these two kinds of nuclei have exercised me for a long time past, not only when examining the fascia dentata, but in the cerebellum and elsewhere, but I am inclined to think that the marked contrast in appearance depends upon a different behaviour towards the staining-reagent, which again depends upon the age or state of activity of the nucleus. It ^^dll be noticed that a large number of the clear nuclei in this specimen, which was taken, as already stated, from a young animal, are in the act of dividing. I am inclined to think that we may safely conclude that the dark and clear nuclei both belong to granules, but that the former are in a DR. A. HILL ON THE HIPPOCAMPUS. 401 resting condition, while the latter are either about to undergo or have recently under- gone division. 4. Oval or triangular nuclei of 10 or 11 /x in long diameter. They are surrounded by cell-bodies, 18 to 20 /x in long diameter, and resemble very closely the small pyramids of the cortex. Measurements of a considerable number of these cells, as well as of small pyramids seen in the cortex in this section, give their nuclei an advantage in length of 1'5 /a, as compared with the small cortical pyraxiiids ; their cell-bodies are also slightly wider and shorter, but I do not think that any conclusions can be drawn from such trifling difierences. Occasionally a much lai-ger pyramid, wide and blunt in form, is to be found in the fascia dentata. Only one such is exposed in the section under description. It measures 22'2 fi by 18"5 /x. It appears to be a displaced cell of the nucleus fasciae dentatse. The ground substance of the fascia dentata constitutes an open network. The method of staining adopted, resolves it, as it does other parts of the cortex, into a soft brown substance, containing blue dots and dashes. I am not prepared, in this paper, to enter into the meaning of this appearance, but will merely point out that within the fascia dentata the brown substance contains more of the indescribably delicate blue tissue than it does elsewhere, and the latter has, in a certain degree, the appearance of lines disposed vertically to the layer. The Nucleus Fascice Dentatce. As the sheet of pyramidal cells turns over beneath the fimbria, a notable alteration in the shape of the cells is apparent. In the middle of the subiculum the cells are rather smaller than elsewhere. In its upper part they are long pyramids 14 /a in transverse diameter. As the fimbria is passed they become equilateral or irregular and stellate, with a diameter of 18 or 19 fx. It becomes in consequence difficult to say in which direction their apical processes, if they have any, point. The longer axis of the most superficial cells is, except at the apex of the nucleus, usually placed tangentially to the fascia dentata. The nuclei of neurogleial cells are scattered over the section in about the same number as in other parts of the cortex. Of other nerve cells there appear to be none. The ground substance is very transparent and uniform, and embeds a remarkably rich plexus of nerve fibres, which cross one another in all directions. Between the nucleus fasciae dentatae and the fascia dentata a large number of delicate fibres are disposed parallel with the surface. MDCCCXCIII. — B. 3 F 402 DR. A. HILL ON THE HIPPOCAMPUS. Termination of the Hijjpocamjncs at its Anterior End. As it approaches the temporal pole the hippocampal formation, considered as a whole, curves inwards towards the median line. It is not sufficient therefore to cut sections in a coronal plane, but the sections must, if they are to be comparable, be gradually inclined forwards on the outer side until they cut the coronal plane at an. angle of about 45°. Misleading results have been obtained by continuing to cut the Human hippocampus coronally right up to its anterior end. A section of the " claw " (" digitationen ") is obtained in this way. Although it expands in front its true antei-ior end lies on the surface of the brain and not within the ventricle. My sections through the anterior end of the hippocampus of the Ox follow one another at varying intervals until the region in which the hippocampus terminates is nearly reached, and then I retain all the sections until the region is passed. In the way in which the hippocampus progressively decreases in size I can recognize no point which calls for particular comment. Until it is very small it presents all the usual features which mark the folding over of the cortex and its ensheathing in fascia dentata. The mesial and lateral limbs of the fascia dentata are about equal in length. The mesial portion of the granular layer cuts the surface. The layer of pyramidal cells is disposed in the usual manner and shows no indication of a fusion with the fascia dentata. The hippocampal formation is continued beyond the extremity of the descending horn. It is marked off from the general surface of the temporal lobe by a groove ; the continuation forwards of the dentary fissure or the fissure about which the uncinate convolution (of microsmatic animals) curves over into its vincus. This groove is to be seen on the mesial surface of the pyriform lobe for a short distance in front of the termination of the hippocampus. In it lie some small blood-vessels. At its extreme anterior end the fascia dentata is continued up the mesial surface of the pyriform lobe for about twice as far as it is a millimetre farther back. It loses its outer limb and is, as it were, disposed flat on the surface instead of being DOUBLED upon ITSELF around the pyramidal cell-layer. The layer of pyramidal cells is no longer folded upon itself, and it and the fascia dentata fuse at their dorsal borders, and are continued up the surface as a very indefinite sheet of small cells, larger than those of the fascia dentata but small for pyramids. The cells are also -rather fusiform than pyramidal, with their long axes disposed tangentially upwards and downwards. • It is to be remembered that we are now in the region of the nucleus amygdaleus -which occupies the anterior end of the rhinencephalon, filling up its temporal ex- tremity (the natiform protuberance) and running forward beneath the frontal region as the pyriform lobe, or crus, of the olfactory bulb. DR. A. HILL ON THE HIPPOCAMPUS. 403 Termination of the Hippocampus behind or above. In the Ox the foldmg-over of the margin which forms the hippocampus extends about half the length of the corpus callosum on its under surface, and then comes to an end somewhat abruptly 3 millims. to 4 millims. before the anterior pillars of the fornix leave the under surface of the corpus callosum. The splenium corporis callosi extends for some distance behind the hippocampus, the interval between the two being occupied by a bulging or tubercle of the gyrus fornicatus (callosal convolution). A section through the posterior part of the corpus callosum shows the fimbria borne at some distance from the corpus callosum on the surface of the reflected cortex. The alveus is very extensive. Its white fi.bres sweep outwards and upwards to join the corpus callosum, as well as downwards into the fimbria ; about the same number of fibres going apparently in each direction. The fascia dentata is shaped like an arrow-head, as seen in section, its hilum is comparatively narrow, its nucleus well-developed, the large nerve-cells being uniformly distributed. A thin sheet of fibres separates it from the surface of the subiculum. The strata granulosum et moleculare end abruptly at the margin of the fimbria, which is, of course, of great size (8 millims. high by 4 millims. wide at its base, as seen in section) in this region. A number of scattered bundles of longitudinal fibres are found inside the sheath of fascia dentata, beneath its upper or mesial limb. The cells of the nucleus fascias dentatse are considerably larger than any of the pyramids to be found in the cortex of the subiculum. They have an average diameter in the hardened tissue of about 40 ju,, and are almost as broad as they are long : whereas the largest cells of the cortex of the subiculum at this level are not more than 12-14 /a in transverse dia.meter. The stratum granulosum in this region contains about one pyramidal cell (diameter 8 or 9 /a) to ten granules, only the diameter of the nuclei of the latter can be determined, since their cell bodies are never so distinctly visible as to be clearly defined from their neighbours. The folding over of the cortex ceases abruptly at the spot indicated, the outer edges of the two folds making an angle of 60° with one another. At neither end, therefore, does the hippocampus show a gradual transition from the general cortex of the brain. Allowing for the shortest distance necessary, so to speak, in ending the fold, the hippocampal formation is well and comparatively evenly developed in its whole extent. 3 F 2 404 DR. A. KILL ON THE HIPPOCAMPUS. Hypeboodon rostratus. (Figs. 2, 7, 8.) I have for some time been acquainted with the peculiarities of the hippocampus in the brains of aquatic Mammals as illustrated by the Porpoise ; on which subject I read a paper at the meeting of the British Association, in Manchester, in 1887. The Porpoise, however, has a rudimentary hippocampus, and my investigations would have been incomplete and conclusions uncertain, had not Mr. Robert Gray, of Edinburgh University and Aberdeen, brought me from the Arctic regions five brains of Hyperoodon and two brains of Monodon in a capital condition, not only for macroscopical, but also for microscopical study. I'o take out the brain of tlie Whale and to detach from it the hippocampus are tasks requiring a very consider- able knowledge of its anatomy, and I am deeply in Mr. Gray's debt for bringing me specimens, removed without injury and so admirably preserved, as to allow of the use of Weigert's staining-method. Certain peculiarities in the general form of the Cetacean brain as well as in the topography of its hippocampal regions need to be explained before an effective com- parison can be made between this absolutely anosmatic and an ordinary osmatic brain. The Cetacean brain is, as has so often been remarked, extremely convoluted, its fissures being of great depth. This renders the orientation of the brain difficult at first, but a very little study enables one to distinguish the primary and secondary from the tertiary fissures. A still more remarkable peculiarity of the Cetacean brain is its wide departure in general form from the ordinary type. The shape of the brain of my three uncut specimens is in each case distorted by the pressure of the muslin in which, as in pudding bags, they were suspended in a tub of spirits. The result was at first sight most perplexing, the whole brain being a blunt cone, its base completely hidden out of view by the orbital region, so that the medulla oblongata seemed to enter the base of the cone between the cerebellum and the frontal region of the cerebral hemispheres. The measurements of the cerebrum were in one specimen : Height, 7|- inches ; length, 5f inches ; transverse diameter, 8f inches. Unlike any osmatic brain, the length of the Cetacean brain would appear from this specimen to be less than either its height or its width. When I compared my specimens with the plates in Kukenthal's ' Walthiere'* I felt convinced that exquisitely as Kukbnthal has figured the brain in this sumptuous monograph, he has doubted the possibility of so wide a departure from the ordinary * ' Vergleicliend-anatoinisclie und entwickelungscliiclitliclie Untersucliungen an Waltliiereii,' von Dr. Willy Kukenthal. Jena, 1889. DR. A. HILL ON THE HIPPOCAMPUS. 405 Mammalian type, and has conventionalized the brain by somewhat flattening it down and opening it out. With a view to determining its natm'al form I prepared casts from the interior of two skulls in the Cambridge Museum, of one of which fig. 8 is a photo-lithograph . The extreme measurements of this cast are : Length, 1 83 millims. ; height, 160 millims. ; transverse diameter, 266 millims. It shows that the brain has the form of a truncated cone, the apex of the cone being directed upwards and but slightly backwards. Its peculiarity in shape has been considerably exaggerated in my spirit-hardened brains. Inside the skull-case the cerebro-spinal axis presents the ordinary S curvature, but in an unusually marked degree. From the foramen magnum to the pons Varolii the medulla oblongata is directed ventralwards. The plane of the base of the brain, as a whole, makes an angle of 160° with the chief axis of the skull. The long axis of the cerebral cone is vertical to its base. This is the form of the anosmatic brain exhibited in the highest degree, and is clearly due to two causes (l) the absence of the portion of the brain connected with the olfactory sense, and (2) the closing up of surrounding parts which have not this olfactory portion to displace and support them. In the long brain of an osmatic animal, as is well seen in the Ox, although it is, of course, still more obvious in the brain of a Carnivore, the convoluted portion of the hemisphere is spread out, as it were, upon the unconvoluted, or almost unconvoluted, rhinencephalon. In the Human brain the rhinencephalon is reduced in size to the com- paratively insignificant hippocampo-uncinate convolution. The Seal closely resembles Man, as will be pointed out subsequently, in the reduction of its olfactory brain. Every gi'adation between the Dog and the Whale, as regards the relative development of the olfactory brain, is to be found in the Mammalian class, and allowing for the total size of the brain, its size relative to the cross-section of the medulla, and other circum- stances which have yet to be discovered, it is probable that the relation of breadth to height to length, or, as it might be expressed, the mean of the length-breadth and length-height indices of the brain would give a fairly accurate numerical expression for the development of olfaction in different species of animals. It must not be supposed, however, that the only parts of the brain unrepresented in anosmatic animals are the rhinencephala. As I have attempted to show elsewhere,* the brain, in its general sense, is a formation of grey matter secondary to the grey matter of the axis or grey matter bordering the central canal. The " peripheral grey tube," as I termed it to distinguish it from the " central grey tube " which contains the primary centres of peripheral nerves, is disposed about the axis, and the loss of a cerebral nerve means the curtailment of both tubes as well as of all intra-cerebral * ' Plan of the Central Nervous System.' Cambridge, 1885. 406 DR. A. HILL ON THE HIPPOCAMPUS. connections between the missing parts of the two tubes. The short brain of the Whale is not only devoid of rhinencephalon, but its central grey matter also is shortened ; the intervening white matter is absent ; the great brain is reduced from an apparatus connected with five senses to an apparatus in which four only are repre- sented, the missing sense being the one which in macrosmatic animals (Carnivora especially) demands the largest share in cerebral organization. I am not prepared, in this paper, to consider what other parts of the brain are absent in the anosmatic Whale ; it is a large subject, which demands extensive and accurate investigation. It is, perhaps, worth while to point out, however, that there is great risk of error in concluding that parts absent in certain brains are directly connected in those animals in which they occur with the missing sensory apparatus ; take the anterior commissure as an example. The anterior commissure is present in the Whale, although extremely thin. It does not, however, follow from this that the a,nterior commissure is a chiasm between the olfactory nerves or tracts. Its enormous size in Monoti-emes and Marsupials precludes the possibility of this as a complete explanation of its function. The anterior commissure is certainly the commissure between the temporo-sphenoidal lobes, whatever other sets of fibres it may contain. The sense of smell has its cerebral representation in these lobes. In the Whale, the temporo-sphenoidal lobes are reduced to a minimal size, and their commissure is correspondingly deficient. The extreme retraction of the temporo-sphenoidal lobe produces an alteration in the disposition of the parts of the brain, which must be thoroughly understood before an attempt is made to cut a series of sections through the hippocampus which will be comparable with the series cut through this region in the brain of the Ox. The posterior limb of the fissure of Sylvius, instead of sloping upwards and backwards, as in other Mammals, passes almost directly upwards towards the apex of the conical brain. This is due to the fact that the temporo-sphenoidal lobe reaches so short a distance forwards on the basal aspect of the brain that the frontal lobe is allowed to roll round, as it were, into the place on the basal aspect usually occupied by the temporo-sphenoidal lobe. The posterior pillar of the fornix, and its continuation the fimbria, are continued in a very open curve backwards and downwards without any tendency to return. If, therefore, a section of the hippocampal region is to be compared with a coronal section of the brain of a Sheep or Dog, it must be made in a plane extending backwards almost horizontally with regard to the plane of the corpus callosum. The body of the fornix and its posterior pillar are adherent to the under surface of the corpus callosum. In the middle part of its course it can be detached, for its fibres are here collected into a compact biindle placed at right angles to those of the corpus callosum. At its posterior part the fibres evidently spread out in a fan on the under DR. A. HILL ON THE HIPPOCAMPUS. 407 surface of the corpus callosum, and the posterior pillar is marked by a groove on either edge, but is otherwise quite indistinguishable from the fan of fibres upon which it rests. If it be compared in this respect with the brain of the Ox already described, the extreme importance of this feature will be at once recognized ; for in the Ox no part of the fornix is really in contact with the corpus callosum. As the hippocampus is approached, the rounded inner margin of the forceps posterior crosses the posterior pillar of the fornix very obliquely, passing into the substance of the hemisphere on its outer side. The length of the hippocampal region is inconsiderable. In a large brain, weighing (in spirit) 1904 grms., in which the longitudinal distance from the genu to the splenium was 36 millims., and the length of the posterior pillar of the fornix (or rather the structure which would be the posterior pillar in an osmatic brain), from the point at which it diverged from its fellow to the attached portion of the fimbria, was 52 millims., the length of the fimbria properly so called — that is to say, of the portion of the fornix which adhered to the hippocampal convolution — was only 11 "5 millims. Indeed, this measurement is considerably in excess of the length of the real hippo- campal region. The posterior pillar of the fornix never leaves the under surface of the corpus callosum, but adheres closely to its posterior wing or tapetum. Instead of forming, for a considerable part of its course, an independent rounded column, it consists merely of fibres, not very obviously collected into a bundle, which, having begun to diverge from the middle line at some distance in front of the posterior border of the corpus callosum, pursues a long oblique course before it gains the margin of the forceps posterior. The length of the folded portion of the edge of the cortex, the hippocampus properly so called, and therefore the region in which alone the posterior pillar should be named the fimbria, did not exceed 9 millims. Of the exact measure- ment I am doubtful, for two reasons : firstly, as will be explained later on, it is very difiicult to determine the exact point at which the cortex ceases to fold over upon itself; and, secondly, the measurements depend upon the estimated thickness of the sections. It may, however, be taken as certain, that the length of the border of the cortex which is folded over in the manner which marks the cornu Ammonis, and caiTies the fimbria, is at the outside 10 millims. Closer measurement is not necessary for purposes of comparison, for, if we compare the dimensions of the region in^the anosmatic Cetacean brain with its dimensions in a megosmatic brain, such as that of the Ox, we find the most striking difference. In an Ox brain weighing (in spirit) 315 grms., the length of the cornu Ammonis, properly so called, was 50 millims. 408 DR. A. HILL ON THE HIPPO CAJVI PUS. Hippocampus of Hyperoodon x-ostratus. Series of sections of this region were made from two separate brains. The second was the more satisfactory and therefore the sections from this series need alone be described. It was not convenient and did not happen to be necessary to make a complete series. The region was however divided into nine blocks, and from each of these a number of sections were taken (a) from the front, ih) from the middle, (c) from the back of the block. It yielded therefore twenty-seven sets of sections by which it may be said to have been completely explored. Some sections from each set were stained in carmine, and others in carmine and subsequently by Weigert's method. Of the WEiGERT-sections some were bleached in ferridcyanide of potassium and others in permanganate of potassium. The hippocampal convolution commences with a sharp point which is deeply buried in a fissure between the flat posterior pillars of the fornix and the convolution which lies on the mesial side of the collateral fissure. To use any terms other than gyrus hippocampi (or uncinatus) and collateral fissure appears to me pedantic. Although the sense of smell is moderately present in Man, its development is so small as compared with the prominence of the same sense in most other animals that his microsmatic brain approaches in form the anosmatic brain of the whale so closely as to leave no possible doubt as to the homologies of these parts. There is no doubt in my mind as to the homology of the ectorhinal fissure of the Dog with the collateral fissure of Man, and consequently of the pyriform lobe and natiform protuberance of the Dog with the uncinate gyrus (or hippocampal gyrus plus the uncus) of Man, but a very lai-ge series of brains is needed to help one to realize such an extreme variation in development of corresponding parts. From its position at the bottom of a deep recess the gyrus hippocampi gradually emerges on to the surface and at the same time steadily increases in width ; or in other words the fimbria (posterior pillar of the fornix) inclines outwards into the descending horn of the lateral ventricle. In sections from block 8 the fimbria appears as a broad flat strap constricted in the middle like a Naples biscuit. In section 6 the outer limb of the biscuit is much narrower, the hippocampal convolution reaches as far outwards as the under side of the bay between the two limbs. The ependymal wall of the ventricle is attached to the lateral border of the outer lunb so that the whole of the fimbria is outside the ventricle. The two rounded ends of the biscuit give place to ridges triangular in section, the mesial ridge being the first to appear. In block 4 (fig. 2) the gyrus hippocampi lies entirely on the mesial surface, the inner ridge is raised on to the top of its convexity ; the outer ridge carries the ependymal wall on its apex. In block 3 the mesial ridge has disappeared, the ependymal ridge is small ; in block 2 the fimbria is completely lost, no ridge marks DR. A. HILL ON" THE HIPPOCAMPUS. 409 the attachment of the ependymal wall. At its first appearance in block 8 the gyrus hippocampi is simply the edge of the cortex. As the gyrus grows, traces of the reflection of the edge of the cortex down its surface are seen, but so extremely indistinct is this reflected portion«that it is made out with great difficulty. It consists only of sr^Qall triangular cells sparsely scattered in a single layer upon the surface of the molecular layer. The sheet reaches almost down to the collateral fissure. A little farther forward a slight condensation of the molecular substance is visible, and it is seen that the boundaries of this area are marked both outside, inside, and below by scattered small pyramidal cells. By block 5 we have therefore the vestiges of a fascia dentata, but so extremely indistinct is this formation that it was only after long searching that T discovered the existence of any differentiation whatever within what seemed at first to be the molecular layer of the hippocampal gyrus. It is not till we reach the front of block 4 that we come across any definite trace of the layer which is in osmatic animals the fascia dentata. In this block the layer begins to ascend the hippocampal gyrus so as to leave a space between its inferior border and the collateral fissure. Its inferior border is in 4a received into a cup of ependyma. So delicate are these transitions that one would like to reproduce a section from both blocks, but it is as well perhaps, since the number of figures must be limited, to reproduce a section from near the back of block 4, from the region that is to say which lies immediately behind the more definite vestige of a fascia dentata. 4b (fig. 2) is the most posterior of the sections taken from near the middle of this block 4. While in this section the folding over of the surface is very indistinct, in the sections immediately in front of it the rudimentary hippocampus, or, rather, the part of the cortex which would be hippocampus if ensheathed by fascia dentata, is as strongly developed as anywhere in the Whale's brain. It shows the mesial wall of the ventricle from the collateral fissure up to the ventricular slit. The fimbria is divided into two ridges, separated by a groove. To the summit of the outer ridge is attached the ependyma- wall of the ventricle, which is involuted to support the choroid plexus. The ventricle is lined by a very thick ependyma, and a similar layer of epithelium covers the surface of the brain, although it is only seen in perfection in the fissures, being, as it were, stretched almost out of existence on the convexity of the con- volutions. The cells are conical ; their nuclei lie close beneath the surface. The cells con- stitute a palisade, but, although their bases are in apposition, they do not constitute a membrane in any proper sense of the word ; indeed, I am inclined to think that gaps are left between the bases of the cells through which the cerebro-spinal or MDCCCXCIII, — B, 3 G 410 DR. A. HILL ON THE HIPPOCAMPUS. subarachnoid fluid is allowed a free communication with the intercellular spaces of the brain-substance. The apex of each ependymal cell is prolonged into a tapering- process, usually crooked, and therefore soon lost from the section, but sometimes straight and traceable for a very long distance into the brain tissue. The greater number of fibres on the ventricular side of the cortex run tangentially, but especially near the surface layers of longitudinal fibres are visible. The layer of fibres is very thick, occupying nearly one-half of the thickness of the ventricular wall. The layer of pyramidal cells, which has made a wide sweep into the region from beneath the collateral fissure, shows just at the level of this fissure a tendency to break mesially towards the surface. This curve is very pronounced, as will be seen directly, in the more anterior sections, where it constitutes the bottom of a cursive Q, of which the upper curve is formed by the reflected cortex. For a very short distance the pyramidal cells are condensed into a thin stratum, foreshadowing the single-celled sheet of the subiculum, but above this, again, they are massed together in many layers. The sheet of pyramidal cells is carried mesially beneath the fimbria vmtil it almost cuts the surface, the last of its cells being only 18 raillims. from it. The molecular layer is very broad in the gyrus hippocampi, but shows, as far as I can see, no peculiarities in structure. Near its surface are seen the few small pyramidal cells which undoubtedly represent the fascia dentata ; none of them exceed 12 /A in diameter, while the pyramids in the subiculum have an average diameter of about 20 fi. In section 3b, which is represented diagrammatically in fig. 7, the hippocampal formation is as well developed as anywhere. The layers of pyramids in the cortex are not concentrated to the same extent as in the subiculum of an osmatic brain, but still they are reduced more nearly into a single sheet in this situation than they are elsewhere. The cortex is folded outwards abrujitly towards the ventricle. It then returns upon itself beneath the fimbria, and shows at its edge a little tendency to fall over towards the aborted fascia dentata. In block 2 the cortex, as well as the aboi^ted fascia dentata, ar-e caught up on to the mesial surface of what would otherwise be the ependyma-wall of the ventricle. The hippocampus is, that is to say, continued forwards beyond the anterior end of the ventricular slit. In block 1 the hippocampus has disappeared. The most remarkable result of the examination of the hippocampal region of the brain of this animal is the discovery that at no spot in the region is there any fascia dentata. In Hyperoodon, in which the olfactory bulb and tract are COMPLETELY ABSENT, THE PASCIA DENTATA TaRINI IS ABSENT ALSO. It may appear at the first moment as if this result were simply a confirmation of Zuckerkandl's generalizations, which are based upon observations made — not, of DR. A. HILL ON THE HIPPOCAMPUS. 411 course, upon a series of sections sucli as I have just described — but upon a superficial examination of the brains of a number of Mammals, including the Dolphin. This is the case in a very limited sense, however, as will be made clear by an analysis of Zuckerkandl's results. Such au analysis will necessarily include reference to certain large questions which I wish in this paper to avoid as far as possible. The subject to which I wish to restrict myself is the anatomical constitution of the hippo- campus, and more particularly an exact determination by means of series of sections of the extent to which the fascia dentata is developed in the marine Mammalia. ZucKEKKANDL commences his summary of the olfactory lobe as follows : — * " Die Rlndentheile des Gehlrnes, welche in dieser Monographie behandelt werden, haben in gesammt Beziehungen zum Centrum des Geruchsoi'ganes. Was bisher als centrale Statte dieses Sinnesorganes angesehen wurde, ist zum grossten Theile im Gyrus fornicatus enthalten, und indem ich vor AUem liber diesen berichte, wird es moglich, zu libersehen, wie weit zur Zeit die Lehre von dem Centralorgane des Geruchsapparates gediehen ist. In der Beschreibung desselben halte ich mich haupt- sachlich an Broca's ausgezeichnete Schilderungen, welche den Gegenstand erschopfend behandeln." Broca's position is extremely clear, but, as I have tried to show elsewhere,! abso- lutely untenable. BrocaI associates the gyrus fornicatus with the rhinencephalic lobe (pyriform lobe, &c.) in a "grand lobe limbique qui peut done etre compard a une raquette dont I'anneau entourant le seuil de Themisphere, est forme en haut par le lobe du corps calleux, en has par le lobe de I'hippocampe et dont la queue est formee par le lobe olfactif." The handle of the racquet is supposed to be formed by the olfactory lobe and its peduncle, while the two strise diverge to embrace the bat ; the mesial stria going to the gyrus fornicatus, the lateral stria to the hippocampal lobe. The great limbic lobe is described and figured as the olfactory region. The proof of this con- ception is supposed to be based upon comparative anatomy, but the only witnesses whose evidence would be of value (the animals, namely, in which the sense of smell is not developed) are rejected on the ground that " la psychologle des Cetaces est actuellement (et pour longtemps sans doute) trop inconnue pour que Ton puisse savoir ou presumer quelles sont ces fonctions cerebrales qui se sont developpees chez eux au dela du degre que I'anatomie permet d'admettre cliez les autres animaux." Broca's great limbic lobe is an anatomical assumption, which is certainly not justi- fied by the relative development of its parts in animals in which the sense of smell is prominent in different degrees. Nor is it supported by the only investigation into * Loc. cit., p. 32. t 'Plan of Central Nervous System,' Cambridge, 1885. J Beoca. " Sur les centres olfactifs," ' Revue d'Anthropologie,' 2" serie, vol. 2, p. 386, and also " Le gTand lobe limbique etla scissure limbique dans la serie des mammiferes," idem, vol. L, p. 385. 3 G 2 412 DR. A. HILL ON THE HIPPOCAMPUS. the functions of the gyrus fornicatus which has been made as yet. Horsley and ScHAFEJa* assign to this region the several areas for tactile sense ; an allocation of function which will, if confirmed, j^rove that the middle portion of the gyrus forni- catus instead of being a portion of a limbic lobe is really the base of the great Roland ic lobe, which extends outwards across the surface of the brain and includes the para- central lobule, parallel convolutions, and operculum, together with a portion of the regions adjoining. If this allocation of function is justifiable, the comparatively modest dimensions of the gyrus fornicatus in the limbless Whales is not to be wondered at. The gyrus fornicatus is, however, veiy far from diminutive in Cetacea. ZucKBRKANDL endorses Broca's view with a qua,lifi.cation,t " Die aufgezahlton Resultate erlangen einen erhohten Werth, weil sie den Ruckschluss gestatten, dass wir die E.inde des Lobus hippocampi und des Stirnendes des Lobus Corporis Callosl als centrale Stiitte des Geruchsnerven anzusprechen haben." ZucKERKANDL then goes on to describe, as already explained (p. 592), that within the ring of the limbic lobe there lies another convolution which constitutes the true " gyrus marginalis " or edge of the mantle. This marginal convolution is well developed in most animals and also in the Human embryo, but in adult Primates and Whales is partially rudimentary inasmuch as it is represented in its dorsal portion by the nerve of Lancisi only. It consists of three segments, the fascia dentata Tarini (gyrus dentatus), gyrus snpracallosus (" Lancisi 'schen Streifen" when atrophied, as in Man), and gyrus geniculi (also included in the strise Lancisii in Man). Just in the same way in which Broca's speculative union into a great limbic lobe of three lobes or portions of the great brain — lobus olfactorius, lobus hippocampi, lobus corporis callosi — falls to the ground when tested by comparative anatomy, so, also, is ZuckerkaivDl's association of fascia dentata, nerve of Lancisi, and gyrus geniculi, although attractive as a speculation, opposed to certain facts. It is extremely desirable that the development of this region which surrounds the foramen of Monro should be studied with such exactitude as would leave the question quite beyond the reach of doubt. Failing precise embryological and especially histo- genetical evidence I feel some unwillingness to refer to the subject at all, but lest it should be supposed that I endorse Zuckerkandl's conclusion with regard to the wider question, I feel bound to point out certain considerations which militate against his association of the fascia dentata, nerve of Lancisi, and gyrus geniculi in a " marginal convolution." These considerations inay be best set forth after a state- ment of GiACOMiNi's views. * HoESLEY and Schafeb. " Record of Experiments on the Functions of the Cerebral Cortex," ' Phil. Trans.,' 1888, B., p. I. Bkown and Schai'Er, ditto, 1888, B., p. 303. Fkance and Schafek, 1889, B., p. 331. t Log. tilt., p. 40. ■ . . DR. A. HILL ON THE HIPPOCAMPUS. 413 GiACOMiNi's* description of this region agrees with Zltckerkandl's in general outline although differing in certain points. 1 have not, as already said, had the advantage of seeing the original paper, but Testut states that he is giving an account of GiACOMiNi's views in his ' Traite d'anatomie humaine ' (pp. 497 and 498), and adds, "J'ai controls sur un grand nombre de cerveaux les recherches du professeur italien ; elles sont exactes." I quote from this book. In the Human brain, according to GiACOMiNi, the fascia dentata terminates at its anterior end, after extending to the most anterior part of the fissure between the uncinate gyrus and its uncus, as follows : — " Arrive \k, il s'infldchit en dedans, sort du sillon et devient de nouveau visible h, I'exterieur ; il contourne alors de bas en haut le face interne du crochet de I'hippocampe et disparait, en s'attdnuant de plus en plus, sur la face ventriculaire de ce crochet. Cette extremity anterieure du corps godronne nous apparalt nette- ment, dans la plupart des cas, sous la forme d'une petite bandelette d'aspect gelati- neux, d'une couleur cendree, large d'un millimetre a un millimetre et demi. Nous I'appellerons, du nom de I'anatomiste qui Fa a la fois decouverte et bien decrite, la bandelette de Giacomint." At its posterior extremity the fascia dentata is said to join the nerve of Lancisi in the following way : — " II change d'aspect, de bossele qu'il etait, il devient lisse et uni ; il change aussi de nom et devient le fasciola cinerea. Sous ce nouvel aspect et sous ce nouvel nom, il se porte obliquement en haut et en dedans vers le bourrelet du corps calleux, le contourne de bas en haut, arrive sur sa face supdrieure et se continue la avec les tract us longitudhiaux de Lancisi, soit avec les tractus medians, soit avec les tractus lat^raux." ZucKEBKANDL and GtACOMiNi agree, therefore, in regarding fascia dentata, nerve of Lancisi, and geniculate gyrus as three portions of the same convolution, the most internal of the annular convolutions, the margin of the porta. They differ as to the connection between the fascia dentata and nerve of Lancisi. According to Zucker- KANDL, the callosal convolution constitutes the intervening segment ; according to GiACOMiisri, the intervening segment is the fasciola cinerea, while the callosal con- volution is a portion of the general cortex. I hesitate to give my own views upon this question, because I have not yet investi- gated the matter, as I hope to do soon, in the only way in which it can be definitely settled, by studying (1) a number of fresh brains, (2) series of sections through this region carefully hardened ; (3) its development ; nor have I paid especial attention to the Human brain. I have, however, made a large number of sections of this region in the brain of the Ox, and also of certain Cetacean brains, and am opposed to the view that the fascia dentata, or tissue which belongs to the ring or convolution, out of which the fascia dentata is developed, is continued on to the dorsal surface of the corpus callosvim on the following grounds : — * GiACOMiNi. " Fascia dentata del grande tippocampo nel cervello umano," ' Giornale della. Reale Accademia di Torino,' 1883. 414 DR. A. HILL ON THE HIPPOCAMPUS. 1. The little convolutions which rest against the splenium corporis callosi are indubitably continuous with the atrophied convolution which lies at the bottom of the callosal fissure, and is known as the stria longitudinalis obtecta, seu lateralis, or nerve of Lancisi, when this last expression is not limited to the stria medialis, or to the white fibres which run along the inner edge of the stria lateralis. The stria lateralis emerges from the callosal fissure, in many cases in Man, in most cases in animals, in which the corpus callosum is relatively smaller. It appeal's again on the surface, either above the genu or only in front of the genu, as the geniculate convolution. These parts, callosal convolution, stria obtecta, supra-callosal and geniculate convo- lutions are parts of the general mantle, and do not resemble the fascia dentata in minute structure. 2. In the Ox (and many other animals) the callosal convolutions, when not strongly developed, very closely resemble in form the corresponding convolutions in any human brain in ^vhich they are unusually well developed ; but their size, both in Mau and animals, varies considerably. In the Ox, however, instead of the fascia dentata ceasing beneath the splenium at the spot at which fasciola cinerea and subcallosal convolutions appear, it accompanies the fimbria to within a few millimetres of the level at which the septum pellucidum intervenes between the fornix and corpus callosum. In other words, the fascia dentata continues its subcallosal course instead of accom- panying the fasciola cinerea to the dorsal surface of the corpus callosum, nor can I see, in sections which I have made through this region, the slightest indication upon the under surface of the corpus callosum of any structure which might be regarded as the fascia dentata returning to the splenium to surmount it and join the nerve of Lancisi. These considerations seem to me to indicate that the corpus callosum in its back- ward extension breaks through a convolution which lies outside the ring from which fascia dentata, fimbria and fornix are developed. It is, so to speak, an accident that the fascia dentata terminates at the level of the callosal convolutions in Man. Turning now to the more special point : Zuckerkandl's observations upon the fascia dentata. In the second edition of his ' Twelve Lectures ' (I quote from the English translation),* Edinger says, " All the convolutions which lie near the margin of the hemisphere — the gyrus fornicatus, the gyrus hippocampi, the stria longi- tudinalis Lancisi, and the fascia dentata — are very strongly developed in animals having highly perfected organs of smell. In those which, like human beings, have small olfactory lobes, they are somewhat atrophied, and in the Dolphin, which has no olfactory lobe, they are totally undeveloped (Zuckerkandl)." * Edingek. 'Twelve Lectures on the Structure of the Central Nervous System.' Translated by ViTTUM and Riggs. Philadelphia, 1890. DR. A. HILL ON THE HIPPOCAMPUS. 415 ZuCKERKANDL in his paper makes the following references to the brain of the Dolphin, in all of which, except the last, he omits to make any mention of the fascia dentata : "Cetaceen. — Es wurde nur der Delphin untersucht. " A. Der verktimraerte lobus hippocampi besitzt einen Haken im Sinne der Primaten ; die den Mandelkern enthaltende Rindenpartie liegt liber dem lobus hippocampi. " B. Die Balkenwindung, " C Die Randwindung, und " D. Die Fimbria fehlen " (p. 19). " Cetaceen. — Untersucht wurde der Delphin. Der gyrus marginalis fehlt bis auf das kurze Stuck des dorsalen Schenkels, der in hohem Grade atrophisch ist " (p. 42). " Bei den Cetaceen, Primaten, vielleicht aber schon bei einzelnen Halbaffen, ist die ventrale Portion der Randwindung relativ schwach, die dorsale hingegen auffallend atrophisch. Die DifPerenz zwischen den beiden Portionen des Randbogens is demnach auf einen atrophischen Process zurilckzufuhren. Am moisten atrophisch ist die Rand- windung bei den Cetaceen und bei den nicht anthropoiden AflPen " (p. 56). " Cetaceen. — Untersucht wurde der Delphin. Von einer Balkenwindung ist keine Spur vorhanden " (p. 67). In his chapter, " Ueber die Bedeutung der bisher beschriebenen Rindentheile und liber das Gehirn des Delphins," we find this sentence : " denn hier findet sich sogar ein Rudiment der fascia dentata mit einem dicken stratum granulosum vor, wie aus der schematischen Zeichnung ersichtlich wird, in welcher der ganze Spitzenantheil des Hakens der Fascia angehort." From the time of Treviranus both comparative anatomists and pathologists have frequently noted the fact that absence or deficiency of the olfactory bulb is associated with deficiency of the hippocampus, using the term in its wider sense. This relation is confirmed by Zuckerkandl, to whom also belongs the merit of correcting the extraordinary errors made by Tiedemann,* who mistook the optic thalamus for the hippocampus ; and STANNius,t who describes and figures (the figures are repro- duced by Zuckerkandl), a hippocampus in the Dolphin formed upon the ordinaiy osmatic plan. Zuckerkandl, however, sets out with a case to prove ; to wit, that the outer arcuate convolution is developed into (l) the fascia dentata, (2) the gyrus supra- callosus (or Lancisi's nerve), (3) the gyrus geniculi ; the inner into the fimbria and fornix. The Dolphin is used as a proof that these three parts are proportionately developed and vary as the olfactory bulb, and the case, as based upon the Cetacea, breaks down on every count. * TiEDEMANN. ' Zeitsclirift f . Physiologie,' vol. 2, 1827. t Stannius. " Ueber den Bau des Uelphingehirnes." ' Abh. a. d. Gebiete der Natiirw.,' Hamburg'. Quoted by ZUCKEEKANDL. 416 DE. A. HILL ON THE HIPPOCAMPUS. 1st. He says that in the Dolphin the fimbria is absent; it is present in all the marine Mammals which I have examined ; the only specimen of the Dolphin's brain which I possess is not sufficiently well preserved for this investigation. 2nd. He says that the callosal convolution (Balkenwindung) is absent. "With a brain which is not folded forward beneath the corpus callosum, and in which (more completely than even in Man) the posterior pillar of the fornix is adherent to the corpus callosum, there is no place for this small convolution on the back of the splenium, or between the splenium and the fornix. 3rd. The gyrus arcuatus (Randwindung) is absent. On the contrary, the stria lateralis presents much the same appearance as in other Mammals. It is present in Hyperoodon, Monodon, and Phocaena, and in the former it escapes from the position in which it is hidden at the bottom of the immensely deep callosal fissure and appears on the mesial surface as a narrow supracallosal convolution which ends in front in a gyrus geniculi. In one specimen in my possession, which has been beautifully pre- served in bichromate of ammonia, the supracallosal and geniculate convolutions have much the same appearance as in the Ox. These convolutions, as well as the callosal (or subsplenial) convolutions, are subject to great variations in development. In the Anatomical Museum, at Cambridge, there is a large model of the anterior portion of the human cerebrum which I made for the purpose of illustrating a paper at about the time at which Zuckerkandl's article on the " Riechcentrum " appeared. I was at the time ignorant of Zuckerkandl's researches, and made the model from a fresh human brain in which the geniculate convolution happened to be remarkably well developed and its connection with the stria lateralis very clear. When Zucker- kandl's article ajopeared I abandoned the paper, finding that the work had already been done ; but the examination of a considerable number of human brains and the comparison of the brains of Man and animals had shown me that the circumcallosal con- volution is exceedingly variable. Its variability is strongly suggestive of alterations in the mesial aspect of the brain, produced by the backward extension of the corpus callosum. The rudimentary condition of the nerve of Lancisi in Man and in Cetucea appears to me to depend not upon their union with the fascia dentata and association through this with the olfactory apparatus, but upon the large size of the corpus callosum. 4th. ZucKERKANDL pays but little attention to the only structure which is absent or rudimentary in Cetacea, namely, the fascia dentata. Porpoise. (Figs. 4, 5, 6, 9, 10, 11.) Although less conical than that of the Whale, the brain of the Porpoise {Phoccena communis) resembles the Whale's brain in general appearance. It is short, and extremely broad as compared with osmatic brains, being almost globular in form. DR. A. HILL ON THE HIPPOCAMPUS. 417 As in the Whale, the front part of the cerebral hemisphere is flexed upon the back part, although the base of the brain is not covered nearly so far as in the Whale. The surface is deeply fissured. With a view to showing clearly the peculiarities in shape, which appear to me to be characteristic of anosmatic brains, I prepared a cast of the inside of the skull of a young Porpoise (fig. 9). It exhibits the extreme diminution of antero-posterior length with corresponding increase in breadth and height. Since the brain fits close to the skull case, such a cast represents accurately the shape of the brain ; as is evident from the appearance of the fissures and convolutions upon all parts of its surface. Its extreme measurements are : length, 80 millims. ; height, 78 millims. ; trans- verse diameter (of the two hemispheres), 132 millims. Ge7ie7'al Form of the Hippocampus of the Porpoise. The fornix is. In the Porpoise, quite distinct from the corpus callosum. Its anterior pillar is well defined. The body, which is small and oval in section, is fused neither with its fellow nor with the corpus callosum. The posterior pillar is free. It is diflScult to say how far the fimbria reaches forwards, for it seems to die away on the uncus instead of being tucked in beneath it in the usual manner (see figs. 5, A, b, c); but the total distance from the point at which the posterior pillar first touches the cortex to the tip of the hippocampal region, in a large adult brain, was only 18 millims. The fimbi'ia is strap-shaped and, in the greater part of its extent, completely covers the reflected margin of the cortex, but I estimate (from my sections) that the length of the reflected portion is about 10 millims. IIippocam,pal Region in the Porpoise studied in a series of sections. In its general features this region corresponds closely in structure with that of the Whale. The reflected layer of cortex is throughout feebly developed, small in all its dimensions, and pressed flat against the cortex of the gyrus hippocampi. Section 9a (fig. 10) shows the form of the hippocampus in its posterior part. The fimbria is flat. It does not, as in osmatic brains, take the form of a distinct rod, but is a flat band which covers the reflected portion of the cortex, and belongs distinctly to the sheet of fibres which lines the ventricle. The pyramidal cells, which show in the gyrus hippocampi a tendency towards an arrangement in several layers, are collected at the point where the cortex folds over into a single layer, the cells in which are fairly large. In the reflected cortex the cells become fewer and smaller as the edge is approached. MDCCCXCIII. — B. 3 H 418 DR. A. HILL ON THE HIPPOCAMPUS. There is xio thickening of the layer into a rod (nucleus fascice dentatae). Nor is there any trace, at this level, of the recejDtion of the edge of the cortex into a scabbard of fascia dentata. Small though it be, the fascia dentata is nevertheless distinct in section 5a (fig, H). It consists, at this level, of some three or four layers of cells closely packed together. As in other animals the nuclei are of two kinds, nuclei of small angular nerve cells, and "granules," conventionally so called, i.e., nuclei which are somewhat smaller and darker than those of the angular cells, and lie in clear spaces with but faint traces of cell bodies in tissues as ordinarily prepared. The layer of fascia dentata, with its nuclei and superficial molecular stratum, does not fuse in any way with the layer of pyramidal cells ; but there is not, on the other hand, any appearance from which the relation of the fascia dentata to the cortex can be inferred — whether a special development of some part of the cortex or a new formation superadded. My sections do not, unfortunately, allow me to state with accuracy the length of the fa.scia dentata ; but I can assert that it is found in the Porpoise for a short distance only (at the outside G millims.), and that it nowhere reaches a greater development than in the section (5a) which I have drawn. Hippocampus of Monodon Monoceros. (Figs. 3, 12.) The brain of the Narwhal resembles that of the bottle-nosed Whale so closely in those particulars, upon which in my account of Hyiicroodon I laid especial stress, that I do not propose to call attention to them again. Its extreme flexion is well shown in the accompanying rough sketch of the brain (fig. 12), as seen from the under side after cutting through the axis in front of the cerebellum. Looking at the sketch on the right side first, Ave see, in order, the corpus callosum (its splenium), the anterior right tubercle of the corpora quadri- gemina, the posterior tubercle, the brain-stem with the fillet and crusta on its side ; and lastly, on the extreme left, the optic chiasm. Owing to its extreme flexion, all these structures can be seen at the same time after the brain-stem has been divided by a single cut. The uncinate convolution is larger than in Hyperoodon. It appears on the surface from beneath the forceps posterior of the corpus callosum, bends round the axis, and curves slightly round the temporal pole into the throat of the Sylviaa fissure. The uncus is completely separated from the rest of the convolution (or as it would be expressed in the more usual terminology, the uncinate convolution is separated from the hippocampal convolution) by a deep fissure. It is oval in form. The fimbria is not seen from the uuder surface of the brain, nor is it visible when DR. A. HILL ON THE HIPPOCAMPUS. ' 419 the temporo-sphenoidal lobe is broken off and looked at from its mesial aspect. It is hidden away within the hiatus ventriculi (choroidal fissure). Among the treasures which Mr. Gray brought me from the Arctic regions were two brains of Monodon. They were hardened in bichromate of ammonia, and it was therefore possible to stain sections made from them in htematosylin after Weigeet's method. Unfortunately, as so often happens, the chromic salt had rendered them extremely friable in texture, brittle almost to disintegration, and great care was therefore needed to fix any part which it was necessary to detach with a syrupy solution of collodion before cutting it. The larger brain must, as I judge from its size as well as from histological evidence, have come from a young animal. The smaller brain is labelled " fo3tal Narwhal." From its size, the complexity of its convolutions and the presence within it of medullated fibres and well formed nerve-cells, it is clear that the foetus must have been at full term. The right hippocampal region was removed from the larger brain and cut into a series of sections. From the smaller brain both hippocampal regions were cut up. I had hoped to find in the fcetal brain evidences of development along the ordinary lines followed in osmatic animals, but could recognize no points which seemed to me to need illustration, and therefore confine my description to the brain of the larger animal. The hippocampus of this animal was cut into ten blocks. Each block into three sets of sections (a, b, and c). Section 6b (fig. 3) is a typical section through the region in which the nearest approach is made to the formation of a hippocampus. In order that the whole section might be shown in a figure of practicable size it is magnified only 9 diameters. This necessitates a slight exaggeration in the distinctness of its details ; cells and fibres are shown with the same distinctness as when viewed through an objective magnifying 50 diameters. It was necessary also to make slight restorations in the section owing to the extreme tendency of the tissue to split, already mentioned. It 'is perfectly accurate, however, as a combination-view of three or four sections. In the upper part of the figure are seen from left to right : — (1.) A part of the optic tract ; this is an excellent landmark in all the sections ; its fibres are much larger than any others which are cut across in the several sections. (2.) A part of the cerebral peduncle (substantia innominata) with a distinct bundle of fibres lying just above the attached border of the mesial wall of the descending horn ; the tainia semicii'cularis. In sections a little further forward, block 5, the taenia semicircularis and fimbria form a common sheet of fibres ; a good many fibres running up and down the wall in the plane of the section. It is also useful to mention as giving a clue to the exact pooition of the sections, that from 3b, forwards, the anterior perforated space is 3 H 2 420 DR. A. HILL ON THE HIPPOCAMPUS. recognizable on the outer side of the ta3nia semicircnlaris owing to the presence in the tissue beneath it of extremely large oval celLs containing a jDatch of bright yellow pigment on one side of the nucleus. These are the peculiarly characteristic cells of the basal optic ganglion of Wagner. The ependyma wall of the ventricle is thick, and affords an excellent opportunity for the stud}' of the modifications which epiblastic cells undergo when reduced in function to connective tissue purposes. On its outer side the connective tissue wall forms a mass of fibrous tissue as thick as a penholder, supporting a considerable number of large and a very great quantity of small tortuous vessels. The appearance which it assumes in ail sections through this region, no matter to what animal the brain belongs, are, of course, largelj' due to contraction in the hardening agents. The plate of velum interpositum is prolonged into two pouches, of which tlie mesial or lower pouch fits into the angle beneath the fimbria, while the upper or outer pouch fits into the pocket formed at the angle between the gyrus hippocampi and cortex, which is dipping into the collateral fissure. The prolongation of the velum interpositum, beneath the fimbria, is a marked feature of all the sections. The fimbria, instead of lying as is usually the case on the summit of the hippocampus, lies altogether in the groove between the crus and the hippocampus, and takes the form of a flattened plate which covers over a prolongation of the descending horn. I have several times in this paper called attention to appearances which are suggestive of a deep folding of the cortex inwards and down- wards (as the reflected cortex), and its return upon itself as the fascia, dentata, fimbria, and ependymal wall. In the sections of the hippocampus of the Narwhal, I thought several times that I had come upon this folding, always to find, however, that it was due to the splitting of the section in the line of a blood-vessel which entered the cortex from the velum interpositum and traversed it from without inwards parallel with and very near to the surface of the summit of the hippocampal gyrus. The presence of these blood-vessels which enter the cortex from the ventricle and run towards the surface in this region is remarkable, since the jDlan of blood-supply of the rest of the gyrus hippocampi, as of other parts of the cortex, is for the vessels to enter from the pial surface. It seems to point strongly to the existence in the embryo of a ventricular pouch which has closed up in the course of development. The fimbria increases in transverse sections very rapidly from before backwards ; retaining throughout, however, its strap-like form. As already mentioned, it is separated from the alveus by a deep groove or pocket. Another small groove marks ofi" the portion of the alveus which lies beneath the fimbria as a separate column, more distinct in the sections behind 6 than in the section figured. The sections show in a very striking manner, owing presumably to tlie fact that they are taken from a young animal, the way in which the fibres of the alveus take origin in the cortex of the hIjDpocampal convolution, pass between its columns of cells. DR. A. HILL ON THE HIPPOCAMPUS. 421 and sweeping inwards beneath the floor of the ventricle, curve round the groove between the alveus and fimbria into the fimbria, in which eventually they assume a longitudinal direction. In sections of osmatic brains the formation of the fimbria from the alveus is obscured by the origin of large numbers of fibres in the hippo • campus proper, if it is allowable to restrict this term to the reflected cortex.* The fibres of the alveus are crossed at right angles by fibres which pass outwards into the white matter which lies beneath the collateral fissure. A few delicate fibres are seen near the surface beyond the border of the pyramidal cell-layer, close to the fimbria. They are also found in the cortex of the gyrus luicinatus, but none are to be seen at the extremity of the gyrus. In structure the cortex of the gyrus uncinatus resembles the cortex in other * As pointed out in the iutrodnction to this jDaper, the nomenclature of the region is in complete confusion. The term "hippocampus" oi- "hippocampus major" is used either for the swelling in the descending horn, or for this -plus the fimbria and fasoia dentata, or the terms " pes hijspocampi " are used for the swelling and " cornu Ammonis " for the fimbria and fascia dentata, or cornu Ammonis includes the whole region, or again it is limited to the swelling in the descending horn. Some definition of these terms is absolutely aece.ssary, but it is hardly possible without rearrangement to a certain extent. I am very unwilling to make any suggestions which may still further add to the confusion by introducing a new use of the words, but seeing that a certain formation of brain is found in osmatic animals which is absent in anosmatic animals, it appears to me most convenient to term the osmatic structure the " hippocampus," this being, so far as I can make out, the use which would have been made of the term by those who first described the region, had they known of the difference between the two classes of animals in this respect. I imagine that Abantius or Halleb would have said: "the Dog has a very large hippocampus while the "Whale has none." If then we use the term for the structure which the Dog possesses and the Whale does not possess, we have, at any rate, a useful anatomical distinction between the two. I should then do away with the term " gyrus hippocampi " altogether, for almost all anatomists confuse it with the gyrus uncinatus, and I should term the whole of the most internal con- vohition from the isthmus gyri fornicati downwards and forwards, the " gyrus uncinatus," with its hook, the " uncus." The ascent from the gyrus uncinatus to the fascia dentata I should term " subiculum hippocampi." The swelling in the descending horn has been known for three hundred years as " the hippocampus ;" but then this term is made to include so much more that it has no definite signification, and there seems no reason against our terming it " pes hippocampi " for the sake of distinction. The term hippocampus would then mean the reflected cortex, where its cells form a single sheet, with its enlargement into the nucleus fascias dentatte, plus the fascia dentata itself, and the fibres derived from this reg-ion which pass into the fimbria. It may be objected that I include under the term hippocampus structures which by most anatomists are known collectively as fascia dentata; but my answer is that it seems to me highly misleading to include under this latter name two entirely different tissues — (I) the piece of the layer of large pyramids which is reflected as the nucleus fascioe dentatas, and (2) the particularly distinct layer of " granules " and molecular substance which eusheaths this rod of pyramidal cells. It would be understood that the fimbria is made up of at least two distinct sets of fibres — (1) the fibres of the alveus which are derived from the uncinate gyrus ; (2) the fibres which take origin in " the hippocampus." 422 DR. A. HILL ON THE HIPrOCAMPUS. regions. As the hipjiocampns is approached the pyramidal cells are massed rather more closely together and reduced to a layer three or four cells deep, but no attempt is made to take up an arrangement in Indian file. At its veiy edge the layer is slightly thickened and turned downwards near the surface. The cells of the reflected portion of the layer are smaller than those in the rest of the cortex. For a short distance downward the cells of the reflected layer are pyramidal, angular and darkly stained ; they then give place to cells which do not stain. In these cells, Avhich are placed pretty close together-, indications of cell bodies are visible ; but as they are not stained, the form of the cell cannot be made out. Their nuclei are large and clear with a distinct nucleolus. They increase in numbers towards the inferior and mesial angle of the convolution. They resemble in appearance the cells, or rather the nuclei of the cells, found in embryonic cortex. In the Narwhal the cortex is reflected in the hippocampal region ; but no fascia dentata is devdoped. Its jjlace is taken by embryonic tissue. In an osmatic animal of the same age the hippocampus is relatively far larger and more conspicuous than it is in the adult. HippocajSipus or Phoca Vitulina. (Fig. 13). The brain of the Seal belongs to a strongly microsmatic type resembling in this respect the brain of Man. When it is viewed from the ventral side its general resemblance in form to the human brain is indeed very striking. It is almost as broad as it is long, and the orbital region of the frontal lobe is large, a considerable poi-tion of the frontal lobe having rolled over on to the basal surface. If the parts in which we are especially interested are examined, it is found that the olfactory bulb and tract are distinct although small. The tract ends in an external " stria " which skirts the outer edge of the anterior perforated space (tuber olfac- torium). I was unable to trace the passage, so distinctly seen in macrosmatic brains, of the external stria on to the surface of the pyriform lobe. The rhinencephalic lobe is small and tapers ra.pidly at its hinder end. The rhinal fissure joins the fissure of Sylvius, and then skirts the outer margin of the pyriform lobe as a deep posterior rhinal or ectorhinal fissure, which does not, however, run on into the equally deep collateral fissure, but is separated from the latter by transverse or bridging convolutions. So striking, however, is the resemblance to the human type, that there appears to be no sufliclent reason for the use of a separate nomenclature for this region. I have already insisted that the pyriform lobe, including its natiform protuberance, is the same part as the uncinate {plus the hippocampal) convolution of human anatomy. Difiicult as it is to recognize the homology between these jaarts in the macrosmatic Dog or Ox and microsmatic Man, the aquatic Mammalia supply all the intermediate steps. The crus of the olfactory bulb, or the pyriform lobe, as it is DR. A. HILL ON" THE HIPPOCAMPUS. • 423 often termed by writers who distinguish between the pyriform lobe and 'the natiform protuberance, is represented in Man by the sujoerficial region which intervenes between the internal and external olfactory strife. Adopting, therefore, the terminology of human anatomy, we should describe the uncinate convolution of the Seal as broad in front but bearing no uncus. The dentary fissure has a remarkable disposition, for, instead of ending between the uncus and the rest of the uncinate convolution, it is carried outwards and forwards around the anterior end of the temporo-sphenoidal lobe as fai' as the temporal pole. In Monodon the isolation of the uncus is still more complete. When I examined the hip^jocampal region with the naked eye, I came to the conclusion that the fascia dentatawas well developed, that it appeared on the surface beneath the fimbria and was continued forwards beyond the spot at which the fimbria disappears in the ventricular slit. Microscopical examination, however, showed this to be a complete mistake. The fascia dentata does not appear on the surface at all, but lies at the bottom of a deep dentary fissure. I had mistaken, as any one who looks at the surface of the brain is likely to do, the narrow roll of cortex which inter- venes between the fimbria and the fascia dentata for the fascia dentata itself. It would be difficult to find a name for this portion of the cortex, which is, so to speak, a convolution turned inside out and exposing its medullary layer on the surface. In the Seal, therefore, the cortex is folded over further than in any other aiaunal I have examined; the upper loop of the S (see fig. 13) almost completing the circle. The fascia dentata is hidden away at the bottom of the upper bay. In consequence apparently of its position between two convolutions, the fascia dentata is much flatter than usual, its two limbs being pressed together so that they meet at an acute angle. In my Section 4b, the hippocampal formation is shown at its greatest development. Its anatomical disposition Js curious, the cortex making a very wide sweep before ending in fascia dentata. It might almost be supposed that the gyrus hippocampi lay at the bottom of the dentary fissure, but the arrangement of the pyramidal cells shows that this is not the case, since, as soon as the cortex enters the fissure, its cells are pressed together into a thin sheet of large pyramids with very long tapering apical processes, arranged with great regularity. The small pyramids disappear from the cortex at the same point. This is the part of the gyrus hippocampi which might with propriety be called the subiculum hippocampi,* if the nomenclature of this region ever becomes entitled to claim anything like definiteness. It would be very useful to have a term which might * The term subiculum cornu Ammonisvel hippocampi is nsecl as synonymous with gyrns hippocampi; which, again, is the name applied by many anatomists to a pai't of the convolution termed by others the gyrus uncinatus. 424 - DR. A. HILL ON THE HIPPOCAMPUS. belong to th« cortex Avhich has not yet entered the fascia dentata, but has undergone this change in the disposition of its layers of cells. The fascia dentata is ill- developed ; it is relatively less extensive than even in Man. Its under (mesial) limb commences very gradually on the upper border of the dentary fissure, nearer to the mesial surface of the hemisphere than does its upper or external limb. Its granular layer appears to be made up of " granules " and small pyramidal cells in about the same relative numbers as in megosmatic brains. In the greater part of its extent the hippocampal region presents the same features as in 4b. Slightly behind this level, however (5b), the mesial limb of the fascia dentata reaches the surface of the brain. In the sections behind this it is seen to be sharply folded upon itself and continued up the surface towards the fimbria for a short distance. The course of its folding is, therefore, as follows : — In the front, where it is deeply sunk in the dentary fissure, it is folded into a V. Behind this, the angle of the V gives place to a curve. Behind this again, the mesial limb is folded upon itself in a fresh place. Suggested Terminology. So much confusion arises from discrepancies in the use of the terms applied to the hippocampal region that I am inclined to suggest, even at the risk of adding to the confusion, a revision of the tei'minology on the following lines : — The hippocampus to mean the whole of the region in which the border of the mantle bears fascia dentata. The most internal convolution, from the inferior end of the gyrus fornicatus (isthmus gyri fornicati) onwards, to be termed gyrus uncinatus ; its hook the uncus. The homologous region in megosmatic brains to be teamed the pyriform lobe ; this term to be synonymous with rhinencephalon, excluding the olfactory bulb. The pyriform lobe to be divided into its ci"us, the natiform protuberance, and its caudex or posterior tapering portion. The term gyrus hippocampi to be discarded. Subiculum to mean the portion of the cortex of the uncinate gyrus in which changes preparatory to the formation of the hippocampus are recognizable, i.e., the concentration of large pyramids and the absence of small ones. The swelling in the floor of the descending horn to be termed the pes hippocampi ; the sheet of white fibres which covers its surface the alveus. The reflected portion of the cortex, or portion which returns upon itself, declivus hippocampi ; it;3 thickened border, nucleus fasciae dentatfe. The expression fascia dentata to be rigidly limited to the layer of small cells (zona granulosa) and its superficial molecular layer. DR. A. HILL ON THE HIPPOCAMPUS. 425 The fimbria hippocampi to be reckoned as a part of " the hippocampus " in the general sense suggested above. Dentary fissure to be the fissure between the subiculum hippocampi and fascia dentata. The term hippocampal fissure, which at present is sometimes used as synonymous with dentary fissure (in the sense just defined) and sometimes for the slit by which the choroid plexus enters the descending horn, to be abandoned. Hiatus ventriculi (rather than choroid fissure, or lateral portion of rima transversa cerebri) to be the name for the slit by which the descending horn would be placed in connection with the subarachnoid space, were it not closed by the ependymal wall of the ventricle. I do not pretend that such a terminology is as satisfactory as it might be made, were it possible to overlook terms the use of which is already more or less fixed by custom, but with the exception of the introduction of the terms caudex and declivus (which are already well-known anatomical words) and the substitution of hiatus ventriculi for the confusing expression fissura choroidea or fissura hippocampi, or misapplied expression fissura dentata, I do not suggest any alteration in the terms now in use, nor any application of these terras in a manner which has not been already adopted by certain anatomists. As our knowledge of the structure of the body becomes more exact, it is necessary that terms used at first in a vague manner should be exactly defined. It is time for the brain to be treated with the same respect as has been long shown to other parts of the body. Hitherto anatomists have felt that, since so little is known with regard to the specialization of its functions, the accurate delimitation of its regions is of little moment. Summary of Conclusions. In Chief. — 1. The fascia dentata is absent from the brains of Hyperoodon rostratus and Monodon monocerus. It is but very slightly developed in Phoccena communis. In Phoca vitulina its size is small. 2. The extension of the fascia dentata in the several members of the Mammalian class varies as the relative development of their olfactory apparatus. Subsidiary. — a. The column of large cells known as the nucleus fasciae dentatse is the extreme margin of the general cortex. h. The granular and molecular layers of the fascia dentata belong to a separate portion of the wall of the fore-brain which has undergone this characteristic develop- ment. c. In osmatic brains the fascia dentata ends on the mesial surface of the brain, slightly in front of the anterior end of the ventricular slit. Towards its anterior MDCCCXCIII. — B. 3 I 426 DR. A. HILL ON THE HIPPOCAMPUS. extremity its mesial and external portions or limbs meet below in an acute angle. At its extreme anterior end the two limbs of the V are opened out into a flat band which lies on the surface. d. At its posterior or upper end the fascia dentata terminates abruptly. It is therefore a long riband, folded into a trough ; if laid out flat, the riband would be found to have a nearly uniform width and a very regular and uniform structure. e. Excluding neurogleial cells, blood vessels, and occasional large pyramids, the stratum granulosum of osmatic brains contains two kinds of cell: (1) "granules," (2) small pyramids. In a typical section from the brain of the Ox I find an average of one pyramid to eight granules. f. The anterior commissure and the fornix vary in thickness as the relative develop- ment of the rhinencephalon, although neither of these structures is absent from anosmatic brains. In anosmatic brains the posterior pillars of the foi-nix are closely adherent to the under surface of the corpus callosum. g. There is no reason for associating the fascia dentata with the striae longitudinales (nervus Lancisii), gyrus supracallosalis, and gyrus geniculi, or for supposing that all these four structures belong to a single organ, which forms a part of the cortical centre for the sense of smell. The fascia dentata is a subcallosal structure ; it alone disappears in completely anosmatic animals. The stria longitudinalis lateralis and the minute convolutions (supracallosal and geniculate) into which it enlarges anteriorly are found in animals destitute of olfactory bulb or tract. h. The relative representation of olfaction in brains of different species is shown by the ratio which the length of the hemisphere bears to its other dimensions. BR. A. HILL ON THE HIPPOCAMPUS. 427 Description of Plates. PLATE 23. Fig. 1. A portion of the fascia dentata from the anterior end of the hippocampus of the Calf. Magnified 500 diameters. It consists of some granules with nuclei which stain darkly, and others with nuclei which do not stain ; several of the latter are in process of division. Small pyramidal cells are also seen amongst the pyramids. The grey matter exhibits a delicate fibrillation. It contains sparse nerve-cells. The blood corpuscles are stained darkly ; they lie in chains in the capillary vessels. Fig. 2. The hippocampus of ^?/perooc?OTO rosiraiMS. Section 4b. Stained with carmine- alum, followed by hsematoxylin after Weigert's method, slightly modified. The fimbria is seen as two ridges, from the larger of which springs the ependymal wall of the ventricle. The cortex almost reaches the surface, where it ends with a blunt border. A blood-vessel is cut across at the edge of the cortex, and beneath this a few small pyramidal cells, which represent the fascia dentata, descend on the surface of the molecular layer of the gyrus uncinatus. Fig. 3. Hippocampus of a young Narwhal (Monodon monoceros). Section 6b. Mag- nified 9 diameters. This section was carried through the region in which the traces of hippocampal formation are more distinct than in any other part. O.T., optic tract; T.S., tsenia semicircularis ; N.W., basal optic nucleus of Wagner; E.W., ependymal wall of descending horn of the lateral ventricle; V.I., velum interpositum ; F., fimbria; H., reflection of the cortex at the hippocampus; G.H., gyrus hippocampi {seu uncinatus) ; A.G., groove for an artery. PLATE 24. Fig. 4. Right hemisphere of the brain of the Porpoise [Plioccena communis) from the basal surface, to show the uncinate convolution (natiform protuberance, 3 I 2 428 . DR. A. HILL ON THE HIPPOCAMPUS. &c.}. The base of the 'tween-brain was injured, and hence details in its structure are not shown. Fig. 5. Hippocampal region of the Porpoise. Magnified f . A. As seen from the (mesial) surface. B. As seen from the lateral ventricle. C. As seen from behind and mesiaily. Fig. 6. Five blocks cut from the hippocampal region of the Porpoise brain. Each of the blocks is di-awn as it appears from the front, the four intervening blocks being omitted. Magnified -|. The sections exhibit the want of development of the reflected portion of the cortex. C.A., the most jDOsterior block, shows the strap-shaped posterior pillar of the fornix as it appears Avhere it first touches the cortex of the uncinate gyrus. P.p.f., posterior pillar of the fornix ; C/i.pl., choroid plexus, collected into a mass. 6c and 6d, the fimbria, fi., reduced to small dimensions, rests on the apex of the reflected portion of the cortex. 6e is carried through the nucleus amygdaleus, in front of the anterior end of the hippocampus. Fig. 7. A semi-diagrammatic view of a section (3b) of the hippocampus of Hyperoodon rostratus, near its anterior end ; showing the double curvature of the cortex and the looped layer of minute scattered cells which occupies the place of the fascia dentata. ]S., ependymal wall of the descending horn of the lateral ventricle ; V. III., lateral ventricle; F.D., vestige of fascia dentata; G.H., gyrus hippocampi, seu uncinatus ; F.C., fissura collateralis. PLATE 25. Fig. 8. Photo-lithograph of a cast of the interior of the skull of Hyperoodon rostratus; showing the conical form of brain which is characteristic of anosmatic animals. One-third natural size. Fig. 9. Photo-lithograph of a cast of the interior of the skull of a young Porpoise. Ttvo-thirds natural size. Fig. 10. Transverse section through the hippocampus of a Porpoise. Section 9a, taken from rather behind the middle of the region. Ch.Pl,, choroid plexus ; F., fimbria ; R.C., reflected cortex. Fig. 11. Transverse section through the hippocampus of a Porpoise. Section 5a, DR. A. HILL ON THE HIPPOCAMPUS. 429 carried through the very limited region in which the reflected cortex appears on the surface, enclosed in a sheath of fascia dentata. Fig. 12. The under surface of the brain of Monodon monoceros (foetus at full term), as it is seen after removal of the hind brain, by a transverse section through the front of the pons Varolii. The drawing shows the extent to which the brain is folded upon itself. O.cli., optic chiasm; C.C., crus cerebri; F., fillet; P.C.Q., posterior tubercle of the corpora quadrigemina ; A, C.Q., anterior tubercle ; U., uncus, completely separated from G.U., gyrus uncinatus seu hippocampi ; Sp.C.C, splenium corporis callosi. Fig. 13. Ti-ansverse section of the hippocampus of a young seal (Phoca vitulina). Section 4b, carried through the region in which the hippocampus is best developed. The fascia dentata is seen lying at the bottom of the dentary fissure. ,^A Phil. Trans. \^Q?,B.Plate 23. /A ®^ * OQ? '^^ T.S. «agaSf ^ •^Si- Phil. Trans. \^<^?>B. Plate ,4. ')fi .f 5b. .-fl./V.f. 6c. CJi.-pl. 6d. 6jE. Phil. Tra7-ts.im?>B. Plate \tis )/ c.c \ • ' X \ 11 ■f-v ir,','^i i,. '"':'''. 1 r; t ^^-/^ ?; .v '^ 1'T.1 I cT.' 1 * 7J COLUMBIA UNIVERSITY LIBRARIES This book is due on the date indicated below, or at the expiration of a definite period after the date of borrowing, as provided by the library rules or by special arrangement with the Librarian in charge. DATE BORROWED DATE DUE DATE BORROWED DATE DUE C28 (167) 60M R27 Btc 1 «1 BINDERS of ^^® Q