which thus gains in thickness till at length it is buried under layers of sand like those now forming over peat and clay along our sinking coasts.

We next consider the incipient metamorphism by which a portion of the clay immediately around the fossils, has been transformed into ironstone nodules, or concretions. The iron held in solution in the waters that permeated, and circulated in, the pores of the clay bed, seems to have been attracted and solidified by the chemical force which had been gathered from the sunlight and stored up in the organic matter of the fossils themselves. This force may be compared to that of the galvanic battery, by which metals are deposited from solution. Observe that the nodules take the form of the fossils they enclose. A leaf has the general form of an oval, it is bent to one side, or one leaf lies across the side of another; and the resulting nodule is oval, curved, or has a lump on one side, as the case may be. It would be instructive to compare them with siliceous geodes, that have begun their growth upon fossil corals, crinoids, shells, etc.

The nodules as they lie in situ, in the shale, appear as though the ferrous carbonate had been attracted with such force as not only to fill the spaces between the particles of clay, but to crowd them apart and thicken the layers. Here are specimens that exhibit the continuity of the strata, from the soft outlying shale through the hard nodules, which thus appear like swellings in the clay. The layers are concavo-convex, not only in the nodule above and below the plane of the leaf, but also in the soft shale above and below the nodule. It is as though the expansion of the nodule had crowded and bent the strata into arches upward and downward.

But it is not probable that the leaf retained its chemical force till the clay was condensed to shale by pressure, nor that the strata in the nodule were expanded and thickened again after such condensation. No doubt the iron was attracted soon after the leaf was buried, when the shale was as yet soft ooze. The layers were then much thicker, even in the nodule, than now; and as the weight of overlying deposits increased, the soft outlying shale being compressed more than that in the denser nodule, the layers above and below it were bent into arches as we find them.

If we compare these biological records with the primitive inscriptions of man, both are written in the picture language of

nature, but these are incomparably more minute and accurate. The human records are found exposed on the crumbling surface of monuments; but these were scaled up in lumps of stone nearly as hard as flint and buried in beds of clay beyond the reach of wear and decay. The tablets of cuneiform from the fatherland of Abraham were done in soft clay and afterwards hardened by baking; these were inscribed on softer clay which has since been hardened, automatically, by the writing itself, or, more properly, by the types which printed the writing. Like the casts of the founder who surrounds his models with moist sand, these, too, are casts, but they are casts of such delicate things as fern leaves and insect wings, moulded in fine clay by the gentle touch of water and afterwards chemically hardened by an iron cement.

But, the age of the most ancient human writings is as a thing of yesterday—utterly insignificant when compared with the immeasurable antiquity of these inscriptions from Mazon Creek. They were formed in the incalculably remote Carboniferous Period, and while the vast Permian, Triassic, Jurassic, Cretaceous and Tertiary beds, which contain the record of nearly all animals above the grade of fish-like reptiles known to geologists, were being deposited, these engraved stones were sealed up in the shale beyond the reach of change or decay. The granite of the Himalayas, Sierras, Andes, and all the loftiest mountain chains, was, for the most part, formed out of the sediment of dried up seas after these fragile ferns and insects were embedded in the clay. In Post Tertiary time the overlying rocks were planed down by moving ice, and clay, gravel and bowlders strewn over the surface. Then came an age of shallow lakes followed by swamps, forming the black prairie soil.

Finally, the waters of the Mazon dug their channel down into the shale, and the softer clay being washed away, these nodules were left in placers in the bed of the stream, mingled with granitic gravel derived from glacial drift.

Being thus exposed to the combined action of air and moisture, the ferrous carbonate is changed to the ferric oxide of iron, and the nodules turn red like boiled lobsters. During this process they often divide spontaneously, exposing the fossils within, and thus they were found by the German1 naturalist who made the first collection. But this change of iron from the blue to the red state

1 Mr. Joseph Even, of Peru, Illinois.

is often attended with solution, resolidification, and final complete separation of the iron from the clay. The iron forms as a red brown shell, and the half of a nodule, with a fern on the side, becomes a geode, enclosing the clay in an ochre-like form in its interior.

These beautiful specimens being mostly exhausted from the gravel beds, the collector must anticipate the denuding forces and dig the concretions out of the shale of the river banks and bottom, and crack them for himself. Specimens got in this way show the finest details of structure, and promise to increase our knowledge of the biology of the lowlands of Carboniferous time.

THE FOSSIL DINOCERATA IN THE E. M. MUSEUM AT PRINCETON, NEW JERSEY. By FRANKLIN C. HILL,1 of Princeton, N. J. [ABSTRACT.]

THE collection of the bones of the Dinocerata, at Princeton, is not large, but choice, and I hope to be able from it and the aid of my blackboard, to give a tolerably correct idea of the family

structure.

Our chief pieces are two skulls, so nearly perfect that our fragments of others enable us to restore them entirely.

One of these has been already described and figured in a Bulletin from the Muscum, under the name of Uintatherium Leidianum. The other, a more recent acquisition, is even more perfect, the left side showing everything from the snout to the occipital condyles. We consider it as a Loxolophodon.

Beginning at the snout we notice that a slight twisting of the specimen throws the right " horn-core," so called, a little above and in advance of the left one, so that both are seen in a side elevation, which is not the case with the other pairs.

These animals have been called "Proboscidians" and have been supposed to have have had trunks, like Tapirs, if not like Elephants. But the nasal bones, as has been pointed out by others,

1 Curator of the Museum.

make half the length of the skull. In the Elephants they are very short, and the width of the skull is greater than the length from occiput to insertion of proboscis. But here we have a length of thirty-eight inches and a width of only ten inches. Granted a proboscis for which there is no "insertion" visible,-what power could it exert at the end of such a lever?

Again, in the proboscidians the premaxillaries are very large and carry the heavy incisors, while here we find them very small, and obviously fitted to carry a pad for the lower incisors to play against, as in the ruminants.

This skull measures nearly thirty-nine inches in extreme length, but among our fragments of other individuals we have a "horncore" which indicates a skull sixty inches in length.

The upper canines of this specimen were in place when it was found, but were too fragile for preservation, excepting the spearshaped points. They were curved much in the manner of the Machairodus canines, but unlike them they had a double-bayonet section and no cutting edge.

Of the lower jaw we have two nearly perfect specimens, and fragments of others, so that we have the entire dentition,-three curiously bilobate incisors, a long diastemma, and six diagonally ridged molars which fully justify Mr. Cope's name "Loxolophodon." From these specimens and Mr. Marsh's plates we are able to restore the lower jaw with confidence.

This dentition seems to indicate a diet of soft roots and stems, rather than of grass, in the collection of which food the upper canines doubtless played an important part, though it is not easy to see just what it was. But the peculiar position of the condyles, they being but little above the line of the molar teeth, and presenting directly backwards, allowed the jaw to fall low enough to clear the points of the canines and thus bring those teeth into action, as has already been pointed out by others.

The descending flap of the lower jaw seems to be for a guard for them, indicating that they were not weapons of offence and defence.

Of the cervix we have the first five bones, and they show that the neck must have been at least seventeen inches long, quite long enough, with the long head, to reach the ground.

The spines of these vertebræ are small and weak, but the size and weight of the head convince me that the first dorsals must have had large, strong spines. Our dorsal vertebræ are from the middle of the column. There were probably about twenty dorsals and three lumbars.

The pelvis and caudal vertebræ which we have belong to Uintatherium Leidianum, and, though not entire, are sufficiently so to show that they were similar to those of Mastodon, excepting that the caudals show a much broader and flatter tail than that animal carried.

Our ribs are mostly fragmentary but are mastodontoid in character. Their number, of course, is as indefinite as that of the vertebræ, and we must borrow the shape of the first pair from Mastodon.

Our scapula is not in good enough condition to show the very peculiar form of this bone. Mr. Cope kindly gave me a proof sheet of his plate of a perfect specimen, which I use in my sketch, leaving the description to him.

The humerus has been so well and fully described by Dr. Leidy, that I shall content myself with drawing it and quoting his remark that "it is unique," as in fact is also the ulna, which with a large olecranon has a columnar shaft, expanding very slightly at the distal end, in which it differs widely from the elephants. The radius is straight, and stands almost perpendicularly before the ulna, taking much of the weight of the humerus on its broad head, somewhat after the manner of Hippopotamus.