ready means of escape by flight or concealment from their enemies. We learn from Geology that this Order began to exist nearly at the same time with the Order of Saurians, and has continued co-extensively with them through the secondary and tertiary formations, unto the present time: their fossil remains present also the same threefold divisions that exist among modern Testudinata, into groups respectively adapted to live in salt and fresh-water, and upon the land.

Animals of this Order have yet been found only in strata more recent than the carboniferous series.* The earliest example recorded by Cuvier, (Oss. Foss. Vol. 5, Pt. 2, p. 525,) is that of a very large species of Sea Turtle, the shell of which was eight feet long, occurring in the Muschelkalk at Luneville. Another Marine species has been found at Glaris, in slate referable to the lower cretaceous formation. A third occurs in the upper cretaceous freestone at Maestricht. All these are associated with the remains of other animals that are marine; and though they differ both from living Turtles and from one another, they still exhibit such general accordance in the principles of their construction, with the conditions by which existing Turtles are fitted for their marine abode, that Cuvier was at once enabled to pronounce these fossil species to have been indubitably inhabitants of the sea.t

*The fragment from the Caithness slate, engraved in the Geol, Trans, Lond. V. iii. Pl. 16, Fig. 6, as portions of a trionyx, is pronounced by M. Agassiz to be part of a fish.

† Plate 25, Fig. 4, represents a Turtle from the slate of Glaris: it is shown to have been marine by the unequal elongation of the toes in the anterior paddle; because, in fresh-water Tortoises, all the toes are nearly equal, and of moderate length; and in land Tortoises, they are also nearly equal, and short; but in all marine species they are very long, and the central toe of the anterior paddle, is by much the longest of all. The accordance with this latter condition in the specimen before us, is at once apparent; and both in this respect and general structure, it ap proaches very nearly to living genera, This figure is copied from Vol.

The genera Trionyx and Emys, present their fossil species in the Wealden fresh-water formations of the Secondary series; and still more abundantly in the Tertiary lacustrine deposites; all these appear to have lived and died under circumstances analogous to those which attend their cognate species in the lakes and rivers of the present tropics. They have also been found in marine deposites, where their admixture with the remains of Crocodilean animals shows that they were probably drifted, together with them, into the sea, from land, at no great distance.*

In the close approximation of the generic characters of these fossil Testudinata, of various and ancient geological epochs, to those of the present day, we have a striking example of the unity of design which has pervaded the construction of animals, from the most distant periods in which these forms of organized beings were also called into existence. As the paddle of the Turtle has at all times been adapted to move in the waves of the sea, so have the feet of the Trionyx and Emys ever been constructed for a more quiescent life in fresh-water, whilst those of the Tortoise have been no less uniformly fitted to creep and burrow upon land.

5, Pt. 2, Tab. 14, f. 4, of the Oss, Foss. of Cuvier. M. Agassiz has favoured me with the following details respecting important parts which are imperfectly represented in the drawing from which Cuvier's engraving was taken, "The ribs show evidently that it is nearly connected with the genera Chelonia and Sphargis, but referable to no known species; the fingers of the left fore paddle are five in number; the two exterior are the shortest, and have each three articulations; and the three internal fingers, of which the middle one is the longest, have each four articulations, as in the existing genera, Chelonia and Sphargis."

* Thus two large extinct species of Emys occur, together with marine shells, in the jura limestone at Soleure. The Emys also and Crocodiles, are found in the marine deposites of the London clay at Sheppy and Harwich; and the former is associated with marine exuviæ at Brussels. Very perfect impressions of small horny scales of Testudinata, occur in the Oolite slate of Stonesfield, near Oxford,

The remains of land Tortoises have been more rarely observed in a fossil state. Cuvier mentions but two examples, and these in very recent formations at Aix, and in the Isle of France.

Scotland has recently afforded evidence of the existence of more than one species of these terrestrial reptiles, during the period of the New red, or Variegated sandstone formation. (See Pl. 1, Sec. 17.) The nature of this evidence is almost unique in the history of organic remains.*

It is not uncommon to find on the surface of sandstone, tracks which mark the passage of small Crustacea and other marine animals, whilst this stone was in a state of loose sand at the bottom of the sea. Laminated sandstones are also often disposed in minute undulations, resembling those formed by the ripple of agitated water upon sand.†

* See Dr. Duncan's account of tracks and footmarks of animals impressed on sandstone in the quarry of Corn Cockle Muir, Dumfries-shire Trans. Royal Society of Edinburgh, 1828.

Dr. Duncan states that the strata which bear these impressions lie on each other like volumes on the shelf of a library, when all inclining to one side: that the quarry has been worked to the depth of forty-five feet from the top of the rock; throughout the whole of this depth similar impressions have been found, not on a single stratum only, but on many successive strata ; i, e. after removing a large slab which contained foot-prints, they found perhaps the very next stratum at the distance of a few feet, or it might be less than an inch, exhibiting a similar phenomenon. Hence it follows that the process by which the impressions were made on the sand, and subsequently buried, was repeated at successive intervals.

I learn, by a letter from Dr. Duncan, dated October, 1834, that similar impressions, attended by nearly the same circumstances, have recently been discovered about ten miles south of Corn Cockle Muir, in the Red sandstone quarries of Craigs, two miles east of the town of Dumfries. The inclination of the strata of this place is about 45° S. W. like that of almost all the sandstone strata of the neighbourhood. One of these tracks extended from twenty to thirty feet in length: in this place also, as at Corn Cockle Muir, no bones of any kind have yet been discovered.

Sir William Jardine has informed Dr. Duncan that tracks of animals have been found also in other quarries near Corn Cockle Muir.

In 1831, Mr. G. P. Scrope, after visiting the quarries of Dumfries,

The same causes, which have so commonly preserved these undulations, would equally preserve any impressions that might happen to have been made on beds of sand, by the feet of animals; the only essential condition of such preservation being, that they should have become covered with a farther deposite of earthy matter, before they were obliterated by any succeeding agitations of the water.

The nature of the impressions in Dumfries-shire may be seen by reference to Pl. 26. They traverse the rock in a direction either up or down, and not across the surfaces of the strata, which are now inclined at an angle of 38°. On one slab there are twenty-four continuous impressions of feet, forming a regular track, with six distinct repetitions of the mark of each foot, the fore-foot being differently shaped from the hind-foot; the marks of claws are also very distinct.*

found rippled markings, and abundant foot tracks of small animals on the Forest marble beds north of Bath. These were probably tracks of Crustacea. See Phil. Mag. May, 1831, p. 376.

We find on the surface of slabs both of the calcareous grit, and Stonesfield slate, near Oxford, and on sandstones of the Wealden formation, in Sussex and Dorsetshire, perfectly preserved and petrified castings of marine worms, at the upper extremity of holes bored by them in the sand, while it was yet soft at the bottom of the water; and within the sandstones, traces of tubular holes in which the worms resided. The preservation of these tubes and castings shows the very quiet condition of the bottom, and the gentle action of the water, which brought the materials that covered them over, without disturbing them.

Cases of this kind add to the probability of the preservation of footsteps of Tortoises on the Red sandstone, and also afford proof of the alternation of intervals of repose with periods of violence, during the destructive processes by which derivative strata were formed.

* On comparing some of these impressions with the tracks which I caused to be made on soft sand, and clay, and upon unbaked pie-crust, by a living Emys and Testudo Græca, I found the correspondence with the latter sufficiently close, allowing for difference of species, to render it highly probable that the fossil footsteps were also impressed by the feet of land Tortoises.

In the bed of the Sapey and Whelpley brooks near Tenbury, circular

Although these footsteps are thus abundant in the extensive quarries of Corn Cockle Muir, no trace whatever has been found of any portion of the bones of the animals whose feet they represent. This circumstance may perhaps be explained by the nature of the siliceous sandstone having been unfavourable to the preservation of organic remains. The conditions which would admit of the entire obliteration of bones, would in no way interfere with the preservation of impressions made by feet, and speedily filled up by a succeeding deposite of sand, which would assume, with the fidelity of an artificial plaster mould, the precise form of the surface to which it was applied.

Notwithstanding this absence of bones from the rocks which are thus abundantly impressed with footsteps, the latter alone suffice to assure us both of the existence and character of the animals by which they were made. Their form is much too short for the feet of Crocodiles, or any other known Saurians; and it is to the Testudinata, or Tortoises, that we look, with most probability of finding the species to which their origin is due.*

markings occur in the Old Red Sandstone, which are referred by the na tives to the tracks of Horses, and the impressions of Patten-rings, and a legendary tale has been applied to explain their history. They are caused by concretions of Marlstone and Iron, disposed in spherical cases around a solid core of sandstone, and intersected by these water causes.

* This evidence of footsteps on which we are here arguing, is one which all mankind appeal to in every condition of society. The thief is identified by the impression which his shoe has left near the scene of his depredations. Captain Parry found the tracks of human feet upon the banks of the stream in Possession Bay, which appeared so fresh, that he at first imagined them to have been recently made by some natives: on examination they were distinctly ascertained to be the marks of the shoes of some of his own crew, eleven months before. The frozen condition of the soil had prevented their obliteration. The American savage not only identifies the Elk and Bison by the impression of their hoofs, but ascertains also the time that has elapsed since each animal had passed. From the Camel's track upon the sand, the Arab can determine whether it was heavily or lightly laden, or whether it was lame,

1