164 VARIATION, DISTRIBUTION, AND EVOLUTION OF THE GENUS PARTULA. 
The statistics of the seven shell characters (table 105) show that the colony 
as a whole is made up of smaller shells than those of Papeiha—a fact that is more 
important for the general summary at a later point. As stated earlier, no shell 
exhibits a pillar tooth. In the next place, the few red shells are collectively larger 
and relatively stouter than the yellow ones, and the aperture is notably larger with 
reference to the whole shell. Hitherto the red shells have been smaller than the 
others. Finally, it appears that the two subordinate classes of yellow shells differ 
in certain of their absolute and proportional measurements, with statistical signifi- 
cance. Unless there is some degree of selective mating, it is difficult to account for 
the observed differences. 
The fecundity of the two classes is not equivalent (table 106), but in view of 
the small number of red adults no great significance can be attributed to the dis- 
crepancy. ‘The statistics of heredity (table 106) are also doubtful as far as the red 
adults are concerned, yet it is interesting to find that these produced no red young 
in the cases actually examined and that the red offspring came from yellow parents. 
TasLe 106.—Partula otaheitana rubescens, Haavini Valley. 
FEecunpliTy. HEREDITY. 
Records No. of | Per cent | No. of | No. of | Total Per cent | Per cent Young,| Young, Total 
| gravid.| gravid. | eggs. | young.}contents.| for all. /for gravid. yellow.| red. ; 
Adults: Adults: 
Yellow 207 177 85.5 264 177 441 Ph sAl 2.5 Yellow..| 174 3 177 
Red.. 7 5 71.4 8 4 12 Sd 2.4 Red .... 4 0 4 
All... 182 181 ; Omen) WA le rererens 
TAUTIRA VALLEY. 
Tautira is the largest valley of the peninsula, Taiarapu or Tahiti iti, and resem- 
bles Papenoo in that its innermost portion consists of the principal crater-basin of 
the whole land-mass. The valley runs outward in a direction a little to the west 
of north, and is traversed by a large river. The boundaries of the whole valley are 
high and well marked. 
Less than 100 examples of rubescens were secured here; while these agree in 
general with the shells of the Haavini colony in ground-colors (plate 27, figs. 45 to 48), 
yet the deep tinge, which is so characteristic a feature of the spire in the shells of 
that valley, is either entirely lacking or greatly reduced in Tautira examples. 
Another difference consists in the preponderance of red (fig. 48) and reddish-orange 
shells (figs. 46 and 47), of which there are 59 as compared with 23 of the yellow type. 
The statistical characters of the three distinguishable color-classes (table 107) 
show remarkable differences between the component groups. ‘The clear red shells 
are the smallest in absolute measures, like their counterparts in the valleys of 
Tahiti nui, although in Haavini the relation was the opposite. In no shell is a 
pillar tooth developed even as a trace. 
The snails secured in this valley could not be expanded, owing to the exigencies 
of travel at the time, for the night’s camp was made on an islet of the coral reef 
