164 
V. //. Blackmcui . 
from a triple nuclear fusion in which two nuclei come from the 
embryo sac and only one from the pollen tube, there must be a 
tendency towards the stronger development of the female character. 
It is also interesting to note that the endosperm characters investi¬ 
gated were all schizogonous in nature, and that the hybrid endosperm 
produced was of a similar hybrid nature to that of the embryo, 
a fact which shows, that the two polar cells (or nuclei) were of the 
same nature as the egg cell (or nucleus) and that the two male cells 
also possessed the “Anlagen” of the same characters. 
The interesting hypothesis of Webber ( 1900 ) that in many cases 
the “flecking” of the endosperm is to be explained by the partheno 
genetic development of the second male nucleus, is shown by Correns 
(1901 b) to be untenable. The endosperm often shows such 
characters that it must, on this theory, have developed partly under 
the influence of the three fused nuclei and partly under the influence 
of the male nucleus alone. 
As was mentioned above, Correns soon found from his experi¬ 
ments that the direct effect of the foreign pollen was confined to the 
embryo and endosperm. This has led him to investigate the very 
puzzling cases of Xenia in which the pollen has been described as 
directly affecting structures other than the embryo and endosperm^ 
i.e. tissues which are not of hybrid origin. He points out that some 
of the cases are to be explained by changes in the endosperm 
and embryo, and that for others there is no really trustworthy 
evidence. The subject cannot be discussed here, but his general 
conclusion is, that there is as little satisfactory evidence for this 
class of Xenia as there is for the existence of true graft-hybrids, or 
of telegony amongst animals. 
The interesting speculative questions as to the means by which 
the schizogonous characters, really their “Anlagen,” are separated in 
the sexual cells, and the time at which it occurs, have been discussed 
by Correns (1900 b, 1901 b) and Strasburger ( 1900 , 1901 ), and brief 
reference must be made to them. Correns, from a consideration of 
the fact that in the simplest case with one pair of differentiating 
characters, the two kinds of gametes are produced in equal numbers, 
puts forward the view that the mechanism of separation is that of a 
nuclear division, physiologically of the nature of Weismann’s 
reducing division, but not necessarily by means of an actual cross- 
division of the chromosomes. This view is fairly easily conceivable 
when there is only one pair of characters to be considered, but 
when the pairs of characters are numerous, and when some are 
