superior. But there is a good fauna of characteristic species whose limits are well defined, and not to be derived, by any known analogy of nature, from the species which existed in former epochs. Let us, however, leave the obscure forms of animal life, and come to the magnificent group of fishes which characterize this series of rocks. On the system of Agassiz, fishes are divided into four great orders-(1.) Placoids, or fishes partially studded with a kind of scaly plates-such are the sharks and rays of the present day; (2.) Ganoids, or fishes encased in scales of hard enamelsuch is the bony pike of America; (3.) Cycloids, or fishes with scales, like those of a common salmon; (4.) Ctenoids, or fishes with jagged scales, like those of a perch. In the fishes of the first and second Order, the skeleton is more or less cartilaginous; but the system is not founded on the condition of the skeleton, for each of the four orders contains both cartilaginous and bony fishes. Taking into account the brain, and the whole nervous, circulating, and generative system, the Placoids stand at the highest point of a natural ascending scale; and the Ganoids are also very highly organized. Now the fish described from the old red sandstone come under these two orders, Placoids and Ganoids. Of the Placoids, the families of Cestracionts and Hybodonts are well known; but during the passage of this Article through the press, eight or ten new genera of Placoids have been found by Agassiz among the fishes brought by Mr Murchison from the old red sandstone of Russia; and among them are several new species of Ctenodus-a genus known before only in the carboniferous system. Let the reader bear in mind that these fishes are among the very highest types of their class; and that we can reason upon them with certainty, because some of them belong to families now living in our seas.

Again, of the Ganoid order, we have in the old red sandstone Cephalaspides, Acanthodians, Dipterians, Sauroids, and Coelacanthes. In none of them is there the most remote affinity to Crustaceans, or any other Articulata. On the contrary, they in many respects make an approach to the higher class of Reptilesfor example, in their dentition, and some of them in their ball and socket jointed vertebræ.

These facts are not new; the greater part of them forming now the common stock of our geologists. They could not be altogether, we should think, unknown to our author; but he turns them to no account. And these fishes, he tells us, have hetorocercal tails, and therefore resemble the embryo of a salmon. Therefore, on his system, they must be low in organic structure. cling to such feeble analogies as these, and to keep out of sight the broad and speaking facts of nature, is about as wise as it would be for the Captain of a man-of-war to clear his decks for action, by throwing overboard his great guns, and then to fight

To

his enemy with the rickety furniture of his cabin. Again, he tells us (pp. 68, 69) that the Cephalaspis very much resembles a trilobite that the Coccosteus has a jaw like the nippers of a lobster, and its mouth opening vertically, contrary to the usual 'mode of the vertebrata'-and that these facts enforce our placing these fishes near the Crustaceans! Now these supposed facts are only blunders and guesses made by the first observers before any good evidence was before them. To repeat these blunders now, is discreditable to the author, and only shows the pertinacity with which he clings to his philosophy of resemblances; er that he is unacquainted with the progress of discovery and the present condition of geology.

5. Carboniferous System.-In England this system begins with great coral reefs, and beds made up of the fragments of radiata, and with beds composed of shells, often with both valves united. The animals must, therefore, in many cases, have died on the spot where we find their remains; and to account for the structure and position of these rocks, (which form some of the grandest features in South Britain,) we must suppose them to have been deposited on the conglomerates of the old red sandstone, or on the broken edges of the still older systems, during long periods of time; and we must further suppose, what is implied in the conditions of the successive beds, that during these periods there were many changes of level between sea and land. Masses of vegetable matter, which grew on the neighbouring land, were drifted among these ancient strata, along with mud and sand, and thus contributed their spoils to some of the lower and less perfect beds of coal. Over them came the rich carboniferous deposits of England, which must be accounted for in some different manner; for they are not marine, and some of them contain a few beds of fresh water shells. We believe they were formed by the repeated submergence of ancient jungles which were not drifted from the flat regions where they grew. In Scotland this series is geologically more perfect. We have deposits of good coal below the carboniferous limestone, and afterwards (as in England) alternating with it. The limestone series is less perfect than in England; but over it we have, in the great Caledonian trough between the Grampians and our frontier chain, an upper coal formation like that of South Britain. In Ireland, the lower division of the system is still more perfect, and probably encroaches on the old red sandstone; but, unfortunately for that island, the upper division is feeble and degenerate.

It is difficult, and perhaps impossible, from want of evidence, to know when land-plants first began to grow. Our author speaks positively, as usual, on this question, and what he tells us is wrong. In the Rhenish provinces, and in the Hartz mountains, we find traces of such plants, and even thin beds of coal,

among rocks abounding with both Devonian and Silurian fossils; and, in Ireland, the carboniferous impress descends so low, that some geologists have already been induced to shift their boundary lines, and to cut off some of the upper bands from the old red sandstone and give them to the carboniferous system. One thing we may, however, affirm, that we know no older types of landplants than those of the rocks we are now describing; and they show us a gorgeous flora. Several hundred species have been described; many more are known, but have not yet been figured; many, no doubt, will be hereafter discovered; and many must be lost for ever.

But what are these old types-these first-fruits of nature's vegetable germs? Are they but rude, ill-fashioned forms, as if 'some of nature's journeymen had made them, and not made them well?' Far otherwise. Among them, we find pine-trees in structure more near to the magnificent pine-trees of the South Sea Islands, than to any other forest trees of the present world. We find also palm-trees and tree-ferns, and other gigantic forms of vegetable life, approaching to the structure of tree-ferns. In this old flora are grand but strange forms, so unlike all living nature, that our best botanists know not in what order of the vegetable kingdom they are to find their place. Our equiseta and lycopodia, and some of our arundinaceous plants, are represented in this old flora by types generically different from living nature; but equal to living types in complexity of structure, and superior to them in the scale of development. To bring it into comparison with any thing in living nature, the old carboniferous flora may be said most nearly to approach the rank and compli cated vegetation of a tropical jungle; where the powers of vegetable life are stimulated to the utmost by great heat combined with great moisture.

These ancient forests performed a most important part for the good of the future inhabitants of the earth; and they may have helped to purge the atmosphere of an excess of carbonic acid, and may so have prepared the surface of the earth for new inhabitants. Insects, we know, were buzzing in the air during this period, (and here again our author is mistaken;) but we believe that no birds cheered these old forests with their song, and that no reptiles were seen crawling on the ground.* Fishes, however,

During the passage of these sheets through the press, we have learned, from the last number of Professor Silliman's Journal, that traces of birds have been found in the carboniferous system of North America. Should this fact be established, it gives us one more argument against the theory of development.

abounded in the seas and salt marshes; and it was during this period that they reached their most perfect organic type. They were the lords and despots of creation, and they had a noble structure in conformity with their high office. Since then the class has greatly increased in its species, but has degenerated to a less noble type. Here, as in the old red sandstone, the Cestracionts and the Sauroids abounded; and the fish of the latter family reached their most complete development both of size and structure. Fishes of both these families are still living in our seas and lakes-the feeble representatives of the ancient race-and they have been submitted to the knife of the anatomist.

Should any one of our readers have been misled by our author's hypothetical descriptions, we entreat him to study the admirable dissertation on the Sauroids, in the second volume of the Poissons Fossiles of Agassiz, and then to look over the list of fossil fishes characteristic of the Paleozoic period. It is there shown to demonstration, that the Sauroids in their general osseous structure, and in the development of their nobler organs, run close upon the class of Reptiles. Yet have they a general structure so peculiar, that no anatomist can confound them with Reptiles, or derive one class from the other, by any known law of organic nature. It is true that all the Paleozoic fishes have heterocercal tails; and some of them (the Cestracionts) have other anatomical arrangements in which they resemble the salmon, while it is in an embryo condition. On this account, as it falls in with his theory, our author degrades the old fishes from their true anatomical place in the scale of nature, and dares to quote Agassiz as giving some confirmation to his views.-(P. 71.) But. he misunderstands Agassiz, and keeps out of sight the whole pith of the argument. Agassiz discards the embryonic theory, because it will not lend itself to the demonstrations of comparative anatomy; for the theory would, in this part of nature's kingdom, lead only to false conclusions, and turn upside down every principle of true arrangement. But we may quote a few words from the dissertation to which we have alluded: Je n'accorderai pas ⚫ une valeur exagérée à l'embryologie, qu'on a trop souvent invoquée, comme un argument sans replique, dans les débats dont il vient d'être question.' Il existe sans doute un type général de conformation embryonique dans toute la série des vertébrés, 'mais nous savons aussi que le type particulier de chaque espèce 'entre de très-bonne heure en conflit avec le plan général.'*

In the whole of our past labours, we never met with passages more worthless and untrue to nature, than those in which

*Poissons Fossiles, Vol. ii. Part ii. P. 34.

the author of 'The Vestiges' gives us his comments on our ancient families of fossil fish. If he had not read the great work of Agassiz, he was not in a condition to enter on the question. If he had read it, and understood it, when he wrote the passages to which we have just referred, then have we a far graver charge to bring against him. But we bring no such charge against him. Like many other men, he shut his eyes to nature, or only took a one-sided view of her; and then brooded over the fantasies of his mind till his dreams became to him as substantial realities: and, under this delusion, he composed his work in solitude,' hoping to give other men the benefit of his visions, and thereby to improve their happiness, (p. 380.) Burnet started with better principles, but he went on guided by a like spirit. He told us of all that had befallen the earth from its creation until now; and he added, with all sincerity, that he could not inform mankind of all that was to happen hereafter to our world, only because he had not leisure for retirement and long meditation.

6. Permian System. Zechstein. Magnesian Limestone Formation. Under these names is designated a series of well-known deposits which succeed the carboniferous rocks, and sometimes pass into them by insensible gradations. In the south of England they are represented by conglomerates, partly composed of the solid and more or less rounded fragments of the older rocks. Here again we have a proof of the long periods of time during which the ancient works of nature were perfected; for the old rocks were solid as they are now, and their organic remains were petrified at the time these conglomerates were forming. In Scotland we know of no good illustrations of this system; but in the north of England it is made up of the following regular series of deposits, tallying almost step by step with the Zechstein of Germany :-(1.) Lower red sandstone; (2.) Marl slate, with many impressions of fish; (3.) Magnesian limestone, admitting of further subdivisions; (4.) Red marl and gypsum; (5.) Upper slaty limestone, partly magnesian; (6.) Upper red marl and gypsum. This last deposit passes into the great red sandstone series of central England. We have copied this succession from a paper published by Professor Sedgwick in the 'Transactions of the Geological Society of London," (Vol. iii., 2d series.) Led by the physical affinities of the system, and following the classification of previous writers, he placed it at the base of the secondary series; but he stated distinctly, that in zoological characters, it had more affinity with the lower systems above described, than with those which came next above it in the ascending scale. Much has been done since, and the results may be stated sufficiently for our purpose in a few lines. (1.) All the fossil plants of the lower red sandstone are, so