288 DIVERGENT EVOLUTION THROUGH SEGREGATION. 
evolution. Any cause that, out of two or more kinds of suecessful 
variations, brings together one kind in such a way as to facilitate 
their breeding together, or to hinder their breeding with those of other 
kinds, is, according to my definition, a cause of segregate breeding; 
and the experience of breeders shows that wherever such causes operate 
divergent evolution is the result, and that the divergence accumulates 
when the process is continued through many generatious. From their 
experiments we learn that any form of segregate breeding persistently 
continued will result in divergent evolution. As any form of natural 
selection in which other than typical forms have the advantage will 
result in monotypic evolution, so any form of segregate generation will 
produce polytypic evolution. I call this the law of cumulative diver- 
gence through cumulative segregation. It is a generalization established 
by the widest experience of mankind in the cultivation of plants and 
the breeding of animals; and any assumption that is not in accord 
with it may be wisely called in question. 
I therefore judge that the advantage or disadvantage of their diver- 
gence, to individuals diverging from the typical form of a species, can 
not be the factor that determines whether the divergence shall be aceu- 
mulated. 
A divergent member of any inter-generating group can not long per- 
petuate its kind, if the divergence is any disadvantage; for the superior 
propagation of the more successful kinds will soon overpower the 
influence of the less successful; and the result will be monotypic evo- 
lution. The case is, however, very different with variations that are 
wholly or partially separated from each other and from the type by 
their divergent adaptation to different kinds of resources, or by any 
other cause. The perpetuation of such variations depends not upon 
any advantage they possess above the type from which they diverge, 
but upon ability to appropriate from the environment sufficient simply 
to maintain existence, and the result is polytypic evolution. In other 
words, of the freely crossing forms of any species it is only those that 
are most successful that are perpetuated; while of forms that are neither 
competing nor crossing, every kind is perpetuated that is not fatally deficient 
in its adaptations. It follows that a form that under present conditions 
maintains only a precarious existence, may, if kept from crossing, main- 
tain its characteristics unimpaired for many generations, and at last, 
through changes in the environment, enter upon a period of great pros- 
perity. Such would be the case with a form depending upon resources 
at first scarce, and afterwards very abundant. 
Again, the individuals of a species that are brought together in their 
attempts to appropriate some new kind of resource, and are thus led to 
breed with each other, and not with the rest of the species, become a 
new inter-generating group in which a new and divergent form of nat- 
ural selection is established, depending on divergent adaptations in 
the organism, without any change in the environment. The gradual 
