THBOUGH CUMULATIVE SEGREGATION. 
267 
fertility, especially in the case of flowers that failed of securing 
a cross. And afterwards, when it came into competition with the 
equally adapted variety from which it was partially protected by 
Segregate Vigour, it would neither be driven out nor lose its 
separate existence in a commingled race. It will be observed 
that we have in such a case Local, Germinal, and Floral Segre- 
gation, each producing partial effects which are enhanced by the 
Segregate Vigour. In order to bring out the relation of these 
factors to each other, let us assume definite values for each. 
Let us suppose that of the flowers are self-fertilized, are 
fertilized with pollen from another flow r er of the same plant, 
are fertilized with pollen from other plants of the same new 
variety, and -Jy are fertilized with pollen from the older variety 
occupying contiguous areas. Therefore the sum of the segre- 
gating influences, which is called the “ Eatio of pure breeding,” 
and is represented by E in Table IT., equals and the “ Eatio 
of cross-breeding,” represented by c in all the tables, equals yy. 
Again, let us suppose that the fertility of the pure breeds is the 
same as that of the half-breeds, but that the superior vigour of 
the former is such that any one of the pure seeds has twice as 
good a chance of germinating, growing to maturity, and producing 
seed as any one of the crossed seeds. The general effect on the 
final result will in that case be the same as if the “ Eatio of 
increase for the pure unions ” (which I call M) equalled 10, 
while the “ Eatio of increase for the cross unions ” (which I call 
m) equalled 5. Turning now to Table V., we can easily find the 
ratio in which the number of pure-breeds will stand to the half- 
breeds, if the conditions continue long ; for in the column in 
which m equals 5 and in the line marked c=yy we find 
which means that the half-breeds will equal the pure-breeds 
multiplied by or by 
Segregate Vigour and Segregate Fecundity beticeen Human Faces. 
My attention has recently been called to the following facts 
relating to the Japanese and Aino races, who have for many 
centuries met under circumstances favourable for interfusion 
without any apparent effect of this kind. I quote from ‘ Me- 
moirs of the Literature College, Imperial University of Japan,’ 
No. 1 : “ The Language, Mythology, and Geographical Nomen- 
clature of Japan viewed in the Light of Aino Studies,” by 
Basil Hall Chamberlain, p. 43 : — 
