404 Scientific Proceedings, Royal Dublin Society. 



were each dominant to the single, and it was expected that either rose or 

 jiea would be dominant the one to the otlier — " that either rose or pea would 

 dominate in the hybrids, and that the F2 generation " (i.e. the second crosses) 

 " would consist of dominants and recessives in the ratio 3 : 1.'" But the 

 expectation was not fulfilled, for, when rose and pea were mated, their first 

 crosses were a new kind of comb — walnut — and tlieir second crosses consisted 

 of four kinds, namely, walnut, rose, pea, and single — also new — in the ratio 

 9:3:3:1. This ratio shows that there are really two pairs of alternative 

 characters concerned in the case. The characters and their factors may be 

 difficult to identify, and, because of this difficulty, they can be represented 

 in the meantime by unknown symbols only. By so representing tliem, we 

 may be able to trace the connections between some of the determinants and 

 to get some idea of their effects. The formula to meet the case is : — 



No single character can be identified, nor can the effect of any factor be 

 told. All that can be said is that walnut results with the concurrence of X 

 and Y, rose with X and ;(/, pea with x and Y, and single with x and y. 

 Nor can it be said how far any factor is responsible for the character 

 produced. How far X or how far Y, for instance, is responsible for walnut 

 there is no evidence to show. And no more can it be said, since each is the 

 result of more than one factor, that rose or pea is dominant tlie one to the 

 other, or that either is dominant to single. What can be said is that a 

 factor in the rose-comb is dominant to a factor in the single, and a factor in 

 the pea is dominant to another factor in the single. 



But from matings between some of these combs and another kind — the 

 Breda — further information can be gathered. The fowl with this comb " is 

 usually spoken of as combless, for the place of the comb is taken by a 

 covering of short bristle- like feathers. In reality it possesses the vestige of a 

 comb in the form of two minute lateral knobs of comb-tissue." - When this 

 comb is mated with rose on the one hand, or single on the other, the first 

 crosses have two points in common. The progeny of the roses are still roses, 

 but split in two ; and the progeny of the singles are still singles, but also split 

 in two. Tlie Breda comb, therefore, carries a factor for splitting or 

 duplicity which is dominant to a factor for non-splitting or simplicity carried 

 by both the rose and single combs. Then, if we represent the duplicity 



' Punnett's Mendelism, 3rd ed., p. 29. - Idem, p. 35. 



