14 VARIATION, DISTRIBUTION, AND EVOLUTION OF THE GENUS PARTULA. 
Showers are so frequent as to diminish the effects of local differences in topography. 
During typhoon weather, when as much as g inches of rain may fall in 24 hours, 
the disturbance extends over such a wide area as to affect all parts of Guam to the 
same degree. Hence the figures for the station of record well serve to indicate the 
conditions of rainfall throughout the island and throughout the whole group. 
VEGETATION. 
In the Mariana Islands as elsewhere, snails of the genus Partu/a are inhabitants 
of the forests and thickets where there is a considerable amount of moisture; they 
are never found upon the grasslands or savannas of low or high altitude. The areas 
of their occurrence are thus directly determined by ecological circumstances of a 
botanical nature. The animals dwell upon the leaves and branches of trees, shrubs, 
and herbs, where they generally remain quiescent during clear daylight; at night 
and under the stimulation of rains during the day, they move actively about and 
seek the ground in ordertofeed. They eat dead and dying plant materials or humus, 
apparently for the sake of the fungoid growths which provide them with sustenance. 
Quite as clearly as in the Society Islands, the intrinsic or specific characters of 
the plants with which Partule are associated are virtually indifferent; the in- 
dispensable requisites are that there shall be a sufficiently high and dense growth to 
provide shade, to conserve moisture, and to effect the production of a rich humus. 
Hence the limits to the areas occupied by Partule are set by the more ultimate 
ecological conditions which determine the distribution of suitable vegetation. A 
forest tree or shrub may harbor the animals, where a plant of the strand, similar in 
habit or identical in species, will be devoid of them. It is true that the animals of 
the Mariana Islands flourish in situations distinctly more open than the species of 
the Polynesian territory could tolerate, but aside from this qualification the general 
principles stated above hold true. 
Safford’s voluminous monograph enumerates the several types of botanical 
associations in Guam according to habitat, and it also specifies the characteristic 
members of each; this work is the basis for the following description. Although 
most of the distinguishable plant assemblages are utterly impossible as habitations 
for Partule, the comprehensive list is most valuable, as it serves to bring into clearer 
relief those which are entirely suitable. The items are here given in an altered 
sequence which places the habitable areas at the end, as follows: coral reefs, man- 
grove swamps, marshes, savannas, strand, inner beach, village environs, abandoned 
clearings, cliffs, rrver borders, and forests. 
With the algal flora of the reefs and with the distinctive plants of the mangrove 
swamps we have no concern. The marsh plants include the reed Trichoon, the fern 
Achrostichum, and occasional trees such as the “hibiscus,” Pariti tiliaceum, as well 
as the cultivated and native aroids, like the “taro” and “elephant-ear.” The 
last-named is a form which may carry Partulz in a suitable forest situation, but 
the snails are never found upon one in a marsh association. 
The savannas are very interesting by virtue of their sharp contrasts with the 
forests in botanical respects, and also because they often abut directly upon the 
