DIVERGENT EVOLUTION THROUGH SEGREGATION. 281 
In either case, the kind that is to propagate is determined by the 
selection, and those that are not to propagate are in some way ex- 
cluded. The process may therefore be called the exclusive breeding 
of certain kinds; and natural selection may be defined as_ the exclusive 
breeding of those better adapted to the environment. 
But if from one stock of horses we wish to develop two distinct 
breeds, one of which shall excel in fleetness and the other in strength 
for carrying or drawing burdens, the result will not be gained by simply 
preventing all that are inferior in strength or fleetness from breeding. 
By this process, which is the exclusive breeding of the desired kinds, 
we should obtain one breed with fair powers of strength and fleetness; 
but the highest results in either respect would not be gained. Such 
experiments show that the exclusive breeding of other than average forms 
causes monotypic evolution, and that to secure divergent or polytypic evolu- 
tion some other principle must be introduced. 
In the case of natural selection, the separation it introduces is between 
the living and the dead, between the successful and the unsuccessful. 
In other words, natural selection is the exclusion of all the forms that 
through lack of adaptation to the environment fail of leaving progeny, 
and therefore in the exclusive generation of the forms that through 
better adaptation to the environment are better able to propagate. 
Variation with the natural selection of other than average forms may there- 
fore account for the transformation of an ancient species into a series of 
successive species, the last of which may now exist in full force; but with- 
out the aid of se-generation it will by no means account for the divergent 
evolution of any one of these species into a family of coexisting species. 
As I have just shown, natural selection is the exclusive generation 
of those better fitted to the environment; and it tends to the modifica- 
tion of species simply through the generation of the better fitted forms, 
while they are prevented from crossing with the less fitted, which fail 
of propagating through their lack of fitness. Now, from the very nature 
of this process, which results from the success and failure of individuals 
in appropriating the resources of the environment, it follows that it 
can not be the cause of separation between the successful competitors, 
and therefore any divergence of character that arises between the differ- 
ent groups of the successful can not be attributed to natural selection. 
Natural selection explains the prevention of crossing between the fitted 
and the unfitted, and shows how the successive generations of a species 
may gradually depart from the original type, becoming in time a differ- 
ent species; but it can not explain the divergences that arise between those 
that have, by the fact of successful propagation, proved their fitness. It 
depends on superiority of adaptation to the environment, and tends to 
produce increasing adaptation; butdivergent kinds of adaptation are not 
necessary conditions for it, and it can not be the cause of inereasing 
divergence between the incipient kinds that otherwise arise. 
