VI. SELECTION AND ISOLATION 383 



Here, the well-known experiment of Castle carried out with hooded 

 rats should be mentioned. He was able, through selection in succes- 

 sive generations, to extend or diminish the black of the colour pattern 

 of the rats. In his experiment it was clearly shown that the selected 

 rats having the extreme colour-pattern tend to push the extremity in 

 successive generations, the colour pattern becoming more and more 

 conspicuous. His observation was spread over about 20 generations of 

 the rats. Another well known example of this kind is Tower's ex- 

 periment with Colorado beetles in which he likewise succeeded in in- 

 creasing the peculiar colour pattern by selection. 



Many other evidences can be presented to show the effect of the 

 selection, but it appears that they were not adequately interpreted, 

 presumably because of the general acceptance of Johannsen's pure- 

 line theory, which, however, the writer is unable to accept as being 

 legitimate. 



The beans with which Johannsen carried out the experiment and 

 also the paramecia with which a similar result was obtained afterwards 

 by another author might have been taking place a gradual change so 

 slowly that the progress of the change was overlooked. It should be 

 noted, however, that even during the experiments of Johannnsen, two 

 mutations, one for higher and one for lower seed-weight, were de- 

 tected ; this may show that his selection would produce high and low 

 seed-weight respectively. Granting that such mutations were only rare, 

 it may also conceivable that the peculiar condition under which the 

 observations were made might remove the factor which was causing 

 the gradual change. It should be remembered that the observations 

 were made with the descendants from a single homozygous organisms 

 exclusively propagating by self-fertilization. Whatever might be the 

 reason why apparently unusual result were obtained for which the 

 pure-line theory was derived, there seems little doubt that his findings 

 were exceptional since as just stated there are many evidences showing 

 the striking effect of the selection. 



However, most examples in which the selection appeared to be 

 effective were commonly interpreted as being due to the accumula- 

 tion of minute gene mutations occurring in a certain direction. Such 

 an interpretation, on the other hand, may be proper as it may not be 

 entirely impossible to consider the gradual alteration of the gene to 

 be an accumulation of minute mutations. 



At any rate, the genes in each individual even under the same 

 environment may perform a change peculiar to the individual according 

 to the individual difference in the structure of the genes, giving rise 

 to a peculiar character. When such a peculiar character becomes to 

 have a survival value, it will be selected by natural selection. Since 



