R. C. PUNNETT 315 



{Text-fig. 1, p. 9). If the dark and light belong to a system of multiple 

 allelomorphs segregating in the normal way we should not expect 

 this approximation, and I gather that it is in explanation of it that 

 Castle suggests the hypothesis of " mutual modification." On the inter- 

 pretation I have suggested there is no need to postulate any such 

 problematical process. The appearance of such grades as 9-14 in the 

 " F^ " generation may be supposed to be due to the fact that some of the 

 parents, especially those belonging to the higher numbered grades, were 

 heterozygous for S as well as for P. From such animals bred together, 

 on the assumption that normal and independent segregation of the 

 several factors occurred, we should expect various intermediates of 

 grades 9-14. Further we should also look for a small proportion of 

 animals tending towards complete pigmentation. For such might be 

 expected to occur when P reacted on a basis of Typical Dutch. On 

 Castle's explanation in terms of multiple allelomorphs, particularly 

 when the hypothesis of " mutual modification " is superadded, it is 

 difficult to understand why these highly pigmented rabbits should 

 appear in the " F^ " generation. 



The result of mating F^ (ex Dark x White Dutch) with White 

 Dutch was the production of two groups of about equal size, but 

 slightly overlapping {Text-fig. 1, p. 9), so that every grade between 4 

 and 17 was represented. On the assumption that the dark and light 

 belong to a system of multiple allelomorphs it was to be expected that 

 the two groups would be sharply separated, and that their mean grades 

 would be close to those of the parental stocks. Here again Castle finds 

 evidence for " mutual modification," but, as in the case of the " Fo " 

 generation, there is no need for this assumption on the interpretation 

 that I have suggested. 



(3) The cross between Tan Dutch and White Dutch {pp. 13~lJf). 

 From this cross the F-^ individuals were intermediate in grade ranging 

 from 6-9, the range of Tan Dutch being 2-5, and that of White Dutch 

 15-17. The Fo, generation comprising 130 individuals, forms a con- 

 tinuous and almost regular curve on grading, with the mode very close 

 to the mean, and with a range of 3-14. Animals ex F^ x White Dutch 

 exhibit a range of 7-17. All of these results are in accord with the 

 supposition that Tan Dutch and White Dutch differ in several factors 

 from one another, such factors being all present in Tan Dutch but 

 absent from White Dutch. Castle supposes that Tan Dutch belongs to 

 the same multiple allelomorphic system as Self-colour, White, and Dark 

 Dutch. The experimental facts appear to me to be entirely opposed to 



