238 



THE AMERICAN NATURALIST [Vol. XL11 



from it, as compared with other feral species, is a notable 

 addition to the difficulty of maintaining such a conclusion. 



As long as species were generally understood to be 

 relatively fixed in characters, their delimitation was rel- 

 atively simple, but the general understanding that all 

 living organisms are descended from others which were 

 different from them has greatly complicated the subject. 



Whether the evolution has been by imperceptible pro- 

 gressive modifications of structure, or by mutations, or 

 by both methods, the result is essentially the same from 

 the practical standpoints of taxonomy; from these stand- 

 points, then, similarity of individuals must remain the 

 consideration to which most weight will be given in 

 taxonomic usage. It has been conclusively proved that 

 many mutants and elementary species or races breed 

 true in enough instances to establish the rule for at least 

 a number of generations; this should not, however, in 

 my opinion, admit them to the category of species, which, 

 though necessarily difficult in delimitation, will still re- 

 main the practical taxonomic groups, recognizing, never- 

 theless, that they are made up of either relatively con- 

 stant or of widely fluctuating elementary components, 

 which, in turn, will presumably yield the species of future 

 geologic ages. 



The recognition of the existence of incipient or ele- 

 mentary species or races within the composition of 

 species, explains, in large part, the multiplication of 

 species and of groups of assumed lower rank, in many 

 of the larger genera, nearly every taxonomist, except the 

 most conservative, having taken more or less part in thus 

 increasing the number of descriptions and of names. 

 They have been variously denominated species, sub- 

 species, varieties, snbvarieties and forms, according to 

 the point of view of the investigator. 



Geographic distribution has been invoked as a very 

 useful aid in determining the limits of s]>ecies. It is a 

 well-recognized fact that certain areas of the earth's sur- 

 face, some large, some small, are characterized by types 



