MEINDELIAN EXPRRIMEXTS WITH FOWLS. 257 



same condition as in the F^, the former all males, the latter all 

 females. Thus we shall have for the males 



W Wp, 2 W P, P P, 



and for the females 



W Wp, 2 Wp P, P P. 



In the males, therefore, the ratio should be one unpigmented, two 

 partially pigmented, and one fully pigmented. It must be ad- 

 mitted that this does not agree with tlie actual results obtained 

 by Ba.teson and Punnett; but if segregation is not always complete, 

 some of the W P's might appear in P^ as fully pigmented and 

 would then be added to the P P's, and the number of partially 

 pigmented and fully pigmented might be equal. 



In the females on my hypothesis there should be one unpig- 

 mented to thi-ee fully pigmented, and none partially pigmented. 

 This again does not agree with the actual figures ; but if segre- 

 gation is incomplete, the P in the Wp P may be afiected by 

 the W, and so a certain number of females might appear as 

 partially pigmented. In any case there is actually a great excess 

 of fully pigmented females over the partially pigmented, whereas 

 according to Bateson and Punnett the numbers should be equal. 



There is another point in Bateson and Punnett's figures of the 

 F„'s from this cross which may be of great importance, although 

 it is not noticed by the authors themselves. According to their 

 iiypothesis the total numbei'S of the sexes should be equal, whereas 

 they are actually 1 40 males to 167 females. Thus there is not only 

 an excess of fully pigmented females over the partially pigmented, 

 but a large excess, more than 19 per cent., of females over males. 

 This excess occurs in ahnost every family, and it seems improbable 

 that it should be accidental. In the F./s from the reciprocal cross 

 Leghorn c? X Silky 2 , on the other hand, there is an excess of 

 males, 75 to 69 females. It would appear that the Leghorn was not 

 only dominant as to character, but prepotent as to sex. I cannot 

 attempt to explain this, but it is not in accordance with Mendelian 

 theory. 



In any case we are justified in saying that the actual results 

 obtained by Bateson and Punnett in the F^ generation do not 

 ao'ree in either of the reciprocal crosses with their assumptions, 

 and that there is nothing in their results to prove that imperfect 

 segreo^ation may not occur in the cross of Silky with Brown 

 Leo-horn. It would scarcely he profitable to compare in detail 

 the results of all the other matings recorded by Bateson and 

 Punnett with those which would be expected on my hypothesis ; 

 the comparisons would be similar to those already given as 

 examples. It is sufficient to point out that all their explanations 

 are founded on the assumption that the pigmentation of the Silky 

 is inherited as an indivisible unit, and that the result of my 

 experiment shows that this assumption is contrary to fact. 



Proc. Zool. See— 1912, N^o. XYII. 17 



