REDUCTION OF VARIABILITY. 139 



logy we could quote several instances of gradual change of char- 

 acters throughout the range of a "Paarungsgenossenschaft" of 

 animals, the case of the gopher, of the field-mice, of the garter- 

 snakes, fox-sparrows. It is significant that animals with greater 

 power of dispersal are more apt to be monotypic throughout 

 the whole of the area into which they are confined through 

 reasons of ecology. Jack-rabbits and coyotes are less liable to 

 produce several local forms within a territory without barriers 

 than the slower moving and less roving animals. 



Where the "type-specimens" are really commoner than "in- 

 termediary stages" these types may be called species in good 

 accordance to my definition of the term. In practice it is be- 

 lieved that a new species upon which a new specific name is con- 

 fered, is a stable group, somewhat variable perhaps, but cen- 

 tering about a mass of individuals all like the type-specimen, 

 and thought to have a vaguely appreciated tendency to remain 

 in this condition. Our definition of species, as groups of organ- 

 isms, so constituted and situated, that they tend, under con- 

 ditions, which promise to be permanent, to reduce automatic- 

 ally their potential variability also defines the species of the 

 taxonomist. 



