182 REPORT OF THE CONFERENCE ON GENETICS, 
happens according to Von Guaita’s theory is that the two factors W and 
A struggle, and neutralise each other so that neither of them is 
manifested in the offspring, leaving the two Ms, which are similar and 
compatible, in sole possession of the field. This theory accounts in a 
very ingenious way for the character of the hybrid: and doubtless some 
elaboration of it could be suggested which would account for the 
reproduction of the three categories OO, OC, and CC in the next 
generation. 
The theory put forward by Bateson two years ago to account for these 
phenomena is that there is in the germ-cell of an albino a factor deter- 
mining albinism, which he calls g, and similarly that there is a factor g’ 
in the waltzer, determining its colour characters, which we have already 
called OC. 
The result of the union of g and g’ is a hybrid g’g, whose character we 
have already denoted by CC. ‘The result of the union of g’ and g—the 
production of a form more different from either of them than they are 
from each other—is said to be a specific result in the sense that the pro- 
duction of water is said to be the specific result of the union of H, and O. 
But the most striking part of this theory is that which refers to the germ- 
cells of the hybrid. According to it, 50 per cent. of the germ-cells of the 
hybrid bear the factor g’ and 50 per cent. g. Now the result of the union 
of two hybrids each containing (50 per cent. g’) + (50 per cent. g) germ-cells 
is the production of offspring falling into the following categories in the 
proportions indicated by the numbers prefixed to them—25 g’q’ (or g’), 
25 g’g, 25 gq’, and 25 gq, or (g'g and gg’ being the same) 25 gq’, 50 gg, and 
25 g—which, the reader will remember, is the actual result. 
This proportion is simply the result of the random union of the 
gametes of the hybrids, and can be illustrated by making pairs of counters 
by taking one of the pair at random from a hat containing red (R) and 
white (W) counters in equal numbers, and the other of them from 
another hat with similar contents. The result of a large number of trials 
will be in percentage 25RR, 25RW, 25WR, and 25WW, or 25RR, 50RW, 
25WW. 
The difference between the above-outlined Mendelian and Weismannian 
theories is that while the former tries to account for the segregation and 
not for the reversion, the latter tries to account for the reversion and not 
for the segregation. It is when we fix our attention on that part of the ~ 
Mendelian theory which refers to the nature of the gametes of the hybrid 
that we see what the doctrine of gametic purity really means, how pro- 
foundly new and definite a thing that theory is, and how widely it differs 
from any other theory of heredity whatsoever. 
Let us imagine that we have one of our hybrids, with its rich brown 
coat and black bead-like eyes, before us; a mouse that we might easily 
mistake for a wild one caught in a trap, if we did not know its parentage. 
According to the particular Mendelian theory we have been discussing, none 
of the gametes of this mouse contain an element representing the character 
of the animal which bears it, namely, g’g. But half of the gametes bear the 
element, g’ and half g. The fact that 50 per cent. of the children of such 
hybrids are like their parents is not due to the presence in the germ-cells of 
their parents of any elements representing their own characters, but to the 
