1 66 Sex in Bryony [ch. 



by the composition of the gametes of which the zygotes are 

 formed. 



Naturally therefore we look for a simpler solution of 

 the problem of sex-determination. As was stated above, it 

 would a priori seem most probable that one sex is hetero- 

 zygous in sex and the other homozygous. I now propose 

 to consider the applicability of this simpler account. 



Two lines of work have been followed, that of experi- 

 mental breeding, and that of cytological study of the germ- 

 cells. Both have led to very positive conclusions, but these 

 conclusions are diametrically opposed, as will be immediately 

 seen. From experimental breeding we are on the whole led 

 to conclude that the types used have females heterozygous 

 in sex (female being dominant) and males homozygous 

 recessives, while in all cases in which cytological evidence 

 is forthcoming it appears that the females are homozygous 

 and the males heterozygous. The meaning of this curious 

 discrepancy will be considered when the facts have been 

 related. 



Evidence from Breeding Experiments. 



The Case of Bryonia. 



By the nature of the case, direct Mendelian experiment 

 cannot be applied. There are however certain indirect 

 lines by which the problem can be approached. The first 

 is that followed by Correns in cross-breeding monoecious 

 or hermaphrodite types with dioecious or bisexual types. 

 In this connection the most striking experiment is that 

 which he made by crossing Bryonia dioica with B. alba 

 which is monoecious. The reciprocal crosses gave with 

 much consistency a surprising difference in results. B. 

 dioica, female, fertilised with the pollen of B. alba gave F^ 

 females, with or (usually) without occasional male flowers. 

 The observed numbers from this mating were 589 females, 

 with 2 males which must be regarded as exceptional. In 

 the reciprocal cross, alba (monoecious) used as female 

 X dioica $ gave jF^ consisting of males and females in 

 approximately equal numbers. In Cambridge we have 

 repeated both experiments and obtained the same results. 

 Unfortunately the hybrids, however produced, are absolutely 



