RESUME. 307 
Rurutu and Tubuai of the Austral Group—in all of which it displays the true 
diagnostic characters of the species with only quantitative differences. Next in 
significance is the absence of hyalina from the other islands of the three groups 
specified. After weighing the evidences for the two alternative views (1) that 
human agencies have been responsible for the introduction of the species from one 
of the three groups to the other two, and (2) that hyalina is a very ancient species 
which had attained its present characteristics before the old and continuous moun- 
tain ranges had been isolated by subsidence so as to become separate islands and 
island groups, judgment is given in favor of the second. 
CHAPTER IV. 
Partula clara Pease is a well-separated species and is typical in its restriction 
to a single island. Its colonies now exist in the valleys of about four-fifths of 
Tahiti, and they display exceedingly significant variations in size, shape, coloration, 
and texture. More than 1,000 adults and adolescents were secured. For the sake 
of a comparison with P. hyalina, the statistical description begins with the general 
treatment of the whole series; subsequently the populations of the major ecological 
regions are compared, and finally the several valley colonies are biometrically 
defined in geographical order. 
When the statements of Garrett and Mayer are taken into full account, the 
results of the present investigation are significant of many evolutionary changes. 
The species was rare in Garrett’s time in the islands (1861-1888) and was found “in 
the upper portions of the valleys in the southwest part of Tahiti.” Mayer (1899) 
found that clara constituted 10 per cent of the population of a certain valley 
(Vaihiria), while in my later experience it formed 23 per cent. In fine, the “rare”’ 
species of Garrett now amounts to 3.4 per cent of the island Partule, to 4.5 per 
cent of the population of the inhabited sector, and to 5.1 per cent of the total number 
of snails in the valleys of its collection. 
Again, the species was earlier restricted to a small southern sector, as Garrett’s 
evidence attests, while now it ranges throughout four-fifths of the whole island. 
Furthermore, in addition to its spontaneous dispersal, it has become differentiated 
into distinguishable subspecies, viz. parva, incrassa, minor, angusta, prima, marmo- 
rata, each of which has its proper territory, though this may be an old or a new-won 
region of the island. Mutation has taken place in some, but not in all, of the locali- 
ties into which migration has extended, and it has occurred also in parts of the 
former territory of occupation. The evidence in summary indicates that this species 
is one which has recently revived and developed after a long period of racial decline. 
CHAPTER V. 
In sharp contrast with the other arboreal species of Tahiti, Partula filosa Pfeiffer 
is restricted to a single valley (Pirai) situated in the drier northwestern sector of 
the island; yet in spite of its rigid confinement to one valley, the species is far 
from invariable, no less than five color classes being distinguishable, together with 
statistical differences between the sub-colonies of the upper and lower portions of 
