76 VARIATION, DISTRIBUTION, AND EVOLUTION OF THE GENUS PARTULA. 
otaheitana, but makes no mention of clara in Orofere, in which valley at the present 
time more than 18 per cent of the adults are clara. Neither the chart nor the verbal 
descriptions refer to clara as existing in the peninsula Taiarapu, in which this species 
now forms 4.28 per cent of my collections of adult animals. Hence, the wider 
range of clara has only recently been attained, like its greater relative abundance. 
II]. In attaining its present wide range, Partula clara has come to exhibit distinct 
variations in the characters of its shell, and to some degree also in coloration. It is 
impossible to obtain a definite basis for comparing these characteristics with the 
qualities of clara in Garrett’s time, although, as regards the matter of coloration, 
it is evident that the three major types were present when the species was discovered. 
Pease speaks of the color of the shell as “corneous or pale chestnut, variegated with 
longitudinal strigations,”’ and of varieties with one, two, or three bands. 
From the detailed descriptions of this memoir, it will be seen that the structural 
characters of the shells vary considerably from valley to valley; the same is true of 
the color characters of colonies taken as wholes, giving due regard to the propor- 
tionate numerical values of the three major color-types. Furthermore, the relative 
abundance of clara varies like the other characters of the several valley colonies. 
It is even less possible than in the case of hyalina to refer the variations noted to 
environmental factors. 
The banded examples prove to be essentially different from one another when 
widely separated valleys are compared, and in each valley the structural characters 
of such shells approximate those of the unbanded associates; this is especially 
marked in the analysis of the Aoua and Vaihiria colonies. It would seem, then, 
that we are not dealing with three color-types differentiated in an original locality, 
which types have subsequently spread in company over neighboring areas, although 
this may be true for closely connected valleys such as those from Taharua to Vai- 
hiria. What we do find is a dispersed series of unbanded forms, which have pro- 
duced characteristic banded mutants in certain but not in all of their valley habitats; 
in the case of marmorata such mutations in color do not present themselves, although 
structural deviations are exhibited by the subordinate types of separated valleys. 
Although the colonies of the valleys now inhabited vary considerably, they 
appear to group themselves about certain common varietal types, which have been 
designated in the foregoing description by propernames. That procedure is perhaps 
not entirely justified by the facts provided by this species alone, for the varieties 
overlap in many cases. But when the many subordinate types of the complex 
species P. otaheitana are described in a subsequent section of this volume, it will 
appear that the condition of partial differentiation into more or less clearly demar- 
cated varieties, presented by clara as a whole, leads to an actual and distinct dis- 
continuity of similarly produced varietal types in a species which far earlier passed 
through the evolutionary stages in which clara now appears. Hence there is ample 
justification for the procedure in this case. 
The six recognizable varieties of Partula clara are defined in statistical terms 
by the figures of table 29, which give the respective mean values of all seven meas- 
