256 MR. J. T. CUNNINGHAM ON 



expectation on my hypothesis are an excess of fully pigmented 

 females and of partially pigmented and unpigmented males, with 

 a deficiency of fully pigmented males. These differences would 

 be explained if segregation were, as 1 have shown it to be in my 

 experiment, incomplete, and at the same time pigmentation tended 

 to develop more in the females. Thus in the females the WW's 

 would be partially contaminated by P, and some of them would 

 be partially pigmented instead of unpigmented, while some of the 

 WP's would appear as fully pigmented instead of partially so. 

 In the males, on the other hand, the PP's would be affected by W, 

 and this would reduce the number of fully pigmented in that sex 

 and increase the number of partially pigmented. 



We have now to consider the F,^ generation from the mating 

 B in which the Silky is the male and the Brown Leghorn the 

 female. According to the formula of Bateson and Punnett the 

 expectation is 3 fully pigmented, 3 partially pigmented, and 

 2 unpigmented in each sex. This expectation is fairly well 

 fulfilled in the males, but in the females there is a great excess of 

 fully pigmented over the partially pigmented : the former are 

 nearly half as many again as the latter, and this the authors call 

 a slight excess. The complete numbers are : — 



According to my hypothesis the F^'s are 



WP c? , Wp P $ , 



and the gametes 



W+P, Wp + P. 



The zygotes therefore will be 



W Wp, W P, Wp P, p p. 



The first and last of these are the same as those which occurred 

 in the original parents, while the second and third are the zygotes 

 of the male and female F/s. This is the ordinary result of 

 Mendelian segregation in the F^ : one pure dominant, one pure 

 recessive, and two heterozygotes. The difficulty is to decide what 

 will be the distribution of the sexes among these zygotes. If we 

 suppose that Wp is always coupled with the female character, 

 then ail the W Wp's and Wp P's will be females, and all the rest 

 males. But this would make not only the W P's but also all the 

 P P's male, and it seems improbable that the pure Silky character 

 in the F, should be confined to one sex. We may conclude that 

 the P P's will be of both sexes and likewise the W Wp's. We 

 may assume that the heterozygotes W P and Wp P will be in the 



