136 AN ATTEMPT TO DESCRIBE THE ANIMALS THAT MADE 



He sees that some of them were impressed by bipeds, others by 

 quadrupeds ; some by thick-toed animals, and others by narrow- 

 toed ; some by three-toed, others by four-toed, and others by five- 

 toed animals ; some by long and narrow heeled, others by short 

 and broad heeled, and others by heelless animals. Nor can he, by 

 any effort of the imagination, conceive how they all were made by 

 a single animal. I never knew a man who attempted to do this. 

 Let any one examine the outline drawings accompanying this paper, 

 and he will be satisfied on this point. Now there must be some 

 very decided characters in these tracks, that produce this conviction 

 of differences in the animals that made them. And why may not 

 these peculiarities be expressed on paper, and thus in fact become 

 the basis of generic and specific characters ? True, they are im- 

 perfect; but so are the characters of a large part of the genera 

 and species of fossil animals and plants. 



In the fourtli place, the feet of animals furnish excellent charac- 

 ters for distinguishing classes, orders, genera, and species. To be 

 satisfied of this point, let any one compare the feet of mammiferous 

 animals with those of reptiles ; or the latter with those of birds ; or 

 among the Mammalia, the feet of the Ruminantia with those of 

 the Carnivora, or Marsupialia ; or, among birds, the feet of the 

 Grallse with those of the Passeres, or Palmipedes ; or the feet of 

 the kangaroo, or Platypus, with those of the tiger or hog ; or those 

 of the Struthio Rhea with those of the eagle, or albatross, or 

 jacana. Indeed, the characters of several of the orders of birds are 

 drawn from their feet. Many other animals could, to a considera- 

 ble extent, be classified on the same basis. When we attempt in 

 the same way to distinguish genera and species, we are met by too 

 many exceptions to make such characters an easy and safe guide. 

 But in the absence of better distinctions, they might be used with 



