11)2 



THE AMERICAN NATURALIST [Vol. XLII 



the Bourbon nose is characteristic of that family. Ergo, 

 the Jews and the Bonapartes are distinct species of Homo ! 



2. True species are incapable of fertile hybridity. The 

 example of the Catalo, the fertile hybrid between Bos 

 americanus and Bos taunts, will suffice. The domestic 

 dog interbreeds freely on the plains with the coyote, and 

 no one doubts the specific validity of Cams latrans, what- 

 ever we may think about coyotes in general. The do- 

 mestic cat, according to Pocock, is a hybrid between the 

 wild cat of England and the wild cat of Egypt, with a 

 distinct tendency to vary along ancestral lines after cen- 

 turies of fertile hybridity. The domestic races of dog 

 freely interbreed, and yet we are quite sure they are the 

 derivatives of several wild species of Canis. May not 

 fertility, as a physiological inheritance, account for the 

 preservation of their distinctive types, notwithstanding 

 man's artificial selection? Is it probable, for instance, 

 that the Boston terrier hybrid will continue longer than 

 the fad of its breeding remains undiminished ? 



3. A species is a type which varies only within narrow 

 limits. The jungle fowl is fairly constant in nature. 

 Its extraordinary variability is seen in the domestic fowls, 

 whether they be derived from a single or several wild 

 species. And the doves are still better examples. A 

 turkey and guinea-fowl, on the other hand, though they 

 have been domesticated for centuries, vary but little from 

 their ancestral types. 



There are other definitions. But, you say, if I accept 

 no definition of species what rules do the taxonomists 

 have who "make" so many thousands of them? We 

 must have some, and we do have them, even if we are so 

 often accused of depending on whim and imagination. 

 And these are mine : Forms of animals which present dis- 

 tinct assemblages of characters, in form, color and ar- 

 rangements of parts under natural conditions, which are 

 recognizable from descriptions and figures, should re- 

 ceive distinctive names and be catalogued, provided, of 

 course that the assemblage of characters includes all onto- 



