658 proceedings of the academy of [oct., 



Racial Variation among Terri^strial Vertebrates. 



Variation in animals may be of several different types: (1) sexual, 

 (2) age and seasonal, (8) dimorphic, (4) individual, albinistic, etc., 

 and (5) specific or racial. 



We have a terse nomenclature by which differences of sex, age, etc., 

 may be denoted, and sometimes, as in birds, quite a complicated 

 terminology by which every plumage is designated (cf. Dwight, Auk, 

 1902, p. 248), but our binomial and trinomial nomenclature is used 

 only for specific or other racial variations. 



Among terrestrial vertebrates racial variation corresponds closely 

 to geographic environment, and in many groups it is very easy to recog- 

 nize the effect of the environment of several different life areas in pro- 

 ducing recognizably distinct races from the same type. 



In birds and mammals this correspondence is most marked, though 

 among them we find some genera much more plastic than others ; the 

 song sparrows {M elosjAza) , for instance, breaking up into a very large 

 number of forms, while the robins {Merula) are remarkably constant 

 over large areas. In birds and mammals the individual variation in 

 size, after making due allowance for age and sex, is exceedingly shght, 

 and the same may be said of color, provided the additional allowance 

 for season is made, so that very slight differences in measurements 

 or in shades of color, which might appear trivial, are really constant 

 and perfectly reliable as indications of the differentiation of a distinct 

 form. 



Among reptiles individual \ariation is very much greater, and 

 geographic races cannot be so clearly designated as in the classes 

 just considered. Measurements are of little significance, except in 

 the relation which one dimension may hold to another; color and scale 

 formula are also subject to grea.t variation. Some species, however, 

 vary much more than others, and in slightly differentiated forms with 

 a high percentage of individual variation extremes may overlap in 

 certain respects, or an occasional individual may revert to an ancestral 

 type in some character or other, without affecting the fact that a race 

 has been differentiated. Such individuals are referable to the same 

 category as geographic intcrgrades in the usually narrow belt, where 

 hfe-areas which have given rise to two forms merge one into the other. 



In the Batrachia individual variation is further complicated by the 

 metamorphosis which is often responsible for the persistence of certain 

 early (larval) characters in the adult, and that many species in this 

 class have been based upon this and other individual variations I 

 have little doubt. 



