ON THE CAUSES OF VARIATION. 493 



nection with the necessities of life. Neither touches the origin of 

 the variation, but both express laws thereof or methods by which 

 it is accumulated. The inherent tendency to vary, whether in 

 external or adaptive structure, or internal or reproductive char- 

 acter, is simply an observed fact, the causes of which we are en- 

 deavoring to analyze. 



Physiological selection is remarkably exemplified in insects, 

 and probably in no other class are the modifications which may be 

 attributed to it more easily studied ; for in no other class are the 

 genitalia of the male so variable or so complex. There has so far 

 been no attempt to homologize the different parts in the different 

 orders of insects, so that they have received different names ac- 

 cording to individual authors. Ordinarily there are two pairs of 

 claspers, themselves very variable, associated with sundry hooks 

 and tufts of hair. There are families, as in the Cecidomyidce, 

 among the Diptera, in which many species are almost, and others 

 absolutely, indistinguishable except by the differences in the male 

 genitalia. In all other orders there are an immense number of 

 forms which can only be distinguished by a careful study of those 

 organs. Descriptive entomology to-day, which does not take ac- 

 count of these organs, is in fact almost valueless, and we must 

 necessarily assume that, where there is differentiation of structure 

 in these important parts, it implies a corresponding modification 

 on the part of some associated female, even where no other differ- 

 entiated characters are to be detected, and upon Romanes's law 

 such must be looked upon as physiological varieties, and will be 

 counted good species in proportion as the differentiation involves 

 other observable characters or as their life-habits determine. 



Sexual Selection. — The part of sexual selection in inducing 

 variation may next be considered. While it is evidently at the 

 bottom of the diversity in sex so common among many animals, 

 it is difficult to see how it can play any very important part in 

 the differentiation of species, except on the hypothesis that the 

 greater the differentiation between the sexes the greater the tend- 

 ency to vary in the offspring. In no class of organism is this 

 factor more notable than in insects, and volumes might be written 

 to record the interesting and curious facts in this class alone. 

 As a general rule it may be said that with insects, as with other 

 animals, it acts chiefly in inducing secondary sexual characteris- 

 tics in the male, and in simplifying the characteristics of the 

 female. Nowhere do we find greater contrasts between the sexes, 

 involving almost every organ, both colorationally and structur- 

 ally. Where color is affected, the greater brilliancy almost 

 always belongs to the male sex, as in birds. So, where song or 

 sound is employed to attract, the sound-organs are either peculiar 

 to, or most highly developed in, the males. As in higher animals, 



