system of rocks, of which the cornstone and conglomerate beds are in the immediate vicinity, and the position and relation of the three to one another easily determinable. The upper or yellow deposit occupies the valley of Stratheden nearly throughout its entire length and breadth, and ranges along the base of the heights of Nydie, Cults, the Lomonds, Binnarty, and the Cleish hills, dipping under the carboniferous lower group, and generally separated by overlying masses of trap. The sandstones, indeed, of both systems, resemble each other so much in color and texture, that in many instances along the line now indicated the trap must be taken as a guide by which to ascertain the qualities and respective positions of the two series. Glenvale, a beautiful ravine which intersects the Lomond range, presents admirable sections of the whole group, in their regular order of superposition and finely displaying their contrasting mineral charac

ters.

ORGANIC REMAINS. These are abundantly distributed in scales, teeth, spines, coprolites, and other remains, and are to be found in every opening and quarry throughout the range of the deposit. It is only in Dura Den, however, that any entire animal forms have as yet been obtained, and these all confined to a portion of the rock not exceeding thirty yards by three in breadth, a narrow trough excavated for the purpose of forming a water-shed to the mill, which stands in the center of the valley. The fossils derived from this single spot consist of four new genera, and seven or eight new species, that have been added to our catalogue of extinct animals. These remains were all in a state of beautiful preservation; the scales and fins are brightly enameled, and contrasted with the matrix in which they are set, the colors are as vivid and glistening as when the animals were sporting in their native element. The specimens, I believe, of the various collections made in this rich depository by different parties were all submitted to the examination of M. Agassiz, who has figured several of them in his "Monograph" on the old red sandstone, but without completing, it is much to be regretted, his descriptions of the various fossils. We give the following abridgment of such descriptions as are contained in the work.

There are two new species of Holoptychius represented, namely, Andersoni and Flemingii, and these are distinguished entirely by the form and tracery of their respective scales. The H. Andersoni is described as a small spindle-shaped (fusiforme) fish, thick and short, and narrowing rapidly toward the tail.

The scales are much less than those of the other species, as deep as they are broad, and resembling in general form the scales of the H. Murchisoni found in Clashbennie. What peculiarly distinguishes them is the figure of the ornaments (le dessin des ornemens) of the surface, which are parallel, horizontal, very marked and distant in the A. and never extending in the striæ to the posterior edge. The scales, again, of H. Flemingii are on the sides of the fish deeper than they are broad, and on the belly they become rounder. Their ornaments are also very distinct in the F., consisting of a system of waving lines, which run horizontally toward the outer edge without any perceptible ramification, while the wrinkles of the scale rise from a series of little hills (collines) ranged parallel over the length of the inner edge, undulating and very close. This specimen is represented as very imperfect. The other is nearly entire, the plates of the head and several of the teeth are well preserved, every scale is in its place, and the fins are only wanting to restore the normal outline of the fish. This fossil has been figured and erroneously described in the author's "Geology of the County of Fife" under the name of gyrolepis (holoptychius now) giganteus, from which, says M. Agassiz, it differs specifically.

From the fossils of this locality has been established the new genus of GLYPTOPOM US, the specimen of which being originally

mistaken by Agassiz for a platygnathus, but since found by him to differ from that genus in several material points. The scales of the platygnathus, for example, are round and imbricated, possessing in this respect all the characteristics of the scales of the cœlacanthes, while on the other hand those of the glyptopomus resemble the scales of the sauroids, which are rhomboidal or square (ou carrées), closely set and never imbricated, as shown in the subjoined illustration.

Moreover, the platygnathes are lengthy (allongés) in the body, likewise long (longue) in the tail, which is furnished with a very strong fin, whereas the body of the glyptopomes is very thick, and the tail short. The ornaments of the scales of G. bear a close affinity to those generally of the coelacanthes. Only one species as yet has been found, the glytopomus minor, and figured in the tab. 26 of the "Monograph" under the name of platygnathus minor.

The glyptopomus minor, says M. Agassiz, found in Dura Den, and of which there is but one specimen, is possessed of a body broad and thick, approaching in form to that of the holoptychius. The fish is lying on the belly, and turned slightly to the left, so that it is the back and right side which are represented in the plate. The head is proportionally small, covered with bones very irregularly carved, presenting a dense and diversified granular aspect. On the side of the head there is a large enameled plate, which shows that the cheek was covered, as in the polypterus,

with one single osseous plate, on the under edge of which was fixed the large masticatory muscle. The scales on the body of the fish are large, high on the sides, and nearly square on the back, where in the middle they form an oblique series converging to an acute angle. The scales are very thick set on the side of each other, and apparently connected only by means of the skin to which they are attached. The enameled surface is not smooth, but rather marked with a fine granulation, which imparts a rich velvet gloss to the scale. Traces only of the fins are preserved, partly of the ventral, partly of the dorsal or caudal, and the rays of which are all apparently short and slender. This specimen forms a part of the author's collection, but inadvertently described as belonging to that of Professor Jameson.

Another genus, established from the fossils of Dura Den by M. Agassiz, is the Pamphractus, of which there are two species, Hydrophilus and Andersoni. These are both in the collection of the author, and have a special history of their own, from which, when read in all its details, it would appear they have suffered as roughly at the hands of geologists in simply determining their class, order, or genus, as they ever did from the physical revolutions amidst which their lot was originally cast.

Before the type of a new and strange form called Pterichthys, had been determined by this learned palæontologist, collectors were everywhere puzzled by the specimens of the animal that, from time to time were casting up. The winged appendages of the sides of the head, as movable fins, had easily given rise to a variety of opinions concerning their true affinities, and which, says M. Agassiz, "have been regarded by the most able naturalists successively as Tortoises, Fishes, Crustacea, and even Coleoptera." The fossils of Dura Den were at first regarded by him as belonging to the type as well as genus Pterichthys, and my specimens were actually returned from Neufchâtel so named-the "broad" and "narrow" species—and the label still remains attached. Meanwhile, five or six species of the genus Pterichthys had been already determined and described by him, from the fossils of Cromarty and Morayshire-these in the collection chiefly of Mr. Hugh Miller; and Mr. Miller being, about the same period, engaged in the preparation of his work, "The Old Red Sand

stone," speedily under the new nomenclature, as he was so opportunely furnished with the materials, gave the public the benefit of M. Agassiz's discovery and version of their true and authentic history. What we had hastily, certainly, but still influenced much in the matter by the judgment of others, referred to the order of Coleoptera, he was enabled at once, upon the inspection of a Dura Den specimen, and from its very striking resemblance to his own, to pronounce to be a Pterichthys. A few pages before he had stated that he could make nothing of the creature, although some specimens of the fossil had been in his possession for a period of nearly ten years; but Now, he was able to record,-“I very lately enjoyed the pleasure of examining the bona fide ichthyolite itself,— one of the specimens of Dura Den, and apparently one of the more entire, in the collection of Professor Fleming. Its character as a Pterichthys I found very obvious." But short-lived, indeed, are all mundane enjoyments. The most intellectual, in the revolutions of science, are not exempt from their general character of vanity. While the two northern sages were thus gazing, in all the raptures of a new discovery, "upon the bona fide ichthyolite itself," the philosopher, under the shelter of the Jura, was doubting, re-examining, and finally correcting, his own first judgment; and, while the virgin pages of "The Old Red Sandstone" had scarcely time to reach their author, the "Monograph" was announcing to the world the determination of a new genus, and that the fossil of Dura Den was a Pamphractus, and no Pterichthys at all.

"I had at first," says Agassiz, "connected with pterichthys the only species known of that genus, by calling it pterichthys hydrophilus, but a more profound study and attentive comparison of that species with the genus coccosteus, have proved that it ought to form a distinct genus, intermediate betwixt pterichthys and coccosteus, which I have named pamphractus, in consequence of the divided form of the carapace. The pectoral fins of pamphractus resemble very much those of the pterichthys in their form, being slender, elongated, and crooked (courbée). But the plates of the carapace are all differently arranged. The central plate is very large (énorme); it covers two-thirds of the whole carapace, and unites the anterior articulation of the head with the carapace. The lateral plates, which acquire so great a development in the