the top of the other, instead of meeting on the median line. But it is now generally admitted that these spines in reality belong to the dermoskeleton, although they are closely related in position and aspect to the real neural spines of the vertebræ.* But there are other projections from the vertebræ of certain fish, which constitute "diverging appendages" of the vertebra itself; such are certain spines, which are to be met with in the salmon and herring, the mackerel tribe and the dolphin, projecting from the ribs near their heads and sometimes diverging from the parapophyses and even from the neurapophyses. The homologue of these diverging appendages may be distinctly recognized in the ribs of birds; each of which, as is well known, is furnished with a process that passes over the ribs next below, and serves to give additional firmness and compactness to the bony case. We shall presently find that the recognition of them in the cranial vertebræ of fishes conducts us, according to Professor Owen, to the true homology of the bones of the extremities.

The fact that the spines which support the dorsal fin, constituting a second row of greater or less extent above the true neural spines, belong to the dermo-skeleton, is extremely well seen in the sturgeons, which have a well-developed osseous endo-skeleton coexisting with a covering of hard enamelled calcareous plates; and to this tribe the philosophical anatomist finds it requisite to make frequent reference, for the determination of the parts that really belong to each division. Here we find the rays upon which the dorsal fin is supported, clearly developed from the dermal plates, which along the middle line of the back shoot upwards and backwards a moderately long spine. From the base of these dermal spines, other spines usually shoot downwards into the intervals of the neural spines; these inverted interneural spines, which are double in the flat-fish, appear to be regarded by Professor Owen as formed by the “vegetative repetition" of the neural spines themselves; but we must take leave to question this determination, for it seems to us much more natural to consider them as portions of the dermo-skeleton passing inwards, -the manner in which they are intercalated among the true neural spines bearing a very strong resemblance to the reception of the fangs of the teeth into the alveolar processes of the jaw. It is not only in the back that we find these additional parts derived from the dermo-skeleton; for just as in the framework of the dorsal fin we find interneural spines and dermoneural spines, so in that of the anal fin we recognize interhæmal spines and dermohæmal spines. The framework of the caudal fin is composed of similar intercalary and dermal spines, superadded to the proper neural and hæmal spines of the proper caudal vertebræ, which have coalesced and been shortened by absorption, in the progress of embryonic development, to form the base of the terminal fin. There is usually an exact correspondence in intimate structure, between these dermal spines and the real bones which support them; this conformity has been urged by Professor Agassiz as an argument against the possibility of drawing a valid distinction in such cases between the parts belonging to the neuro-skeleton and those which appertain to the dermal envelope. But such a conformity exists likewise between the undoubted tegumentary scales of the lepidosteus

* In the herring, moreover, we find a sort of sternum formed by dermal plates, which are articulated to the ends of the ribs.

and its true internal bones; so that the presence of true bony structure must not be regarded as of itself proving that the part which exhibits it belongs to the neuro-skeleton; this being only determinable by its connexions and by the history of its development.

We shall now follow Professor Owen, as briefly as may be consistent with clearness, through his view of the craniology of the osseous fishes; in which group, owing to the permanent distinctness of a large proportion of the elements of which the skull is composed, the vertebral structure may be recognized more clearly than in any other.

"The bones of the skull are primarily divided in anthropotomy into those of the cranium and those of the face; but the proportions which these divisions bear to one another in man are reversed in fishes. According to this binary classification, the facial series in fishes includes an extensive system of bones-the hyoid-of which part only, viz., the styloid element,' is admitted into the skull by the anthropotomist, who describes it as a process of the 'temporal bone.' This very 'temporal,' moreover, is originally and essentially an assemblage of bones, which are always distinct in fishes and reptiles; and the squamous part, which enters so largely into the formation of the cranial cavity in man and most mammals, has no share in its formation in the lower vertebrata. The two classes of cranial and facial bones having been originally founded upon the exclusive study of the most peculiarly and extremely modified skull in the whole vertebrate series—that of man, their characters, as might be expected, are artificial, and applicable to the same bones in only a small proportion of the vertebrata; the unity of the plan pervading the organization of which, it is the business of the anatomist, properly so called, to demonstrate.

"The bones of the skull of fishes are primarily divisible into those of

A. Neuro-skeleton;

B. Splanchno-skeleton;
c. Dermo-skeleton.

"The bones of the neuro- or proper endo-skeleton are arranged here, as in the rest of the body, in a series of horizontally succeeding segments; each segment consisting of an upper (neural) and a lower (hæmal) arch, with a common centre, and with diverging appendages. As the bones respectively entering into the formation of these segments are the same in relative position, and nearly the same in number, as in the typical vertebræ of the trunk-the excess arising from subdivision of the peripheral elements-the same term ought to be extended to those cranial segments which has been usually restricted to their neural arches, in which the typical characters of the vertebra are least departed from." (pp. 86-8.)

Before proceeding with the enumeration of the parts entering into the composition of the cranial vertebræ, we must stop to notice that Professor Owen fixes the number of these vertebræ (which has been a subject of much discussion) at four; and that he justifies this by reference to the encephalon, which consists of four primary divisions, succeeding each other horizontally in a linear series. These are,-1, the medulla oblongata, with the superimposed cerebellum, recently termed by Vogt the epencephalon; 2, the third ventricle, with its upper (pineal) and lower (hypophysial) prolongations, and the superimposed optic lobes, or the mesencephalon; 3, the cerebrum proper, or prosencephalon; and 4, the olfactory ganglionic prolongation of the cerebrum, or rhinencephalon. Though subject to various degrees of anchylosis, the cranial vertebræ of fishes always agree in number with these primary divisions of the encephalon, and are named by Professor Owen in accordance with them.

Fig. IV. Disarticulated bones of the cranial vertebræ, and sense-capsules, in Cod-fish; the hæmal arches (H,H) and appendages in diagrammatic outline.

"Each cranial vertebra, or natural segment of the skull, is divided into a neural arch, with which the centrum and parapophyses are always more immediately connected; and a hemal arch, with its appendages.

"The neural arches are:

I. Epencephalic arch (Figs. IV and V, 1, 2, 3, 4);
II. Mesencephalic arch (Figs. IV and VI, 5, 6, 7, 8);
III. Prosencephalic arch (Figs. IV and VII, 9-12);
IV. Rhinencephalic arch (Figs. IV and VIII, 13-15).
"The hæmal arches are:

i. Scapular, or scapulo-coracoid (Fig. IV, in outline, 50-52);
ii. Hyoid or stylo-hyoid (Ditto, 25, 38-43);

iii. Mandibular or tympano-mandibular (Ditto, 25-32);

iv. Maxillary, or palato-maxillary (Ditto, 20-22).

"The appendages of the hamal arches are:

1. The pectoral (Fig. IV, in outline, 54-57);

2. The branchiostegal (Ditto, 44);

3. The opercular (Ditto, 34-37);

4. The pterygoid (Ditto, 23-24).

"The bones of the splanchno-skeleton constitute:

The ear-capsule or petrosal, and otolite (Fig. IV, 16, 16');
The eye-capsule or sclerotic, and pedicle (Ditto, 17);

The nose-capsule or ethmoid, and turbinal (Ditto, 18-19);
The branchial arches (Ditto, 45-49).

"The bones of the dermo-skeleton are:
Supra-temporals (Fig. IV, 71);
Supra-orbitals (Ditto, 72);
Sub-orbitals (Ditto, 73, 73');

Labials (Ditto, 74)."

(pp. 88-9.)

In most

We shall now examine each of these divisions separately. osseous fishes, the bones encompassing, or in vertebral relation with, the epencephalon, and thus forming the neural arch of the first or occipital vertebra, are six in number, as shown in Fig. V. These are called, in

ichthyology, the basi-occipital (1), the ex-occipitals (2, 2), the supra-occipital (3), and the par-occipitals (4, 4). No one can have any difficulty in recognizing in the basi-occipital the centrum or body of a vertebra, since it articulates posteriorly with the body of the atlas; and bearing in mind the relations of these bones to the nervous cord, it is obvious that the exoccipitals are the neurapophyses, the par-occipitals the parapophyses, and the supra-occipital (which very frequently possesses a prominent keel or ridge superiorly) the neural spine. Among the variations in form presented by these bones in different members of the class, are some which indicate in a very striking manner their conformity to the vertebral elements of the spinal column. On the other hand, various degrees of anchylosis are met with, which, by uniting two or more pieces, apparently reduce the number of parts of the vertebra; and in the polypterus the elements of the neural arch are all united into one piece, which corresponds to the occipital bone of man. In the early condition of the latter, we find seven ossific centres; one for the basilar portion or body of the vertebra, two for the condyles or neurapophyses, two below the crucial ridge for the parapophyses, and two above for the triangular portion which forms the summit of the bone and represents the neural spine. The supra-occipital, its homologue in fishes, is itself divided by a median suture in the lepidosteus, showing that it is in like manner formed from two centres of ossification.

32

п

The neural arch of the second cranial vertebra surrounding the mesencephalon, is composed of seven bones, of which the parietal elements in the human cranium are the largest, hence this vertebra is termed the parietal vertebra. Its body or centrum is formed by the basi-sphenoid (Fig. VI, 5); its neurapophyses are the bones termed the ali-sphenoids (6); its parapophyses are the mastoid bones (8); whilst its spine is formed by the parietals (7), which in the fish are comparatively small, in accordance with the small size of the encephalon, whose upper portion they are to protect. The names given to these bones sufficiently indicate the parts of the human cranium with which they are homologous. The basisphenoid is always united by continuous ossification with the pre-sphenoid (9), which is considered by Professor Owen as the centrum of the third or frontal vertebra; and the fact that the whole basi-pre-sphenoid is developed from a single centre of ossification is shown by him not to afford any sufficient objection to this homology, since many other cases exist in which bones that are elsewhere undoubtedly distinct, are in like manner repre

Fig. VI. Disarticulated neural arch of parietal vertebra, viewed from behind; from

the Cod.

sented by a single bone developed from one ossifying centre. The elements of the occipital and parietal vertebræ are so formed as to leave a large cavity, or otocrane, for the lodgment of the proper acoustic capsule; this cavity, which is analogous to the orbital cavity for the lodgment of the eye, is excavated in the ex-occipital, par-occipital, ali-sphenoid, and mastoid bones, with the addition in some instances of the parietal and supra-occipital. The acoustic capsule is either cartilaginous or osseous; when in the latter state it is known as the petrosal bone; and although it coalesces with the elements of the neuro-skeleton in higher animals to form the temporal bone, yet we think that Professor Owen is perfectly justified in regarding it as in itself a portion of the splanchno-skeleton, like the sclerotic capsule of the eye.

n

I

The neural arch of the third or frontal vertebra, which surrounds the prosencephalon of fishes, has for its centrum (as already stated) the pre-sphenoid (Fig. VII, 9); its neurapophyses are the orbito-sphenoids (10), whose essential functions are the protection of the sides of the prosencephalon, and the transmission of the optic nerve; the post-frontals (12) form its parapophyses; whilst the frontal (11), which is oftener divided by a median suture than a single bone, obviously constitntes its spine. We thus see the exceedingly complex nature of the human sphenoid bone; since, independently of the pteregoid processes, whose representatives have not yet come before us, the upper portion entering into the walls of the cranial cavity is represented

12

frontal vertebra, viewed from behind; from the Cod.

Fig. VII. Disarticulated neural arch of in the fish by the basi-pre-sphenoid, the ali-sphenoids, and the orbito-sphenoids, which enter into the composition of two distinct vertebræ. All these are anchylosed into one bone in the polypterus. The circle of bones which completes the axis of the skull anteriorly,

from behind; from the Cod.

and protects the olfactory ganglia, may be regarded as the neural arch of the fourth or nasal vertebra. Its body, however strange this may appear to the mere anthropotomist, formed by the vomer (Fig. VIII, 13); which, instead of being a narrow plate that occupies scarcely any space on either side of the median plane, is here a broad thick bone, whose aspect presents no difficulty in its recognition as the centrum of a vertebra. The neurapophyses

arch of the nasal vertebra, viewed are formed by the prefrontals (14), which defend and support the olfactory prolongations of the cerebral axis, and bound the orbits anteriorly; and the spine is formed by the nasal bone (15), which is usually