NATURAL VARIATIONS 135 



In the first case we are dealing with the fundamental 

 question of Predominance, and the variations may be called 

 Natural Variations ; in the second case the question is 

 directly related to the action of Environment, and the 

 variations are essentially Acquired Variations. This dis- 

 tinction is a most important one. It is, however, to be 

 noted that the latter variations, when fixed and " trans- 

 mitted," as it is clear they can be, become characters in the 

 descendants of the organism first exhibiting them ; and 

 their reappearance will be subject to natural variation. 



Natural Variations. 

 These might be roughly grouped in three classes. 



I. The offspring inevitably varies from its parents. 

 II. The offspring of different parents inevitably vary 



from each other. 

 III. The successive offspring from two parents while 



never exactly resembling the parents, never 



exactly resemble each other. 



Did we hold that such a thing as a unicellular Individual 

 obtains in Nature, we should conclude that any variations 

 it ever exhibited were the direct result of altered Environ- 

 ment ; and further, that if the environment after alteration 

 remained absolutely constant, the variations would pass on 

 as fixed characters to the organism's descendants. Where, 

 however, the Individual results from a sexual process of 

 some sort, there is, the acquisition of variations apart, 

 inevitable natural variation. For the uniting gametes have 

 their own special potential characters (based on actual 

 special characters), and as the Individual's somatic tissues 

 evolve from the strife for the reproduction of both uniting 

 elements, the characters of both parents will be exhibited 

 by the offspring. The proportional representation will, of 

 course, vary, and special factors can influence greatly the 

 degree of parental resemblance as a whole. But the inevit- 

 able result always is that the offspring varies in many ways, 

 some recognisable, some not, from both its parents. 



Where the Individual type is unisexual the offspring 

 may be said to exactly resemble one parent, and to vary 

 from the other, in the matter of Sex. This we would regard 



