Studies on variation and selection. ItQ 



Is there any possibility, that, notwithstanding these very con- 

 clusive experiments, it will be found, that in other cases, it will never- 

 theless be possible to modify a genetic factor by selection? A priory, 

 there is something to be said for this possibility, though it seems 

 probable that the invariability of the nature of genes is the same 

 in all organisms. In any case, only such experiments should be able 

 to prove the contrary, in which the material was at least as pure 

 as in the experiments showing the immutability of genes. 



It is easy to see, that in the first years of Genetics, when we 

 only as yet studied the "unit-character-genes", and it was not known 

 that some genes have only very little influence on the development, 

 the fact that selection may modify all sorts of characters gradually, 

 was easily held to prove, that it was possible that genes were changed 

 in this way. 



Castle experimented with partially albinistic rats. In these rats, 

 the pigment occurs on definite areas on the body. In those who 

 have the least of it, the anterior half of the body is pigmented, with 

 the exception of a white blaze and white forefeet. When there is a 

 little more of the body pigmented, first the white blaze disappears, 

 the next have a median dorsal stripe, which in other rats extends 

 down upon the sides of the body, in some reaches about half-way 

 down, whereas the still more heavy pigmented ones only have the 

 belly and feet white. Next come rats, which have only a narrow white 

 stripe on the under side, and white feet, which stripe becomes a spot 

 in some others and finally disappears, and lastly there are some rats 

 which only have one white foot and such as are pigmented all over. 



We find that most wild Mus norvegicus have some white, in 

 fact, the only completely pigmented rats we have ever seen, we have 

 bred ourselves. The darkest pigmented rats are generally called 

 "Irish" in the fancy, the lightest "Hooded". Don'caster, who has 

 extensively interbred Irish and Hooded rats, states, that only once 

 he has hesitated to class a rat as Hooded or as Irish. Personally 

 we have had many rats which we would be tmable thus to classify. 

 Generally, when Irish and Hooded rats are crossed, the hybrids are 

 Irish, and when mated inter se, these give again a number of Hooded 

 young, about 25 %. 



From these facts. Castle concludes, that the "Hooded character 

 is a unit-character". This he clearly means to be understood to mean, 

 that Hooded rats differ from Irish rats in one genetic factor only, 

 and that all Hooded rats belong to one genotype. 



