EVOLUTION THROUGH CUMULATIVE SEGREGATION. 
189 
Divergent Evolution through Cumulative Segregation. By Eev. 
John Thomas Gulick. (Communicated by Alfred Russel 
Wallace, E.L.S.) 
[Read 15th December, 1887.] 
Introduction. 
In my study of Sandwich-Island terrestrial mollusks my atten- 
tion was early arrested by the fact that wide diversity of allied 
species occurs within the limits of a single island, and in 
districts which present essentially the same environment. As 
my observations extended, I became more and more impressed 
with the improbability that these divergences had been caused 
by differences in the environment. It was not easy to prove that 
sexual selection had no influence; but, owing to the very low 
grade of intelligence possessed by the creatures, it seemed im- 
possible that the form and colouring of the shells should be the 
result of any such process. I was therefore led to search for 
some other cause of divergent transformation, the diversity of 
whose action is not dependent on differences in nature external 
to the organism. 
I found strong proof that there must be some such principle, not 
only in the many examples of divergence under uniform activities 
in the environment, but in the fact that the degrees of diver- 
gence between nearly allied forms are roughly measured by the 
number of miles by which they are separated, and in the fact 
that this correspondence between the ratios of distance and 
the ratios of divergence is not perceptibly disturbed by passing 
over the crest of the island into a region where the rainfall is 
much heavier, and still further in the fact that the average 
size of the areas occupied by the species of any group varies, 
as we pass from group to group, according as the habits of the 
group are more or less favourable to migration. I perceived that 
these facts could all be harmonized by assuming that there is some 
cause of divergence more constant and potent than differences in 
nature external to the organism ; and that the influence of this 
cause was roughly measured by the time and degree of separation. 
During the summer of 1872 I prepared two papers in which 
these facts and opinions were presented. One of these, entitled 
“ The Variation of Species as related to their Geographical Dis- 
tribution, illustrated by the Achatinellinae,” was published in 
LINN. JOURN. — ZOOLOGY, VOL. XX. 16 
