THE ORIGIN OF VARIATION 21 



available food and the temperature at which development 

 occurs, but also by numerous genetic factors. It seems also to 

 be the case in some of the naturally occurring strains oiDaphnia 

 studied by Woltereck (1908) ; e.g. the low-helmed form from 

 the Lund See could be easily transformed into a high-helmed 

 form, but the apparently similar variant (mutant E) of the 

 Frederiksburg See could not be modified by the same conditions. 



These facts are of some importance. In the minds of most 

 workers there is a general idea that animals live in a variety 

 of places and are exposed to a diversity of environmental factors 

 that produce a great amount of merely somatic modifications 

 — that all animals are in varying degrees plastic and receive 

 a more or less marked amount of modification from the food 

 they eat, the soil on which they live, and so on, and that much 

 variation is without moment in evolution, because it is not 

 heritable. The assumption that animals are plastic is no 

 doubt a sound one ; but each case ought to be considered on 

 its own merits and tested by experiment. 



In practice what is done, in taxonomy at least, is to 

 proceed by no particular principle except some such idea as 

 that, if a short form of a marine Gastropod (e.g.) is found in 

 brackish water, it is a ' stunted ' (somatic) form. The result 

 is that species and their variation are described according to 

 the systematist's very varying knowledge of experimental 

 work. This is, of course, a matter of systematic procedure ; 

 but it is important, as to a certain extent the work of the 

 systematist is taken as evidence of the plasticity of animals. 

 As we suggest later (p. 55) we do not know if this plasticity is 

 actually without evolutionary significance. Moreover, most 

 workers would probably agree that more of the alleged 

 fluctuations are hereditary than was at one time supposed. 



The role of intrinsic and extrinsic factors in the production 

 of fluctuations deserves considerably more attention than it 

 has yet received. Investigations are often carried out under 

 insufficiently standardised conditions and there is a consequent 

 tendency to attribute variation to unknown differences in the 

 environment. Again, there is usually a considerable probability 

 that the species studied are genetically very diverse. The two 

 loopholes so provided are quite sufficient to prohibit much 

 generalisation. It would, however, be a matter of some interest 

 to discover how far. variation can be eliminated by rearing 



