exists in the structure of these two parts, and, above all, in the disproportion of their dimensions and forms, which would offer a distinct resistance to the pressure to which the animals must have been exposed. The body most resembles that of certain fishes of the family lepidoides, but differs considerably in the position of the fins, and the singular scales with which it is covered. Its form is that of an elongated spindle, swelling out on the anterior parts and narrowing insensibly to the end of the tail, which is proportionately very slender, since its diameter does not exceed a quarter of the width of the body, near the nape of the neck.

"The scales of the body are of very peculiar form, and quite unlike those of any other genera. On each side there is only one range of plates, high and narrow, inserted transversely in the middle of the sides, whilst on the edge of the back and the belly there are series of little scales, disposed obliquely to the extremities of those on the sides. The analogies in the structure of the scales appears to confirm the position assigned to the Qonoidonts and Siluroids, in the order Ganoids following the accipensers or sturgeons. The exterior surface of these scales is ornamented with undulated furrows, disposed in the direction of their greatest diameter."

This description relates chiefly to a beautiful and nearly perfect specimen in the possession of Mr. Lyell; other species are remarkable for a still greater departure from the general form

of fishes.

Of Cephalaspis Lloydii, M. Agassiz remarks—

"Of the four species, this is the one the head of which, at first sight, least resembles that of a fish. In the specimens where the superior surface is preserved, it is difficult to divest oneself of the idea that these fossils are the shells of some molluscous animal; for so completely regular are the striæ, that they perfectly resemble the lines of growth of the testacea.

"In conclusion, M. Agassiz remarks that the structure of the head in this genus resembles singularly that of the shell of crustaceans, which possess also an exterior coloured layer, under which is found a layer of a granular structure; and then a layer of lamellar structure; and it was not without long deliberation that I decided on considering several specimens as the heads of cephalaspis, rather than the scales of some unknown crustacean."-Murchison, p. 589.

We have quoted this long and accurate description, partly because it has relation to one of the most extraordinary forms of animal life with which the researches of Geology have made us acquainted, and partly, also, as an instance of the beautiful clearness and perspicuity with which most of the new genera described in Mr. Murchison's work are defined and explained. Technical language must, of course, be employed in such minute description, but the general idea is given with great force and accuracy, and without any unnecessary use of scientific terms. Besides this remarkable genus of fishes, the fauna of the

Devonian system presents not very much of popular interest. We have, indeed, the goniatites of the mountain limestone displaced by a new form of the ammonite, simpler in the walls of separation than any other form, and having the siphuncle on the inner instead of the outer margin of the shell, thus increasing the links in this interesting, and now rare, family. In the beds of Devonshire there occur abundance of shells allied to the terebratula, a few univalves, and many corals. These most of them differ from the shells and corals both of the mountain limestone and Silurian; although out of a list of twenty-eight shells referred to by Mr. Murchison as met with in the new red sandstone, as many as seven are identified with Silurian species. The fossils of the whole system are, however, described and figured in a recent part of the transactions of the Geological Society, where the able memoir on this subject, by Messrs. Sedgwick and Murchison, will stand as a lasting record of their great and important discovery.

"the

Before concluding, we wish to make a few remarks on Report," to which we have already referred, and the object of which we have already, in a few words, declared.

The great advantage of a work like this of Mr. De la Beche's is to mark down at a given moment the actual Geological condition of a district-to state the opinion formed after a sufficiently careful and minute survey with regard to the stratification, the superposition of rocks, the nature of the various deposits, their extent, and the advantages derived or injuries suffered, by their being as they are found to be. Just as in geography, a map of an unknown district is the first step to improvement, and however imperfect, will soon lead to more accurate knowledge; or, as in any science, to have the limits of knowledge and ignorance, or error, definitely drawn, is a great step towards advance in that science-so, in Geology, the publication of a report, professing to embody the details of Geological knowledge of an important district, cannot but do good, if it only draws attention to the very faults of the work. Geologists are well aware what a great step was gained when the first volume of the Geology of England and Wales was published by Messrs. Conybeare and Philips not that the work was perfect or sufficient, but it enabled people to see where observation was wanted; where mistakes had been made; and where, above all, certain facts might be shown: this is of itself a great step; and if the effect of the Ordnance Geological survey were merely to collect and put on record the scattered knowledge hitherto accumulated by the efforts of private individuals, even then a great benefit might be hoped for and good be fairly anticipated to accrue to science.

But much more has been attempted, and, so far as the work has proceeded, much, very much more has been done. By the appointment of Mr. De la Beche and a surveying staff to assist him—by the personal examination of a good Geologist, aided by authority and power, we have obtained not merely an abstract of what was known already with regard to the interesting district of the South-West of England, but a useful addition to that knowledge in the economic details connected with it.

As bearing upon the immediate subject of this article, the Report, indeed, is not quite so satisfactory as could be wished. Mr. De la Beche may have laboured much, but others have been more successful; and we have not been able to make so much use of his work as if our object had been the elucidation of some other branches of Geological science. There is, however, much valuable detail even in this department, and some ingenious speculative enquiry concerning the possible nature and position of the land of those early periods, and the cause and circumstances of the accumulation of the carbonaceous series.

Not contented, however, with the evidence of fossils with regard to the age of this carbonaceous series, and unwilling, it would seem, to give up entirely all his preconceived opinions, our author, (see chap. v. containing a "general view of the carbonaceous rocks of the district,") brings forward many authorities and employs much argument to prove, if possible, that in the carboniferous deposits of the most ancient date, the fossils are identical, or nearly so, with those of the mountain limestone series. Whether the explanation of this apparent anomaly be not generally to be sought for in a reference to the Devonian system, we will not now stop to enquire. Certainly the fact that the carbonaceous rocks of Devonshire pass insensibly into the Devonian system, and the carboniferous formations of Wales in a somewhat similar manner into the old red sandstone, would indicate that here, at all events, the evidence of superposition is in favour of the two being identical, and, according to the conclusion that Professor Sedgwick has come to, that the "coal measures themselves are really represented in Devonshire: this, however, is a point to which Mr. De la Beche would seem to attach some importance, though under the circumstances there appears to us but little ground for discussion. The separation of two systems from the older rocks, and the changes of classification in Geology thus introduced, is, beyond all doubt, a subject of the greatest importance to science, and in its consideration are involved several points of interest. Among them there is, however, one

of which we have not as yet said much; but which is worthy of careful attention and consideration: we allude to the extensive range of the same species of fossils in the older rocks, and the assumed persistance and even re-appearance of certain species of shells and crustacea.

There has long been a feeling among Geologists, even the most eminent, that the evidence of fossils is not alone sufficient to classify in Geology, and the possibility has been suggested, that in those early periods when, to all appearance, the creation of animals upon our globe was yet recent, there might have been causes in operation which, through a long series of ages and during the accumulation of vast thicknesses of strata, would produce certain species at one time, and then, after an apparent cessation of their existence, might cause them to appear again.

That in all rocks of more recent periods such a phenomenon appears not to have happened-that, so far as we know from accurate and definite knowledge, there is no instance of the same species having ever been re-created, and that so exuberant is the creative power that, in our own seas, we frequently find analogous forms, but never identical species indigenous in two districts, either of land or water, widely separated by natural barriers. All this, it is true, does not amount to positive argument against such an hypothesis, or, at least, only to an argument of small weight. It is, however, an undeniable fact, that this idea has been gradually giving way before the advance of definite Geological and Zoological knowledge; and that whenever the actual work of field Geology has been fairly undertaken, the results have always hitherto been such as to strengthen our dependence on fossils, and weaken the influence that mere mineral character might have on our decision.

This is eminently the case in each of the two systems before us, and we find as marked a distinction in the general character of the organic remains of each of these two new groups as occurs in the case of any of the principal subdivisions of the secondary rocks. We are willing to acknowledge that even yet there is much doubt with regard to certain species; but it seems to us that the prima facie improbability is so great of any interference in a law which, being first grounded on observed facts, all subsequent experience has tended to verify, that we look with great caution upon evidence tending to invalidate it. When, moreover, this evidence is doubtful in character, and, as in many cases, not capable of the accuracy of investigation which is absolutely essential to decide positively on a specific form, we must be allowed to hesitate long before giving way to the belief that

it is only in more recent formations that fossils are absolutely characteristic.

The Devonian and Silurian systems, too, while they tend much to prove that it is only safe, even in these early formations, to look to organic remains for the subdivisions of classification, point out yet more clearly the total inadequacy of mineral structure to give any correct and definite notion on this important point. Let it not be forgotten, that so long as mineral structure was allowed to influence the Geologist in judging of the age of the rocks of Devonshire and Cornwall, not only was no approximation made towards any useful and true result, but the wildest hypotheses were indulged in, and the most unphilosophical views promulgated and received. The mere occurrence of roofing-slate in the district under consideration, was supposed to stamp it at once as contemporaneous with the slate rocks of North Wales, of Cumberland, and of various parts of Germany. The granite was declared primitive; and, in one remarkable case, in spite of direct evidence to the contrary occurring at every turn, the granite, the killas or clay-slate of Cornwall, and the numerous bands of fossiliferous limestone of that county, were all ranked together as the various parts of one great miscellaneous formation. It was the examination of fossils which first excited the attention of those distinguished Geologists whose labours we have been tracing. It was the careful comparison of fossils that first gave the hint upon which the superstructure of the Devonian system was erected; and it was the conclusions arrived at by the Geological Naturalist, and not by the Mineralogical Geologist, that ultimately decided the question, and gave a convincing and crowning proof of the reality of the existence of such a system. We desire not, indeed, to elevate the Naturalist at 'the expense of the Geologist, for his office is not and cannot be, so far as Geology is concerned, the principal one. His deductions are of the utmost importance, but they are so only when in the hands, and under the guidance, of the field Geologist. this position his labours are invaluable; and the greater the advance that has been made hitherto in Geology, the more important, the more conclusive, and the more satisfactory, has his assistance been found, and the more useful have his suggestions and decisions proved.

In

On this subject Mr. Murchison and Mr. De la Beche are a little at variance; and as it is one of very great interest at the present time, we have been tempted to say more than by some, perhaps, may seem required. So much, however, depends on the due admission of zoological evidence in Geology, and on the