100 University of Michigan 



fall in their near vicinity. The number of whorls (and the 

 altitude) appears to be inversely proportional to the amount 

 of exposure to the dry trade-winds; this factor probably acts 

 through increase in the rate of evaporation, which would de- 

 crease the length of the active periods of the cerions. All of 

 the lots, which have a mean number of whorls that approaches 

 12, are from the base of escarpments, usually on the lee side 

 of the larger hills. 



Typical C. uva uva probably comes from the hills behind 

 Willemstad. This lot (Table XI) happens to give a mean size 

 and the mean number of whorls near those of all of the lots 

 taken together. Although the change in form, due to the num- 

 ber of whorls and the resultant difference in altitude, is very 

 conspicuous (Plate XVIII), it is too variable to be of any 

 racial importance ; in fact, aperture formation appears to be 

 hastened by injury. The actual size of the shells seems more 

 important, and the extreme lots deserve recognition as ecologi- 

 cal forms. The most dwarfed shells may be called form 

 diablensis, new (fig. xviii-A2), with the top of Ronde Klip 

 (C12b) as the type locality. The largest shells also have the 

 heaviest sculpture, and may be included in the form hatoensis, 

 new (fig. xviii-F6), with the eastern escarpment of Seroe 

 Spelonk, near Landhuis Hato (Clld) as the type locality. 



C. uva uva is mainly restricted to the limestone outcrops, 

 but in a few places invades the borders of the central region 

 of older rocks. In these localities (Cb6, 10), a rather large 

 portion of the shells show a tendency to reduce the sculpture 

 (figs. xviii-C6, D5), and even the sculptured shells are usually 

 rather slender. Practically smooth shells occur, and have been 

 described as C. uva desculptum; the exact type locality of 

 this very conspicuous form is unknown, but it is probably 

 somewhere around Sint Anna Baai (Cb6b). 



