Recent Advances in the Study of Heredity. 13 
lingering faith in the okl doctrine. They have, however, done what 
is better: they have narrowed down the issue by making it possible 
to deal with smaller and more specific problems. For instance, 
thanks to the knowledge which we now possess of the existence and 
(to some extent) of the nature of separately heritable unit-characters, 
I have been able to furnish a definite answer to the specific 
question whether there is anything like ancestral contribution 
within the limits of a single unit-character. 
But to return to the change in opinion which we have seen is 
taking place: the persistence of the contributional view amongst 
those who are not professedly Mendelian is most clearly proved by 
their attitude to two corollaries which follow from the interpretation 
which Mendel put on the facts which he observed. 
The first of these concerns the ancestry of the parent forms 
used in the cross, and asserts that the offspring of the yellow hybrids 
produced by crossing a yellow with a green Pea will consist of 75% 
yellow and 25% green Peas whatever the ancestry of the tivo parent 
forms is, provided only that they are both homozygous. 
The second concerns the posterity of the cross and asserts that 
the greens produced in F 2 are as pure as the pure greens started 
with, and, what is more remarkable still, that a green in (say) F 5 
is equally pure. Those who still cling to the contributional view 
have grown familiar with the fact that such extracted greens do 
breed true ; but they do not regard this fact as final proof of the 
Mendelian interpretation of this case, because they suspect that it 
may still be possible for these “green” germ-cells, which produce 
only greens, to contain an admixture of the yellow character without 
that character being manifested in the individuals to which they 
give rise. 
The reader will remember that the result of the experiment 
proved that the ancestry of one of the two forms (both of which 
were homozygous) involved in the cross did not affect the ratios in 
F 2 in the way in which the theory of ancestral contributions would 
lead us to expect; and, further, that it was demonstrated that an 
extracted “ green ” belonging to F 5 is as pure a green as one which 
has bred true for many generations. It was, I think, unwise to 
express the conclusion to which I was led by the results of this 
experiment, in the statement that “ ancestry does not make a 
difference,” a statement which, as I have already pointed out, does 
not mean anything unless we make it clear whether we are referring 
to individuals or masses. To a biometrician, who is accustomed to 
