432 
PACIFIC EXPLORATION: H. A. PILSBRY 
the shore zone. Each species has its definite territory, sometimes one, 
sometimes several adjacent ridges. They bear living young. If their 
dispersal were effected to any extent by birds, one would expect to 
iind sporadic colonies, at least in a few cases, instead of the logical and 
definite distribution which is actually found. ^ Inter-island drift upon 
trees, here as throughout Polynesia, is excluded by the absence of large 
streams to transport such materials to the shore. The present currents 
and trades would not drift from island to island. Finally, the conditions 
on the shore zone are such that a forest snail thrown upon the strand 
could not exist. 
The Pleistocene deposits of land shells, though only now beginning 
to be explored, show that even then the island faunas were more closely 
related than at present. Numerous relatively large species were com- 
mon to two or more islands, while at the present day there^are none. 
All the facts are exphcable on the theory that the islands have been 
successively isolated by subsidence. Otherwise the distribution of the 
snails seems enigmatic. 
It is obvious that speculation upon the origin and migrations of the 
Pacific faunas, such as I have been allowing myself to indulge in, will 
have a vastly broader basis when we have adequate knowledge of the 
faunas of all the principal high islands. Hypotheses which now seem 
reasonable may be confirmed, or they may be definitely excluded from 
further consideration. All the high islands of the Caroline, Austral 
and Hervey groups should be investigated. In Samoa and the Mar- 
quesas we certainly know on]y a fragment of the fauna. Outlying 
islands, such as Rapa, specially need examination. On the Ameri- 
can side, Juan Fernandez and Mas-a-fuera call for investigation. 
At present it may be said that we have a good, though by no means 
complete knowledge of the snails of the Hawaiian Islands, which resi- 
dent naturahsts are zealously working on. In the Society Islands, 
through the work of Andrew Garrett, we have a fair knowledge. In 
Fiji, Tonga and scattered islands of other groups, it is likely that most 
of the large forms are known, the small and minute scarcely at all. Work 
on land snails has been sporadic, in large part done by chance visits 
of naturalists whose chief interest was in other subjects. Everywhere 
material must be preserved for anatomical study, as the affinities of 
many land snails can be worked out only from the soft parts, and at 
present we are guessing at the taxonomy in some of the families. Out- 
side of Hawaii but fit tie material has been collected for anatomical study. 
A word as to the 'low islands.' A group of small or minute land 
snails has been adapted to atolls and the shore zone of high islands. 
