PARTULA OTAHEITANA. a7 
figure and 41 exists because some of the yellow adults with yellow young only have 
mated with DR snails, but their young are so few as to lack the expected red types. 
The transfer of 4 of the third group in (2) is indicated, and the same degree of 
correction may be made for the first group, viz, yellow adults with red young only. 
The formula thus becomes: 
(3) Yellow adults: red young, 2; red and yellow young, 15; yellow young only, 30=47 
It may be repeated that these figures stand for the three classes of the mates of 
the RR yellow snails. 
Continuing the test, we may next determine the number of DD and DR indi- 
viduals in the red class. From (3) we have learned that they are as 2:15, and hence, 
in the 28 actual specimens, there are 3 DD and 25 DR. The former will have red 
young only, no matter what their mates might be. ‘The latter will have red young 
only in % of the 25 DR cases, or in one instance, while in the remaining 24 both 
kinds of young are expected. Theoretically, then, the expected numbers in the 
two groups of gravid red adults would be: 
(4) Red adults: red young only, 4; red and yellow young, 24=28 
Actually, when the figures of the second and third groups of (1) are combined for 
stated reasons, and when the transfer of } of the 13 individuals of the first group is 
also made, again for indicated reasons, the figures are as follows: 
(5) Red adults: red young only, 9.5; red and yellow young, 18.5=28 
The difference between theory and observation amounts to 5.5 cases among 
28, or 19.6 per cent. 
We may now proceed to deal with the returns on the assumption that yellow 
color is dominant, using the terms of (1) and (2) without rearranged formulation. 
Beginning with the red class assumed to be recessive, we find that RRXRR is 
expected in 37 per cent of 28 cases, or 10; the excess of 13 over 10 is due to the cases 
where yellow as well as red young should be expected if the average numbers of 
young were greater. Having corrected the number for the third class by trans- 
ferring 3, or about one-fourth, to the middle group, we also correct the figures for 
red adults bearing yellow young to the same fractional amount of 4, or 2, and 
(1) becomes: 
(6) Red adults: yellow young, 6; yellow and red young, 12; red young, 1o=28 
Hence we discover the proportionate numbers of DD and DR yellow adults, 
which are as 6:12 or as 1:2. Testing by working out the expected results in 
the case of gravid yellow adults, to be compared with the empirical observations, 
one-third, or 16, would be DD mated with DD, DR, and RR, and would conse- 
quently have all yellow young. Two-thirds, or 31, would be DR, with three kinds 
of mates as follows: 
DRXDD z+ of 31= 7 cases, yellow young only. 
DRXDR $1 of 31=13 cases, yellow and red young. 
DRXRR 3+ of 31=11 cases, yellow and red young. 
