180 LOCK : STUDIES IN PLANT BREEDING 



rp x rp gave 9rp + 3r + 3p + Is ; rp x s gave equal 

 numbers of rp, r, p, and s ; and in a particular case true 

 Malay x single gave the same result. 



The authors express these results by supposing a " com- 

 pound allelomorph " rp, which splits up giving off equal 

 numbers of gametes rp, r, p, and s ; in a manner identical 

 with that suggested by Bateson* in his criticism of De Vries,f 

 who had explained similar proportions obtained with colours 

 of the flowers in Antirrhinum in a purely Mendelian 

 fashion. 



It appears to me that the formula of De Vries (which is also 

 that of Mendel) is to be preferred in both these cases, and that 

 the idea of " resolution of a compound allelomorph " is quite 

 unnecessary in either of them. 



Quite recently Bateson and Punnett have published a sug- 

 gestion as to the nature of the " walnut " comb J of which the 

 following paragraph (which was written however before the 

 appearance of their note) may serve as a summary. 



If we write : r= Ab ; p=aB ; s =ab. A and a being a pair 

 of allelomorphs of which A is dominant, and B and b in like 

 manner, the artificial Malay will be ABab and the pure Malay 

 A BAB. The whole phenomenon then resolves itself into a 

 simple case of dihybridism. 



This view is further supported by the following consideration. 

 The single comb being the primitive form , it is almost neces- 

 sary to suppose that both rose and pea have arisen from it 

 separately by mutation ; there seems therefore to be no inherent 

 difficulty in treating rose— single and pea— single as separate 

 pairs of allelomorphs ; and the rp appearance would then be 

 induced by the simultaneous presence of A and B, whilst in the 

 absence of both of them the primitive single comb would be 

 • ■xpected to appear. 



* Bateeon, W. Note on the resolution of compound charnrtors by 

 t toss- brooding. Proc. (Jamb. Phil. Bee. XII., p. 69. 



t Do Vries, H. Mutations-theorio J I., p. I'J7. 



t Batoson, W., and Punnett, R.C., in Proc. Camb. Phil. Soc, Octo- 

 ber 2 • 190 



