50 VARIATION. 



necessary. If we can only see one link and the weight which 

 hangs from the chain, we may feel sure that this link is in itself 

 responsible for holding up the weight, and if we are fortunate 

 enough to see two or three we may think that a cooperation of 

 these two or three is necessary and sufficient. The difficulty in 

 our analysis comes in because of the circumstance, that as long 

 as a gene is actively doing its work we cannot study it as 

 such. We can on the other hand, not study a gene apart from an 

 organism carrying it, we can only compare the development of 

 two organisms, of which one had the gene in its germ, whereas 

 it was lacking from that of the other. So long as a group of 

 plants or animals are all pure for a given gene, there is no pos- 

 sibility of studying it. 



In such a process as pigment- formation, we know in some 

 animals about ten different genes which all play their own role 

 in the process, and we may study the effect of each of these 

 genes on the development of animals which have different com- 

 binations of other genes. But we have no idea of the number of 

 genes actually concerned with the production of pigment. In 

 the first place we cannot study any genes which are not absent 

 from at least one individual, and in the second place we can- 

 not know in how far other genes which we have actually been 

 studying in their effect upon things other than pigmentation, 

 may influence these processes. 



Ordinarily, we must see that variation, the diversity, which 

 we observe in a group of organisms may be either due to a differ- 

 ence in genotype, or to a variation in the environment, or to a 

 combination of the two causes of variability. A population, 

 consisting of animals or plants which have either one or the 

 other of two genotypes may vary continuously, because of the 

 fact that both bio-types are modified in both directions by the 

 environment. But if we are dealing with a population in which 

 all the individuals have the same combination of genes, the 

 same genotype, all variation in this population must be due to 

 the action of the environment exclusively And in such a popu 

 lation we can see if evolution of some kind is still possible. 



