CHAPTER I. 
THE MARIANA ISLANDS AND THEIR ECOLOGICAL 
CHARACTERISTICS. 
GENERAL CONSIDERATIONS. 
While the present volume is complete in itself as an investigation of the nature 
and ecology of the species of Partula which occur in the Mariana Islands, obviously 
there are certain features of the oceanic realm and certain characteristics of the 
animals themselves that are fundamental for the consideration of the specific area 
in question and of its own peculiar types. The earlier volume on the species of 
Tahiti comprises an account of the distribution of these animals in islands of the 
Pacific Ocean; it is necessary to re-state here only a few of the basic facts which are 
particularly important. 
The land gastropods of the genus Partula are restricted to Oceania in its widest 
sense, although they do not occur everywhere throughout this great area. They 
are absent from the Hawaiian Islands, but the classic Achatinellide of that group 
are their near relatives. With few exceptions, the snails are found only on the larger 
“high” islands of volcanic structure, or on composite islands consisting of limestone 
as well as of volcanic materials. Although the animals crawl about upon the vege- 
tation of suitable areas, or more rarely upon the ground, apparently they subsist 
upon the mycelia of fungi which grow on decaying plant materials. The areas of 
habitation in the islands of their occurrence are conditioned by a complex of environ- 
mental circumstances, such as moisture, shade, higher temperature, and altitude. 
The favorable nature of Oceania for studies upon variation and distribution 
can not be over-emphasized. The following quotation from the volume on the 
Tahitian species (oc. cit., pp. 13, 14) presents the case concisely: 
The situation being what it is, it would be impossible to find a more ideal combina- 
tion of circumstances for the investigation of the members constituting a definite bio- 
logical group, and of their geographical distribution. The total area of occurrence is 
large, exceeding that of the United States; within this, the habitable bodies of land are 
separate islands, more or less distant from one another, associated in lesser or greater 
numbers in groups that lie relatively near or far apart. Hence the degrees of geographical 
relationship are marked with extraordinary distinctness, without any question of inter- 
mediate connections that intervene between comparable ecological regions of a single 
continent. Furthermore, it has long been known that within the confines of a solitary 
island, the areas suitable for the existence of Partule are more or less isolated valleys, 
whose differing forms may be analyzed in correlation with their geographical and 
topographical proximity. In brief, then, the valleys and their diverse species constitute 
elements of a primary order, to be compared with one another; such elements taken 
together form an island-complex, which, as an element of a secdnd order, may be con- 
trasted with a similar complex of another island in the same association; uniting the 
several islands into a combination of a tertiary grade—the island group and its species— 
this may be investigated as a whole in relation to other combinations of the same status 
in different parts of the whole area. The circumstances are such, then, as to give an 
unusual interest and significance to the investigation of the genus Partu/a and its dis- 
tribution, on the basis of a systematic and detailed analysis not ordinarily possible 
when dealing with similar problems of zoogeography. 
3 
