PARTULA FRAGILIS. at 
specimens to make it certain that fragilis and quadrasi are the same. It is unfor- 
tunate that neither of the two earlier authors published figures of the shells, for the 
type is peculiar and unmistakable, differing sharply from other species, even in the 
very young or adolescent condition. 
Partula fragilis is interesting on many accounts, but above all on account of a 
character that is little short of astonishing; zt produces eggs and young snails before 
the parental shell attains the lip of the aperture as the conventional sign of its repro- 
ductive maturity. No other species displays this feature, so far as my knowledge 
goes; it is so distinctive that it deserves all the prominence that can be given to it. 
The total number of completely formed or “adult” specimens taken in all 
localities was 36, together with a single “dead”’ shell in the garden of Government 
House at Agafia, to which place it had undoubtedly been brought on a cluster of 
Neottopteris from some distant forest. The absolute and relative numbers of P. 
fragilis in the several localities and regions of Guam have been enumerated in the 
census table (table 4). The species seems to prefer the limestone table-lands of 
the northern part; it is absent from the ample collections taken in the central and 
eastern portions of the southern half. Its occurrence on the high ground of Umatac 
Madog possibly indicates some degree of climatic similarity between the northern 
plateau and the higher forested slopes of the volcanic mountains in the south. 
The relative frequency of fragilis in the small collection from Oco is high, but 
this does not necessarily indicate a correspondingly great abundance in absolute 
numbers, in comparison with the situation at Barrigarda and elsewhere. Undoubt- 
edly the center of the fragilis population is in the neighborhood of Oco; but the 
high percentage in question really signifies an interesting difference between fragilis 
and the associated forms in a physiological respect, namely, in its readier response 
to increased moisture and rainfall. When the visit was made to Oco the day was at 
first clear, hot, and dry; soon, however, a heavy tropical downpour occurred, and 
within a few minutes Partule emerged from the secluded places where they had 
taken refuge from the strong light. The fragilis individuals were quicker than 
radiolata and gibba in their responses to the changed conditions, and hence relatively 
more of their kind were taken. It is interesting in this connection to note that 
precisely similar observations were made in Huahine, a member of the Society 
Islands, when P. arguta and P. annectens behaved in the same way, in the restricted 
area about Tepare Valley, where only they exist. 
Another character of fragilis worthy of special note is the pellucid quality of 
the shell. The soft body within is boldly maculated, and its pattern shows clearly 
through the semi-transparent shell. In this respect fragilis resembles young 
P. clara of Tahiti, P. turgida of Raiatea, and a few other species. Despite the fact 
that the favored habitat is the northern area, whose rock is calcareous, the shell is 
diaphanous and not influenced by the quality of the soil, either directly or in- 
directly; the specimens from the volcanic region of Umatac Madog are no lighter in 
weight, and, conversely, the associated gibba are no heavier in the limestone terri- 
tories than they are on the volcanic heights. The quality of texture is clearly 
determined by congenital factors. 
