174 VARIATION, DISTRIBUTION, AND EVOLUTION OF THE GENUS PARTULA. 
15.5, and the resulting young would all be red in color. The second combination 
of DRX DR would occur in | of 88 instances, or 40.5, and the offspring would be 
red and yellow; it is true that the proportions among the young would be 3 red to 
1 yellow, and therefore if only 1 or 2 were present in the parental brood-pouch, 
the chances are in favor of the former color being displayed, and hence the real 
genetic constitution of the known parent would not be indicated. Finally, the third 
combination of DRX RR would come about in # of 88 cases, or 32, and again the 
offspring would be red and yellow, here in equal numbers. Summing up, on theory 
there should be: 
(5) Red adults: red young only, 48.5; red and yellow young, 72.5 =121 
although we know that one qualification certainly must be made and another is 
probable; the first is that some of the red adults with red young only would undoubt- 
edly produce yellow young as well, for they are DR mated with DR, and not DD, 
while the second is that if the relations are the same as in Helix, some of the appar- 
ently yellow young produced by the parents of the second class of (5) might be red 
in later life. To some extent, the corrections to be made, if all of the facts about the 
constitution of the red parents were known, would counterbalance one another. 
Before a comparison of the empirical observations of (1) with the theoretical 
figures of (5) can be made, the former need first to be combined, and then to be 
corrected in one particular. The 16 red adults that bear yellow young only are, by 
assumption, DR mated with DR, whose young are too few on the average to show 
both color-types in the offspring generation; hence they are to be assigned to the 
second class of red parents. In the next place, some of the red parents with 
red young would undoubtedly produce yellow young also, if their products were 
sufficiently numerous; hence they too must be transferred to the second class. 
The only discoverable basis for estimating them is the fractions similarly transferred 
in the treatment of the yellow class, namely, % of 98, or 32; this fractional number 
is only approximate, but it is the only one that can be determined on the basis of 
analytical results. Making the indicated adjustments the empirical figures are: 
(6) Red adults: red young only, 66; red and yellow young, 55=121 
The departure from theory amounts to 17.5 out of 121 cases, or 14.4 per cent. 
B. Yellow color assumed to be dominant.—We may now make the same kind 
of analysis, assuming the contrary value for the yellow color-factor, namely, that 
it is dominant to red. Rearranging the gravid adults of the two series (1) and (2), 
the classification is as follows: 
(7) Yellow adults: yellow young only, 40; yellow and red young, 16; red young only, 19 =74 
(8) Red adults: yellow young only, 16; yellow and red young, 7; red young only, 98=121 
As the red adults are RR by assumption, the yellow adults will be DD+DR 
in an unknown proportion. 
The 121 red adults will have mated with DD, DR, and RR adults, according 
to the chance possibilities. Union with RR would occur in 64 per cent of all cases, 
which follows from the fundamental data of table 114; that is, 77 are of this kind, 
and they would bear red young only (RR RR). Empirically, however, we find 98 
