CONNECTICUT RIVER SANDSTONE. 55 



drag upon the ground, and the body of the creature was therefore but slightly 

 elevated. The position and trailing movements of the feet suggest some analogy 

 to existing testudinate types, but the character of the feet forbids this inference. 

 Among the sandstone fossils there are really none that indicate tortoises, although 

 a different opinion has formerly been held. The evidence upon which this infer- 

 ence rests, consists in the frequent occurrence of parallel grooves or furrows (PI. 

 36), broadly separated, that have been produced by the feet and legs of an 

 animal moving over soft mud, probably beneath water. The discovery of the 

 fine specimen drawn upon Plate 35 explains the cause of these double rows of 

 furrows (PL 36), for it may be supposed, that if the animal to which the impres- 

 sions (PL 35) were due, sank deeply into the soft ground, the dragging of each 

 set of feet would produce the impressions of the deep grooves that have been 

 supposed to indicate the existence of tortoises. Although in these cases the 

 impress of the feet are not preserved, their places are accurately marked, occurring 

 in regular alternation from side to side. If a tortoise were thus to sink, its 

 solid body would plow through the mud and leave a distinct trace, a con- 

 dition I have never seen. 



In the present state of information, it is impossible to comprehend the 

 analogies of the animal making the impressions upon Plate 35. 



Dr. Hitchcock describes the animal that made the impressions represented on Plate 

 35 under the name of Tarsodactylus caudatus. The impressions figured on Plate 

 36, he refers to a species of his genus Helcura. 



The original of Plate 35 is in possession of Amherst College. 



PLATE XXXVII. 



The footprint that forms the subject of this Plate is remarkable for magni- 

 tude, being seventeen inches in length and eleven in breadth. It indicates the 

 most colossal of all the sandstone animals. Tridactylous footprints occur of equal 

 length, but for solid, massive proportions, this is unequalled. The impress is 

 very perfect, and shows the osseous divisions of the foot. There are four ponder- 

 ous toes, joined in contact to the heel. They do not materially differ in length ; 

 the two central are longest, and project further forward than the lateral, which, 

 in turn, project furthest behind. They are symmetrically arranged, are in con- 



