date as the old red sandstone; a view which has since been verified beyond the possibility of question by the discovery, in Russia, of two series-one resembling our old red sandstone conglomerate, and the other our devonian gritty and slaty beds-both evidently belonging to one period-both fossiliferous, and each containing fossils, by which the identity of these rocks with the beds of our own country, both in Devonshire and Scotland, is placed beyond doubt.

In Belgium, and in other parts of Europe, the passage from the silurian rocks to those of newer date is perfectly unbroken; and even in Wales it is not easy, nor is it always possible, to distinguish so accurately between the two as to state where the lower series terminates, and which is to be considered the lowest. member of the overlying group.

The line of demarcation between these strata being thus slightly marked in some places, the naturalist is enabled to trace the gradual transition of the animals characteristic of the one into those of the other series; but in the corals, the encrinites, the trilobites, and even in the shells, this is often difficult, although there is on the whole a considerable difference in the general appearance of a group of the fossil remains of Invertebrata taken from the upper and lower series of strata. An example of this difference is seen in the annexed figure of a remarkable bivalve shell (16, 17), not uncommon in some of the rocks of the period we are now considering, but altogether confined to that period. The peculiar form and magnitude of the hinge teeth (fig. 17), and the beauty of the shell, render it worthy of notice.

But, although the differences of this kind are not

such as need detain us here, there is another change of a far more striking character. I mean the intro

duction and speedy increase of the great natural class of fishes, whose remains are comparatively rare even in the uppermost silurian rocks, but which become extremely abundant in those beds immediately superjacent. The description of these fishes will be the chief subject of the remaining part of the present chapter. Fig. 18

Fig. 19

All the fishes at present known to exist in the waters which cover our globe may, with comparatively few exceptions, be grouped naturally and properly in two divisions, the one containing those species whose scales

are jagged on the outer edge like the scales of the perch (fig. 18), and the other those whose scales are smooth and simple at the margin, like the scales of a herring or salmon (fig. 19). To the peculiarities thus alluded to, might be added many others derived from the minute anatomy of the fishes. Fig. 20

Fig. 21

Ganoid Scale-BONY PIKE. (Lepidosteus.)

Placoid Scale-EXTINCT RAY. (Spinacorhinus polyspondylus.)

The exceptions to this arrangement are comprised within a few natural families, of which the sturgeon, the Siluride or cat-fish, the bony pike of the North American lakes (fig. 20), and some others, form one group; and the saw-fish, the rays (fig. 21), and the sharks another. These two groups were naturally looked on as of comparatively small importance, so long as only the existing species of fishes were known, for they contain, with the exception of the sharks and rays, but very few species, and these are neither abundant nor widely spread.

When, however, it was discovered by M. Agassiz, on looking carefully at the numerous species of fish whose fragments are found fossil in the older rocks, that all these, without a single exception, belonged to one or the other of the two groups alluded to, it became necessary to reconsider the subject of the classification of fishes, and learn, if possible, the nature and

extent of the difference that existed between those of the earlier seas and the present time.

The result of this, and the conclusions arrived at by a careful and minute study of the natural history and anatomy of fishes, has been lately laid before the public by M. Agassiz, a naturalist whose great acuteness of observation and knowledge of the structure of fishes, have long been well known to the scientific world, and were appreciated by Cuvier, who left in his hands the papers he had himself accumulated on the subject of Ichthyology.

According to M. Agassiz, fishes may be collected into four natural orders, two of which have been already mentioned as including most of the recent fishes, while each of the other two groups, so rarely met with at present, contains species representatives of another order, equally important and well marked, and in former times represented almost to the exclusion of any species of the two orders now so abundant. The first of the two groups, that of which the sturgeon and the bony pike are characteristic, is called GANOID; while the other, containing the sharks and rays, is known as the PLACOID Order *. Of these, the Placoid seems to have been the first introduced, but the Ganoid was that which attained its greatest development in the ancient seas.+

Ganoid, from the Greek yavos (ganos), splendour; the scales of these fishes being generally coated with polished enamel, and often exhibiting a very brilliant lustre. Placoid, from λağ (plax), a plate or slab; be

cause the skin of the animals of this

order is irregularly covered with

plates, studded often with enamel. (See figures 20, 21.)

The remaining two groups are called respectively Ctenoid (KTEVOS, ctenos, a comb) and Cycloid (KUкλoç, cyclos, a circle), from the shape and structure of the scale. (See figures 18, 19.)

The tribe of existing Placoid fishes most resembling those whose remains are found fossil, is that of which the sharks are the well-known representative. These powerful and rapacious animals, which are at this day the tyrants of the deep, seem to have been, when first introduced, of small size, and were accompanied by some few species of the next or Ganoid order. Only nine species of these shark-like monsters have yet been determined with certainty from the silurian and devonian rocks; and of these, two only are from the former.

It is chiefly the Ganoid fishes whose remains are handed down to us in the old red sandstone and other rocks of that period. Sixty distinct species of these fish have been mentioned, and almost all of them are known from British specimens. Most of them are remarkable for exhibiting strange peculiarities of shape, approximating them in some intances to the structure of the lower order of animals, combined with some apparent affinities to the class of reptiles.

The most remarkable group of these fishes contains several genera, three of which will require special notice. They are the Cephalaspis (or bucklerheaded, fig. 22), the Pterichthys (or wing fish, fig. 23), and the Coccosteus* (fig. 24), so called from the berry-like tubercles with which its bony scales are covered.

The most extraordinary part of the first of these fishes, "the buckler-headed," is the head from which

* All these names are derived from the Greek. They are thus obtained-Cephalaspis, кɛþaλǹ) (cephale), a head; aσrıç (aspis), a shield or buckler. 2. Pterichthys, πTEрov (pteron), a wing; 1x0vs (ichthys), a fish. 3. Coccosteus, KOKKоg (coccos), a berry; oσTεov (osteon), a bone.