282 VARIATION, DISTRIBUTION, AND EVOLUTION OF THE GENUS PARTULA. 
The data are scanty, but the analysis is interesting because the characters are 
the same that were investigated in the independent cases of affinis and sintstrorsa. 
First assuming that banding is recessive, the numbers of DD, DR, and RR come 
out as 7:13:45. When the test is carried out in the case of the plain snails, the 
expected numbers are 7 : 10, and the empirical numbers are 8 : 9 for the two groups 
of parents with plain young only and parents with both kinds of young. The 
difference amounts to 5.9 per cent of the cases involved. 
Under the contrary assumption, the proportion of the three genetic classes 
would be 2 : 10:5, which seems more natural, but the independent test gives a 
departure of observation from theory to the extent of 16.9 per cent of the cases. 
Were it not for the clear case of affinis and the probably similar relation in the 
sinistrorsa series analyzed, the present result might not be significant as to the reality 
of a Mendelian nature of this phenomena. 
PARTULA OTAHEITANA CRASSA ABERRANS SUBVAR. NOV.—TAAPUNA VALLEY. 
Taapuna is the most northern valley of the western sector that is suitable as a 
habitation for Partule. It is true that two or three small gullies occur in the interval 
between this valley and Tahaa Point, at the extreme northeast, but snails are either 
absent from them or very infrequently found. ‘Taapuna is inhabited by otaheitana 
snails belonging to the present primary variety, crassa, but the colony exhibits 
several remarkable features that justify its recognition as a distinct section of this 
division on the same plane with occidentalis; the quaternary term aberrans is given 
to it for the sake of brevity. Only a few snails were taken in 1906 (17 in all), but 
in 1909, 208 adults were secured, mainly from the interior part of the valley. Here 
P. nodosa is represented by a flourishing association of the variety composita similar 
to that of Punaruu—a fact of some interest in view of the difference between the 
otaheitana crassa of the latter valley and the peculiar colony of Taapuna. 
The colony as a whole is characterized, among other things, by its color- 
composition. Disregarding certain specimens that were decolorized by imperfect 
preservative, there remain 180 adults, to be classified as follows: 
cestatawlightlytbanded 2.2] 5 er tse ee eee 4= 2.2 per cent. 
Rlainse itches (yellowish) sae ae eee er eee 20=II.I per cent. 
Plain, medium (yellow-brown, red-brown)................ 73 =40.5 per cent. 
Planydark*(darkibrown)s:28 28 8.2. eee eee 83 =46.1 per cent. 
The shells of the first-named class are much like their correspondents of Punaruu 
in the lightness of the revolving band (plate 34, figs. 12 and 13). The light “plain” 
shells (plate 34, figs. 3 to 5) are somewhat similar to the ‘“‘yellow” class of crassa 
occidentalis from Papehue and Aoua, not only in the richer ground-color, but in its 
smoothness and lack of strigations as well. The class called “medium” (plate 34, 
figs. 6 to 8) comprises a novel color-form not hitherto encountered; here the ground 
tint is an even red-brown or an even yellowish-brown, entirely devoid of strigation. 
Finally the “dark” class comprises a few fine specimens corresponding to phea 
confluens of typical crassa colonies, but the more numerous shells display a rich, 
even shade of brown, sometimes of a very deep hue (plate 34, figs. 9 to 11); these 
