against each other and form a columnar epithelium, one cell thick, over the whole surface of the egg, which is thus divided into a central yolk made up of large spherical cells filled with yolk, and a surface layer which may, for convenience, be called a blastoderm, although it gives rise only to the ectodermal and mesodermal organs of the embryo, while the yolk cells give rise to the endodermal epithelium of the intestine and liver, and also, in all probability, to part of the mesoderm.

As soon as the so-called blastoderm is formed it excretes from its outer surface the so-called protoderm or chitenous embryonic cuticle, which is marked by polygonal casts of the ends of the cells, from which it is soon separated and thrown off.

Origin of the Mesoderm. An area of the blastoderm now becomes many cells thick and forms the primitive cumulus noticed by Osborn. In sections it bears a close resemblance to a corresponding part of the spider's egg, as figured by Balfour. It is five or six cells thick at the centre, and gradually thins out until, at the edges, it is only one cell thick. Our sections do not enable us to decide whether it is formed by division of the blastoderm cells or by the addition of new cells from the central yolk, and it may possibly be derived in part from each of these sources.

The ventral nervous system arises as two parallel thickenings of the germinal area, separated from each other by a median ingrowth, which grows into the surface of the yolk, where it spreads out on each side to form the mesoderm, which appears to receive additions from the yolk cells. It soon splits on each side to form a splanchnic layer on the surface of the yolk, and a somatic layer which covers the nervous system and lines the surface epithelium. As the appendages grow out the somatic layer and cavity extend into them, but soon become obliterated in all parts of the body by the development of a network of mesodermal connective tissue corpuscles, as shown at the anterior end of the body in the figure.

The yolk at first forms a continuous mass of cells which almost fill the entire body of the embryo, but it soon becomes divided into yolk somites, by the development of paired mesodermal processes or dissepiments which grow upward from the ventral to the dorsal surface, from points just outside the bases of the limbs. The division of the yolk into somites which is so well shown in Packard's figures, is the effect of the growth of these processes, which give rise to the muscles of the limbs and ento-sternum, while the latter is itself formed by a thickening of the splanchnic mesoderm on the ventral surface of the yolk.

The Origin of the Digestive Tract. At the time when the first limb-buds appear the oral invagination also appears, between and a little in front of the first pair, although the mouth soon travels backwards, as shown in the figure. The oral invagination grows upwards and forwards, as shown in the figure, and its inner end bends back upon itself, and pushes in to the yolk. It consists of an epithelial lining which soon becomes thrown into longitudinal folds, or ridges, upon which a thick chitinous cuticle is developed ; and of an investing sheath of mesoderm, which is derived from the somatic layer, and from the muscles of the œsophagus and stomach.

Our sections give the history of the endoderm with clearness which leaves little to be desired, and Limulus is remarkable for the fact that there are no endodermal structures whatever at the time of hatching. The limbs, muscles, heart, nervous system and sense organs of the newlyhatched Limulus are well developed, but so far as its endoderm is concerned it is not a larva but an embryo, and it lives a free active life and even undergoes one moult while its endoderm is in an embryonic condition. The figure shows a median longitudinal section of an embryo just before hatching. There is no anus or rectum, the oral invagination is well advanced, and has essentially its adult form, but it ends blindly, and the intestine is represented by a solid mass, k, of yolk cells, flattened by pressure, and inclosed in a splanchnic layer of mesoderm, l, which is several cells thick along the ventral middle line, where it is to become the ento-sternum, but elsewhere it is only one cell thick.

As the figure shows, the intestinal region is sharply marked out in a median longitudinal section, but this is not true of transverse sections, for the yolk extends to the lateral edges of the carapace, when it is not interrupted by the muscular dissepiments, and it is not until some time after hatching that the lateral liver-lobes are cut off from the median intestine proper, by the growth of muscles from the sides of the ento-sternum to the dorsal surface.

In its origin the liver is part of the intestine, which is, morphologically speaking, a great flat pouch, reaching to the edges of the body, and imper

fectly divided, by the growth of the muscles, into an axial intestine and two lateral hepatic pouches, each of which is again divided up into the liver tubes in the same way.

At the time of hatching both intestine and liver are solid and made up of yolk cells, inside a sheath of mesoderm. At the stage shown in the figure the yolk contains great numbers of nuclei, which are most abundant near the surface, where they are much more numerous than the yolk cells. A few days after hatching the cells themselves become smaller and more numerous, and arrange themselves in a single layer around a central space, which is formed by the absorption of the yolk. At first these cells are filled with yolk globules, like the yolk cells, which they resemble in everything except size, but after the first moult the yolk substance gradually disappears, from in front backwards, and the cells become converted into the ordinary epithelial lining of the intestine. The endodermal lining of the liver tubes is formed in the same way, but more slowly.

The anal invagination does not make its appearance until after the intestine is formed, and the rectum is very short.

The Nervous System. The ventral nerve cord originates as a pair of parallel thickenings of the blastoderm, which become free from in front backwards, and become divided up into ganglia with longitudinal and transverse commissures. The commissure between the ganglia of the first pair of appendages is in front of the mouth, and there is a distinct pair of ganglia for the seventh pair of appendages. The brain is formed, in part, from the anterior ends of the nerve cords, and in part from two independent ectodermal ingrowths.

The Eyes. Ray Lankester has recently shown that there is a fundamental difference in the structure of the lateral and the median eyes, inasmuch as the sensory cells of the former lie in and are part of the surrounding ectoderm, while the median eyes consist of a vitreous body formed from the ordinary ectoderm cells, and inside this, a retina which is distinct from the adjacent ectoderm. This structural difference corresponds to a fundamental difference in the origin of the two kinds of eyes. The lateral eyes originate, very early in the life of the embryo, by specialization of the ectoderm cells, while the sensory portion of the median eyes is invaginated from the ectoderm of the ventral surface, on the middle line, between the divergent optic lobes of the brain, as shown at h in the figure.

General Conclusions.—To the interesting question-what other embryonic history most closely resembles that of Limulus, there can at present be but one answer. So far as our present knowledge goes the embryonic history of Limulus finds its closest parallel in the embryology of the Arachnida.

In his paper on the development of spiders, Balfour gives the following summary of the points of difference from the crustacea. The mesoblast is formed by a thickening of the median line of the ventral plate, while in most crustacea it is budded off from the walls of an invagination which gives rise to the mesenteron. It becomes divided into somites, the lumen of which is continued into the limbs.

There is generally in the crustacea an invagination which gives rise to the mesenteron, and the proctodaeum is usually found before or not later than the stomodaeum, and both, especially the proctodaeum, are very long and usually give rise to the greater part of the alimentary tract, while the mesenteron is usually short. In the spider, as in the tracheata in general the mesenteron is very long, and the proctodaeum is always short, and is formed later than the stomodaeum.

In all these particulars Limulus agrees with the spider, and no one can compare Balfour's figures of the spider's egg with our sections, without noticing the striking resemblance in nearly every detail of structure. He failed to trace out the origin of the endoderm, but his figures and the description show that in this respect also there is, in all probability, the closest agreement with Limulus.

We have shown that the liver and intestine are, as regards their origin, parts of a large pouch, imperfectly separated by the fissures formed by mesodermal dissepiments, and that the endodermal epithelium of both is formed from the yolk cells, and Balfour says of the spider, that it would not be unwarrantable to conclude, from his observations, that the yolk compartments formed by the mesodermic septa are the hepatic coeca, each of which is enclosed in a layer of splanchnic mesoderm, while its endodermal wall is derived from the yolk cells, and that there can be very little doubt that the endoderm of the intestine owes its origin to cells derived from the yolk.

Observations upon the Chemical Composition and Coagulation of the Blood of Limulus polyphemus, Callinectes hastatus, and Cucumaria sp. By W. H. HOWELL.

The blood of Limulus polyphemus as it flows from the heart is at first of a milky white color, but soon changes on exposure to the air to a dirty bluish white. Almost immediately after escaping the blood begins to clot, though the coagulation is never so complete as to cause it to set into a solid mass. Sometimes the clot consists of isolated masses of fibrin which settle to the bottom, at other times it forms a more or less continuous mass, which gradually contracts forcing out the serum.

All efforts to prevent the coagulation by cooling with ice or saturation with neutral salts were unsuccessful. The serum obtained after removing the clot possesses a strong alkaline reaction and is of a deep blue color by transmitted light, becoming opaque whitish blue by reflected light.

Specimens of this serum neutralized with 1 per cent. acetic acid and heated in a double water bath, showed the presence of four distinct albumins, coagulating, respectively, at 58°-60° C., 68°-70° C., 74°-75° C., and 78°-80° C. The first three of these albumins are completely precipitated after being heated for five or ten minutes, but the last albumin shows the very remarkable peculiarity that the process of heating has to be repeated some fifteen or twenty times before the albumin is completely removed. Each time that the liquid is heated a well marked precipitate occurs, which does not appear to increase when the liquid is kept at 80° for even half an hour. The filtrate from this precipitate is perfectly clear, and, when heated again to 80°, gives an exactly similar precipitate, care being taken to keep the liquid neutral by the addition of a drop or two of 1 per cent. acetic acid after every one or two precipitations.

Efforts made to separate these albumins were not successful. The serum gives the following reactions with the neutral salts. Ammonium sulphate or sodium sulphate added to saturation completely precipitates all the albumins. Magnesium sulphate added to saturation gives a strong precipitate containing portions at least of all four albumins. The filtrate after this saturation could never be obtained perfectly clear. If serum was heated to 75° C., to get rid of the first three albumins, and then saturated with MgSO4, a partial precipitation was still produced. Sodium chloride added to saturation precipitates partially the albumins coagulating at 68° C. and 75° C. The clear blue serum when exposed to the air gives a flocculent precipitate containing portions of the albumins coagulating at 75° C. and 80° C. Serum diluted ten times with water gives a precipitate containing the

albumins of 70° C. and 80° C. Passing CO, through the diluted serum does not increase the precipitate, and if the precipitate formed by dilution is filtered off, and the liquid then treated with CO, no further precipitation results. Serum dialysed through parchment paper gave a precipitate containing the albumins of 70° C. and 80° C. The water used for dialysis was rain water collected in a cistern, since distilled water could not be obtained. Neutralization of the serum with 1 per cent. acetic acid gave a small precipitate having a coagulating temperature of 80° C. When the serum was made distinctly acid with the acetic acid, 1 per cent., very nearly all of the albumin was precipitated in an insoluble form, the portion not precipitated being converted to acid albumin.

From a consideration of these reactions, it is plain that the albumins of limulus serum all bear a closer resemblance to paraglobulin, or, at least, to the globulin family, than to any other known form of albumin. Their precipitation by MgSO4, dilution with water, dialysis, or simple exposure to air are all reactions that place them among the globulins.

Investigation of the blue coloring matter of the serum tended to support the view that it is a compound of copper with an albumin, similar, in a general way, to the compound of albumin and iron in haemoglobin. The color is associated chiefly with the albumin coagulating at 80° C. That the copper of the serum is united with the albumin was shown by the fact, that, when the albumins were precipitated by ammonium sulphate, no copper could be detected in the filtrate, while the precipitate after incineration gave plain evidence of its presence. The respiratory value of this compound is not very evident as far as the limulus is concerned. The blood in the heart and arteries does not possess a blue color; the blue serum does not change its color when treated with CO2, and when hermetically sealed it disappears only after many hours. Examined with the spectroscope it shows no definite absorption bands: though a large portion of the blue end of the spectrum is cut off.

The coagulum of limulus blood is formed entirely from the union of the corpuscles. Shreds of freshly formed fibrin show this very plainly, and, when the process of coagulation is watched under the microscope, beautiful fibrin networks are seen to form, starting from the corpuscles. These networks are easily stained and resemble on a large scale the fibrin network described, for man and other mammals, as originating from the blood-discs. Details of this process must be left for a subsequent paper.

Chemical examination of the fibrin shows that it is practically identical with typical mammalian fibrin, although formed directly from the substance of the corpuscles. Treated with 10 per cent. MgSO, or NaCl solution for many hours, at a temperature of 35°-40° C., it is only slightly dissolved. Treated with 1 per cent. HCl it swells up but dissolves only to a slight extent. Treated with 1 per cent. NaOH it is easily soluble, the solution being percipitated when neutralized with 1 per cent. acetic acid. The blood of the common edible crab of our coast, callinectes hastatus, was submitted to a series of experiments similar to those given for the blood of limulus. When shed, the blood coagulates with great rapidity, and this could not be prevented by the action of cold or neutral salts. The clot differs markedly from that of limulus blood in that it is much more firm, the blood setting into a solid mass. The serum separated from the clot is of a clear blue-green color that shows no opalescence by reflected light. It has a strong alkaline reaction. When carefully neutralized with 1 per cent. acetic acid and heated, the serum gives evidence of only two albumins, one coagulating at about 55° C., the other at 68°-70° C. The latter albumin shows the same peculiarity as that described for the albumin of limulus blood coagulating at 80° C., requiring some twenty or more different heatings before it is completely removed. In some specimens a minute quantity of the albumin required heating to 80° C.

Both albumins are completely precipitated by ammonium sulphate and also by magnesium sulphate. The precipitation by the latter salt is peculiar in that the filtrate from the precipitate first formed, at first clear, soon deposits a new precipitate, and this is repeated a third time. Saturation of the serum with sodium chloride throws down part of the albumin coagulating at 70° C.

Simple dilution of the crab's serum does not cause a precipitate, unless the serum is first neutralized. CO2 when passed through the serum soon decolorizes it, the color reappearing when shaken with air. If the serum is first diluted a precipitate is formed by the CO2.

The clot formed from the crab's blood is also made directly by the union of the corpuscles. This is not so evident as in the limulus blood, but in

specimens allowed to clot under the microscope, and then stained, the process can be clearly seen. The corpuscles send out long and very delicate processes which form a close network binding the corpuscles together.

The fibrin of crab's blood is much more soluble in solutions of neutral salt than that of limulus blood. It is partially soluble in solution of common salt containing from two to ten per cent. of the salt. The solution of the fibrin in 2 per cent. NaCl coagulates at 71°-72° C. Hydrochloric acid 1 per cent. swells up the fibrin but dissolves it only to a slight extent. Caustic soda 1 per cent. readily dissolves the fibrin.

The liquid of the perivisceral cavity and water-vascular system of one of our common holothurians possesses only one albumin, viz., the haemoglobin contained in the red blood corpuscles. Specimens of the perivisceral liquid allowed to stand in watch crystals give an imperfect coagulum, the corpuscles settling to the bottom as a membranous-like sediment. The supernatant liquid contains no albumin at all. The imperfect coagulation is formed by the fusion of the amoeboid white corpuscles or of thick pseudopodia which arise from them. The red corpuscles exhibit no tendency to adhere to one another, though they become entrapped in the masses of fused white corpuscles.

The study of the coagulation of the blood of these three invertebrates, taken together with the work that has been done on other forms, seems to indicate that amongst the invertebrates generally any coagulation that may occur is caused by the direct union of corpuscular elements. The work of the last few years upon the coagulation of mammalian blood shows that here too the fibrin is apparently formed directly from one kind of corpuscle, the so-called blood-discs. The striking resemblance in the method of union of the corpuscles to form the clot, in mammalian blood and the blood of limulus, for instance, and the close similarity in chemical properties of the fibrin produced, suggest that the formation of fibrin in the two cases may result from an essentially similar series of changes.

The Presence of Haemoglobin in the Echinoderms. By W. H. HOWELL.

Haemoglobin has been found in the blood of certain species of several of the divisions of the invertebrates, e. g., annelids, insects, molluscs, but up to the present time no satisfactory example of its presence among the Echinoderms has been discovered. Foettinger described in one of the ophiurians, ophiactes virens, certain red corpuscles contained in the water vascular system which, upon examination with the micro-spectroscope, gave two absorption bands similar to those of oxy-haemoglobin; from this he concluded that the coloring matter of these corpuscles is haemoglobin. But other pigments, turacin, for instance, are known to give very similar if not identical bands, so that this single observation can not be considered as demonstrating the presence of haemoglobin.

During the summer's work at the marine laboratory, the writer's attention was directed by Mr. H. F. Nachtrieb to a holothurian, whose perivisceral liquid is in some cases of a bright red color. Examination of this liquid and of the contents of the water-vascular system showed that the coloring matter has most of the properties of haemoglobin, though certain differences in its chemical reactions have led me to believe that it is not identical with the haemoglobin of vertebrate blood.

The water-vascular system and the body cavity contain, besides colorless amoeboid corpuscles, a large number of oval, nucleated, bi-convex blood discs of a pale red color. When a specimen of this water-vascular liquid is caught in a watch crystal, these corpuscles soon settle to the bottom, forming a membranous-like sediment. If the supernatant liquid is decanted, and the sediment treated with water and then filtered, a beautiful blood-red solution is obtained. This solution when examined with the spectroscope gives the two oxy-haemoglobin bands; the wave length of the middle of each band was determined, and was found to be identical with that of the corresponding band of a solution of oxy-haemoglobin of human blood of the same strength. Addition of Stokes' reducing solution causes the two bands to disappear, and brings out the single band of reduced haemoglobin. By shaking the solution with air the two oxy-haemoglobin bands can again be obtained. The corpuscles when treated with glacial acetic acid and salt give well marked haemin crystals, though these crystals do not form so readily as with vertebrate blood. When incine

rated and tested with ferrocyannide of potassium they give the reaction for iron.

These properties are quite sufficient to show that this coloring matter is a compound similar to ordinary haemoglobin, and justify its right to the title of haemoglobin. In two respects, however, it differs from the haemoglobin of vertebrate blood. An aqueous solution when heated coagulates at 58°-60° C., giving a heavy brown precipitate, the filtrate from which is perfectly free from albumin. Haemoglobin solution from vertebrate blood, on the other hand, coagulates between 70°-80° C. A solution of this haemoglobin is also precipitated by the addition of dilute, 1 per cent. acetic acid, which is not true of vertebrate haemoglobin.

It appears then that we have here a compound similar in structure and in some of its most essential properties to the haemoglobin of vertebrate blood, but differing from it slightly in the albuminous portion of the molecule. It seems quite possible that similar differences may be found when the haemoglobin of other invertebrates is more carefully examined. All efforts to obtain haemoglobin crystals were unsuccessful.

The respiratory value of this pigment cannot be doubted. It performs its functions not only through the ambulacral feet scattered over the surface of the body, but also through the respiratory tree, the ramifications of which are bathed in the perivisceral liquid, The respirations of the animal when undisturbed are very regular, consisting usually of three inspirations followed by a single prolonged expiration, the whole respiratory act being repeated from three to four times a minute. Efforts to increase the respiratory rhythm by heating the sea water surrounding the animal were not successful.

On the Existence of a Post-oral Band of Cilia in Gasteropod Veligers. By J. PLAYFAIR MCMURRICH.

The question as to the phylogeny of the Mollusca is as yet undecided, though recent researches indicate a relationship between this group and that of the Annelida. The discovery of the peculiar forms Neomenia, Proneomenia, and Chatoderma, gave a strong impulse to this idea on account of their similarity in some respects to what obtains in the Polyplacophora, but it is not safe to argue a direct descent from these forms, or even to imagine that they come into the ancestral line at all. It is quite possible that they and the Polyplacophora are offshoots from the direct stem, and probably we must look more to the embryological history of the Gasteropods for light on the subject. Attempts have been made, notably by Hatschek (Studien ü. Entwickelungsgesch. d. Anneliden, Wien, 1878), to trace a relationship between the Polygordius larva of the Annelida and the Molluscan larva, and thus to throw any relationship which may exist between the two groups back to a very early period in their evolution.

If it can be shown that there is considerable similarity between the larvæ of the two groups, and if the differences which do exist can be explained as adaptations to new conditions, the presumption as to the genetic relations between the larva will be greatly strengthened. As regards the arrangement of the cilia, which is the only point to be dealt with in this note, we have in the Polygordius larva a strong præoral locomotive band, a more delicate postoral nutritive band, and a still less apparent ciliated region lying between these two bands and leading into the ciliated mouth. The identity of the cilia of the velum of the Gasteropod larva with the first of these, has been frequently noticed; they form a strong præoral band, occasionally double, and differing from the band of the Polygordius larva only in the extent of its development, and in its incomplete closure in many cases dorsally. This latter point of difference does not, however, hold throughout; the former may be explained by the necessity for a more powerful locomotive apparatus than is required for the Annelida, caused by the presence of a shell, a structure which appears very early in the life-history of the Mollusca. A postoral band has never as yet been described for the Prosobranchs. Several observers have called attention to the presence of a single band of cilia behind the cilia of the velum, and have regarded it as nutritive in function, and the object of this note is to call attention to the fact that this band passes across the ventral surface of the larva behind the mouth, and is, therefore, quite comparable to the pastoral band of the Trochophore. My attention was first called to this fact in the larvæ of Crepidula fornicata, and I was afterwards able to confirm it in those of Fulgur carica, in a species of Neptunea, in two Prosobranch Veligers as yet undetermined, and in the

Opisthobranch Montaguia sp.? In the undetermined Veligers the velum was produced into four long flattened arms, round the margins of which were the strong locomotor cilia. On the under surface of the arms, running parallel to and not very remote from the locomotor cilia, was the finer band of nutritive cilia, the transparency of the arms and their size rendering it very apparent, and it could without much difficulty be traced across the ventral surface of the body immediately behind the mouth. Dr. W. K. Brooks informs me that he noticed the existence of this postoral band some time ago, and was then inclined to attribute some phylogenetic importance to it, but being occupied with other investigations he did not follow up his observations, and refrained from publication. Haddon also has described and figured it for certain Opisthobranchs, but does not seem to have observed it in the Prosobranchs he studied.

The region between these two bands is occupied by numerous very fine cilia, which, as in the Polygordius larva, are continuous with those lining the mouth opening and the œsophagus. The arrangement of cilia which is to be found in the typical Annelid larva is therefore almost exactly reproduced in the Gasteropod Veliger.

Arguing from ontogeny, a phylogenetic history of the Gasteropods somewhat as follows may be constructed. They and the Annelida have had their origin in a Trochophore. In the Gasteropods this ancestor developed a a univalve shell, represented by the larval shell so often replaced as development proceeds by another more highly ornamented and more complicated in structure. The development of this shell, by increasing the specific gravity of the animal, rendered the simple præoral cilia of the Trochophore insufficient for active locomotion, and the extent of the band was increased by the region of the body on which it occurred being as it were pulled out laterally, the characteristic velum being thus produced. Perhaps, too, in the presence of the shell, a reason can be found for the absence of metameric segmentation in the Gasteropods.

The Rhythm and Innervation of the Heart of the Sea Turtle. By T. WESLEY MILLS, Lecturer on Physiology, McGill University, Montreal, Canada.

This investigation was undertaken partly as a continuation of previous work on the sea turtle, a short account of which had already been published in the Journal of Physiology, but chiefly as a continuation of my work on Chelonian heart physiology in general. A paper of mine, on the terrapin's - heart, has recently appeared in Nos. 4 and 5, Vol. VI, of the journal referred to; but the whole of my work on the Chelonians is intended to furnish a systematic comparison from a physiological point of view of several of the genera and species of the Chelonians. It is thought no such systematic comparison has ever before been attempted for physiology, though it is constantly being done in morphology.

The investigation on the terrapin was carried on in the Biological Laboratory at Baltimore; those on the sea turtle, alligator and fish, at the Marine Laboratory at Beaufort, N. C. Only a very brief account of this work is furnished here, the papers in extenso having been sent for publication in the Journal of Physiology of Cambridge, England, in which also due acknowledgment is made of my indebtedness to the authorities and teachers of the Johns Hopkins University for facilitating these investigations.

In all of them the direct method of observation has been used and the heart has been experimented upon mostly in situ. For electric stimulation a Du Bois induction coil, fed by one Daniell's cell has been used, except in the case of one set of experiments on the alligator, in which a Bunsen's cell was substituted.

In order to insure the specimens of the sea turtle used being in the best possible condition, those not used at once on being caught were kept on the sea shore in a "turtle pen," so arranged as to admit of free ingress and egress of water. They were also fed on crabs, their natural diet. In all 20 specimens were employed, and of three species Chelonia Caretta (Loggerhead), Chelonia imbricata (Hawksbill), Chelonia Midas (Green Turtle). By way of comparison experiments have also been made on a limited number of specimens of the land tortoise (pyxis).

The Sympathetic System of Nerves in the Marine Turtles. It is very remarkable that Chelonia Mudas in its cervical and thoracic sympathetic should very closely resemble the terrapin, but differ widely from C. Caretta and

Imbricata. In the latter the sympathetic in the neck runs widely apart from the vagus and almost equals it in size. Its superior and middle cervical ganglia are ill-defined cordiform swellings, while the lower cervical and the first thoracic ganglia are fused together to form the Ganglion cardiacum basale. The latter is very large and gives off upwards many strong branches to the Brachial plexus, and downward to the lungs, etc., and probably to the heart.

From the middle cerv. gang. a strong branch passes to the heart along the vagus. This is a cardiac accelerator. The corresponding branch in the terrapin is much less defined and often wholly wanting.

Chelonia Midas differs greatly from C. Caretta and Imbricati, resembling very closely in its sympathetic system the terrapin. The sympathetic in the neck is very much smaller than the vagus, its ganglia well marked and it is often more or less united (though easily separable), with the vagus. But a great difference is the absence of fusion of the lower cerv. gang. and the first thoracic. These ganglia are in the terrapin often connected by an Annulus Vieussenii.

Sympathetic Cardiac Accelerators. Stimulation of the branch from the middle cervical ganglia referred to above leads in the sea turtle with greater constancy than in the terrapin to acceleration of the rate, and especially augmentation of the force of the beat of the heart. Also stimulation of the gang, card. basal. or the main chain between it and the first associated metamere below leads to acceleration and augmentation. The same laws apply to this as to vagus acceleration.

The Results of the Stimulation of the Vagus. The vagus when stimulated may arrest the auricle and ventricle in the manner described for the alligator. In no Chelonian thus far examined is the heart arrested by the vagus through the reduction of the force of the beat to zero, as in the frog. As in the terrapin unilateral vagus effects are comparatively common, i. e., a strength of current sufficing to arrest one auricle, for example, when the corresponding vagus is stimulated, has much less effect on its fellow.

Comparative Effect of each Vagus. A large number of experiments have given results very closely resembling those obtained for the terrapin i, e., in the great majority of instances the R. vagus has greater power, especially in maintaining the heart in stand-still or continued stimulation, than the left, but in both terrapin and sea turtle this difference appears to be less than in the land tortoise. This difference in the vagi seems to extend to other families of cold-blooded animals, and is to be explained not by a difference in the number of inhibitory fibres in each vagus, nor to inequality in the distribution of them, but to the fact that the right pulsatile venous area (R. sinus and veins), to which the R. vagus is mostly distributed, is the chief or dominating part in the driving machinery of the heart, for arrest of the heart is practically dependent on arrest of the sinus. In one case, in the sea turtle prolonged alternate stimulation of the vagi gave perfect cardiac inhibition for more than six hours. This is the longest case of heart standstill of this kind yet recorded for any animal. Structure of the Chelonian Heart. Between the sinns and the conspicuous auricle there is a part of the heart somewhat different in appearance, structure, and physiological qualities from either the sinus or auricle proper. A similar structure is found in some fishes. Especially have my experiments shown that the capacity for independent rhythm in this part is greater than in the so-called “bulged” part (Gaskell), of the auricles, I have therefore thought it well to name this part sinus extension and consider it a separate part of the heart. It is in reality more allied functionally to the sinus than to the auricle proper.

Very frequently in the Chelonians and especially in the marine turtles, stimulation of the vagus with a weak current suffices to arrest the auricles proper; in that case the contraction wave of the sinus is conducted along the sinus extension to the ventricle. The same occurs in the alligator and the fish. The Law of Inverse Proportion. My work on the various genera of Chelonians and the alligator have shown conclusively that whether the vagus or the sympathetic cause the cardiac acceleration and augmentation, one law invariably applies, viz: That the increase in the rate and force of the heartbeat after stimulation of the vagus or accelerating sympathetic, is always inversely as the rate and force at the time of stimulation, i. e., the slower and weaker the heart, the greater the increase. The vagus seems to be the most constant and powerful cardiac augmentor known to us.

Faradisation of the heart directly has given for the sea turtle results analagous to those obtained in other Chelonians and the fish, but in the sea turtle there seems to be less effect, especially as regards dilation around the area stimulated.

The fact that arrest of the sinus by direct stimulation is not possible when the heart nutrition has much suffered (and especially, therefore, its nerves), and that the dilating effects are like those produced by vagus stimulation, &c., favors the view that the results of faradisation are not due to direct stimulation of the heart muscle but accomplished mediately through its nerves. The light colored areas, which I have pointed out are seen in all cases at the exact points of contact of the electrodes, are due to direct effects on the muscle (contraction).

Spontaneous Rhythm. The order in which spontaneous R. most readily arises and is best maintained is: sinus, S. extension, auricle, ventricle. A large number of experiments on the ventricle expressly, of the sea turtle, show that in the latter, as in all other Chelonians thus far examined, the V. has in most cases a certain capacity for independent R., but that apart from all forms of stimulation this R. is never very marked, though it may last for hours. This seems to be greater in the land tortoises than in other Chelonians.

The ventricle of the sea turtle is characterized by great sensitiveness as compared with that of other Chelonians, hence its spontaneous R. may be greatly increased by slight stimulation. Results, like those of Gaskell, obtained by suspending the heart, attaching recording levers and feeding it through its own system of vessels, without regard to the normal blood pressure therein, must not be considered as those of pure spontaneous rhythm, for such methods furnish stimulation. In my experiments the heart was kept surrounded with nutriment and covered so as to provide a "moist chamber," but it remained in situ. The V. was separated by ligature in most cases, by section in a very few. Is there a Depressor Nerve in the Chelonians? Blood pressure experiments on the terrapin and the sea turtle have shown that no nerve with the characters of a physiological depressor exists in this family. Certain fine nerves in the neck of the sea turtle have on stimulation given results of a peculiar and puzzling kind. They have been inconstant in action, sometimes giving rise to acceleration, sometimes to retardation of the cardiac rhythm, or to both-now one, and now the other.

Stimulation of the Cerebral end of one Vagus, the medulla and other vagus being intact, in all the Chelonians I have examined, has usually produced cardiac arrest. In the sea turtle, in one case, this was followed by decided after-acceleration, but the different genera of Chelonians and even different species and individuals show variation in this respect, as also in the degree to which the heart can be reflexly inhibited; the terrapin and C. Midas resembling each other most and giving the best marked results. In fact this research has confirmed the truth of the law that with anatomical resemblances are usually associated physiological ones.

Evolution of Function. The experiments on the sea turtle have shown that where the cardiac nutrition suffers considerably, the L. auricle may even be quiescent when the right is still beating well; also that the ventricle dies in a certain segmental order, the last part to get rigid being on its right side; thus for some time before death an earlier condition (from a developmental point of view) is established, viz: Reduction to sinus, one auricle and a simplified ventricle. It is thus seen that the order of death for the different parts of the heart indicates its history, the oldest parts have greatest vitality. Further, the greater size and importance of the R. (part of the) sinus, of the R. auricle, &c., have also a relation to the order of acquisition. In the only fishes having an L. auricle, the Dipnoï, this part is very small and insignificant as compared with the right.

Anomalous Results. Stimulation of the liver in certain cases in the sea turtle, when reflex inhibition was being studied, has given results analagous to those obtained in the alligator, and especially in the fish, i. e., the effect has not been pure inhibition, but preliminary acceleration with or without after-retardation. The subject is of great interest though very puzzling in the present state of knowledge. It is further considered in the account of the investigation on the alligator,

On the Physiology of the Heart of the Alligator. By T. WESLEY MILLS.

The animals experimented upon belonged to the species Alligator Mississippiensis. The heart in the Crocodilia, with its two auricles and paired ventricles, though showing much resemblance to lower forms and retaining

the pulsatile sinus venosus, both in its general appearance and in its action, approximates sufficiently to that of the higher vertebrates to suggest on superficial examination the heart of a mammal or bird (with slower action). The blood, too, is more highly oxidized than in the Chelonians, so that altogether the circulatory system shows physiological as well as anatomical advance. With the exception of Gaskell's short paper on the crocodile, (Journal of Physiology, Vol. V., No. 1), nothing has been published on the heart physiology of this group of animals.

The work of the present writer, while it confirms Gaskell's conclusions as regards the cardiac accelerator, is wholly at variance with his views as to the functions of the vagus. The vagus in the Crocodilia, at least in the alligator, is not a pure cardial depressor, but is on the contrary a powerful cardiac augmentor.

The results of the stimulation of the vagus may be thus stated: (1.) Stimulation of the vagus with a weak, interrupted current may weaken the cardiac beat with or without arrest of the auricles; the latter may be arrested and give rise to a brief stop of the ventricles.

(2.) With a stronger current the sinus may be so weakened as to lead to arrest of the auricles and ventricles; or the sinus may be arrested wholly, in which case the auricles and ventricles invariably cease to beat.

(3.) When the cardiac beat recommences it may be in the order, sinus, sinus extension, ventricles; or, sinus, auricles, S. extension, ventricles, i. e., the auricles may remain quiescent as in the Chelonians and fishes when all the rest of the heart is beating.

(4.) The rhythm after vagus stand-still may be (a) without acceleration, or (b) accelerated.

The augmentation in the force of the beat is more marked than acceleration in the rate. Both rate and force follow, as in the Chelonians, the law of inverse proportion.

Comparison of the Vagi and Results of their Prolonged Alternate Stimulation. The vagi in the alligator, as in the Chelonians, have not as a rule equal power in causing and maintaining cardiac inhibition; the right as in the other cold blooded animals examined being more effective. Prolonged stimulation of the vagi alternately leads to corresponding lengthened cardiac arrest.

Accessory Vagi. Certain small nerves are in the alligator given off from the Glossopharyngeal shortly after its exit from the skull, proceed downwards, apart from the vagus, and pass beneath the trachea over the vessels to the heart.

Stimulation of these nerves has led to similar results to those furnished by stimulation of the vagus, i. e., retardation of the rate, weakening of the beat and after acceleration. Hence they have been called by me accessory vagi. There seem to be nerves of somewhat similar function in the sea turtle.

Peculiar Cardiac Inhibition followed by Acceleration. Special attention is called to the following experiment which is believed to be unique in physiology. In a small alligator with the whole brain destroyed for some time, both vagi divided and dead throughout the greater part of their course (stimulation not producing cardiac arrest), a sharp tap over the liver and stomach with a dissecting forceps caused cardiac arrest of brief duration, then slowed irregular rhythm followed by acceleration of a very pronounced kind (from 40 to 50 beats). Here then were the usual phenomena of reflex vagus inhibition, as when the vagi and medulla are intact. This experiment was repeated three times. It does not seem possible to explain this unparalleled result by present theories. I conclude that the impulses passed through the sympathetic system of nerves and that probably other inhibitory fibres than those of the vagus were concerned, and that accelerating fibres were also involved. It is also possible to conceive that terminations of the vagi were in some way reached by these impulses, but in any case the results are new to physiology, the only published case at all resembling it being Marshall Hall's experiment on the eel's stomach (Todd's Cyclopedia of Anat. and Phys., article "Heart.")

Cardiac Augmentors. As described by Gaskell there is in the Crocodilia, from the Ganglion of the eleventh mefamere of the sympathetic chain, a strong well-defined branch passing to the heart.

Stimulation of this nerve has given rise to (1) acceleration following the law of inverse proportion, which seems applicable to all kinds of acceleration. (2.) Decided augmentation of the force of the beat. This is more marked than the acceleration in rate, and in fact may disguise the effects of the nerve, for no actual acceleration of beat may follow,