It is true that it still exhibits great expansion in a vertical direction, and to a superficial observer, if examined without dissection, might seem to be constructed on the same principles; but, on examining the skeleton of one of these cartilaginous fishes, it will be found that the vertebral column is continued uninterruptedly into the upper half of the generally furcate tail, whilst the lower division of the caudal fin is entirely made up of supplementary rays, appended to the inferior aspect of the caudal vertebrae. Possessing this form of the tail, the transition is by no means abrupt from these highly-organized fishes to the Saurian reptiles, with which, as we shall afterwards see, they exhibit many remarkable affinities.

(1748). If in the highest Heterogangliata we found that, in addition to the tegumentary skeleton, or shelly covering, so extensively met with among the Mollusca, the first appearances of an internal osseous system became recognizable, we are not on that account to imagine that, as soon as bones become developed internally, the cuticular secretions hitherto denominated shell at once disappear, but, on the contrary, must be prepared to expect that, in some form or other, calcareous armour deposited by the skin should still be met with. In Pishes, the coexistence of an internal and of an external skeleton is undeniable; and having already described the former, which has been aptly enough called the endoskeleton, it remains for us in the next place to examine the latter or eoooskeleton, which, as we shall soon perceive, forms no unimportant part of the anatomy of the class under consideration.

(1749). The most usual form of the cuticular covering of Fishes is that of imbricated scales, with which the whole exterior of the body is compactly encased, as in a suit of armour. Such an investment is admirably adapted to their habits and economy. The dense and corneous texture of the scales, impermeable to water, defends their soft bodies from maceration, while, from their smooth, polished exterior and beautiful arrangement, they ensure the least possible resistance from the surrounding medium as the fish glides along.

(1750). Examined separately, each scale is found to be partially imbedded in a minute fold of the living and vascular cutis, to which its under surface is adherent. Every scale is, in fact, made up of superimposed laminae of horny matter secreted by the cutis, precisely in the same way as the shelly covering of a mollusk; and by maceration the different layers may readily be separated, - the smallest and most superficial being of course the first formed, while the largest and most recent are those nearest to the surface of the living skin: as far as relates to the mode of growth, therefore, there is the strictest analogy between the scale of a fish and shell. Various are the forms under which these scales present themselves to the ichthyologist: sometimes, as in the Eel, they are thinly scattered over the surface of a thick and slimy cutis; more generally they form a close and compact imbricated mail; in the Pipe-fishes (Syngnaihidae) the whole body is covered with a strong armour composed of broad and thick calcareous plates; and in the Coffin-fishes (Ostracionidae) the integument is converted into a strong box made up of polygonal pieces anchylosed together, so that the tail and fins alone remain moveable.

(1751). The Sturgeon is covered with broad shield-like plates. The skin of the Shark is densely studded with minute sharp spines of almost crystalline hardness; and in many Skates, as in the Thornback, similar cuticular appendages, but of more considerable dimensions, are distributed over the back and tail, forming very efficient defensive weapons.

(1752). But cutaneous spines, although while in a rudimentary condition they are obviously mere extraordinary developments of scales, may occasionally become of sufficient size and importance to make them convertible to various unexpected uses; and when thus exaggerated in their dimensions and appropriated to distinct offices, they assume so much of the character of true bone, that it is no longer easy to demonstrate their real nature, more especially as they then become in many cases really articulated by means of very perfect joints with different pieces of the endoskeleton properly so called.

(1753). Let us examine this important subject with a little attention, and we shall soon perceive how closely the endoskeleton and the exo-sJceleton may become connected, not to say interchangeable, with each other. There is no possibility of mistaking the spines and tubercles upon the back of a common Skate for anything but cuticular appendages, secreted in the same manner as scales from the surface of a vascular pulp; but in the Fireflare (Trygon joastinaca), where, instead of the scattered hooks of the former species, we find a single sharp and serrated spine projecting like a bayonet from the upper surface of the root of the tail, the analogy between this formidable and bone-like organ and an epidermic structure becomes apparently more remote, and, did we not know that the fish possessing such a weapon had no ossified bones internally, we might be tempted to regard this appendage as a process derived from the endoskeleton.

(1754). The spines of the common Stickleback (Gasterosteus) are indubitable derivations from the cuticle; but they are fastened by moveable articulations to the sides of the body, and are raised or depressed by means of muscles inserted into their bases. Advancing one step further, we find in Silurus the first ray of the pectoral fins enormously developed and forming a strong serrated weapon of a very formidable description, which although both in shape and structure exactly comparable to the spine upon the tail of the Fireflare, is nevertheless connected by a most beautiful and perfect joint with the bones of the shoulder, so that it might easily be regarded as forming part of the endoskeleton, did not its peculiar structure indicate its real nature.