204 
R. H. Compton. 
sterile with B, fertile with Q ; No. iv will he fertile with B, sterile 
with G, and No. viii will be fertile with both B and G. The ratio of 
types would therefore be 3 BG : 3 Bg ; 1 bG : 1 bg, instead of 
equality. The discrepancy would be increased the greater the 
number of inhibitory factors acting: thus for four such factors the 
ratio of types would be 9 BG : 3 Bg : 3 bG : 1 bg. 
These difficulties are recognised by Correns, who only advances 
the theory tentatively in the absence of full evidence. 
VIII.— Appearance of Self-Fertile Races. 
One consequence of the theory may be pointed out. At every 
fertilisation there must arise individuals homozygous for the 
absence of one or more of the inhibitory substances. Therefore 
it is theoretically probable that occasionally an individual homo¬ 
zygous for the absence of all the factors would be produced. Such 
an individual would be self-fertile. It may be recalled that Darwin 
(1876) as a result of repeated self-pollinations of partially self-sterile 
plants through a series of generations obtained ultimately, in the 
cases of Ipomcea purpurea and Mimulus luteus, highly self-fertile 
plants. It is suggested that this phenomenon may be due to the 
weeding-out of the partial inhibitors through constant in-breeding, 
and that the self-fertile plants are the ultimate recessives. In some 
such way the self-fertile race of Reseda odorata may have arisen 
from the self-sterile. 1 
IX.— Self-Sterility and Sex. 
It is possible that the phenomena of self-sterility are more 
widely distributed than is generally realised. Indeed it may be 
said that they are intimately connected with sex differentiation. 
The current theories of sex for the most part leave unexplained the 
singular fact that male gametes do not fuse with male, nor female 
with female. 3 In the higher animals and plants the gametes and 
accessory sexual structures are highly specialised, and the inability 
of like to fuse with like is not so surprising at first sight. In the 
isogamous Algas and Fungi, however, it is not by any means an 
expected phenomenon : it is not obvious why the gametes produced 
by one and the same individual of certain species of Spirogyra , or 
1 The objection that in Reseda self-fertility is dominant is met by the 
reminder that the “ inhibitory factor ” may ex liypothesi be lack of nutrition or 
stimulus for the pollen-tubes:—a view of the matter favoured, as we have 
seen, by the chemical evidence and the analogy with immunity. 
3 Such a process takes place, however, in the parthenogenetic oogonia of 
Humaria grantdata, where potential ? nuclei fuse in pairs : and something similar 
occurs in the fusion of polar nuclei in the Angiosperm embryo-sac. 
