PARTULA OTAHEITANA. 213 
Taking all of these facts into account, it would seem that the dextral animals of 
Apirimaue are really afinis, despite their aberrant dimensional characteristics. 
Fecundity (table 156) is not unusual. ‘The data of inheritance of the color- 
characters in Oopu (table 156) show the interconnection of the two types. The 
6 young of the Apirimaue adults were all brown, plain, and dextral like their 
parents. 
Tasie 156.—Partula otaheitana affinis. Oopu and Apirimaue Valleys. 
FEcuNDITY. Herepiry (Oopu VALLEY). 
Records No. of | Per cent | No. of| No. of | Total | Average | Average Young,| Young, Total 
arcs. gravid.| gravid. | eggs. |young. contents. for gravid.) for all. plain. | banded. oe 
Oopu..... 36 32 88.8 41 27 68 1.89 2 VA Adults: 
Apirimaue. 6 4 66.6 3 3 6 1.50 1.00 Plains 22 1 23 
Banded... 25 4 4 
shotaltraneier 22 5 27 
THE MENDELIAN INHERITANCE OF THE COLOR-PATTERN. 
In the foregoing specific descriptions of the several colonies of affinis, two prin- 
cipal types of coloration have been found in almost all cases, namely, the so-called 
plain and the banded; the former is always more numerously represented, and some- 
times it is the exclusive kind, even in cases where several hundreds of snails have 
been taken, asin Papenoo. The genetic interrelationships of the main color-classes, 
where both occur in the same association, are demonstrated by the data of inherit- 
ance, heretofore presented in a tabular form which shows to what extent the young 
snails produced by one type of parent represent the same class or display the alterna- 
tive form of coloration. 
Now it is desirable to treat the observed facts in another way, to see if a Men- 
delian order of heredity is followed by the plain and banded characters, viewed as 
alternatives. The problem, then, is precisely the same as in the earlier case of 
rubescens, where the contrasted yellow and red colors were considered. Likewise 
the same difficulties are encountered, especially that arising from the low average 
numbers of young produced by the gravid adults. In very few instances do the 
banded adults constitute a satisfactorily large percentage of the population, and 
when they do, their offspring are relatively few; yet the method devised in dealing 
with rubescens, when employed in the favorable cases here, gives results that are in 
general accordance with the Mendelian interpretation, even though they can not 
be considered as finally established, for until or unless actual breeding experiments 
can be carried out the last proof can not be given. 
The colonies of only five valleys offer favorable material, namely, those of 
Tuauru, Farapa, Papeiha, Utuufai, and Faone. In Paraura the banded type forms 
a considerable percentage of a large collection, but the data are such as to indicate 
a late appearance of the bands; therefore the young snails of this kind can not be 
recognized as such. The Tehoro colony must be eliminated because the anomala 
form of coloration is not a striped or banded type; the others are useless for one 
