phical distribution. In the Hawaiian Islands snails of the family of
Achatinellidae occur in great abundance, and like the lizards of the
Galapagos Islands different species occur on the different members of the
group. Within the confines of one and the same island, they vary from
valley to valley, and the correlation between their isolation in
geographical respects and specific differences on the other hand, first
pointed out by Gulick, makes this tribe of animals classical material. In
Polynesia and Melanesia are found close relatives of the Achatinellidae,
namely, the Partulae, which are thus in relative proximity to the
Achatinellidae and not on the other side of the world. Furthermore, the
Partulae are not alike in all of the groups of Polynesia where they occur;
the species of the Society Islands are absolutely distinct from those of
the Marquesas, Tonga, Samoan, and Solomon Islands, although they agree
closely in the basic characters that justify their reference to a single
genus. The geological evidence tells us that these islands were once the
peaks of mountain ranges rising from a Pacific continent which has since
subsided to such an extent that the mountain tops have become separate
islands. Thus the resemblances between Hawaiian and Polynesian snails, and
the closer similarities exhibited by the species of the various groups of
Polynesia, are intelligible as the marks of a common ancestry in a
widespread continental stock, while the observed differences show the
extent of subsequent evolution along independent lines followed out after
the isolation of the now separated islands. The principle may be worked
out in even greater detail, for it appears that within the limits of one
group diverse forms occupy different islands, evolved in different ways in
their own neighborhoods; while in one and the same island, the populations
of the different valleys show marked effects of divergence in later
evolution, precisely as in the case of the classic Achatinellidae of the
Hawaiian Islands.

The broad and consistent principle underlying these and related facts is
this: _there is a general correspondence between the differences displayed
by the organisms of two regions and the degree of isolation or proximity
of these two areas_. Thus the disconnected but neighboring areas of the
Galapagos Islands and South America support species that resemble each
other closely, for the reasons given before; long isolated areas like
A