184 MR. J. T. CUNNINGHAM UN RESULTS OF A 



size of crest is, however, n fact worth recording, and is anything 

 but simple dominance in the heterozjgote. The single comics I 

 obtained appeared to be pure recessives, but there were sca,rcely 

 any singles in the later generations. 



Similar remarks apply to leg-feathering. There are many 

 cases of total absence, and these may l)e regarded as pure 

 I'ecessives. But, again, there are individual variations, such as 

 very slight, slight, moderate, and complete. Tliei-e are also three 

 cases, F, I 5 d' , F, III 2 $ , and F- lY 1 6 , in which there was only 

 an infinitesimal trace of feathering on the legs. The parents of 

 the last both had " clean " legs, i. e. no leg-feathering at all, and 

 all the others of F. lY were destitute of feathers on the legs. 

 The case is interesting as showing that a very minute degree 

 of a, dominant character may appear in the offspidng of two 

 recessives. In previous experiments with Japanese Long-tailed 

 fowls I noticed a similar minute trace of feathei'S in chicks of 

 this breed, which normally has no feathers on the legs. These 

 facts tend to support the view that miiuite degrees of a chai'acter 

 may occur in individuals which are not heterozj^gous, and that 

 such a character as complete leg-feathejing is not necessarily an 

 indivisible unit. 



The irregularity of dominance in the double hallux \\i\s 

 mentioned in my previous paper. The later generations show 

 that the heredity of this character is so irregular that it is 

 impossible to distinguish the recessive from the heterozygote. 

 The parents of F., Ill, for example, both liad normal toes on both 

 feet, and of the eleven chicks ten had noi'mal toes and one had 

 the double hallux on both feet. It is evident that one or both of 

 the parents was heterozygous for this character with the normal 

 character dominant. If the chick with double hallux is a 

 recessive, both of the parents must have been heterozygous, and 

 in that case the number of recessives should have been 1 : 3 

 instead of 1 : 10. 



The parent of F^I and of F^ II all had normal toes, yet in the 

 foi-mer brood two out of six had the double hallux, in the latter 

 all seven had normal toes. The female parent of Fg VI had a 

 double hallux on both feet, the male parent had normal toes, and 

 .six of the seven surviving chicks had a double hallux on both 

 feet, but in one of them the hallux was only slightly cleft, and 

 in the 7tli the hallux was double only on the left foot. This 

 case would agree with theoiy if the double hallux in the one 

 parent was a pure dominant and the normal feet in the other 

 pui-e recessive. In other cases it i.s impossible to tell whether a 

 normal is pure recessive or heterozygous. In several cases both 

 parents with normal toes give chicks all normal, e.g'. F^VIII, 

 F3IV, F, V, F,XI ; yet F3III, the parents of which were both 

 from F. VIII. produced three chicks with normal toes to four with 

 the double hallux. It would seem not merely that the normal 

 may be dominant, but that segregation sometimes occurs and 

 sometimes does )iot. 



