Section 2. Comparing the general build, of a dog-fish with that of a rabbit, we notice the absence of a distinct neck, and the general conical form; the presence of a large tail, as considerable, at first, in diameter as the hind portion of the body, and of the first importance in progression, in which function the four paddle-shaped limbs, the lateral fins, simply co-operate with the median fin along the back for the purpose of steering; and, as a consequence of the size of the tail, we note also the ventral position of the apertures of the body. The anus, and urinary and genital ducts unite in one common chamber, the cloaca. Behind the head, and in front of the fore fin (pectoral fin), are five gill slits (g.s.) leading from the pharynx to the exterior. Just behind the eye is a smaller and more dorsal opening of the same kind, the spiracle (sp.). On the under side of the head, in front of the mouth, is the nasal aperture (olf.), the opening of the nasal sac, which, unlike the corresponding organ of the air-frequenting vertebrata, has no internal narial opening. There is, however, a groove running from olf. to the corner of the mouth, and this, closing, in the vertebrate types that live in air and are exposed to incessant evaporation of their lubricating secretions, constitutes the primitive nasal passage. The limbs are undifferentiated into upper, lower, and digital portions, and are simply jointed, flattened expansions.

Section 3. The skin of the dog-fish is closely set with pointed tooth-like scales, the placoid scales, and these are continued over the lips into the mouth as teeth. Each scale consists of a base of true bone, with a little tubercle of a harder substance, dentine, capped by a still denser covering, the enamel. The enamel is derived from the outer layer of the embryonic dog-fish, the epiblast, which also gives rise to the epidermis; while the dentine and bony base arise in the underlying mesoblast, the dermis. A mammalian tooth has essentially the same structure: an outer coat of enamel, derived from epiblast, overlies a mass of dentine, resting on bone, but the dentine is excavated internally, to form a pulpcavity containing blood-vessels and nerves. Most land animals, however, have teeth only in their mouths, and have lost altogether the external teeth which constitute the armour of the dog-fish. Besides the teeth there perhaps remain relics of the placoid scales in the anatomy of the higher vertebrata, in the membrane bones. How placoid scales may have given rise to these structures will be understood by considering such a bone as the vomer of the frog. This bone lies on the roof of the frog's mouth, and bears a number of denticles, and altogether there is a very strong resemblance in it to a number of placoid scales the bony bases of which have become confluent. In the salamander, behind the teeth-bearing vomers comes a similar toothed parasphenoid bone. The same bone occurs in a corresponding position in the frog, but without teeth. In some tailed amphibians the vomers and splenials are known to arise by the fusion of small denticles. These facts seem to point to stages in the fusion of placoid bases, and their withdrawal from the surface to become incorporated with the cranial apparatus as membrane bones, a process entirely completed in the mammalian type.

Section 4. The alimentary canal of the dog-fish, is a simple tube thrown into a Z shape. The mouth is rough with denticles, and has a fleshy immovable tongue on its floor. In the position of the Eustachian tube there is a passage, the spiracle (sp.), running out to the exterior just external to the cartilage containing the ear. The pharynx communicates with the exterior through five gill slits (g.s.), and has, of course, no glottis or other lung opening. There is a wide oesophagus passing into a U-shaped stomach (st.), having, like the rabbit's, the spleen (sp.) on its outer curvature. There is no coiling small intestine, but the short portion, receiving the bile duct (b.d.) and duct of the pancreas (pan.), is called the duodenum (d'dum.). The liver has large left (L.lv.) and right lobes, and a median lobe (M.lv.), in which the gall bladder (g.bl.) is embedded. The next segment of the intestine is fusiform, containing a spiral valve (Figure 4), the shelf of which points steeply forward; it is sometimes called the colon (co.). It is absorptive in function and probably represents morphologically, as it does physiologically, the greater portion of the small intestine. A rectal gland (r.g.) opens from the dorsal side into the final portion of the canal (rectum).

Section 5. The circulation presents, in many respects, an approximation to the state of affairs in the developing embryos of the higher types. The heart (Figure 3, Sheet -14- {Error in First Edition} [16]) is roughly, Z shape, and transmits only venous blood. It lies in a cavity, the pericardial cavity (P.c.c.), cut off by a partition from the general coelome. At one point this partition is imperfect, and the two spaces communicate through a pericardio-peritoneal canal (p.p.c.), which is also indicated by an arrow (p.p.) in the position and direction in which the student, when dissecting, should thrust his "seeker," in Figure 1 Sheet 15. A sinus venosus (s.v. in Figure 3, Sheet 16) receives the venous trunks, and carries the blood through a valve into the baggy and transversely extended -auricle- [atrium] (au.), whence it passes into the muscular ventricle (Vn.), and thence into the truncus arteriosus. This truncus consists of two parts: the first, the conus or pylangium (c.a.), muscular, contractile, and containing a series of valves; the second, the bulbus or synangium (b.a.), without valves and pulsatile. In the rabbit both sinus and truncus are absent, or merged in the adjacent parts of the heart.

Section 6. From the bulbus there branch, on either side, four arterial trunks, the first of which forks, so that altogether there are five afferent branchials (a.br.) taking blood to be aerated in the gills, here highly vascular filamentary outgrowths of the internal walls of the gill slits.

{Lines from Second Edition only.} [There are altogether nine vascular outgrowths (demi-branchs), one on each wall of each gill slit except the last, on the hind wall of which there is none. (In the spiracle is a miniature demibranch, the pseudo-branch. This suggests that the spiracle is really a somewhat modified gill slit.)]

Four efferent branchials (e.br.) carry the aerated blood on to the dorsal aorta (d.ao.). A carotid artery runs forward to the head, and a hypo-branchial artery supplies the ventral side of the pharyngeal region. There are sub-clavian, coeliac, mesenteric, and pelvic arteries, and the dorsal aorta is continued through the length of the tail as the caudal artery (Cd.A.).

Section 7. A caudal vein (Cd.V.), bringing blood back from the tail, splits behind the kidneys (K.), and forms the paired renal portal veins (r.p.v.), breaking up into a capillary system in the renal organ. A portal vein brings blood from the intestines to the liver.

Section 8. Instead of being tubular vessels, the chief veins of the dog-fish are, in many cases, irregular baggy sinuses. Three main venous trunks flow into the sinus venosus. In the median line from behind comes the hepatic sinus (H.S.); and laterally, from a dorsal direction, the Cuvierian sinuses (C.S.) enter it. These, as the student will presently perceive, are the equivalents of the rabbit's superior cavae. They receive, near their confluence with the sinus venosus, the inferior jugular vein (I.J.V.). At their dorsal origin, they are formed by the meeting of the anterior (A.C.S.) and posterior (P.C.S.) cardinal sinuses. The anterior cardinal sinus -is, roughly, the equivalent of the internal jugular vein-, lies along dorsal to the gill slits (g.s.), and receives an orbital sinus from the eye. The posterior cardinal sinus receives a sub-clavian vein (s.c.v.) and a lateral vein (L.V.), and fuses posteriorly with its fellow in the middle line. This median fusion is a departure from the normal fish type. It must not be confused with the inferior cava, which is not found in the dog-fish, the [right] posterior cardinals representing the rabbit's azygos vein. A simplified diagram of the circulation of a fish is given in Figure 2, Sheet 16, and this should be carefully compared with the corresponding small figure given of the vascular system of our other types.

{Lines from Second Edition only.} [The blood of the dog-fish resembles that of the frog.]

Section 9. The internal skeleton, as we have said, is entirely cartilaginous, and only those parts which are pre-formed in cartilage in the skeletons of the higher types are represented here. The spinal column consists of two types of vertebrae, the trunk, bearing short, distinct, horizontally-projecting ribs (r.), and the caudal. The diagrams of Figure 5 [(Sheet 18)] are to illustrate the structure of the centrum of a dog-fish vertebra; C is a side view, D a horizontal median section, A and B are transverse sections at the points indicated by -B and A- [A and B] respectively in Figure C. -(By an unfortunate slip of the pen in the figure, A was substituted for B; section A corresponds to line B, and vice versa.)- The vertebrae are hollowed out both anteriorly and posteriorly (amphi-coelous), and a jelly-like notochord runs through the entire length of the vertebral column, being constricted at the centres of the centra, and dilated between them. The neural arch above the centrum, and containing the spinal cord, is made up of neural plates (n.p.), and interneural plates (i.n.p.), completed above by a median neural spine (n.s.). In the caudal region, instead of ribs projecting outwardly, there are haemal processes, inclined downwards and meeting below, forming an arch, the haemal arch, containing the caudal artery and vein— the vein ventral to the artery— and resembling the neural arch, which contains the spinal cord above, in shape and size.

Section 10. The pectoral limb and girdle (Figure 4, Sheet 16) have only a very vague resemblance to the corresponding structures in the rabbit. The girdle (g.) is a transverse bar lying ventral to the pericardial wall, and sending up a portion (sc.), dorsal to the attachment of the limb, which answers to the scapula and supra-scapula of the forms above the fish. Three main cartilages, named respectively the pro- (p.p.), meso- (m.p.), and meta-pterygium, form the base of the limb. With these, smaller cartilaginous plates, rods, and nodules articulate, and form a flattened skeletal support for the fin.

Section 11. The pelvic girdle and limb (Figure 2, Sheet 15) are similar in structure, but the pro-pterygium and meso-pterygium are absent, and the cartilage answering to the meta-pterygium goes by the name of the basi-pterygium. In the male, but not in the female, the pelvic fins are united behind the cloaca, and there are two stiff grooved copulatory organs, the claspers (cl. in Figure 1), which have a cartilaginous support (cl.c.). These claspers form the readiest means of determining the sex of a specimen before dissection.

Section 12. The skull consists of a cartilaginous cranium, and of jaw and visceral arches. The cranium persists throughout life, in what closely resembles a transitory embryonic condition of the higher types. There is a nasal capsule (na.c.), a brain case proper, and lateral otic (auditory) capsules (ot.c.) containing the internal ear. (This should be compared with the frog's embryonic skull.) The upper jaw has a great bar of cartilage, the palato-pterygoid, as its sole support; the arch of premaxilla, maxilla, jugal, and squamosal— all membrane bones— is, of course, not represented. In the frog this bar of cartilage is joined directly to the otic capsule by a quadrate portion, but this is only doubtfully represented in the dog-fish by a nodule of cartilage in the pre-spiracular ligament (p.s.). The lower jaw is supported, by Meckel's cartilage (M.C.). The hyoid arch consists of two main masses of cartilage, the hyomandibular (h.m.), and the ceratohyal (c.h.); the former of these is tilted slightly forward, so that the gill slit between it and the jaw arch is obliterated below, and the cartilage comes to serve as the intermediary in the suspension of the jaw from the otic mass. There are five branchia[l] arches, made up pharyngo-, epi- and cerato-branchials, and the ventral elements fuse in the middle line to form a common plate of cartilage. Outside these arches are certain small cartilages, the extra branchials (ex.b.) which, together with certain small labials by the nostrils and at the sides of the gape, probably represent structures of considerably greater importance in that still more primitive fish, the lamprey. The deep groove figured lateral to the otic capsule is the connecting line of the orbital and anterior cardinal sinuses; the outline of the anterior cardinal sinus in this figure and in Figure 1 is roughly indicated by a dotted line.

Section 13. Figure 3a is a rough diagram of the internal ear— the only auditory structure of our type (compare Rabbit, Sheet 7). To dissect out the auditory labyrinth without injury is a difficult performance, but its structure may be made out very satisfactorily by paring away successive slices of the otic mass. Such a section is shown by Figure 3b; through the translucent hyaline cartilage the utriculus and horizontal canal can be darkly seen. The ductus endolymphaticus (vide Rabbit) is indicated by a dotted line in our figure. It is situated internal to the right-angle between the two vertical canals, and reaches to the surface of the otic capsule.

Section 14. The brain shows the three primary vesicles much more distinctly than do our higher types. The fore-brain has large laterally separated olfactory lobes (rh.), there are relatively small "hemispheres" (pr.c.), the stalk of the pineal gland tilts forward, and the gland itself is much nearer the surface, being embedded in the cartilage of the brain case, and the pituitary body is relatively very large, and has lateral vascular lobes on either side. Following the usual interpretation of the parts, we find optic lobes (op.l.) as the roof of the mid-brain, and behind a very large, median, hollow, tongue-shaped cerebellum (c.b.). The medulla is large, and certain lateral restiform tracts (r.t.) therein, which also occur in the higher types, are here exceptionally conspicuous.

Section 15. The dog-fish has ten pairs of cranial nerves, corresponding to the anterior ten of the rabbit very closely, when we allow for the modification the latter has suffered through the conversion of some part of the spiracular cleft to an eardrum, and the obliteration of the post-hyoid branchial slits.

The first and second nerves are really brain lobes, and nerves of the special senses of smell and sight respectively.

The third (oculomotor), the fourth (patheticus), and the sixth (abducens) are distributed to exactly the same muscles of the eyeball as they are in the rabbit.

The fifth nerve, has, in the dog-fish, as in the rabbit, three chief branches. V.2 and V.3 fork over the mouth just as they do in the mammal; V.1 passes out of the cranium by a separate and more dorsal opening, and runs along a groove along the dorsal internal wall of the orbit, immediately beneath a similar branch of VII., which is not distinct in the rabbit. The grooves are shown in the figure of the cranium, Sheet 18; the joint nerve thus compounded of V. and VII. is called the ophthalmic (oph.). It is distributed to the skin above the nose and orbit. When the student commences to dissect the head of a dog-fish he notices over the dorsal surface of the snout an exudation of a yellowish jelly-like substance, and on removing the tough skin over this region and over the centre of the skull he finds, lying beneath it, a quantity of coiling simple tubuli full of such yellowish matter. These tubuli open on the surface by small pores, and the nerves terminate in hair-like extremities in their lining. These sense tubes are peculiar to aquatic forms; allied structures are found over the head and along a lateral line (see below) in the tadpole, but when the frog emerges from the water they are lost. They, doubtless, indicate some unknown sense entirely beyond our experience, and either only possible or only necessary when the animal is submerged.

In addition to the ophthalmic moiety mentioned above, the seventh nerve has a vidian branch (vid.) running over the roof of the mouth, and besides this its main branches fork over the spiracle, just as V. forks over the mouth, and as IX. and X. fork over gill clefts. This nerve in the rabbit is evidently considerably modified from this more primitive condition.

The eighth is the auditory nerve, as in the rabbit.

The ninth nerve forks over the first branchial cleft.

The tenth nerve is easily exposed by cutting down through the body wall muscles over the gill clefts, into the anterior cardinal sinus (A.C.S.). It gives off (a) branches forking over the posterior four gill slits, (b) a great lateral nerve running inward, and back through the body-wall muscle, and connected with a line of sense organs similar to those in the head, the lateral line, and (c) a visceral nerve curving round to the oesophagus and stomach. In dissection it becomes very evident that the tenth nerve is really a leash of nerves, each one equivalent to the ninth.

We may here call the attention of the reader to the fact of the singular resemblance of V., VII., IX., and the factors of X. That each has a ventral fork, we have already noticed. Each also (?IX.) has a dorsal constituent connected with the sense organs of the skin. The vidian branch of VII., however, is not evidently represented in the others.

Section 16. The coelom of the dog-fish is peculiar— among the types we treat of— in the possession of two direct communications with the exterior, in addition to the customary indirect way through the oviduct. These are the abdominal pores (a.p.) on either side of the cloaca in either sex. They can always be readily demonstrated by probing out from the body cavity, in the direction indicated by the arrow (a.p.) in Figure 1, Sheet 15. They probably serve to equalize the internal and external pressure of the fish as it changes its depth in the water, just as the Eustachian tubes equalize the pressure on either side of the mammal's tympanic membrane.

Section 17. The musculature of the dog-fish body is cut into V-shaped segments, the point of the V being directed forward. The segments alternate with the vertebrae, and are called myomeres. Such a segmentation is evident, though less marked, in the body wall muscles of the frog, and in the abdominal musculature of the rabbit and other mammals it is still to be traced.

Section 18. The uro-genital organs of the female dog-fish (Figure 1, Sheet 17) consist of an unpaired ovary (ov.), paired oviducts (o.d.), enlarged at one point to form an oviducal gland (o.d.g.), kidneys (k.), with ureters (ur.) uniting to form a urinary sinus (u.s.) opening into the cloaca by a median urinary papilla separate from the oviducal openings. The eggs contain much yolk, and, like those of the fowl, are very large; like the fowl, too, one of the ovaries is suppressed, and it is the right ovary that alone remains. The two oviducts meet in front of the liver ventral to the oesophagus, and have there a common opening by which the ova are received after being shed into the body cavity. The eggs receive an oblong horny case in the oviduct; in the figure such a case is figured as distending the duct at e. The testes of the male (T. in Figure 2) are partially confluent in the middle line. They communicate through vasa efferentia (v.e.) with the modified anterior part of the kidney, the epididymis (ep.), from which the vas deferens (v.) runs to the median uro-genital sinus (u.g.s.), into which the ureters (ur.) also open. The silvery peritoneum (lining of the body cavity) covers over the reddish kidneys, and hides them in dissection.

Section 19. Figure 3, Sheet 17, is a generalized diagram of the uro-genital organs in the vertebrata; M.L. is the middle line of the body, G. is the genital organ, Pr. is the pronephros, or fore kidney, a structure which is never developed in the dog-fish, but which has functional importance in the tadpole and cod, and appears as a transitory rudiment in the chick. A duct, which is often spoken of as the pronephric duct (p.d.), and which we have figured under that name, is always developed. Anteriorly it opens into the body cavity. It is also called the Mullerian duct, and in the great majority of vertebrata it becomes the oviduct, uniting with its fellow, in the case of the dog-fish, ventral to the oesophagus. In the male it usually disappears; the uterus masculinus of the rabbit is still very generally regarded as a vestige of it. Kolliker has shown, however, that this interpretation is improbable. Ms. is the mesonephros, some or all of which becomes the epididymis in the male of types possessing that organ, and is connected with G. by the vasa efferentia. Mt., the metanephros, is, in -actual fact- [the frog], indistinguishably continuous with Ms., and is the functional kidney, its duct (metanephric duct) being either undifferentiated from the mesonephric (as is the case with the frog) or largely split off from it, as in the dog-fish, to form the ureter.

Section 20. The correspondence of the male organs of the dog-fish with those of the rabbit, will be more evident if the student imagine—

(a) the testes, vasa efferentia, and epididymis of each side to shift posteriorly until they reach a position on either side of the cloaca; and

(b) The uro-genital apertures, instead of meeting dorsally and posteriorly to the anus, to shift round that opening and meet anteriorly and ventrically to it.

Section 21. This completes our survey of this type. Except where we have specified differences, the general plan of its anatomy follows the lines of the other vertebrate types described.



Questions on the Dog-Fish

1. Describe the alimentary canal of the dog-fish, and compare it with that of the rabbit in detail.

2. Compare the coelom of the dog-fish and rabbit.

3. Draw diagrams to illustrate the course of the circulation in the dog-fish.

4. (a) Describe fully the heart of a dog-fish. (b) Compare it with that of a rabbit.

5. Give an account of the respiratory apparatus of the dog-fish.

6. Draw diagrams of a dog-fish vertebra, and compare the centrum with that of a rabbit.

7. Compare the vertebral column of the dog-fish and rabbit.

8. Draw diagrams of the limbs and limb-girdles of the dog-fish. Compare the pectoral with the pelvic fin.

9. Draw diagrams of (a) the male and (b) the female urogenital organs of the dog-fish. (c) Compare them carefully with those of the rabbit.

10. Compare the circulation in the kidney of dog-fish and rabbit.

11. Give an account of the cranio-facial apparatus of the dog-fish. State clearly what representation of this occurs in the frog and in the rabbit.

12. Give drawing (a) from above, (b) from the side, of the dog-fish brain.

13. State the origin and the distribution of the fifth, seventh, ninth, and tenth cranial nerves in the dog-fish.

14. Compare, one by one, the cranial nerves of the dog-fish with those of any higher vertebrate, as regards their origin and their distribution.

15. Describe the auditory organ of the dog-fish. What parts are added to this in the higher type?

16. Draw the cloaca (a) of a male, (b) a female dog-fish.

17. (Practical.) Demonstrate in a dog-fish the pathetic nerve, the opening between pericardium and coelom. the abdominal pores, and the ureter.



-Amphioxus_

1. Anatomy

Section 1. We find in Amphioxus the essential vertebrate features reduced to their simplest expression and, in addition, somewhat distorted. There are wide differences from that vertebrate plan with which the reader may now be considered familiar. There are no limbs. There is an unbroken fin along the median dorsal line and coming round along the ventral middle line for about half the animal's length. But two lowly vertebrates, the hag-fish and lamprey, have no limbs and a continuous fin. There is, as we shall see more clearly, a structure, the respiratory atrium, not apparently represented in the true vertebrate types, at least in their adult stages. There is no distinct heart, only a debateable brain, quite without the typical division into three primary vesicles, no skull, no structures whatever of cartilage or bone, no genital ducts, no kidneys at all resembling those of the vertebrata, no pancreas, no spleen; apparently no sympathetic chain, no paired sense organs, eyes, ears, or nasal sacs, in all of which points we have striking differences from all true vertebrata; and such a characteristic vertebrate peculiarity as the pineal gland we can only say is represented very doubtfully by the eye spot.

Section 2. The vertebral column is devoid of vertebrae; it is throughout life a rod of gelatinous tissue, the notochord (Figure 1, n.c.), surrounded by a cellular sheath. Such a rod is precursor to the vertebral column in the true vertebrates, but, except in such lowly forms as the lamprey, is usually replaced, partially (e.g., dog-fish) or wholly (as in the rabbit) by at first cartilaginous vertebrae whose bodies are derived from its sheath. Further, while in all true vertebrata the notochord of the developing young reaches anteriorly at most to the mid-brain, and is there at its termination enclosed by the middle portion of the skull, in Amphioxus it reaches far in front of the anterior extremity of the nervous system, to the end of the animal's body.* On this account the following classification is sometimes made of those animals which have a notochord:—

-Chordata_ (= Vertebrata, as used by Lankester).

1. Having the notochord reaching in front of the brain. Cephalochorda = Amphioxus.

2. Having the notochord reaching anteriorly to the mid-brain, a brain of three primary vesicles and a skull. Craniata = all "true vertebrata": fishes, amphibia, reptiles, birds, and mammals (Vertebrata of Balfour).

3. Having the notochord confined to the tail. Urochorda = the ascidians, or sea-squirts, certain forms of life only recently recognised as relatives of the vertebrata.

* The anterior end of the notochord in the developing rabbit or dog lies where the middle of the basisphenoid bone is destined to be.

Section 3. Figure 1, Sheet 19, shows the general anatomy of Amphioxus. We recognise four important points of resemblance to the earlier phases of the higher and the permanent structure of the lower members of the vertebrata, and it is these that justify the inclusion of amphioxus in this volume. In the first place there is the—

-Notochord_. In the next, just above it (at s.c.) we find—

-A Dorsal Tubular Nervous Axis_, the spinal cord. Thirdly, the pharynx (ph.) is perforated by—

-Respiratory Slits_, though these, instead of being straight slashes, are modified from a U-shape [slant very much forward and are much more numerous than in any true vertebrate.]. -And-, Fourthly, there is, as we shall see, a—

-Vertebrate Type of Circulation_. [And finally the body-wall muscles are divided into—]

[-Myomers_.]

Section 4. The alimentary canal of Amphioxus commences with an "oral cavity," not represented in our vertebrata, surrounded by a number of cirri, or tentacles, supported by a horny substance which seems to be chitin, a common skeletal material among invertebrates. A velum (v.) forms a curtain, perforated by the mouth and by two smaller hyoidean apertures, between the oral cavity and the pharynx (ph.). "Pharyux" is here used in a wider sense than in the true vertebrata; it reaches back close to the liver, and is therefore equivalent to pharynx + oesophagus + a portion or all of the stomach. The [so-called] hyoidean apertures are not equivalent to the similarly-named parts of the vertebrata. Behind the pharynx the intestine (int.) runs straight out to the anus (an.), which opens not in the middle line, as one might expect, but in the left side! The liver lies usually on the creature's right, and instead of being a compact gland, is simply bag-like.

Section 5. The circulation is peculiarly reduced (Figure 2). The cardiac aorta (c.ao.) lies along the ventral side of the pharynx, and sends branches up along the complete bars between the gill slits. There is no -distinct- heart, but the whole of the cardiac aorta is contractile, and at the bases of the aortic arches that run up the bars there are contractile dilatations that assist in the propulsion of the blood. Dorsal to the pharynx, as in fishes, there is a pair of dorsal aorta (d.ao.) that unite above the liver (compare the frog, for instance), and thence run backward as a median dorsal aorta (d.ao.'). A portal vein (p.v.) bring blood back from the intestine (and apparently from the whole posterior portion of the animal) to the liver. Thence hepatic veins (hep.) take it to the cardiac aorta.

{Lines from First Edition only.} -When we remember that in the embryonic vertebrate the heart is at first a straight tube, this circulation appears even more strikingly vertebrate in its character than before.-

Section 6. The coelom, or body cavity, of Amphioxus lies, of course, as in the vertebrata, between the intestinal wall and the body walls, and, just as in the vertebrata, it is largely reduced where gill slits occur. But matters are rather complicated by the presence of an atrial cavity round the pharynx, which is not certainly represented in the vertebrata, and which the student is at first apt to call the body cavity, although it is entirely distinct and different from that space. The mutual relation of the two will become apparent after a study of Figures 10, 11, 12 (Sheet 21). Figure 10 gives diagrammatically a section of a very young stage of Amphioxus; P is the pharynx portion of the alimentary canal, coe. is the coelom surrounding it at this stage here as elsewhere; mt.c. are certain lymph spaces, the metapleural canals, between which a small invagination (i.e., a pushing-in), at., of the outer epidermis occurs; n.c. is the notochord, and s.c. the spinal cord. The gill slits, by which P. communicates with the exterior, are not shown. Next Figure 11 shows the invagination (at.) pushing its way in, and cut off from the exterior by a meeting of the body wall below. Note that at. is a portion of the animal's exterior thus embraced by its body, and that its lining is therefore of the same material as the external integument. In Figure 12, at. is developing upward, so that the true body hangs into it. Now imagine the gill slits perforated, as shown by the double-headed arrow in Figure 12. Figure 3, on Sheet 20, is a less diagrammatic representation of a cross-section of the pharyngeal region (vide Figure 1, Sheet 19). The student should compare Figure 3, Sheet 20, and Figure 12, Sheet 21. The atrium and metapleural canals are easily recognised in both. In Figure 3 the coelom is much cut up by the gill slits, and we have remaining of it (a) the dorsal coelomic canals (d.c.c.) and (b) the branchial canals (br.c.) in the bars between the slits. The atrial cavity remains open to the exterior at one point, the atrial pore (at.p.).

Section 7. The method of examining cross-sections is an extremely convenient one in the study of such a type as Amphioxus. The student should very carefully go over and copy the six sections on Sheet 20, comparing Figure 1 as he goes. He should do this before reading what follows. One little matter must be borne in mind. These figures are merely intended to convey the great structural ideas, and they are considerably simplified; they must not be regarded as a substitute for the examination of microscopic sections. [He will notice a number of rounded masses from the body wall. The] -For instance, the body-wall- muscles of Amphioxus are arranged in bundles bent sharply in an arrow shape, the point forward. -A number of these bundles are cut in any one section, and so the even shading of our diagrams, if they professed to be anything more than diagrams, should be broken up into masses.- These -bundles, we may mention-, are called myomeres, and they are indicated in Figure 1 by lines pointing acutely forward. [Several are consequently cut in any transverse section (Sheet 20), and these are the rounded masses he sees.] Similar myomeres, similarly situated, are found in fish, behind the head, and, less obviously, they occur with diminishing importance as the scale of the vertebrata is ascended.

Section 8. If we compare the nervous system of amphioxus with that of any vertebrate, we find at once a number of striking differences. In the first place, the skeletal covering of it, the cranium and the neural arches of vertebrae, are represented only by a greatly simplified connective tissue. In the next, a simple and slight anterior dilatation alone represents the brain. A patch of black pigment anterior to this (e.s.) may or may not be what its name implies an eye-spot. There is a ciliated funnel, c.f. (Figure 1, Sheet 19), opening on the left side, which has been assumed to be olfactory in its functions, and in the mouth chamber a ciliated pit (c.p.), which may, or may not, be an organ of taste. The ventral fissure of the spinal cord is absent. The dorsal nerves are without ganglia, and do not come off in pairs, but alternately, one to the left, then one to the right, one to the left, one to the right, and so on. The ventral nerves are very short, more numerous than the dorsal, and never unite with these latter to form mixed nerves.

The student will observe that here, just as in the case of the ciliated funnel and anus, the Amphioxus is not strictly symmetrical, but twisted, as it were, and so departs from the general rule of at least external bilateral symmetry obtaining among the vertebrates. It habitually lies on one side in the mud of the sea bottom, and it is probable that this external asymmetry is due to this habit, so that too much classificatory importance must not be attached to it. The soles and other related fish, for instance, are twisted and asymmetrical, through a similar specific habit, to such an extent that both eyes lie on one side of the animal.

Section 9. No kidney on the vertebrate pattern is found, but the following structures have, among others, been suggested as renal organs:—

(a) Certain canals, the brown tubes of Lankester (b.t.L., Figure 2, Sheet 19), a pair of pigmented tubes opening into the atrium at the hind end of the pharynx, lying forward along by the dorsal coelomic canals, and having an internal opening also.

(b) Certain tubuli described by Weiss as situated in a series along the upper corners of the atrial cavity, and communicating, after the fashion, of the "nephridia" of the earthworm, with the coelom and with the exterior (or, rather, with that portion of the animal's exterior enclosed in by the atrial wall; compare Section 6).

(c) The general epithelial lining of the atrium.

The reproductive organs (Figure 4, Sheet 20, g.) are masses of cells situated in an isolated part of the coelom in the atrial folds, and, having no ducts, their contents must escape into the atrium by rupture of the body-wall. Thence they escape either by gill-slits, pharynx and mouth, or, more generally, through the atrial pore. The animals, like all the vertebrata, are dioecious, i.e., male or female.

Section 10. The endostyle (end.), in Figures 3 and 4, is a ciliated path or groove on the under side of the pharynx, which is generally supposed to represent the thyroid gland of vertebrates. The vertebrate thyroid, early in development, is certainly an open and long narrow groove in the ventral side of the pharynx. The hyper-pharyngeal groove (h.p.) has been in the past compared to the pituitary body, but there is little doubt now that this structure is represented by the ciliated pit.

Section 11. The student is advised to revise this chapter before proceeding, and to schedule carefully the anatomical features under the headings of (1.) distinctly vertebrate characters, (2.) characters contrasting with the normal vertebrate structure, (3.) facts of doubtful import, with the suggestions given in the text written against them.



2. The Development of Amphioxus

Section 12. The development of amphioxus, studied completely, is at once one of the most alluring and difficult tasks in the way of the zoologist; but certain of its earlier and most obvious fasts may very conveniently be taken into consideration now.

Section 13. The phenomena of the extrusion of polar bodies and fertilization are treated of later, and will, therefore, not be considered now. We will start our description with an egg-cell, which has escaped, of course, since there are no genital ducts, by rupture of the parent, has been fertilized by the male element, and is about to develop into a young amphioxus. It is simply a single cell, with some power of amoeboid motion, a single nucleus and nucleolus; and in amphioxus its protoplasm is clear and transparent. Frequently ova are loaded with granules of food store (yolk), which enable the young animal to go far with its development before it is hatched and has to begin fending for itself. Such an ovum as that of our present type, however being devoid of such yolk (alecithal = without yolk), necessitates a very early start in life, and, for reasons too complicated to state fully here, the development in such a case is considered particularly instructive and primitive by zoologists.

Section 14. The first thing to be seen in the developing cell is a deepening circular groove (Figure 1, Sheet 21), which divides the ovum into two parts. Another groove then cuts at right angles to this subdividing the two into four (Figure 2). Another groove, at right angles to both the former, follows, making the four eight (Figure 3). And so subdivision goes on. The whole process is called segmentation or cleavage.

Section 15. At the end of segmentation we get a hollow sphere of small cells, the cells separating from one another centrally and enclosing a cavity as the process proceeds. This is the blastosphere, shown diagrammatically in Figure 4, and of which an internal view, rather truer to the facts of the case as regards shape, is given as Figure 5. The central cavity is the segmentation cavity (s.c.).

Section 16. Invagination follows (Figure 6). In this process a portion of the blastosphere wall is the tucked into the rest, as indicated by the arrow, so that a two-layered sack is formed. The space ar. is the archenteron, the primordial intestine, and its mouth is called, the blastopore (bp.). The outer layer of this double-walled sac is called the epiblast. For the present we will give the inner lining no special term. The young amphioxus has, at this stage, which is called the gastrula stage, a curious parallelism with such a lowly form as the Hydra of our ditches. This latter creature, like the gastrula, consists essentially of two layers of cells, an outer protective and sensory layer, and an inner digestive one; it has a primordial intestine, or archenteron, and its mouth is sometimes regarded as being a blastopore. All animals that have little yolk, and start early in life for themselves, pass through a gastrula stage, substantially the same as this of amphioxus.

Section 17. The anus is perforated later near the region occupied at this stage by the blastopore. Hence the anterior end of the future amphioxus, the head end, is pointing towards the Figure 6, and the letters ep. are marked on the side which will be dorsal.

Section 18. Figure 7 i. is a dorsal view of the gastrula at a somewhat later stage, and here indications of distinctly vertebrate relationships already appear. Figure 7 ii. is a cross-section, its position, being shown by cross-lines in 7 i. and 6. Note first that the epiblast along the mid-dorsal line is sinking in to form what is called the neural plate (n.p.), and simultaneously on either side of it rise the neural folds (n.f.). Now, at Figure 8, a slightly later stage is represented, and at 9 i. the inturned part is separated from the general external epiblast as the spinal cord. The remainder of the epiblast constitutes the epidermis.

Section 19. Reverting to Figure 7 ii., along the dorsal side of the archenteron a thickening of its wall appears, and is gradually pinched off from it to form a cellular rod, lying along under the nervous axis and above the intestine. This is the notochord (compare Figures 8 and 9).

Section 20. Finally, we note two series of buds of cells, one on either side of the archenteron in Figure 7 ii. In 8 these buds have become hollow vesicles, growing out from it, the coelomic pouches. They are further developed in 9; and in 9 ii., which is a diagrammatic figure, they are indicated by dotted lines. They finally appear to (? entirely) obliterate the segmentation cavity— they certainly do so throughout the body; and their cavities are in time cut off from the mesenteron, by the gradual constriction of their openings. In this way the coelom (body cavity) arises as a series of hollow "archenteric" outgrowths, and ms. becomes the alimentary canal. mt.c., the metapleural canals, probably arise subsequently to, and independently of, the general coelomic space, by a splitting in the body-wall substance.

Section 21. Hence, in considering the structure of amphioxus, we have three series of cells from which its tissues are developed:—

1. The epiblast.

2. Walls of the coelomic pouches, which form (a) an inner lining to the epiblast, (b) an outer coating to the hypoblast, and (c) the mesentery (m.), by which the intestine is supported. This is the mesoblast.

3. The lining of the mesenteron, or hypoblast.

From the epiblast the epidermis (not the dermis), the nervous system (including the nerves), and the sensory part of all sense organs are derived. From the mesoblast the muscles, the dermis genital and excretory organs, circulatory fluid and apparatus, any skeletal structures; and all connective tissue are derived. The mass of the body is thus evidently made of mesoblast. The hypoblast is the lining of the intestine and of the glands which open into it; and the material of the notochord is also regarded, as hypoblast.

Section 22. Figure 9 ii. shows all the essential points of the structure of amphioxus. Epiblast is indicated by a line of dashes, mesoblast by dots, and hypoblast, dark or black. The true mouth is formed late by a tucking-in of epiblast, the stomodaeum (s.d.), which meets and fuses with the hypoblast, and is then perforated. The position of this mouth is at the velum. The formation of the atrium has been described. The metapleural folds run forward in front of the velum, as the epipleurs (ep. in Sections 1 and 2), and form an oral hood (b.c.), around which the tentacles appear, and which is evidently not equivalent to the vertebrate mouth cavity, but in front of and outside it. The anus is formed by a tucking in, the proctodaeum, similar to the stomodaeum.

Section 23. The formation of the respiratory slits is complicated, and difficult to describe, but, since investigators have still to render its meaning apparent, it need not detain the elementary student.*

* See Balfour's Embryology, Volume 2, and Quarterly Journal of Microscopical Science March, 1891.



Questions on Amphioxus

1. Draw diagrams, with the parts named, of the alimentary canal of (a) amphioxus, (b) any craniate; (c) indicate very shortly the principal structural differences between the two.

2. Describe, with a diagram, the circulation of amphioxus. Compare it with that of the craniata.

3. Draw from memory transverse sections, of amphioxus (a) in the oral region, (b) through the pharynx, (c) just anterior, and (d) just posterior to atrial pore.

4. Describe fully the coelom of amphioxus, and compare it with that of the frog in regard to (a) development, (b) its relation to other organs in the adult.

5. Compare the atrial cavity and coelom of amphioxus. To what series of cavities in the frog are the metapleural canals to be compared?

6. Describe the notochord of amphioxus, and point out its differences from the vertebrate notochord.

7. Describe, with diagrams, the nervous system of amphioxus, and compare its nervous axis, in detail, with that of a vertebrate.

8. Compare the genital organs of amphioxus with those of a higher vertebrate.

9. What structures have been regarded, as renal organs in amphioxus?

10. What is a gastrula? With what lower type has the gastrula been compared? Discuss the comparison.



-Development_

The Development of the Frog

Section 1. We have now to consider how the body of the frog is built up out of the egg cell, but previously to doing so we must revert to the reproductive organs of our type.

Section 2. In the testes of the male is found an intricate network of tubuli, the lining of which is, of course, an epithelium. The cells of this epithelium have their internal borders differentiated into spermatozoa, which, at a subsequent stage, are liberated. A spermatozoon (Figure 3, Sheet 13, sp.) is a rod-shaped cell containing a nucleus; in fact, consisting chiefly of nucleus, with a tail, the flagellum, which is vibratile, and forces the spermatozoon, forward by its lashing. The spermatozoa float in a fluid which is the joint product of the testes, anterior part of the kidney, and perhaps the prostate glands.

Section 3. In the ovary, the ova are formed, and grow to a considerable size. They are nucleated cells, the nucleus going by the special name of the germinal vesicle and the nucleolus the germinal spot. The ova prey upon the adjacent cells as they develop. The protoplasm of the ovum, except at that part of the surface where the germinal vesicle lies, is packed with a great amount of food material, the yolk granules. This yolk is non-living inert matter. An ovum such as this, in which the protoplasm is concentrated towards one pole, is called telolecithal.

Section 4. After the ovum has finished its growth, and elaborated the yolk within itself, a peculiar change occurs in the small area free from yolk— the animal pole, in which the germinal vesicle lies. This germinal vesicle divides, and one moiety is budded off from the ovum. The ovum has, in fact, undergone cell division into a very large cell containing most of its substance, and a small protoplasmic pimple surrounding half of its nucleus. The disproportion is so great between the two cells, that the phenomenon does not at first suggest the idea of cell division, and it is usually described as the extrusion of the first polar body. There follows a second and similar small cell, behind the first, the second polar body. Since the nucleus of the ovum has divided twice, it is evident that the nucleus remaining now in the ovum is a quarter of the original nucleus. Very little protoplasm is given off with the polar bodies; they play no further part in development, but simply drop off and disappear. Not only in the frog's ovum, but in all vertebrata, two polar bodies are given off in this way before the sexual process occurs. Their exact meaning has been widely discussed. It is fairly evident that some material is removed from the nucleus, which would be detrimental to further developments, and the point debated is what is the precise nature of this excreted material. This burning question we can scarcely deal with here.

Section 5. But here we may point out that in all cells the function of the nucleus appears to be to determine growth and division. It is the centre of directive energy in the cell.

Section 6. Fertilization is effected by a spermatozoon meeting with the ovum. It fuses with it, its nucleus becoming the male pro-nucleus. This and the female pro-nucleus, left after the extrusion of the polar cells, move towards each other, and unite to form the first segmentation nucleus.

Section 7. The ovum next begins to divide. A furrow cutting deeper and deeper divides it into two; another follows at right angles to this, making the two four, and another equatorial furrow cuts off the animal pole from the yolk or vegetative pole. (See Sheet 22, Figures 1, 2, and 3.) And so segmentation (= cleavage) proceeds, and, at last, a hollow sphere, the blastosphere (Figure 4) is formed, with a segmentation cavity (s.c.). But, because of the presence of the yolk at the vegetative pole of ovum, and of the mechanical resistance it offers to the force of segmentation, the protoplasm there is not nearly so finely divided— the cells, that is to say, are much larger than at the animal pole. The blastosphere of the frog is like what the blastosphere of amphioxus would be, if the future hypoblast cells were enormously larger through their protoplasm being diluted with yolk.

Section 8. The next phase of development has an equally curious resemblance to and difference from what occurs in the case of the ova of animals which do not contain yolk. In such types (e.g., amphioxus) a part of the blastosphere wall is tucked into the rest, and a gastrula formed by this process of invagination. In the frog (Figure 5) there is a tucking-in, but the part that should lie within the gastrula, the yolk-containing cells, are far larger than the epiblast (ep.) which should, form the outer layer of cells. Hence the epiblast can only by continual growth accommodate what it must embrace, and the process of tucking-in is accompanied by one of growth of the epiblast, as shown by the unbarbed arrow, over the yolk. This stage is called the gastrula stage; ar. is the cavity of the gastrula, the archenteron; b.p. is its opening or blastopore. Such a gastrula, formed mainly by overgrowth of the epiblast, is called an epibolic gastrula, as distinguished from the invaginate gastrula of amphioxus. The difference is evidently entirely due to the presence of yolk, and the consequent modification of invagination in the former case.

Section 9. Comparing the two gastrulas, it is not difficult to see that if we imagine the ventral wall of the archenteron of amphioxus to have its cells enormously enlarged through the mixing of yolk with their protoplasm, we should have a gastrula essentially like that of the frog.

Section 10. Figure 6 shows a slightly later ovum than Figure 5, seen from the dorsal side. b.p. is the blastopore. In front of that appears a groove, the neural groove, bordered on either side by a ridge, the neural fold (n.f.). This is seen in section in Figure 7; s.c. is the neural groove; n.f., as before, the neural fold. The neural folds ultimately bend over and meet above, so that s.c. becomes a canal, and is finally separated from the epiblast to form the spinal cord. Below the neural groove a thickening of the dorsal wall of the archenteron appears, and is pinched off to form a longitudinal rod, the precursor of the vertebral column, the notochord, shown in Figure 7 (n.c.), as imperfectly pinched off.

Section 11. Simultaneously, on either side of the notochord appear a series of solid masses of cells, derived mainly by cell division from the cells of the wall of the archenteron, and filling up and obliterating the segmentation cavity. These masses increase in number by the addition of fresh ones behind, during development, and are visible in the dorsal view as brick-like masses, the mesoblastic somites or proto-vertebrae (Figure 6, i., ii., iii.). In Figure 7, these masses are indicated by dotting. In such a primitive type as amphioxus these mesoblastic -somites- [masses] contain a cavity, destined to be the future body cavity, from the first. In the frog, the cavity is not at first apparent; the mesoblast at first seems quite solid, but subsequently what is called the splitting of the mesoblast occurs, and the body cavity (b.c. in Figure 7) appears. The outer mesoblast, lying immediately under the epiblast, constitutes the substance of the somatopleur, and from it will be formed the dermis, the muscles of the body wall, almost all the cartilage and bone of the skeleton, the substance of the limbs, the kidneys, genital organs, heart and bloodvessels, and, in short, everything between the dermis and the coelom, except the nervous system and nerves, and the notochord. The inner mesoblast, the mass of the splanchnopleur, will form the muscle and connective tissue of the wall of the alimentary canal, and the binding substance of the liver and other glands that open into the canal.

Section 12. Figure 8 is one which we reproduce, with the necessary changes in each plate of embryological figures given in this book, so that the student will find it a convenient, one for the purpose of comparison. The lines of dashes, in all cases, signify -epiblast- [hypoblast] , the unbroken black line is -hypoblast-, [epiblast] dotting shows mesoblast, and the shaded rod (n.c.) is the notochord. c.s. is the spinal cord; br.1, br.2, br.3 are the three primary vesicles which constitute the brain, and which form fore, mid, and hind brain respectively. I. is the intestine and Y. the yolk cells that at this early stage constitute its ventral wall.

Section 13. Figure 9 gives a similar diagram of a later stage, but here the blastopore is closed. An epiblastic tucking-in at st., the stomodaeum pre-figures the mouth; pr., the proctodaeum, is a similar posterior invagination which will become the anus. Y., the yolk, is evidently much absorbed. Figure 10 is a young tadpole, seen from the side. The still unabsorbed yolk in the ventral wall of the mesentery gives the creature a big belly. Its mouth is suctorial at this stage, and behind it is a sucker (s.) by which the larvae attach themselves to floating reeds and wood, as shown in the three black figures below.

Section 14. We may now consider the development of the different organs slightly more in detail, though much of this has already been approached. The nervous system, before the closure of the neural groove, has three anterior dilatations, the fore-, mid-, and hind-brains, the first of which gives rise by hollow outgrowths to two pairs of lateral structures, the hemispheres and the optic vesicles. The latter give rise to the retina and optic nerve as described in {Development} Section 40.

Section 15. The hypoblastic notochord is early embraced by a mesoblastic sheath derived from the protovertebrae. This becomes truly cartilaginous, and at regular intervals is alternately thicker and thinner, compressing the notochord at the thicker parts. Hence the notochord has a beaded form within this, at first, continuous cartilaginous sheath. This sheath is soon cut into a series of vertebral bodies by jointings appearing through the points where the cartilage is thickest and the notochord most constricted. Hence what remains of the notochord lies within the vertebral bodies in the frog; while in a cartilaginous fish, such as the dog-fish, or in the embryonic rabbit, the lines of separation appear where the notochord is thickest, and it comes to lie between hollow-faced vertebrae. Cartilaginous neural arches and spines, formed outside the notochordal sheath, enclose the spinal cord in an arcade. The final phase is ossification. As the tadpole approaches the frog stage the vertebral column in the tail is rapidly absorbed, and its vestiges appear in the adult as the urostyle.

Section 16. The development of the skull is entirely dissimilar to that of the vertebral column. It is shown on Figures 1 and 8, Sheet 14; and in the section devoted to the frog's skull a very complete account of the process is given. The process of ossification is described under the histology of the Rabbit.

Section 17. The origin of the circulatory and respiratory organs is of especial interest in the frog. In the tadpole we have essentially the necessities and organization of the fish; in the adult frog we have a clear exposition of the structure of pigeon and rabbit. The tadpole has, at first, a straight tubular heart, burrowed out in somatic mesoblast, and produced forward into a truncus arteriosus. From this arise four afferent branchial arteries, running up along the sides of the four branchial arches, and supplying gills. They unite above on either side in paired hyper-branchial arteries, which meet behind dorsal to the liver, to form a median dorsal aorta. Internal and external carotid arteries supply the head. These four afferent branchial arches are equivalent to the first four of the five vessels of the dog-fish. At first, the paired gills are three in number, external, and tree-like, covered by epiblast (Figures 10 and 11, e.g.), and not to be compared to fish gills in structure, or in fact -with- [to] any other gills within the limits of the vertebrata. Subsequently (hypoblastic) internal gills (int.g., Figure 12), strictly homologous with the gills of a fish, appear. Then a flap of skin outside the hyoid arch grows back to cover over the gills; this is the operculum (op. in Figures 11 and 12, Sheet 22), and it finally encloses them in a gill chamber, open only by a pore on the left, which resembles in structure and physiological meaning, but differs evidently very widely in development, from the amphioxus atrium. At this time, the lungs are developing as paired hollow outgrowths on the ventral side of the throat (Figure 12, L.). As the limbs develop, and the tail dwindles, the gill chamber is obliterated. The capillary interruptions of the gills on the branchial arches (aortic arches) are also obliterated. The carotid gland occupies the position of the first of these in the adult. The front branchial arch here, as in all higher vertebrata, becomes the carotid arch; the lingual represents the base of a pre-branchial vessel; the second branchial becomes the aortic arch. The fourth loses its connection with the dorsal aorta, and sends a branch to the developing lung, which becomes the pulmonary artery. The third disappears. A somewhat different account to this is still found in some text-books of the fate of this third branchial arch. Balfour would appear to have been of opinion that it gave rise to the cutaneous artery, and that the third and fourth vessels coalesced to form the pulmocutaneous, the fourth arch moving forward so as to arise from the base of the third; and most elementary works follow him. This opinion was strengthened by the fact that in the higher types (reptiles, birds, and mammals) no fourth branchial arch was observed, and the apparent third, becomes the pulmonary. But it has since been shown that a transitory third arch appears and disappears in these types.

Section 18. The origin of the renal organ and duct has very considerable controversial interest.* In Figure 13, Sheet 22, a diagrammatic cross-section, of an embryo is shown. I. is the intestine, coe. the coelom, s.c. the spinal cord; n.c. the notochord, surrounded by n.s., the notochordal sheath, ao. is the dorsal aorta. In the masses of somatic mesoblast on either side, a longitudinal canal appears, which, in the torpedo, a fish related to the dog-fish, and in the rabbit, and possibly in all other cases, is epiblastic in origin. This is the segmental duct, which persists, apparently, as the Wolffian duct (W.D.). Ventral to this appears a parallel canal, the Mullerian duct (M.D.), which is often described as being split off from the segmental duct, but which is, very probably, an independent structure in the frog. A number of tubuli, at first metamerically arranged, now appear, each opening, on the one hand, into the coelom by a ciliated mouth, the nephrostome (n.s.), and on the other into the segmental duct. These tubuli are the segmental tubes or nephridia. There grows out from the aorta, towards each, a bunch, of bloodvessels, the glomerulus (compare Section 62, Rabbit). These tubuli ultimately become, in part, the renal tubuli, so that the primitive kidney stretches, at first, along the length of the body cavity from the region, of the gill-slits backward. The anterior part of the kidney, called the pronephros, disappears in the later larval stages. Internal to the kidney on either side there has appeared a longitudinal ridge, the genital ridge (g.r.), which gives rise to testes or ovary, as the case may be.

* In the discussion whether the vertebrata have arisen from some ancestral type, like the earthworm, metamerically segmented, and of fairly high organization, or from a much lower form, possibly even from a coelenterate. Such a discussion is entirely outside the scope of the book, though its mention is necessary to explain the importance given to these organs.

Section 19. The student should now compare the figures on Sheet 17. In the male, tubular connections are established between the testes and the middle part of the primitive kidney (mesonephros). These connections are the vasa efferentia (v.e.), and the mesonephros is now equivalent to the epididymis of the rabbit. The Wolffian duct is the urogenital duct of the adult, and the Mullerian duct is entirely absorbed, or remains, more or less, in exceptional cases.

In the female, the Mullerian duct increases greatly in length— so that at sexual maturity its white coils appear thicker and longer than the intestine— and becomes the oviduct; the Wolffian duct is the ureter, and the mesonephros is not perverted in function from its primary renal duty.

Section 20. Tabulating these facts—

In the adult male: Pronephros disappears. The Mullerian duct (? = pronephric duct) disappears. Mesonephros = Epididymis; its duct, the urogenital. Metanephros and duct, not clearly marked off from Mesonephros. (Compare Dog-fish, Section 19.)

In the adult female: Pronephros disappears. The Mullerian duct, the oviduct. Mesonephros and Metanephros, the kidney, and their unseparated ducts, the ureters.

Section 21. Hermaphrodism (i.e., cases of common sex) is occasionally found among frogs; the testis produces ova in places, and the Mullerian duct is retained and functional. The ciliated nephrostomata remain open to a late stage of development in the frog, and in many amphibia throughout life. Their connection with the renal tubuli is, however, lost.

Section 22. The alimentary canal is, at first, a straight tube. Its disproportionate increase in length throws it into a spiral in the tadpole (int. Figure 11), and accounts for its coiling in the frog. The liver and other digestive glands are first formed, like the lungs, as hollow outgrowths, and their lining is therefore hypoblastic. The greatest relative length of intestine is found in the tadpole, which, being a purely vegetable feeder, must needs effect the maximum amount of preparatory change in its food.



The Development of the Fowl

Section 23. The frog has an ovum with a moderate allowance of yolk, but the quantity is only sufficient to start the little animal a part of its way towards the adult state. The fowl, on the contrary, has an enormous ovum, gorged excessively, with yolk, and as a consequence the chick is almost perfected when it is hatched. The so-called yolk, the yellow of an egg, is the ovum proper; around that is a coating of white albumen, in a shell membrane and a shell. At either end of the yolk (Figure 1, y.) twisted strands of albuminous matter, the chalazae (ch.) keep the yolk in place. The animal pole is a small grey protoplasmic area, the germinal area (g.a.), on the yolk.

Section 24. We pointed out that the presence of the yolk in the frog's egg led to a difference in the size of the cells at the animal and vegetable poles. The late F.M. Balfour, borrowing a mathematical technicality, suggested that the rate of segmentation in any part of an ovum varies inversely with the amount of yolk. In the fowl's egg, except just at the germinal area, the active protoplasm is at a minimum, the inert yolk at a maximum; the ratio of yolk to protoplasm is practically infinity, and the yolk therefore does not segment at all. The yolk has diluted the active protoplasm so much as to render its influence inappreciable. The germinal area segments, and lies upon the yolk which has defeated the efforts of its small mingling of protoplasm to divide. Such a type of segmentation in which only part of the ovum segments is called meroblastic. If we compare this with the typical blastosphere of the lower type, we see that it is, as it were, flattened out on the yolk. This stage is shown in section in the lower figure of Figure 1. b.d., the blastoderm, is from this point of view, a part of the ripped and flattened blastosphere, spread out on the yolk; s.c. is the segmentation cavity, and y. the yolk.

Section 25. There is no open invagination of an archenteron in the fowl, as in the frog—, the gastrula, like the blastosphere, stage is also masked. But, in the hinder region of the germinal area, a thick mass of cells, grows inward and forward, and, appearing in the dorsal view of the egg as a white streak, is called the primitive streak (p.s.). By a comparison of the figures of frog and fowl the student will easily perceive the complete correspondence of the position of this with the blastopore of the frog. The relation of the two will be easily understood if we compare the fowl's archenteron to a glove-finger under pressure— its cavity is obliterated— and the frog's to the glove-finger blown out. The tension of the protoplasm, straining over the enormous yolk, answers to the pressure. The gastrula in the fowl is solid. The primitive streak is, in fact, the scar of a closed blastopore. As we should expect from this view of its homology, at the primitive streak, the three embryonic layers are continuous and indistinguishable (Figure 2). Elsewhere in the blastoderm they are distinctly separate. Just as the yolk cells of the frog form the ventral wall of the intestine, so nuclei appear along the upper side of the yolk of the fowl, where some protoplasm still exists, and give rise to the ventral hypoblastic cells. By conceiving a gradually increasing amount of yolk in the hypoblastic cells in the ventral side of the archenteron, the substantial identity of the gastrula stage in the three types, which at first appear so strikingly different, will be perceived. Carry Figures 4 and 5 of the frog one step further by increasing the size of the shaded yolk and leaving it unsegmented, and instead of ar. in 5 show a solid mass of cells, and the condition of things in the fowl would at once be rendered.

Section 26. Figure 3a of the fowl will conveniently serve for comparison with Figure 7 of the frog. The inturning of the medullary groove is entirely similar in the two cases. The mesoblast appears as solid mesoblastic somites. In the section above Figure 4 this layer is shown as having split into somatopleur (so.) and splanchnopleur (spch.). Figure 3 answers to Figure 6 of the frog, and Figure 4 is a later stage, in which the medullary groove is beginning to close at its middle part. The clear club-shaped area around the embryo (a.p.) is the area pellucida; the larger area without this is the area opaca (a.o.), in which the first bloodvessels arise by a running together and a specialization of cells. The entire germinal area grows steadily at its edges to creep over and enclose the yolk.

Section 27. So far, the essential differences between the development of fowl and frog, the meroblastic segmentation, absence of a typical gastrula, and the primitive streak, seem comprehensible on the theory that such differences are due to the presence of an enormous amount of yolk. Another difference that appears later is that, while the tadpole has an efficient pronephros, the fowl, which has no larval (free imperfect) stages in its life history, has the merest indication of such a structure.

Section 28. Another striking contrast, due to, or connected with, this plethora of yolk, is the differentiation of a yolk sac (= umbilical vesicle) and the development of two new structures, the amnion and allantois, in the fowl. If the student will compare Figure 10 of the frog, he will see that the developing tadpole encloses in its abdomen all the yolk provided for it. This is a physical impossibility in the fowl. In the fowl (Figure 2, Sheet 24) the enormous yolk (Y.) lies outside of the embryo, and, as the cells of the germinal area grow slowly over it, umbilical bloodvessels are developed to absorb and carry the material to the embryo. In the case of an embryo sinking in upon, as it absorbs, this mass of nutritive material, a necessity for some respiratory structure is evident. From the hinder end of the fowl's intestine, in a position corresponding to the so-called, urinary bladder of the frog, a solid outgrowth, the allantois, which speedily becomes hollow, appears. Early stages are shown in Figures 1 and 2, Sheet 24 (al.); while the same thing is shown more diagrammatically on Sheet 23, Figure 6 (all.). This becomes at last a great hollow sac, which is applied closely to the porous shell, and the extent of which will be appreciated by looking at Figure 5, Sheet 24, where the allantois is shaded. Allantoic bloodvessels ramify thickly over its walls, and aeration occurs through the permeable shell.

Section 29. The nature of the amnion will be understood by following Figures 4b, 5, and 6 on Sheet 23. The three embryonic layers are indicated by broken lines, dots, and black lines, just as they are in the frog diagrams. Not only is the embryo slowly pinched off from the yolk sac (y.s.), but, as the yolk is absorbed beneath it, and it grows in size, it sinks into the space thus made, the extra-embryonic somatopleur and epiblast rise up round it as two folds, which are seen closing in 5, and closed in 6, over the dorsal side of the young chick. In this way a cavity, a., lined by epiblast, and called the amniotic cavity, is formed. Dorsal to this, in 6, comes a space lined by somatic mesoblast, and continuous with p.p., the pleuro-peritoneal cavity, or body cavity of the embryo. Outside this, again, is a layer, of somatopleur internally and epiblast externally, the false amnion (f.a.), which is continuous with the serous membrane (s.m.) enclosing the rest of the egg. The student should, carefully copy these diagrams, with coloured pencils or inks for the different layers, and should compare them with the more realistic renderings of Figures 2, 5, and 8, Sheet 24.

Section 30. The heart in the fowl appears first as a pair of vessels, which unite to form a straight trunk in the median line, as the flattened-out embryo closes in from the yolk. The way in which this straight trunk is thrown, first of all, into the S shape of the fish heart, and then gradually assumes the adult form, is indicated roughly by Figure 3. In one respect the development of the heart does not follow the lines one would expect. Since, between the fish and the higher form comes the condition of such an animal as the frog, in which the auricles are divided, while there is only one ventricle, we might expect a stage in which the developing chick's heart would have one ventricle, and a septum between the auricles. But, as a matter of fact, the ventricles in fowl and rabbit are separated first, and the separation of the auricles follows, and is barely complete at birth.

Section 31. Two vitelline veins from the yolk sac (v.v.) flow into the heart from behind, as shown in Figure 1. A later more complete and more diagrammatic figure of the circulation is seen in Figure 7. At first there are two anterior cardinal (a.c.), and two posterior cardinal veins (p.c.) uniting to form Cuvierian sinuses (c.s.) that open into the heart just as in the dog-fish. But later the inferior cava is developed and extends backward, the posterior cardinals atrophy, the Cuvierian sinuses become the superior cavae, and the anterior cardinals the internal jugular veins. The vitelline veins (v.v.) flow, at first, uninterruptedly through the liver to the inferior cava, but, as development proceeds, a capillary system is established in the liver, and the through communication, the ductus venosus, is reduced— at last— completely. Bearing in mind that the yolk is outside the body in the fowl and inside it in the frog, the vitelline veins of the former have a considerable resemblance in position, and in their relation to the portal vein, to a portion of the single anterior abdominal vein. Blood is taken out to the allantois, however, by the arteries of the latter type.

Section 32. Five aortic arches are generally stated to appear altogether in the fowl, but not simultaneously. The first two, the mandibular and the hyoid vascular arches, early disappear, and are not comparable to any in the frog. The third is the first branchial arch, and, like the corresponding arch in the frog, forms the carotid artery; the second branchial is the aortic arch; and what has hitherto been regarded as the third (the fifth arch, i.e.) the pulmonary artery. A transitory arch, it is now known, however, appears between the second branchial and the last, and it is therefore the fourth branchial arch which is the pulmonary, just as it is in the frog.

Section 33. Blood, it may be mentioned, first appears in the area vasculosa, the outer portion of the area opaca. Embryonic cells send out processes, and so become multipolar; the processes of adjacent cells coalesce. The nucleus divides, and empty spaces appear in the substance of each of the cells.

In this way, the cavities of the smaller vessels and capillaries are formed, and the products of the internal divisions of the cells become the corpuscles within the vessels. The red blood corpuscles of the rabbit, it may be added, are nucleated for a considerable portion of embryonic life. Larger vessels and the heart are burrowed, as it were, out of masses of mesoblast cells. The course of the blood in the embryo is by the veins to the right auricle, thence through the imperfection of the auricular septum already alluded to, into the left auricle. Then the left ventricle, aortic arches (for the future pulmonary artery is in communication by a part presently blocked, the ductus arterious, with the systemic aorta), arteries, capillaries, veins. The liver capillary system and the pulmonary system only become inserted upon the circulation at a comparatively late stage.

Section 34. With the exception of the reduction of the pronephros, what has been said of the development of the frog's nervous system, renal and reproductive organs, and skeleton, applies sufficiently to the fowl for our present purposes. The entire separation of Wolffian and Mullerian ducts from the very beginning of development is here beyond all question (vide Section 18). But the notochord in the fowl is not so distinctly connected with the hypoblast, and so distinct from the mesoblast, as it is in the lower type, and no gills, internal or external, are ever developed. The gill slits occur with a modification due to the slitting and flattening out of the embryo, already insisted upon; for, whereas in the tadpole they may be described as perforations, in the fowl they appear as four notches between ingrowing processes that are endeavouring to meet in the middle line.



The Development of the Rabbit

Section 35. The early development of the rabbit is apt to puzzle students a little at first. We have an ovum practically free from yolk (alecithal), and, therefore, we find it dividing completely and almost equally. We naturally assume, from what we have learnt, that the next stages will be the formation of a hollow blastosphere, invagination, a gastrula forming mesoblast by hollow outgrowths from the archenteron, and so on. There is no yolk here to substitute epiboly (Section 9) for invagination, nor to obliterate the archenteron and the blastopore through its pressure.

Yet none of these things we have anticipated occur!

We find solid mesoblastic somites, we find primitive streak, allantois and amnion, features we have just been explaining as the consequence of an excess of yolk in the egg. We even find a yolk sac with no yolk in it.

Section 36. A solid mass of cells is formed at the beginning, called a morula, Figure 1. In this we are able to distinguish rather smaller outer layer cells (o.l.c.), and rather larger inner layer cells (i.l.c.), but these cells, in their later development, do not answer at all to the two primitive layers of the gastrula, and the name of Van Beneden's blastopore (V.B.b.), for a point where the outer layer of cells is incomplete over the inner, only commemorates the authorship of a misnomer. The uniformity, or agreement, in the development of our other vertebrate types is apparently departed from here.

{Illustration: Development Section 36.}

Section 37. As the egg develops, however, we are astonished to find an increasing resemblance to that of the fowl. A split occurs at one point between outer layer and inner layer cells, and the space resulting (Y in Figure 2) is filled by an increasing amount of fluid, and rapidly enlarges, so that presently we have the state of affairs shown in 3, in which the inner layer cells are gathered together at one point on the surface of the ovum, and constitute the germinal area. If, with Hubrecht, we regard the outer layer cells as an egg membrane, there is a curious parallelism between this egg and the fowl's the fluid Y representing the yolk; and the inner layer cells the cells of the fowl's germinal area.

At any rate, the subsequent development goes far to justify such a view. The inner cells split into epi-, meso-, and hypo-blast, like the blastoderm in the fowl; there is a primitive streak and no blastopore; an amnion arises; the yolk sac, small and full of serous fluid, is cut off just as the enormous yolk of the fowl is cut off; and an allantois arises in the same way. There is no need to give special diagrams— Figures 3, 4b, 5, and 6 of the fowl will do in all respects, except proportion, for the development of the rabbit. The differences are such as we may account for, not on the supposition that the rabbit's ovum never had any yolk, but that an abundant yolk has been withdrawn from it. The nutrition of the embryo by yolk has been superseded by some better method. The supposition that the rabbit is descended from ancestors which, like the birds and reptiles, laid eggs with huge quantities of yolk, meets every circumstance of the case.

Section 38. But the allantois and yolk sac of the rabbit, though they correspond in development, differ entirely in function from the similar organs of the fowl. The yolk sac is of the very smallest nutritive value; instead of being the sole source of food, its contents scarcely avail the young rabbit at all as nourishment. Its presence in development is difficult to account for except on the supposition, that it was once of far greater importance. At an early stage, the outgrowing allantois, pushing in front of it the serous membrane, is closely applied to the lining of the mother's uterus. The maternal uterus and the embryonic allantois send out finger-like processes into each other which interlock, and the tissue between the abundant bloodvessels in them thins down to such an extent that nutritive material, peptones and carbohydrates, and oxygen also, diffuse freely through it from mother to foetus,* and carbon dioxide, water, and urea from the foetus to the mother. The structure thus formed by the union of the wall of the maternal uterus, allantois, and the intermediate structures is called the placenta. Through its intermediation, the young rabbit becomes, as it were, rooted and parasitic on the mother, and utilizes her organs for its own alimentation, respiration, and excretion. It gives off CO2, H2O, and urea, by the placenta, and it receives O and elaborated food material through the same organ. This is the better method that has superseded the yolk.

* The embryo.

Section 39. In its later development, the general facts already enunciated with regard to the organs of frog and fowl hold, and where frog and fowl are stated to differ, the rabbit follows the fowl. In the circulation the left fourth vascular arch (second branchial) gives rise to the aortic arch; in the right the corresponding arch disappears, except so much of it as remains as the innominate artery. The azygos vein (Chapter 3) -is a vestige of- [is derived from] the right posterior cardinal sinus. Both pulmonary arteries in the rabbit are derived from the left sixth vascular arch (= fourth branchial). Compare Section 32. The allantois altogether disappears in the adult fowl; in the adult mammal a portion of its hollow stalk remains as the urinary bladder, and the point where it left the body is marked by the umbilicus or navel. The umbilical arteries become the small hypogastric arteries on either side of the urinary bladder. There is no trace of a pronephros at all in the rabbit.

Section 40. We may note here the development of the eye. This is shown in Figure 4, Sheet 24. A hollow cup-shaped vesicle from the brain grows out towards an at first hollow cellular ingrowth from the epidermis. The cavity within the wall of the cup derived from the brain is obliterated, [and the stalk withers,] the cup becomes the retina, and -its stalk- [thence fibres grow back to the brain to form] the optic nerve. The cellular ingrowth is the lens. The remainder of the eye-structures are of mesoblastic origin, except the superficial epithelium of the cornea. The retinal cup is not complete at first along the ventral line, so that the rim of the cup, viewed as in Figure 1, r., is horseshoe shaped. -Hence the optic nerve differs from other nerves in being primitively hollow.- In all other sense organs, as, for instance, the olfactory sacs and the ears, the percipient epithelium is derived, from the epiblast directly, and not indirectly through the nervous system. These remarks apply to all vertebrate types.

Section 41. The supposition, that the general characters of the rabbit's ovum were stamped upon it as an heritage from a period when the ancestors of the mammals were egg-laying reptiles, is strengthened by the fact that the two lowest and most reptile-like of all the mammalia, the duck-billed platypus and the echidna, have been shown to depart from the distinctive mammalian character, and to lay eggs. And, in further confirmation of this supposition, we find, in tracing the mammals and reptiles back through the geological record, that in the Permian and Triassic rocks there occur central forms which combine, in a most remarkable way, reptilian and mammalian characteristics.

Section 42. In conclusion, we would earnestly recommend the student to see more of embryological fact than what is given him here. It is seeing and thinking, much more than reading, which will enable him to clothe the bare terms and phrases of embryology with coherent knowledge. In Howes' Atlas of Biology there is a much fuller series of figures of the frog's development than can be given here, and they are drawn by an abler hand than mine can pretend to be. There is also an Atlas d'Embryologie, by Mathias Duval, that makes the study of the fowl's development entertaining and altogether delightful. Such complete series as these are, from the nature of the case, impossible with the rabbit. Many students who take up the subject of biology do so only as an accessory to more extended work in other departments of science. To such, practical work in embryology is either altogether impossible, or only possibly to a very limited extent. The time it will consume is much greater, and the intellectual result is likely to be far less than the study of such plates as we have named.



The Theory of Evolution

Section 43. We have now considered our types, both from the standpoint of adult anatomy and from embryological data; and we have seen through the vertebrate series a common structure underlying wide diversity in external appearance and detailed anatomy. We have seen a certain intermediateness of structure in the frog, as compared with the rabbit and dog-fish, notably in the skull and skeleton, in the circulation, in the ear, and in the reduced myomeres; and we have seen that the rabbit passes in these respects, and in others, through dog-fish- and frog-like stages in its development, and this alone would be quite sufficient to suggest that the similarities of structure are due to other causes than a primordial adaptation to certain conditions of life.

Section 44. It has been suggested by very excellent people that these resemblances are due to some unexplained necessity of adherence to type, as though, the power that they assume created these animals originally, as they are now, coupled creative ability with a plentiful lack of ideas, and so perforce repeated itself with impotent variations. On the other hand, we have the supposition that these are "family likenesses," and the marks of a common ancestry. This is the opinion now accepted by all zoologists of repute.

Section 45. It must not be for a moment imagined that it is implied that rabbits are descended from frogs, or frogs from dog-fish, but that these three forms are remote cousins, derived from some ancient and far simpler progenitor. But since both rabbit and frog pass through phases like the adult condition of the dog-fish, it seems probable that the dog-fish has remained more like the primordial form than these two, and similarly, the frog than the rabbit.

Section 46. Hence we may infer that the mammals were the last of the three groups, of which we have taken types, to appear upon the earth, and that the fishes preceded, the amphibia. Workers in an entirely independent province, that of palaeontology, completely endorse this supposition. The first Vertebrata to appear in the fossil history of the world are fishes; fish spines and placoid scales (compare dog-fish) appear in the Ordovician rocks. In the coal measures come the amphibia; and in the Permo-triassic strata, reptile-like mammals. In the Devonian rocks, which come between the Silurian and the coal measures, we find very plentiful remains of certain fish called the dipnoi, of which group three genera still survive; they display, in numberless features of their anatomy, transitional characters between true fish and amphibia. Similarly, in the Permian come mammal-like reptiles, that point also downward to the amphibia. We find, therefore, the story told by the ovum written also in the rocks.

Section 47. Now, when this fact of a common ancestry is considered, it becomes necessary to explain how this gradual change of animal forms may have been brought about.

Section 48. Two subcontrary propositions hold of the young of any animal. It resembles in many points its parent. It differs in many points from its parent. The general scheme of structure and the greater lines of feature are parental, inherited; there are also novel and unique details that mark the individual. The first fact is the law of inheritance; the second, of variation.

Section 49. Now the parent or parents, since they live and breed, must be more or less, but sufficiently, adapted to their conditions of living— more or less fitted to the needs of life. The variation in the young animal will be one of three kinds: it will fit the animal still better to the conditions under which its kind live, or it will be a change for the worse, or it is possible to imagine that the variation— as in the colour variations of domesticated cats— will affect its prospects in life very little. In the first case, the probability is that the new animal will get on in life, and breed, and multiply above the average; in the second, it is probable that, in the competition for food and other amenities of life, the disadvantage, whatever it is, under which the animal suffers will shorten its career, and abbreviate the tale of its offspring; while, in the third case, an average career may be expected. Hence, disregarding accidents, which may be eliminated from the problem by taking many cases, there is a continual tendency among the members of a species of animals in favour of the proportionate increase of the individuals most completely adapted to the conditions under which the species lives. That is, while the conditions remain unchanged, the animals, considered as one group, are continually more highly perfected to live under those conditions. And under changed conditions the specific form will also change.

Section 50. The idea of this process of change may be perhaps rendered more vivid by giving an imaginary concrete instance of its working. In the jungles of India, which preserve a state of things which has existed for immemorial years, we find the tiger, his stripes simulating jungle reeds, his noiseless approach learnt from nature in countless millions of lessons of success and failure, his perfectly powerful claws and execution methods; and, living in the same jungle, and with him as one of the conditions of life, are small deer, alert, swift, light of build, inconspicuous of colour, sharp of hearing, keen-eyed, keen-scented— because any downward variation from these attributes means swift and certain death. To capture the deer is a condition, of the tiger's life, to escape the tiger a condition of the deer's; and they play a great contest under these conditions, with life as the stake. The most alert deer almost always escape; the least so, perish.

Section 51. But conditions may alter. For instance, while most of these deer still live in the jungle with tigers, over a considerable area of their habitat, some change may be at work that thins the jungle, destroys the tigers in it, and brings in, let us say, wolves, as an enemy to the deer, instead of tigers. Now, against the wolves, which do not creep, but hunt noisily, and which do not spring suddenly upon prey, but follow by scent, and run it down in packs, keen eyes, sharp ears, acute perceptions, will be far less important than endurance in running. The deer, under the new conditions, will need coarser and more powerful limbs, and a larger chest; it will be an advantage to be rough and big, instead, of frail and inconspicuous, and the ears and eyes need not be so large. The old refinements will mean weakness and death; any variation along the line of size and coarseness will be advantageous. Slight and delicate deer will be continually being killed, rougher and stronger deer continually escaping. And so gradually, under the new circumstances, if they are not sufficient to exterminate the species, the finer characteristics will be eliminated, and a new variety of our old jungle deer will arise, and, if the separation and contrast of the conditions is sufficiently great and permanent, we may, at last, in the course of ages, get a new kind of deer specifically different in its limbs, body, sense organs, colour, and instincts, from the deer that live in the jungle. And these latter will, on their side, be still continually more perfected to the jungle life they are leading.