128 



LOWER VERTEBRATES. 



no species closely related to any of those found in Europe, but among the species 

 of the great basin of Utah and California are numerous forms closely allied to 

 species found in Europe and northern Asia. Similar resemblances between the fauna 

 of western America and Asia have been noticed in other groups as well as in fishes. 



Southward, the Cyprinidse have extended their range in America as far as central 

 Mexico, but none exist in Central America, South America, or Cuba. While they 

 abound in all parts of Europe and Asia, very few have entered the waters of Africa. 

 Many species have a very wide range of distribution, while numerous others are con- 

 fined to a single river basin. Such confined species are evidently products of differ- 

 ences in environment. 



In Europe, different writers, especially Siebold and Gtinther, have described many 

 natural ' liybrid ' foims among the Cyprinidse, and certain writers have gone so far as to 

 be able to detect at once the parentage of such hybrids even to the indication of which 

 is the male aild which the female parent. I have had no opportunity to study any of 

 these alleged hybrids in life, and so can have no opinion as to the case in Europe. I am 

 sure, however, that the American species behave differently, as in all my somewhat 

 extensive exjjerience as a collector of these fishes, I have never yet found a specimen 

 which I had the slightest reason to regard as a hybrid. 



Fig. 84 — SemotUus buUarls, fall-fish, silver chub. 



Intermediate species and varying species are of course numerous, and many of 

 them are extremely inconstant in their characters. It is, however, quite illogical to 

 ascribe these peculiarities to hybridism. While crosses between fishes which we con- 

 ceive to be distinct species, may doubtless sometimes occur in nature, I do not believe 

 that, either in America or Europe, they are so frequent as to make any confusion 

 in classifications. In other words, I believe that the great majority of alleged hybrid 

 Salmonidsa and Cyprinidse in Europe rest either on individual variations, or else on 

 real specific differences, and that Siebold's hypothesis of general hybridization is 

 unwarranted by the facts. 



Among the American Cyprinidse the following are some of the most noteworthy 

 genera : Canipostoma, the stone-roller, differs from all other known vertebrates in 

 having its very long intestines, which are six or seven times the length of the body, 

 wound in a spiral of some fifteen coils around the air bladder. The species feed on 

 mud, and swarm in all small brooks of the western and southern states in the spring. In 

 this genus, as already noticed, the nuptial tubercles of the male are more developed 

 than in any other, often covering the entire head, body, and fins. 



Another remarkable American genus is Exoglossum, the single species of which is 

 found in the rivers of the middle states. In this genus the two rami of the lower 



