ADAPTIVE VARIATIONS. 389 



of all the changes would be rather in the direction of 

 adaptation to the new surroundings, than in that of 

 non-adaptation. It is not to be supposed for a moment 

 that every one of a group of organisms exposed to 

 new conditions of life will become better adapted to 

 them than any one of the group had been originally; 

 but merely that the characters of the group will, on an 

 average, become better adapted than they had been be- 

 fore. Doubtless many instances can be thought of in 

 which the effect produced by a change of environment 

 has no appearance of being in the least adaptive, but 

 this may be due to our ignorance of what constitutes an 

 adaptation. For instance, it may be asked in what 

 way a starved animal is better adapted to semi-starva- 

 tion than a well-nourished one? It is, of course, less 

 adapted in that it has, stored up in its body, less food 

 material such as fat and glycogen on which it can 

 live, but it is obviously better adapted in that its metab- 

 olism is considerably smaller than that of a well- 

 nourished animal; i. e., it actually lives on considerably 

 less food. Again, it may be asked in what way a dusky 

 coloured Polyommatus phlccas is better adapted to a 

 warm climate than a copper-coloured one, and vice 

 versa with reference to a cold climate? Possibly there 

 is nothing adaptive about the colour of the wing scales, 

 but doubtless it would be found that, on an average, the 

 dusky butterflies could withstand a greater degree of 

 heat than the coppery ones, and the coppery ones a 

 greater degree of cold. Hence the change would, on 

 the whole, be in the direction of adaptation. 



It is probable that somatic variations, by reason of 

 their adaptation to changed surroundings, are of very 



