VIII., fig. 3.) It seems a piece of exquisite shell-work, such as we sometimes see on the walls of a grotto. There are two dorsals the posterior, immediately over the tail, and directly opposite the anal fin; the anterior, somewhat higher up than the ventrals; and all the fins are of great size. The anterior edge of each is formed of a strong spine, round as the handle of a halbert, and diminishing gradually and symmetrically to a sharp point. Though formed externally of solid bone, it seems to have been composed internally of cartilage, like the bones of some of the osseous fishes those of the halibut, for instance; and the place of the cartilage is generally occupied in the stone by carbonate of lime. The membrane which formed the body of the fin was covered, like that of the Cheiracanthus, with minute scales, of the same scallop-like pattern with the rest, but of not more than one sixth the size of those which cover the creature's sides and back. Imagine two lug-sails stiffly extended between the deck of a brigantine and her two masts, the latter raking as far aft as to form an angle of sixty degrees with the horizon, and some idea may be formed of the dorsals of this singular fish. They were lug-sails, formed not to be acted upon by the air, but to act upon the water. None of my specimens show the head; but, judging from analogies furnished by the other families of the group, I entertain little doubt that it will be found to be covered, not by bony plates, but by minute scales, diminishing, as they approach the snout, into mere points. In none of the specimens does any part of the internal skeleton survive.

My collection contains the remains of yet another fish of this group, which was unfurnished with a name only a few months ago, but which I first discovered about five years since. (See Plate VIII., fig. 2.) It is now designated the

Diplacanthus; and, though the smallest ichthyolite of the formation yet known, it is by no means the least curious. The length from head to tail, in some of my specimens, does not exceed three inches; the largest fall a little short of five. The scales, which are of such extreme minuteness that their peculiarities can be detected by only a powerful glass, re semble those of the Cheiracanthus; but the ridges are more waved, and seem, instead of running in nearly parallel lines, to converge towards the apex. There are two dorsals, the one rising immediately from the shoulder, a little below the nape, the other directly opposite the anal fin. The ventrals are placed near the middle of the belly. There is a curious mechanism of shoulder-bone involved with a lateral spine and with the pectorals. The creature, unlike the Cheiracanthus, seems to have been furnished with jaws of bone: there are fragments of bone upon the head, tubercled apparently on the outer surface; and minute cylinders of carbonate of lime running along all the larger bones, where we find them accidentally laid open, show that they were formed on internal bases of cartilage. But the best marked characteristic of the creature is furnished by the spines of its fins, which are of singular beauty. Each spine resembles a bundle of rods, or, rather, like a Gothic column, the sculptured semblance of a bundle of rods, which finely diminish towards a point, sharp and tapering as that of a rush. (See Plate VIII., fig 4.)* The rest of the fin presents the appearance of a mere scaly membrane, and no part of the internal skeleton appears. Perhaps this last circumstance, common to all the ichthyolites of the formation, if we except the families of the Coccosteus

* Agassiz reckons four species of Diplacanthus D. longispinus, D. striatulus, and D. striatus.

D. crassispinus,

and Pterichthys, may throw some light on the apparently membranous condition of fin peculiar to the families of this order. What appears in the fossil a mere scaly membrane attached to a single spine of bone, may have had in the living animal a cartilaginous framework, like the fins of the dogfish and thorn-back, that are amply furnished with rays of cartilage though, of course, all such rays must have disappeared in the stone, like the rest of the internal skeleton. Unquestionably, the caudal fin of the two last described fossils must have been strengthened by some such internal framework; for, as they differ from the other fins, in being unprovided with osseous spines, they would have formed, without an internal skeleton, mere pendulous attachments, altogether unfitted to serve the purposes of instruments of motion. There may be found in the bony spines of all this order direct proof that, had there been an internal skeleton of bone, it would have survived. The spines run deep into the body, as a ship's masts run deep into her hulk; and we can see them standing up among the scales to their termination, in such bold relief, that, from a sort of pictorial illusion, they seem as if fixed to the creature's sides, and foreshortened, instead of rising in profile from its back or belly. (See Plate VIII., fig. 1.) The observer will of course remember, that, in the living animal, the view of the spine must have terminated with the line of the profile, just as the view of a vessel's mast terminates with the deck, though the mast itself penetrates to the interior keel. Now, it must be deemed equally obvious, that, had the vertebral column been of bone, not of cartilage, instead of exhibiting no trace, even the faintest, of having ever existed, it would have stood out in as high relief as the internal buts or stocks of the spines. And such are the general characteristics of a few of the ichthyolites of this