Form and Function: A Contribution to the History of Animal Morphology

CHAPTER X

THE EMBRYOLOGICAL CRITERION

Pander's work of 1817 was the forerunner of an embryological period in which men's hopes and interest centred round the study of development. "With bewilderment we saw ourselves transported to the strange soil of a new world," wrote Pander, and many shared his hopeful enthusiasm. K. E. von Baer's Entwickelungsgeschichte was by far the greatest product of this time, but it stands in a measure apart; we have in this chapter to consider the lesser men who were Baer's contemporaries, friends, followers or critics.

It was largely a German science, this new embryology, and its leaders were all personally acquainted. Pander, von Baer and Rathke were on friendly terms with one another; von Baer dedicated his master-work to Pander; Rathke dedicated the second volume of his Abhandlungen to von Baer. Interest in the new science was, however, not confined to Germany. In Italy, Rusconi commenced in 1817 his pioneer researches on the development of the Amphibia with a Descrizione anatomica degli organi della circolazione delle larve delle Salamandre aquatiche (Pavia), in which he traced the metamorphoses of the aortic arches. This was followed in 1822 by his Amours des Salamandres aquatiques (Milan), and in 1826 by his memoir Du développement de la grenouille (Milan). In this last paper he described how the dark upper hemisphere of the frog's egg grows down over the lower white hemisphere and leaves free only the yolk plug; he observed the segmentation cavity and the archenteron, but thought that the former became the alimentary canal; he observed and interpreted rightly the formation of the medullary folds. The circular blastopore in the frog in later years often went by the name of the anus of Rusconi.

In France Dutrochet[183] investigated the fœtal membranes in various vertebrate classes; Prévost and Dumas studied the very earliest stages of development in birds, mammals and amphibia (Ann. Sci. nat., ii., iii., 1824, xii., 1827).

A little later came Dugès' studies of the osteology and myology of developing amphibia (1834),[184] and Coste's careful researches into the early developmental history of mammals.[185]

It was in 1825 that Heinrich Rathke (1793-1860), published his famous discovery of gill-slits in the embryo of a mammal,[186] a discovery which aroused considerable interest, and greatly stimulated embryological research. He describes how in a young embryo of a pig he saw four slits in the region of the neck, going right through into the œsophagus. They were separated by partitions which he called Kiemenbogen (gill-arches), and immediately in front of the first gill-slit lay the developing lower jaw. He compared these gill-slits with those of a dogfish. We reproduce his drawing of the pig-embryo (Isis, Pl. IV., fig. 1).

Later in the same year Rathke discovered gill-slits in the chick,[187] in this case finding only three. He described growing out from in front of the first slit a structure which he compared to the operculum or gill-cover of a fish.

These discoveries were confirmed and extended for the chick[188] by the embryologist Huschke, a pupil of Oken. Like Rathke, he found only three indubitable gill-slits, but he noticed that the body-wall in front of the first gill-slit was really composed of two arches, which were on the whole similar to the gill-arches. The hinder of these two seemed to him to be a horn of the hyoid, the front one, which was bent at an angle, to be the rudiment of the upper and lower jaws (p. 401). Between these two arches he found an opening, just as between two gill-arches a gill-slit. This opening led into the mouth-cavity, and according to Huschke it became the external ear-passage. He discovered also three pairs of aortic arches in close relation with the gill-arches, so close indeed, that he did not hesitate to call them gill-arteries, and to recognise their resemblance with the aortic arches of fish. He traced, in part at least, the metamorphosis which these aortic arches undergo. This part of his discovery he developed in fuller detail in a paper of 1828,[189] in which he gave some excellent figures.

Shortly after Huschke's first paper, von Baer published his views and observations on this subject in a short memoir in Meckel's Archiv.[190] In this paper he confirmed Rathke's discovery, and described the slits and arches in the dog and the chick. Both Rathke and he found gill-slits in the human embryo about this time (p. 557). There were generally present, he found, four gill-slits, and, as Rathke had suggested, the first gill-arch became the lower jaw. Von Baer also confirmed Rathke's discovery of the operculum, assigning it, however, to the second gill-arch. He refused to accept Huschke's derivation of the auditory meatus from the first gill-slit. Von Baer saw what had escaped Rathke and Huschke, that there were, not three nor four, but as many as five aortic arches.

In his view of the metamorphosis of the aortic arches in the chick the first two pairs disappeared completely, the third pair gave rise to the arteries of the head and the fore-limbs, the right side of the fourth arch became the aorta, the left half of the fourth and the right half of the fifth arch became the pulmonary arteries, while the left half of the fifth arch disappeared. This schema, which for the last three arches was the same as Huschke's, von Baer upheld for the chick even in the second volume of his Entwickelungsgeschichte (p. 116); he rectified it, however, for mammals in the same volume (p. 212), deriving both pulmonary arteries from the fifth arch, and the aorta from the fourth left. He fully recognised the great analogy of the embryonic arrangement of gill-arches and gill-arteries in Tetrapoda with their arrangement in fish (i., pp. 53, 73).

Huschke, in a paper of 1832,[191] chiefly devoted to the development of the eye, figured and described the developing upper and lower jaws, and maintained against von Baer that the first slit turns into the auditory meatus and the Eustachian tube.

These were the first papers of the embryological period. Before going on to discuss the principles which guided embryological research during the next ten or twenty years it is convenient to note what were the main lines of work characterising the period.

The typical figure of the period is Rathke, who produced a great deal of first-class embryological work. He was, even more than von Baer, a comparative embryologist, and there were few groups of animals that he did not study. His first large publication, the Beiträge zur Geschichte der Thierwelt (i.-iv., Halle, 1820-27), contained much anatomical work in addition to the purely embryological; he commenced here his series of papers on the development of the genital and urinary organs, continued in the Abhandlungen zur Bildungsund Entwickelungs-Geschichte des Menschen und der Thiere (i., ii., Leipzig, 1832-3). A fellow-worker in this line was Johannes Müller, whose Bildungsgeschichte der Genitalien (Düsseldorf) appeared in 1830.

In a memoir on the development of the crayfish which appeared in 1829,[192] Rathke found in an Invertebrate confirmation of the germ-layer theory propounded by Pander and von Baer. He was greatly struck by the inverted position of the embryo with respect to the yolk. In following out the development of the appendages he noticed how much alike were jaws and legs in their earliest stage, and how this supported Savigny's contention that the limbs of Arthropods belonged to one single type of structure. In his paper (1832) on the development of the fresh-water Isopod, Asellus,[193] Rathke returns to this point. Commenting on the original similarity in development of antennæ, jaws and legs, he writes, "Whatever the doubts one may have reserved as to the intimate relation existing between the jaws and feet of articulate animals after the researches of Savigny on this subject and mine on developing crayfish, they must all fall to the ground when one examines with care the development of the fresh-water Asellus" (p. 147 of French translation).

Further comparative work by Rathke is found in the two volumes of Abhandlungen and in a book, Zur Morphologie, Reisebemerkungen aus Taurien (1837), which contains embryological studies of many different types, including a study of the uniform plan of arthropod limbs. Later on Rathke devoted himself more to vertebrate embryology, producing among other works his classical papers on the development of the adder (1839), of the tortoise (1848), and of the crocodile (1866). He laid the foundations of all subsequent knowledge of the development of the blood-vascular system in a series of papers of various dates from 1838 to 1856. The diagrams in his paper on the aortic arches of reptiles (1856) were for long copied in every text-book.

Rathke was a foremost worker in another important line of embryological work, the study of the development of the skeleton and particularly of the skull. We shall discuss the history of the embryological study of the skull in some detail below; meantime, we note the two other important lines of research which characterise this period. One is the intensive study of the development of the human embryo, a study pursued by, among others, Pockels, Seiler, Breschet, Velpeau, Bischoff, Weber, Müller, and Wharton Jones.[194] The other important line—the early development of the Mammalia—was worked chiefly by Valentin,[195] Coste,[196] and, above all, by Bischoff, whose series of papers[197] was justly recognised as classical.

What interests us chiefly in the work of this embryological period is, of course, the relation of embryology to comparative anatomy and to pure morphology. The embryologists were not slow to see that their work threw much light upon questions of homology, and upon the problem of the unity of plan. Von Baer, we have seen, recognised this clearly in 1828; Rathke, in one of his most brilliant papers, the Anatomische-philosophische Untersuchungen über den Kiemenapparat und das Zungenbein (Riga and Dorpat, 1832), used the facts of development with great effect to show the homology of the gill-arches and hyoid throughout the vertebrate series; Johannes Müller made great use of embryology in his classical Vergleichende Anatomie der Myxinoiden (i. Theil, 1836), and, according to his pupil Reichert, firmly held the opinion that embryology was the final court of appeal in disputed points of comparative anatomy;[198] Reichert himself in a book of 1838 (Vergleichende Entwickelungsgeschichte des Kopfes der nackten Amphibien) discussed the two different methods of arriving at the "Type"—the anatomical method of comparing adults, and the embryological method of comparing embryogenies. Of the embryological method, he says, "Its aim is to distinguish during the formation of the organism the originally given, the essence of the type, and to classify and interpret what is added or altered in the further course of development. Embryologists watch the gradual building up of the organism from its foundations, and distinguish the fundament, the primordial form, the type, from the individual developments; they reach thus, following Nature in a certain measure, the essential structure of the organism, and demonstrate the laws that manifest themselves during embryogeny" (p. vi.). The embryologists, influenced in this greatly by von Baer, gradually felt their way to substituting for the "Archetype" of pure morphology what one may perhaps best call the embryological archetype. How the transition was made we can best see by following out the course of discovery in one particular line. We choose for this purpose the development of the skull, a subject which excited much interest at this time and upon which much quite fundamental work was done, particularly by Rathke and Reichert.

Following up his discovery of gill-slits and arches in the embryos of birds and mammals, Rathke in two papers of 1832[199] and 1833[200] worked out the detailed homologies of the gill-arches in the higher Vertebrates. He describes how in the embryo of the Blenny there is a short, thick arch between the first gill-slit and the mouth. A furrow appears down the middle of the arch dividing it incompletely into two. In the anterior halves a cartilaginous rod is developed which is connected with the skull; these rods become on either side the lower jaw and "quadrate." In the posterior halves two similar rods are formed which develop into the hyoid. The hyoid is at first connected with the skull, but afterwards frees itself and becomes slung to the "quadrate." From the hinder edge of the hyoid arch grows out the membranous operculum, in which develop later the opercular bones and branchiostegal rays. The upper jaw is an independent outgrowth of the serous layer.

The serial homology of the lower jaw and quadrate with the hyoid and with the true gill-arches was thus established in fish, and Rathke had little difficulty in demonstrating a similar origin of lower jaw and hyoid in the embryos of higher Vertebrates. He could even, as we have noted before, find the homologue of the operculum in a flap which grows out from the hyoid arch in the embryo of birds.

But Rathke could not altogether shake himself free from the transcendental notion of the homology of jaws with ribs, and this led him to draw a certain distinction between the first two and the remaining gill-arches, by which the homology of the former with the ribs was asserted and the homology of the latter denied. He thought he could show that the skeletal structures (lower jaw, "quadrate," and hyoid) of the first two arches were formed in the serous layer, just like true ribs, and like them in close connection with the vertebral skeletal axis. The other, "true," gill-arches appeared to him to be formed in the mucous layer, in the lining of the alimentary canal. They had no direct connection with the vertebral column, and seemed therefore to belong to what Carus[201] had called the visceral or splanchno-skeleton. He did not, however, let this distinction hinder him from asserting the substantial homology of all the gill-arches inter se, the first two included.

Rathke's discoveries relative to the development of the jaws, the hyoid and the operculum, enabled him to make short work of the homologies proposed for them by the transcendentalists. He could prove from embryology that the jaws were not the equivalent of limbs, as so many Okenians believed. He could reject, with a mere reference to the facts of development, Geoffroy's comparison of the hyoid and the branchiostegal rays in fish with sternum and ribs. He could show the emptiness of the attempts made by Carus, Treviranus, de Blainville and Geoffroy, to establish by anatomical comparison the homologies of the opercular bones, for he could show that these bones were peculiar to fish, and were scarcely indicated, and that only temporarily, in the development of other Vertebrates.[202] He did not, however, himself realise the relation of the ear-ossicles to the gill-arches, though he knew that Spix and Geoffroy were quite wrong in homologising them with the opercular bones in fish. He described, it is true, the development of the external meatus of the ear and the Eustachian tube from the slit which appears between the first and the second arch, as Huschke had done before him; he described, in confirmation of Meckel, the "Meckelian process" of the hammer running down inside the lower jaw; but the discovery of the true homologies of the ear-ossicles was not made until a year or two later by Reichert.

In his further study of the development of Blennius viviparus, Rathke observed some important facts about the development of the vertebral column and skull. He found that the vertebral centra were first formed as rings in the chorda-sheath, which give off neural and hæmal processes. The vertebra later ossifies from four centres. The chorda (notochord) is prolonged some little way into the head, and the base of the cranium is formed by the expanded sheath, which reaches forward in front of the end of the notochord. This cranial basis shows a division into three segments, in which Rathke was inclined to see an indication of three cranial vertebræ. (It turned out that this division into three segments did not really exist, and Rathke later acknowledged that he had made an error of observation.) The side walls of the skull grow out from this base and form a fibrous capsule for the brain. The cranial section of the chorda itself shows no sign of segmentation; but later on the cranial portion of the chorda-sheath ossifies, like the vertebræ, from several centres. The vomer, which, in the classical form of the vertebral theory of the skull, was the centrum of the fourth, or foremost, cranial vertebra, does not, according to Rathke, develop in continuity with the cranial basis and the chorda sheath, but develops separately in the facial region.

Von Baer, like Rathke at this time, was also to some extent a believer in the vertebral theory of the skull. In his second volume (1834, pub. 1837) he holds that the development of the skull, as the sum of the anterior vertebral arches, is in general the same as that of the other neural arches, and is modified only by the great bulk of the brain (Entwickelungsgeschichte, ii., p. 99). He had, however, some doubts as to the entire correctness of the vertebral theory, doubts suggested by a study of the developing skull. "In the course of the formation of the head in the higher animals, something additional is introduced which does not originally belong to the cranial vertebræ. At first we see the vertebration in the hinder region of the skull very clearly. Afterwards it becomes suddenly indistinct, as if some new formation overlaid it" (i., p. 194).

Even more clearly is his doubt expressed in his paper on Cyprinus. "Upon the formation of the vertebral column only this need be said, that at this stage the notochord is very clearly seen, and the upper and lower arches and spinous processes are visible right to the end of the tail, but the separation into vertebræ ceases abruptly where the back passes into the head. I do not hesitate to assert that bony fish, too, have at this stage an unsegmented cartilaginous cranium (as cartilaginous fish have all their life), the prominences and hollows of which constitute its only resemblance with the vertebral type" (1835, p. 19).

A convinced supporter of the vertebral theory was Johannes Müller, who, in his classical memoir on the Myxinoids,[203] discussed at some length the relation between the development of the vertebræ and the development of the skull. His memoir is principally devoted to comparative anatomy, but in treating of the skeleton he pays much attention to development. He describes the formation of the vertebræ in elasmobranch embryos; for the facts regarding other Vertebrates he relies largely on work by Rathke (Blennius, 1833) and Dugès (1834). He recognises as the basis of his comparisons the homology of the notochord in all vertebrate embryos with the persistent notochord which forms the chief part or the whole of the vertebral column in the Cyclostomes. The notochord possesses an inner and an outer sheath and the outer sheath is continuous with the basis cranii (p. 92). It is in the outer sheath that the vertebræ develop—from four separate pieces, in fish at least, plus an additional element which helps to form the centrum. The skull of Vertebrates consists, according to Müller, of three vertebræ, whose centra are the basioccipital, the basisphenoid and the presphenoid. Other bones besides those belonging to the vertebræ are present, but this formation out of three vertebræ gives the essential schema for the skull. Now the brain capsule, like the sheath of the spinal cord, is a development from the outer sheath of the notochord. If the skull consists of vertebræ we should expect the centra of the skull-vertebræ to develop in the outer sheath at the sides of the cranial section of the notochord as two separate halves, just as do the bodies of the vertebræ; we should expect further the cartilaginous side-walls of the cranium to develop in the membranous brain-sheath just as the neural arches develop in the membranous sheath of the spinal column. In Rathke's discovery (!) of a segmentation of the basis cranii into three parts, and of the isolated formation of the vomer, Müller sees a confirmation of his view that the skull is composed of three and not four vertebræ. But there is nothing in Rathke's observations to support the idea that the centra of the cranial vertebræ are formed from separate halves. Müller has to be content with a reference to the state of things in Ammocoetes (which, by the way, he did not know to be the young of Petromyzon). In the simple skull of Ammocoetes the base is formed chiefly by two cartilaginous bars lying more or less parallel with the longitudinal axis of the skull and embracing with their hinder ends the cranial portion of the notochord.

These bars, declares Müller, are clearly the still separate halves of the pars basilaris cranii, and represent the divided centra of the two hinder cranial vertebræ. To complete the parallel between the development of the skull and of the vertebræ, it would have been necessary to show that the side walls of the cranium developed in a similar manner from separate pieces. Müller could not prove this point from the available embryological data, and indeed the facts which he did use had to be twisted to suit his theory. A curious apparent confirmation of his idea that the centra of the cranial vertebræ are formed from separate halves was supplied in 1839 by Rathke's discovery of the trabeculæ in the embryonic skull of the adder.

The next big step in the study of the development of the skull was taken by a pupil of Müller, C. B. Reichert, who showed in his work very distinct traces of his master's influence. Reichert's first and most important contribution to the subject was his paper on the metamorphosis of the gill, or, as he called them, the visceral arches in Vertebrates,[204] particularly in the two higher classes. Reichert describes the similar origin in embryo of bird and mammal (pig) of three "visceral" arches. These arches stand in close relation to the three cranial vertebræ which Reichert, like Müller, distinguishes. He makes the retrograde step of admitting only three aortic arches, and he is not inclined to consider the three visceral arches as equivalent to the gill-arches of fish—in his opinion they have more analogy with ribs, though differing somewhat from ribs in their later modifications. The visceral arches are processes of the visceral plates (von Baer), which grow downwards and meet in the middle line, leaving between one another and the undivided body wall three visceral slits opening into the pharynx. The first visceral process is different in shape from the others, for it sends forward, parallel with the head and at right angles to its downward portion, an upper portion in which later the upper jaw is formed. The other two processes are straight. From the hinder edge of the second visceral arch there develops, as Rathke had seen, a fold which is comparable with the operculum of fish. The first slit develops externally into the ear-passage, internally into the Eustachian tube, and in the middle a partition forms the tympanic ring and tympanum. Inside each of the visceral processes on either side a cartilaginous rod develops. In the first process this rod shows three segments, of which the first lies inside that portion of the process which is parallel with the head. This upper segment forms the foundation for the bones of the upper jaw. The lowest segment of the cartilaginous rod becomes Meckel's cartilage, and on the outer side of this the bones of the lower jaw are formed. The middle segment becomes in mammals the incus (one of the ear-ossicles), and in birds the quadrate. Meckel's cartilage, which was discovered by Meckel[205] in fish, amphibians and birds, is a long strip of cartilage which runs from the ear-ossicle known as the hammer in mammals,[206] to the inside of the mandible. Reichert shows how this relation comes about.

Fig. 9.—Meckel's Cartilage and Ear-ossicles in Embryo of Pig. (After Reichert.)

The hammer, according to his observations on the embryo of the pig, is simply the proximal end of Meckel's cartilage, which later becomes separated off from the long distal portion (see Fig. 9). The third ear-ossicle of mammals, the stapes, comes not from the first arch but from the second. The cartilaginous rod of the second arch segments like the first into three pieces. Of these the uppermost disappears, the middle one, which lies close up to the labyrinth of the ear, becomes the stapes, and the lowest becomes the anterior horn of the hyoid. The stapes forms a close connection with the hammer and the incus. In birds, where there is a single ear-ossicle, the columella, the middle piece of arch I forms, as we have seen, the quadrate, by means of which the lower jaw is joined to the skull. The proximal end of Meckel's cartilage, which in mammals forms the hammer, here gives the articular surface between the lower jaw and the quadrate. The columella is formed from the middle piece of the three into which the cartilage of the second arch segments. It is, therefore, the homologue of the stapes in mammals. The third arch takes a varying share, together with the second, in the formation of the hyoid apparatus.

In this paper Reichert made a distinct advance on the previous workers in the same field—Rathke, Huschke, von Baer, Martin St Ange, Dugès. Huschke was indeed the first to suggest that both upper and lower jaws were formed in the first gill-arch. But both von Baer and Rathke[207] held that the upper jaw developed as a special process independent of the lower jaw rudiment, and the actual proof that the upper jaw is a derivative of the first visceral arch seems to have been first supplied by Reichert. His brilliant work on the development of the ear-ossicles founded what we may justly call the classical theory of their homologies. His views were attacked and in some points rectified, but the main homologies he established are even now accepted by many, perhaps the majority of morphologists.

In a paper of 1838 on the comparative embryology of the skull in Amphibia,[208] Reichert added to his results for mammals and birds an account of the fate of the first and second visceral arches in Anura and Urodela.

The first visceral arch, he found, gave in Amphibia practically the same structures as in the higher Vertebrates. Its skeleton segmented, as in mammals and birds, into three parts; the upper part gave rise to the palatine and pterygoid in Anura, but seemed to disappear in Urodeles, where the so-called palatine and pterygoid developed in the mucous membrane of the mouth; the middle part gave, as in birds, the quadrate, which formed a suspensorium for both arches; the lower part, as Meckel's cartilage, formed a foundation for the bones of the lower jaw. Of arch II., the lower part became the horn of the hyoid, the upper part had a varying fate. In some Anura it formed the ossicle of the ear (homologue of the columella of birds and the stapes of mammals), in others it disappeared. In reptiles the upper segment of the second arch formed, as in birds, the columella.

The account of the metamorphoses of the visceral arches in Amphibia forms only a small part of Reichert's memoir of 1838, the chief object of which was to discover the general "typus" of the vertebrate skull, and to follow out its modifications in the different classes. Von Baer had shown that the generalised type appeared most clearly in the early embryo; Reichert therefore sought the archetype of the skull in the developing embryo. He brought to his task the preconceived notion that the skull could be reduced to an assemblage of vertebræ, but he saw that comparative anatomy alone could not effect this reduction; he had recourse, therefore, to embryology, hoping to find in the simplified structure of the embryo clear indications of three primitive cranial vertebræ (p. 121, 1837).

In the head he distinguished two tubes, the upper formed by the dorsal plates, the lower by the ventral or visceral plates. Both of these tubes were derived from the serous or animal layer (cf. von Baer, supra, p. 118). The walls of the lower tube were formed by the visceral processes, within which later the skeleton of the visceral arches developed. The walls of the upper tube formed the bones and muscles of the cranium proper. The facial part of the head was formed by elements from both upper and lower tubes. The dorsal tube showed signs of a division into three cranial vertebræ (Urwirbeln, primitive vertebræ). In mammals and birds, as Reichert had shown in his 1837 paper, the three cranial vertebræ were indicated by transverse furrows on the ventral surface of the still membranous skull (see Fig. 10, p. 148).

Even in mammals and birds, however, the positions of the eye, the ear-labyrinth, and the three visceral arches were the safest guides to the delimitation of the cranial vertebræ (pp. 134-138, 1837). In Amphibia generally there were no definite lines of separation on the skull itself. "At this stage," he writes of the cartilaginous cranium of the frog, "we find no trace of a veritable division into vertebræ in the cartilaginous trough formed by the basis cranii and the side parts. On the contrary, it is quite continuous, as it is also in the higher Vertebrates during the process of chondrification" (p. 44, 1838). The vertebræ in the membranous or cartilaginous skull could be delimited in Amphibia by the help of the eye and the ear-labyrinth, which lie more or less between the first and second, and the second and third vertebræ, but, above all, by the vesicles of the brain.

As in the higher Vertebrates, the visceral arches are associated with the cranial vertebræ as their ventral extensions, being equivalent to the visceral plates which form the ventral portion of the "primitive vertebræ" or primitive segments of the trunk.

If the three cranial vertebræ are not very distinct in the early stages of development when the skull is still membranous or cartilaginous, they become clearly delimited when ossification sets in. Three rings of bone forming three more or less complete vertebræ are the final result of ossification. The composition of these rings is as follows:—

The other bones of the skull are not included in the vertebræ, and this is in large part due to the fact that the sense capsules are formed separately from the cranium (p. 29, 1838). The ear-labyrinth, it is true, fuses indissolubly with the cranium at a later period, but the bones which develop in its capsule are not for all that integral parts of the primitive cranial vertebræ. This point, it is interesting to note, had already been made by Oken in his Programm (1807). But many of the bones developed in relation to the sense organs can find their place in the generalised embryonic schema or archetype of the vertebrate skull, for they are of very constant occurrence during early development.

Having arrived at a generalised embryonic type for the vertebrate skull, of which the fundamental elements are the three cranial vertebræ and their arches, Reichert goes on to discuss the particular forms under which the skull appears in adult Vertebrates. He accepts in general von Baer's law that the characters of the large groups appear earlier in embryogeny than the characters of the lesser classificatory divisions. "When we observe new and not originally present rudiments in very early embryonic stages, as, for instance, that for the lacrymals, the probability is that they belong to the distinctive development of one of the larger vertebrate groups. From these are to be carefully distinguished such rudiments as arise later during ossification, mostly as ossa intercalaria, in order to give greater strength to the skull in view of the greater development of the brain, etc.; the latter give their individual character to the smaller vertebrate groups, and comprise such bones as the vomer, the Wormian bones, the lowermost turbinal, etc." (p. 63, 1838).

He did not accept the Meckel-Serres law of parallelism. He recognised the great similarity between the unsegmented cartilaginous cranium of Elasmobranchs, and the primordial cranium of the embryos of the higher Vertebrates, but he did not think that the cranium of Elasmobranchs was simply an undeveloped or embryonic stage of the skulls of the higher forms. Rather "do the Holocephala, Plagiostomata, and Cyclostomata appear to us to be lower developmental stages individually differentiated, so that the other fully differentiated Vertebrates cannot easily be referred directly to their type" (p. 152, 1838). The skull of these lower fishes is itself a specialised one; it is an individualised modification of a simple type of skull. And this holds good in general of the skulls of the lower Vertebrates—they are individualised exemplars of a simple general type, not merely unmodified embryonic stages of the greatly differentiated skulls of the higher Vertebrates (p. 250, 1838). Differentiation within the vertebrate phylum is therefore not uniserial, but takes place in several directions. Reichert describes two sorts of modifications of the typical skull—class modifications and functional modifications. The causes of the modifications which characterise classificatory groups are unknown; the second class of modifications occur in response to adaptational requirements.

Reichert's two papers are of considerable importance, and Müller's remark in his review[209] of them is on the whole justified. "These praiseworthy investigations supply from the realm of embryology new and welcome foundations for comparative anatomy" (p. clxxxvii.).

The development of the skull was, however, more thoroughly worked out by Rathke, and with less theoretical bias, in his classical paper on the adder.[210] This memoir of Rathke's is an exhaustive one and deals with the development of all the principal organ-systems, but particularly of the skeletal and vascular. He confirmed in its essentials Reichert's account of the metamorphoses of the first two visceral arches, describing how the rudiment of the skeleton of the first arch appears as a forked process of the cranial basis, the upper prong developing into the palatine and pterygoid, the lower forming Meckel's cartilage, while the quadrate develops from the angle of the fork. The actual bone of the upper jaw (maxillary) develops outside and separate from the palato-pterygoid bar. The cartilaginous rod supporting the second visceral arch divides into three pieces on each side, of which the lower two form the hyoid, the uppermost the columella. Like Reichert he held the visceral arches to be parts of the visceral plates, containing, however, elements from all three germ-layers—the serous, mucous, and vessel layers.

The first gill-slit, or, as Rathke here prefers to call it, pharyngeal slit, closes completely in snakes and in Urodeles. It forms the Eustachian tube in all other Tetrapoda. As regards the vertebræ, Rathke describes them as being formed in the sheath of the chorda from paired rudiments, each of which sends two branches upwards, and two branches downwards. The two inner pairs of processes coalesce round the chorda, and later form the centrum; the upper outer pair meet above the spinal column; the lower outer pair form ribs. The odontoid process of the axis vertebra is the centrum of the atlas (p. 120). The formation of vertebral rudiments begins close behind the ear-labyrinth, but in front of this the chorda-sheath gives origin to a flat membranous plate which afterwards becomes cartilaginous. This plate reaches forward below the third cerebral vesicle as far as the infundibulum. The notochord ends in this plate, which is the basis cranii, just at the level of the ear-labyrinth. In no Vertebrate does the notochord extend farther forward (p. 122). The basis cranii gives off three trabeculæ. The middle one is small and sticks up behind the infundibulum; it is absent in fish and Amphibia, and soon disappears during the development of the higher forms. The lateral trabeculæ are long bars which curve round the infundibulum and reach nearly to the front end of the head. Together they are lyre-shaped. The cranial basis and the trabeculæ are formed, like the vertebræ, in the sheath of the notochord, and the only differences between the two in the early stage of their development are that the formative mass for the cranial basis is much greater in amount than that for the vertebræ, and that the cranial basis by means of its processes, the trabeculæ, reaches well in front of the terminal portion of the notochord (p. 36). The capsule for the ear-labyrinth develops quite independently of the cranial basis and the notochord. It resembles on its first appearance, in form, position, composition, and connections, the ear-capsule of Cyclostomes, and so do the ear-capsules of all embryonic Vertebrates (p. 39). It manifests clearly the embryonic archetype, ... "there exists one single and original plan of formation, as we may suppose, upon which is built the labyrinth of Vertebrates in general" (p. 40). When ossification sets in, the ear-capsule forms three bones, of which two fuse with the supraoccipital and exoccipitals.

During the formation of the ear-capsule the cranial basis develops from a plate to a trench, for in its hinder section the side parts grow up to form the side walls of the brain, in exactly the same way as the processes of the vertebral rudiments grow up to enclose the spinal column (pp. 122, 192). The foundations of the skull are now complete, and ossification gradually sets in.