DIVERGENT EVOLUTION THROUGH SEGREGATION. 285 
sity of adaptation, there we find the most decided divergences in the 
organic forms. That is, where separation and divergent selection have 
long acted, the results are found to be the greatest. The first and 
third of these propositions will probably be disputed by few, if by any. 
The proof of the second is found wherever a set of closely allied organ- 
isms is so distributed over a territory that each species and variety 
occupies its own narrow district, within which it is shut by barriers 
that restrain its distribution, while each species of the environing types 
is distributed over the whole territory. The distribution of terrestrial 
mollusks on the Sandwich Islands presents a great body of facts of 
this kind. 
SELECTION OF EVERY KIND INSUFFICIENT TO ACCOUNT FOR DIVER- 
GENT EVOLUTION. 
Though I have no reason to doubt the importance of sexual selection 
in promoting the transformation of many species, I think I ean show 
that unless combined with some separative or segregative influence 
that prevents free intercrossing, it can avail nothing in producing a 
diversity of races from one stock. In the nature of its action sexual 
selection is simply exclusive. It is the exclusive breeding of those 
better fitted to the sexual instincts of the species, resulting from the 
failure to breed of the less fitted. It therefore indicates a method of 
separation between the better fitted and the less fitted; but it gives 
no explanation of separation between those that are equally successful 
jh propagating. 
I maintain that in a great number of animal species there are sexual 
and social instincts that prevent the free crossing of clearly marked 
races; but as these segregative instincts are rarely the cause of failure 
to propagate, and since when they are the cause of failure the failure is 
as likely to fall on one kind as on another, 1 conclude that the segre- 
gate breeding resulting from these instincts can not be classed as either 
sexual or social selection. Reflexive selection in all its forms is, like 
natural selection, the result of success and failure in vital processes 
through which the successful propagate without crossing with the 
unsuccessful; but it in no way secures the breeding in separate groups 
of those that are successful in propagating. The exclusion of certain 
competitors from breeding is a very different process from the separa- 
tion of the suecessful competitors into different groups that are pre- 
vented from inter-crossing, and whose competition even is often limited 
to the members of the same group. Sexual selection, like other forms 
of reflexive selection, can extend only as far as members of the same 
species act on each other. If the individuals of the two groups have 
through difference in their tastes ceased to compete with each other 
in seeking mates, they are already subject to different and divergent 
forms of sexual selection; and is there any reason to attribute this 
