CHAPTER XX.

PEDIGREE AND HISTORY OF THE ANIMAL KINGDOM.

III. Vertebrate Animals.

Not one of the natural groups of organisms—which, we have designated as tribes, or phyla, on account of the blood-relationship of all the species included in them—is of such great and exceeding importance as the tribe of Vertebrate Animals. For, according to the unanimous opinion of all zoologists, man also is a member of the tribe; and his whole organization and development cannot possibly be distinguished from that of other Vertebrate animals. But as from the individual history of human development, we have already recognized the undeniable fact that, in developing out of the egg, man at first does not differ from other Vertebrate animals, and especially from Mammals, we must necessarily come to the conclusion, in regard to the palæontological history of his development, that man has, historically, actually developed out of the lower Vertebrata, and that he is directly derived from lower mammals. This circumstance, together with the many high interests which, in other respects, entitle the Vertebrata to more consideration than other organisms, justifies us in examining the pedigree of the Vertebrata and its expression in the natural system, with special care.

Fortunately, the records of creation, which must in all cases be our guide in establishing pedigrees, are especially complete in this important animal tribe, from which our own race has arisen. Even at the beginning of our century Cuvier’s comparative anatomy and palæontology, and Bär’s ontogeny of the Vertebrate animals, had brought us to a high level of accurate knowledge on this matter. Since then it is especially due to Johannes Müller’s and Rathke’s investigations in comparative anatomy, and most recently to those of Gegenbaur and Huxley, that our knowledge of the natural relationships among the different groups of Vertebrata has become enlarged. It is especially Gegenbaur’s classical works, penetrated as they are throughout with the fundamental principles of the Theory of Descent, which have demonstrated that the material of comparative anatomy receives its true importance and value only by the application of the Theory of Descent, and this in the case of all animals, but especially in that in the Vertebrate tribe. Here, as everywhere else, analogies must be traced to Adaptation, homologies to Transmission by Inheritance. When we see that the limbs of the most different Vertebrata, in spite of their exceedingly different external forms, nevertheless possess essentially the same internal structure; when we see that in the arm of a man and ape, in the wing of a man or a bird, in the breast fins of whales and sea-dragons, in the fore-legs of hoofed animals and frogs, the same bones always lie in the same characteristic position, articulation and connection—we can only explain this wonderful agreement and homology by the supposition of a common transmission by inheritance from a single primary form. On the other hand, the striking differences of these homologous bodily parts proceed from adaptation to different conditions of existence. (Compare Plate IV.)

Ontogeny, or the individual history of development, like comparative anatomy, is of especial importance to the pedigree of the Vertebrata. The first stages of development arising out of the egg are essentially identical in all Vertebrate animals, and retain their agreement the longer, the nearer the respective Vertebrate animal forms, when fully developed, stand to one another in the natural system, that is, in the pedigree. How far this agreement of germ forms, or embryos, extends, even in the most highly developed Vertebrate animals, I have already had occasion to explain (vol i. pp. 306-309). The complete agreement in form and structure, for example, in the embryos of a man and a dog, of a bird and a tortoise, existing in the stages of development represented on Plates II. and III., is a fact of incalculable importance, and furnishes us with the most important data for the construction of their pedigree.

Finally, the palæontological records of creation are also of especial value in the case of these same Vertebrate animals; for their fossil remains belong for the most part to the bony skeleton, a system of organs which is of the utmost importance for understanding their general organization. It is true that here, as in all other cases, the fossil records are exceedingly imperfect and incomplete, but more important remains of extinct Vertebrate animals have been preserved in a fossil state, than of most other groups of animals; and single fragments frequently furnish the most important hints as to the relationship and the historical succession of the groups.

The name of Vertebrate Animals (Vertebrata), as I have already said, originated with the great Lamarck, who towards the end of the last century comprised under this name, Linnæus’ four higher classes of animals, viz. Mammals, Birds, Amphibious animals, and Fishes, Linnæus’ two lower classes, Insects and Worms, Lamarck contrasted to the Vertebrata as Invertebrata, later also called Evertebrata.

The division of the Vertebrata into the four classes above named was retained also by Cuvier and his followers, and in consequence by many zoologists down to the present day. But in 1822 Blanville, the distinguished anatomist, found out by comparative anatomy—which Bär did almost at the same time from the ontogeny of Vertebrata—that Linnæus’ class of Amphibious animals was an unnatural union of two very different classes. These two classes were separated as early as 1820, by Merrin, as two main groups of Amphibious animals, under the names of Pholidota and Batrachia. The Batrachia, which are at present (in a restricted sense) called Amphibious animals, comprise Frogs, Salamanders, gilled Salamanders, Cæcilia, and the extinct Labyrinthodonta. Their entire organization is closely allied to that of Fishes. The Pholidota, or Reptiles, on the other hand, are much more closely allied to Birds. They comprise lizards, serpents, crocodiles, and tortoises, and the groups of the mesolithic Dragons, Flying reptiles, etc.

In conformity with this natural division of Amphibious animals into two classes, the whole tribe of Vertebrate animals was divided into two main groups. The first main group, containing Amphibious animals and Fishes, breathe throughout their lives, or in early life, by means of gills, and are therefore called gilled Vertebrata (Branchiata, or Anallantoida). The second main group—Reptiles, Birds, and Mammals—breathe at no period of their lives through gills, but exclusively through lungs, and hence may appropriately be called Gill-less, or Vertebrata with lungs (Abranchiata, or Allantoida). However correct this distinction may be, still we cannot remain satisfied with it if we wish to arrive at a true natural system of the vertebrate tribe, and at a right understanding of its pedigree. In this case, as I have shown in my General Morphology, we are obliged to distinguish three other classes of Vertebrate animals, by dividing what has hitherto been regarded as the class of fishes into four distinct classes. (Gen. Morph. vol. ii. Plate VII. pp. 116-160.)

The first and lowest of these classes comprises the Skull-less animals (Acrania), or animals with tubular hearts (Leptocardia), of which only one representative now exists, namely, the remarkable little Lancelet (Amphioxus lanceolatus). Nearly allied to this is the second class, that of the Single-nostriled animals (Monorrhina), or Round-mouthed animals (Cyclostoma), which includes the Hags (Myxinoida) and Lampreys (Petromyzonta). The third class contains only the genuine Fish (Pisces): the Mud-fishes (Dipneusta) are added to these as a fourth class, and form the transition from Fish to Amphibious animals. This distinction, which, as will be seen immediately, is very important for the genealogy of the Vertebrate animals, increases the original number of Vertebrate classes from four to eight.

In most recent times a ninth class of Vertebrata has been added to these eight classes. Gegenbaur’s recently published investigations in comparative anatomy prove that the remarkable class of Sea-dragons (Halisauria), which have hitherto been included among Reptiles, must be considered quite distinct from these, and as a separate class which branched off from the Vertebrate stock, even before the Amphibious animals. To it belong the celebrated large Ichthyosauri and Plesiosauri of the oolitic and chalk periods, and the older Simosauri of the Trias period, all of which are more closely allied to Fish than to Amphibious animals.

These nine classes of Vertebrate animals are, however, by no means of the same genealogical value. Hence we must divide them, as I have already shown in the Systematic Survey on p. 133, into four distinct main-classes or tribes. In the first place, the three highest classes, Mammals, Birds, and Reptiles, may be comprised as a natural main-class under the name of Amnion animals (Amnionata). The Amnion-less animals (Anamnionata), naturally opposed to them as a second main-class, include the four classes of Batrachians, Sea-dragons, Mud-fish, and Fishes. The seven classes just named, the Anamnionata as well as the Amnionata, agree among one another in numerous characteristics, which distinguish them from the two lowest classes (the single-nostriled and tubular-hearted animals). Hence we may unite them in the natural main group of Double-nostriled animals (Amphirrhina). Finally, these Amphirrhina on the whole are much more closely related to those animals with round mouths or single nostrils than to the skull-less or tube-hearted animals. We may, therefore, with full justice class the single and double-nostriled animals into one principal main group, and contrast them as animals with skulls (Craniota), or bulbular hearts (Pachycardia), to the one class of skull-less animals, or animals with tubular hearts. This classification of the Vertebrate animals proposed by me renders it possible to obtain a clear survey of the nine classes in their most important genealogical relations. The systematic relationship of these groups to one another may be briefly expressed by the following table.

The only one representative of the first class, the small lanceolate fish, or Lancelet (Amphioxus lanceolatus) (Plate XIII. Fig. B), stands at the lowest stage of organization of all the Vertebrate animals known to us. This exceedingly interesting and important animal, which throws a surprising light upon the older roots of our pedigree, is evidently the last of the Mohicans—the last surviving representative of a lower class of Vertebrate animals, very rich in forms, and very highly developed during the primordial period, but which unfortunately could leave no fossil remains on account of the absence of all solid skeleton. The Lancelet still lives widely distributed in different seas; for instance, in the Baltic, North Sea, and Mediterranean, where it generally lies buried in the sand on flat shores. The body, as the name indicates, has the form of a narrow lanceolate leaf, pointed at both extremities. When full grown it is about two inches long, of a white colour and semi-transparent. Externally, the little lanceolate animal is so little like a vertebrate animal that Pallas, who first discovered it, regarded it as an imperfect naked snail. It has no legs, and neither head, skull, nor brain. Externally, the fore end of the body can be distinguished from the hinder end only by the open mouth. But still the Amphioxus in its internal structure possesses those most important features, which distinguish all Vertebrate animals from all Invertebrate animals, namely, the spinal rod and spinal marrow. The spinal rod (Chorda dorsalis) is a straight, cylindrical, cartilaginous staff, pointed at both ends, forming the central axis of the internal skeleton, and the basis of the vertebral column. Directly above the spinal rod, on its dorsal side, lies the spinal marrow (medulla spinalis), likewise originally a straight but internally hollow cord, pointed at both ends. This forms the principal piece and centre of the nervous system in all Vertebrate animals. (Compare above vol. i. p. 303.) In all Vertebrate animals without exception, man included, these important parts of the body during the embryological development out of the egg, originally begin in the same simple form, which is retained throughout life by the Amphioxus. It is only at a later period that the brain develops by the expansion of the fore end of the spinal marrow, and out of the spinal rod the skull which encloses the brain. As these two important organs do not develop at all in the Amphioxus, we may justly call the class represented by it, Skull-less animals (Acrania), in opposition to all the others, namely, to the animals with skulls (Craniota). The Skull-less animals are generally called tubular-hearted (Leptocardia), because a centralized heart does not as yet exist, and the blood is circulated in the body by the contractions of the tubular blood-vessels themselves. The Skulled animals, which possess a centralized, thick-walled, bulb-shaped heart, ought then by way of contrast to be called bulbular-hearted animals (Pachycardia).

Animals with skulls and central hearts evidently developed gradually in the later primordial period out of those without skulls and with tubular hearts. Of this the ontogeny of skulled animals leaves no doubt. But whence are these same skull-less animals derived? It is only very lately that an exceedingly surprising answer has been given to this important question. From Kowalewsky’s investigations, published in 1867, on the individual development of the Amphioxus and the adhering Sea-squirts (Ascidia) belonging to the class of mantled animals (Tunicata), it has been proved that the ontogenies of these two entirely different looking animal-forms agree in the first stage of development in a most remarkable manner. The freely swimming larvæ of the Ascidians (Plate XII. Fig. A) develop the undeniable beginning of a spinal marrow (Fig. 5 g) and of a spinal rod (Fig. 5 c), and this moreover in entirely the same way as does the Amphioxus. (Plate XIII. Fig. B.) It is true that in the Ascidians these most important organs of the Vertebrate animal-body do not afterwards develop further. The Ascidians take on a retrograde transformation, become attached to the bottom of the sea, and develop into shapeless lumps, which when looked upon externally would scarcely be supposed to be animals. (Plate XIII. Fig. A.) But the spinal marrow, as the beginning of the central nervous system, and the spinal rod, as the first basis of the vertebral column, are such important organs, so exclusively characteristic of Vertebrate animals, that we may from them with certitude infer the true blood relationship of Vertebrate with Tunicate animals. Of course we do not mean to say by this, that Vertebrate animals are derived from Tunicate animals, but merely that both groups have arisen out of a common root, and that the Tunicates, of all the Invertebrata, are the nearest blood relations of the Vertebrates. It is quite evident that genuine Vertebrate animals developed progressively during the primordial period (and the skull-less animals first) out of a group of worms, from which the degenerate Tunicate animals arose in another and a retrograde direction. (Compare the more detailed explanation of Plates XII. and XIII. in the Appendix.)

Out of the Skull-less animals there developed, in the first instance, a second low class of Vertebrate animals, which still stands far below that of fish, and which is now represented only by the Hags (Myxinoida) and Lampreys (Petromyzonta). This class also, on account of the absence of all solid parts, could, unfortunately, as little as the Skull-less animals leave fossil remains. From its whole organization and ontogeny it is quite evident that it represents a very important intermediate stage between the Skull-less animals and Fishes, and that its few still existing members are only the last surviving remains of a probably very highly developed animal group which existed towards the end of the primordial period. On account of the curious mouth possessed by the Hags and Lampreys, which they use for sucking, the whole class is usually called Round-mouthed animals (Cyclostoma). The name of Single-nostriled animals (Monorrhina) is still more characteristic. For all Cyclostoma possess a simple, single nasal tube, whereas, in all other Vertebrate animals (with the exception of the Amphioxus) the nose consists of two lateral halves, a right and a left nostril. We are therefore enabled to comprise these latter (Anamnionata and Amnionata) under the heading, double-nostriled animals (Amphirrhina). All the Amphirrhina possess a fully developed jaw-skeleton (upper and under jaw), whereas it is completely wanting in the Monorrhina.

Apart also from the peculiar nasal formation, and the absence of jaws, the Single-nostriled animals are distinguished from those with double nostrils by many peculiarities. Thus they want the important sympathetic nervous system, and the spleen which the Amphirrhina possess. Of the swimming bladder, and the two pairs of legs—which all double-nostriled animals have, at least in their embryonic conditions—not a trace exists in the Single-nostriled animals, which is the case also in the Skull-less animals. Hence, we are surely justified in completely separating the Monorrhina, as we have separated the Skull-less animals, from the Fishes, with which they have hitherto been erroneously classed.

We owe our first accurate knowledge of the Monorrhina, or Cyclostoma, to the great zoologist, Johannes Müller of Berlin; his classical work on the “Comparative Anatomy of the Myxinoida” forms the foundation of our modern views on the structure of the Vertebrate animals. He distinguished two distinct groups among the Cyclostoma, which we shall consider as sub-classes.

The first sub-class consists of the Hags (Hyperotreta, or Myxinoida). They live in the sea as parasites upon other fish, into whose skin they penetrate (Myxine, Bdellostoma). Their organ of hearing has only one annular canal, and their single nasal tube penetrates the palate. The second sub-class, that of Lampreys, or Prides (Hyperoartia, or Petromyzontia) is more highly developed. It includes the well-known Lamperns, or Nine-eyes, of our rivers (Petromyzon fluviatilis), with which most persons are acquainted. They are represented in the sea by the frequently larger marine or genuine Lampreys (Petromyzon marinus). The nasal tube of these single-nostriled animals does not penetrate the palate, and in the auricular organ there are two annular canals.

PEDIGREE OF VERTEBRATES

All existing Vertebrate animals, with the exception of the Monorrhina and Amphioxus just mentioned, belong to the group which we designate as Double-nostriled animals (Amphirrhina). All these animals possess (in spite of the great variety in the rest of their forms) a nose consisting of two lateral halves, a jaw-skeleton, a sympathetic nervous system, three annular canals connected with the auricular sac, and a spleen. Further, all Double-nostriled animals possess a bladder-shaped expansion of the gullet, which, in Fish, has developed into the swimming bladder, but in all other Double-nostriled animals into lungs. Finally, in all Double-nostriled animals there exist in the youngest stage of growth the beginnings of two pairs of extremities, or limbs, a pair of fore legs, or breast fins, and a pair of hinder legs, or ventral fins. One of these pairs of legs sometimes degenerates (as in the case of eels, whales, etc.), or both pairs of legs (as in Cæciliæ and serpents) either degenerate or entirely disappear; but even in these cases there exists some trace of their original beginning in an early embryonic period, or the useless remains of them may be found in the form of rudimentary organs. (Compare above, vol. i. p. 13.)

From all these important indications we may conclude with full assurance that all double-nostriled animals are derived from a single common primary form, which developed either directly or indirectly during the primordial period out of the Monorrhina. This primary form must have possessed the organs above mentioned, and also the beginning of a swimming bladder and of two pairs of legs or fins. It is evident, that of all still living double-nostriled animals, the lowest forms of sharks are most closely allied to this long since extinct, unknown, and hypothetical primary form, which we may call the Primary Double-nostriled animals (Proselachii). We may therefore look upon the group of primæval fish, or Selachii, to which the Proselachii probably belonged, as a primary group, not only of the Fish class, but of the whole main-class of double-nostriled animals.

The class of Fish (Pisces) with which we accordingly begin the series of Double-nostriled animals, is distinguished from the other six classes of the series by the swimming bladder never developing into lungs, but acting only as a hydrostatic apparatus. Agreeing with this, we find that in fish the nose is formed by two blind holes in front of the mouth, which never pierce the palate so as to open into the cavity of the mouth. In the other six classes of double-nostriled animals, both nostrils are changed into air passages which pierce the palate, and thus conduct air to the lungs. Genuine fish (after the exclusion of the Dipneusta) are accordingly the only double-nostriled animals which exclusively breathe through gills and never through lungs. In accordance with this, they all live in water, and both pairs of their legs have retained the original form of paddling fins.

PEDIGREE OF THE NON-AMNIONATE CRANIOTA.

Genuine fish are divided into three distinct sub-classes, namely, Primæval fish, Ganoid fish, and Osseous fish. The oldest of these, where the original form has been most faithfully preserved, is that of the Primæval fish (Selachii). Of these there still exist Sharks (Squali), and Rays (Rajæ), which are classed together as cross-mouthed fishes (Plagiostomi), and the strange and grotesquely formed Sea-cats, or Chimæracei (Holocephali). These primary fish of the present day, which are met with in all seas, are only poor remains of the prevailing animal groups, rich in forms, which the Selachii formed in the earlier periods of the earth’s history, and especially during the palæolithic period. Unfortunately all Primæval fish possess a cartilaginous, never a completely osseous skeleton, which is but little, if at all, capable of being petrified. The only hard parts of the body which could be preserved in a fossil state, are the teeth and fin-spikes. These are found in the older formations in such quantities, varieties, and sizes, that we may, with certainty, infer a very considerable development of Primæval fish in those remote ages. They are even found in the Silurian strata, which contain but few remains of other Vertebrata, such as Enamelled fish (and these only in the most recent part, that is, in the upper Silurian). By far the most important and interesting of the three orders of Primæval fish are Sharks; of all still living double-nostriled animals, they are probably most closely allied to the original primary form of the whole group, namely, to the Proselachii. Out of these Proselachii, which probably differed but little from genuine Sharks, Enamelled fish, and the present Primæval fish, in all probability, developed in one direction, and the Dipneusta, Sea-dragons, and Amphibia in another.

The Ganoid, or Enamelled fish (Ganoides), in regard to their anatomy stand midway between the Primæval and the Osseous fish. In many characteristics they agree with the former, and in many others with the latter. Hence, we infer that genealogically they form the transition from Primæval to Osseous fish. The Ganoids are for the most part extinct, and more nearly so than the Primæval fish, whereas they were developed in great force during the entire palæolithic and mesolithic periods. Ganoid fish are divided into three legions according to the form of their external covering, namely, Mailed, Angular-scaled, and Round-scaled. The Mailed Ganoid fish (Tabuliferi) are the oldest, and are directly allied to the Selachii, out of which they originated. Fossil remains of them, though rare, are found even in the upper Silurian (Pteraspis ludensis of the Ludlow strata). Gigantic species of them, coated with strong bony plates, are found in the Devonian system. But of this legion there now lives only the small order of Sturgeons (Sturiones), including the Spade-sturgeons (Spatularidæ), and those Sturgeons (Accipenseridæ) to which belong, among others, the Huso, which yields isinglass, or sturgeon’s sound, and the Caviar-sturgeon, whose eggs we eat in the shape of caviar, etc. Out of the mailed Ganoid fish, the angular and round-scaled ones probably developed as two diverging branches. The Angular-scaled Ganoid fish (Rhombiferi)—which can be distinguished at first sight from all other fish by their square or rhombic scales—are at present represented only by a few survivors, namely, the Finny Pike (Polypterus) in African rivers (especially the Nile), and by the Bony Pike (Lepidosteus) in American rivers. Yet during the palæolithic and the first half of the mesolithic epochs this legion formed the most numerous group of fishes. The third legion, that of Round-scaled Ganoid fish (Cycliferi), was no less rich in forms, and lived principally during the Devonian and Coal periods. This legion, of which the Bald Pike (Amia), in North American rivers, is the only survivor, was especially important, inasmuch as the third sub-class of fish, namely, Osseous fish, developed out of it.

Osseous fish (Teleostei) include the greater portion of the fish of the present day. Among these are by far the greater portion of marine fish, and all of our fresh-water fish except the Ganoid fish just mentioned. This class is distinctly proved by numerous fossils to have arisen about the middle of the Mesolithic epoch out of Ganoid fish, and moreover out of the Round-scaled, or Cycliferi. The Thrissopidæ of the Oolitic period (Thrissops, Leptolepis, Tharsis), which are most closely allied to the herrings of the present day, are probably the oldest of all Osseous fish, and have directly arisen out of Round-scaled Ganoid fish, closely allied to the existing Amia. In the older Osseous fish of the legion called Physostomi, as also in the Ganoides, the swimming bladder throughout life was connected with the throat by a permanent air passage (a kind of windpipe). This is still the case with all the fish belonging to this legion, namely, with herrings, salmon, carp, shad, eels, etc. However, during the chalk period this air passage, in some of the Physostomi, became constricted and closed, and the swimming bladder was thus completely separated from the throat. Hence there arose a second legion of Osseous fish, the Physoclisti, which did not attain their actual development until the tertiary epoch, and soon far surpassed the Physostomi in variety. To this legion belong most of the sea fish of the present day, especially the large families of the Turbot, Tunny, Wrasse, Crowfish, etc., further, the Lock-jaws (Plectognathi), Trunk fish, and Globe-fish and the Bushy-gills (Lophobranchi), viz., Pipe-fish, and Sea-horses. There are, however, only very few Physoclisti among our river fish, for instance, Perch and Sticklebacks; the majority of river fish are Physostomi.

Midway between genuine Fish and Amphibia is the remarkable class of Mud-fish, or Scaly Sirens (Dipneusta, or Protopteri). There now exist only a few representatives of this class, namely, the American Mud-fish (Lepidosiren paradoxa) in the region of the river Amazon, and the African Mud-fish (Protopterus annectens) in different parts of Africa. A third large Salamander-fish (Ceratodus Fosteri) has lately been discovered in Australia. During the dry season, that is in summer, these strange animals bury themselves in a nest of leaves in the dry mud, and then breathe air through lungs like the Amphibia. But during the wet season, in winter, they live in rivers and bogs, and breathe water through gills like fish. Externally, they resemble fish of the eel kind, and are like them covered with scales; in many other characteristics also—in their internal structure, their skeleton, extremities, etc.—they resemble Fish more than Amphibia. But in certain features they resemble the Amphibia, especially in the formation of their lungs, nose, and heart. There is consequently an endless dispute among zoologists, as to whether the Mud-fish are genuine Fish or Amphibia. Distinguished zoologists have expressed themselves in favour of both opinions. But in fact, owing to the complete blending of characteristics which they present, they belong neither to the one nor to the other class, and are probably most correctly dealt with as a special class of Vertebrata, forming the transition between Fishes and Amphibians. The still living Dipneusta are probably the last surviving remains of a group which was formerly rich in forms, but has left no fossil traces on account of the want of a solid skeleton. In this respect, these animals are exactly like the Monorrhina and the Leptocardia. However, teeth are found in the Trias which resemble those of the living Ceratodus. Possibly the extinct Dipneusta of the palæolithic period, which developed in the Devonian epoch out of primæval fish, must be looked upon as the primary forms of the Amphibia, and thus also of all higher Vertebrata. At all events the unknown forms of transition—from Primæval fish to Amphibia—were probably very like the Dipneusta.

A very peculiar class of Vertebrate animals, long since extinct, and which appears to have lived only during the secondary epoch, is formed by the remarkable Sea-dragons (Halisauria, or Enaliosauria, also called Nexipoda, or Swimming-footed animals). These formidable animals of prey inhabited the mesolithic oceans in great numbers, and were of most peculiar forms, sometimes from thirty to forty feet in length. From many and excellently preserved fossil remains and impressions, both of the entire body of Sea-dragons as well as of single parts, we have become very accurately acquainted with the structure of their bodies. They are usually classed among Reptiles, whilst some anatomists have placed them in a much lower rank, as directly allied to Fish. Gegenbaur’s recently published investigations, which place the structure of their limbs in a true light, have led to the surprising conclusion that the Sea-dragons form quite an isolated group, differing widely both from Reptiles and Amphibia as well as from Fish. The skeleton of their four legs, which are transformed into short, broad, paddling fins (like those of fish and whales) furnishes us with a clear proof that the Halisauria branched off from the main-stock of Vertebrata at an earlier period than the Amphibia. For Amphibia, as well as the three higher classes of Vertebrata, are all derived from a common primary form, which possessed only five toes or fingers on each leg. But the Sea-dragons have (either distinctly developed or in a rudimentary condition as parts of the skeleton of the foot) more than five fingers, as have also the Selachians or Primæval fish. On the other hand, they breathed air through lungs, like the Dipneusta, although they always swam about in the sea. They, therefore, perhaps, in conjunction with the Dipneusta, branched off from the Selachii, but did not develop into higher Vertebrata; they form an extinct lateral line of the pedigree, which has died out.

The more accurately known Sea-dragons are classed into three orders, distinct enough one from the other, namely, Primæval Dragons, Fish Dragons, and Serpent Dragons. The Primæval Dragons (Simosauria) are the oldest Sea-dragons, and lived only during the Trias period. The skeletons of many different genera of them are met with in the German limestone known as “Muschel-kalk.” They seem upon the whole to have been very like the Plesiosauria, and are, consequently, sometimes united with them into one order as Sauropterygia. The Serpent Dragons (Plesiosauria) lived in the oolitic and chalk periods together with the Ichthyosauria. They were characterised by an uncommonly long thin neck, which was frequently longer than the whole body, and carried a small head with a short snout. When their arched neck was raised they must have looked very like a swan; but in place of wings and legs they had two pairs of short, flat, oval-paddling fins.

The body of the Fish Dragons (Ichthyosauria) was of an entirely different form; these animals may be opposed to the two preceding orders under the name of Fish-finners (Ichthyopterygia). They possessed a very long extended body, like a fish, and a heavy head with an elongated, flat snout, but a very short neck. Externally, they were probably very like porpoises. Their tail was very long, whereas it was very short in the members of the preceding orders. Also both pairs of paddling fins are broader and show very different structure from that seen in the other two orders. Probably the Fish Dragons and Serpent Dragons developed as two diverging branches out of the Primæval Dragons; but it is also possible that the Plesiosauria alone originated out of the Simosauria, and that the Ichthyosauria were lower off-shoots from the common stock. At all events, they must all be directly, or indirectly derived from the Selachii, or Primæval fish.

The succeeding classes of Vertebrata, the Amphibia and the Amniota (Reptiles, Birds, and Mammals), owing to the characteristic structure which they all exhibit of five toes to each foot, may all be derived from a common primary form, which originated from the Selachii, and which possessed five toes on each of its four limbs. When we find a less number of toes than five, we can show that the missing ones must have been lost in the course of time by adaptation. The oldest known Vertebrata with five toes are the Batrachias (Amphibia). We divide this class into two sub-classes, namely, mailed Batrachians and naked Batrachians, the first of which is distinguished by the body being covered with bony plates or scales.

The first and elder sub-class of Amphibia consists of the Mailed Batrachians (Phractamphibia), the oldest land living Vertebrata of which fossil remains exist. Well-preserved fossil remains of them occur in the coal, especially of those with Enamelled heads (Ganocephala), which are most closely allied to fish, namely, the Archegosaurus of Saarbruck, and the Dendrerpeton of North America. There then follow at a later period the gigantic Labyrinth-toothed animals (Labyrinthodonta), which are represented in the Permian system by Zygosaurus, but at a later period, more especially in the Trias, by Mastodonsaurus, Trematosaurus, Capitosaurus, etc. The shape of these formidable rapacious animals seems to have been between that of crocodiles, salamanders, and frogs, but in their internal structure they were more closely related to the two latter, while by their solid coat of mail, formed of strong bony plates, they resembled the first animals. These gigantic mailed Batrachians seem to have become extinct towards the end of the Triassic period. No fossil remains of mailed Batrachia are known during the whole of the subsequent periods. However, the still living blind Snakes, or Cæciliæ (Peromela)—small-scaled Phractamphibia of the form and the same mode of life as the earth-worm—prove that this sub-class continued to exist, and never became completely extinct.

The second sub-class of Amphibia, the naked Batrachia (Lissamphibia), probably originated even during the primary and secondary epochs, although fossil remains of them are first found in the tertiary epoch. They are distinguished from mailed Batrachia by possessing a naked smooth, and slimy skin, entirely without scales or coat of mail. They probably developed either out of a branch of the Phractamphibia, or out of the same common root with them. The ontogeny of the three still living orders of naked Batrachia—-the gilled Batrachia, tailed Batrachia, and frog Batrachia—distinctly repeats the historical course of development of the whole sub-class. The oldest forms are the gilled Batrachia (Sozobranchia), which retain throughout life the original primary form of naked Batrachia, and possess a long tail, together with water-breathing gills. They are most closely allied to the Dipneusta, from which, however, they differ externally by the absence of the coat of scales. Most gilled Batrachia live in North America: among others of the class is the Axolotl, or Siredon, already mentioned. (Compare above, vol. i. p. 241.) In Europe the order is only represented by one form, the celebrated “Olm” (Proteus anguinus), which inhabits the grotto of Adelsberg and other caves in Carinthia, and which, from living in the dark, has acquired rudimentary eyes which can no longer see (vol. i. p. 13). The order of Tailed Batrachia (Sozura) have developed out of the gilled Batrachia by the loss of external gills; the order includes our black and yellow spotted land Salamander (Salamandra maculata), and our nimble aquatic Salamanders (Tritons). Many of them—for instance, the celebrated giant Salamanders in Japan (Cryptobranchus Japonicus)—still retain the gill-slits, although the gills themselves have disappeared. All of them, however, retain the tail throughout life. Tritons occasionally—when forced to remain in water always—retain their gills, and thus remain at the same stage of development as gilled Batrachia. (Compare above, vol. i. p. 241.) The third order, the tailless or frog-like Batrachia (Anura), during their metamorphosis, not only lose their gills, with which in early life (as so-called tadpoles) they breathe in water, but also the tail with which they swim about. During their ontogeny, therefore, they pass through the course of development of the whole sub-class, they being at first Gilled Batrachia, then Tailed Batrachia, and finally Frog-like Batrachia. The inference from this is evidently, that Frog-like Batrachia developed at a later period out of Tailed Batrachia, as the latter had developed out of Gilled Batrachia which originally existed alone.

In passing from the Amphibia to the next class of Vertebrata, namely, Reptiles, we observe a very considerable advance in the progress of organization. All the double-nostriled animals (Amphirrhina) up to this time considered, and more especially the two larger classes of Fish and Batrachia, agree in a number of important characteristics, which essentially distinguish them from the three remaining classes of Vertebrata—Reptiles, Birds, and Mammals. During the embryological development of these latter, a peculiarly delicate covering, the first fœtal membrane, or amnion, which commences at the navel, is formed round the embryo; this membrane is filled with the amnion-water, and encloses the embryo or germ in the form of a bladder. On account of this very important and characteristic formation, we may comprise the three most highly developed classes of Vertebrata under the term Amnion-animals (Amniota). The four classes of double-nostriled animals which we have just considered, in which the amnion is wanting (as is the case in all lower Vertebrate animals, single-nostriled and skull-less animals), may on the other hand be opposed to the others as amnion-less animals (Anamnia).