THE MUTATION THEORY 337 



6. The direction of mutation in a given gene is, however, preferen- 

 tial, occurring oftener in some directions than in others. This is well 

 illustrated in the studies on variegated corn and four-o'clocks, and on 

 the bar eye and white eye and other series in Drosophila. 



7. The mutability and preferential direction may themselves be- 

 come changed through mutation, as illustrated by some of the same 

 cases. 



8. The mutations do not ordinarily occur in two or more different 

 genes at once. In only two instances in Drosophila have mutations 

 been found in two different, separated 1 genes in the same line of cells 

 of one individual. But a recurrent case, apparently of this kind, has 

 recently been described in oats, by Nillson-Ehle. 



9. Not only does the mutation usually involve but one kind of 

 gene — it usually involves but one gene of that kind in the cell. That is, 

 the allelomorphs mutate independently of one another, just as totally 

 different genes do. There is evidence for this derived from corn, 

 Portulaca, and Drosophila. 



10. Mutations are not limited in their time of occurrence to any 

 particular period of the life history. This has been proved in the above 

 mentioned studies on mutable plants, in Drosophila, and in other cases. 



11. Genes normal to the species tend to have more dominance than 

 the mutant genes arising from them. This is very markedly the case 

 in Drosophila, where even the relatively few mutant genes that have 

 been called dominant are very incompletely so, and might more justly 

 be called recessive. In other organisms, the same condition of things 

 is strongly suggested, although the direct data on occurrence of muta- 

 tions is as yet too meagre to allow of certainty. 



12. Most mutations are deleterious in their effects. This applies 

 not only to the organism as a whole but also to the development of any 

 particular part: the delicate mechanisms for producing characters 

 are more likely to be upset than strengthened, so that mutations 

 should more often result in apparent losses or retrogressions than in 

 "progressive" changes. This is both an a priori expectation and a 

 phenomenon generally observed. 



13. Mutations with slight effects are probably more frequent than 

 those with more marked effects. This must not be understood as re- 

 ferring to the different mutations of each given gene, but it applies in a 

 comparison of the mutations occurring in different genes. Thus, there 



1 Contiguous genes may be affected in the rare cases known as "deficiencies," 

 found by Bridges and Mohr. 



