18 ON THE WECFOOT CHARACTER IN PIGEONS, 



If we take the ISToi-mal-Web matings, as in experiments 2,6, 

 9, 10, and 13, we find the total results to be 11 Normals to 

 10 Webs. In this case were the Normals heterozygotes we 

 should expect equality, which is practically the result attained. 



Lastly, when we mate Webs to Webs, as has been done in 

 Exps. 1, 5, 8, 11, 12, and 14, we should expect Webs only, and 

 this result was only attained in one case (Exp. 8), where only two 

 birds both webbed were reared. In the other five cases together 

 15 birds were reared, 12 Normal and 3 Webbed. The proportions 

 are almost exact for a DR x DR mating, which, of course, we 

 know they cannot be ; but it is not without significance to note 

 tha,t in four out of these five experiments the matings Avere a 

 cross between the mixed strain and Mr. Staples Browne's strain. 



These, then, are the facts, and at present it seems difficult 

 to reconcile them with the Mendelian theory, although they 

 certainly seem to show that the inheritance of this character is 

 to a cei'tain extent in a,ccoi-dance with that theory. 



According to previous and contemporary work on this character, 

 no individual should have had a normal foot. If, however, the 

 Normals that have appeared be considered as heterozygotes, then, 

 considering the small numbers, the results from the matings are 

 not greatly at variance with what we should expect. 



As regards the Webs (in these experiments), when mated with 

 Normals they behave as recessives, but when mated with other 

 Webs of either the mixed or of Staples Browne's strain they 

 produce Normals to Webs in a ratio closely approximating 3:1. 

 Such a result in the case of no less than five matings seems to 

 show pretty conclusively that the joining of the two strains has 

 produced a factor disturbing the normal course of the Mendelian 

 inheritance. 



It must be carefully noted that each strain by itself breeds 

 true according to Mendel's laws and that the disturbing factor is 

 not brought in from one side only, but is due to the intermixture 

 of the two strains. 



Mention should perhaps be made of a suggestion by Prof. Bate- 

 son, to whom I wrote on the subject. This was that the webbed 

 foot was possibly a double character and that the web between 

 digits 2 and 3 had a separate inheritance from that between digits 

 3 and 4. Now, if we look at Exp. 1 in this light, we find that 

 in I the male is very slightly webbed (practically normal) and 

 that in | he is fully webbed. In the female, on the other hand, 

 the webbing in | is very full and in | extremely small. 



We could thus suppose the male in Exp. 1 to be | N (W) 

 I WW, female § WW | N (W) ; this would then give in F, 

 ^ Noi'mals and Webs equally, | Normals and Webs equally, so 

 that we might expect pure Normals (| N | W, | W | N) and pure 

 Webs in equal numbers — as a matter of fact, 4 were Normals and 

 IfNfW. 



This suggestion, however, becomes untenable for other reasons. 

 If we examine the birds I bred for Mr. Staples Browne (Exp. 14, 



