240 THE DESCENT OF MAX. [Part I. 



been important, they would long ago have been either 

 fixed and preserved, or eliminated. In this respect man 

 resembles those forms, called by naturalists protean or 

 polymorphic, Avhich have remained extremely variable, 

 owing, as it seems, to their variations being of an indif- 

 ferent nature, and consequently to their having escaped 

 the action of natural selection. 



We have thus far been bafHed in all our attempts to 

 account for the differences between the races of man ; but 

 there remains one important agency, namely, Sexual Selec- 

 tion, which appears to have acted as powerfully on man 

 as on many other animals. I do not intend to assert that 

 sexual selection will account for all the differences be- 

 tween the races. An unexplained residuum is left, about 

 Avhich we can in our ignorance only say that, as individ- 

 uals are continually born with, for instance, heads a little 

 rounder or narrower, and with noses a little longer or 

 shorter, such slight differences might become fixed and 

 uniform, if the unknown agencies which induced them 

 were to act in a more constant manner, aided by long-con- 

 tinued intercrossing. Such modifications come under the 

 provisional class, alluded to in our fourth chapter, which 

 for the want of a better term have been called sponta- 

 neous variations. Nor do I pretend that the effects of 

 sexual selection can be indicated with scientific precision ; 

 but it can be shown that it would be an inexplicable fact 

 if man had not been modified by this agency, which has 

 acted so powerfully on innumerable animals, both high 

 and low in the scale. It can further be shown that the 

 differences between the races of man, as in color, hairi- 

 ness, form of features, etc., are of the nature which it 

 might have been expected would have been acted on by 

 sexual selection. But in order to treat this subject in a 

 fitting manner, I have found it necessary to pass the 

 whole animal kingdom in review ; I have therefore dc- 



