IN THE TROPICS. 3*21 
this case there must be failure of dominance of blue in the 
case of some 50 per cent, of the heterozygotes. 
The point can be tested by sowing the white grains and 
once more pollinating by the recessive, i.e,, white. Blue and 
white grains from cob No. 5, which contained 76 per cent, of 
white grains, were sown separately, and pollination was 
effected by a white dent form. 
If Mendel’s Law of segregation of allelomorphs in equal 
numbers holds good, the expectation is that 50 out of 76 of 
the white grains will prove themselves to be pure white 
homozygotes, whilst the remaining 26 will be heterozygotes ; 
and we may expect the blue colour to appear among the 
offspring of the latter. 
The actual offspring of the white grains was as follows (a 
single cob being gathered from each plant) :— 
23 cobs with white grains exclusively. 
4 cobs with from one to four blue grains and the 
remainder white. 
10 cobs with from 1*7 to 25*7 per cent, of blue grains, 
i.e., 12 or more in each cob. 
Assuming that the four cobs with from 1 to 4 blue grains 
are cases of accidental xenia (as is perhaps most probable, 
since 1 to 6 blue grains appeared upon some 10 per cent, of 
the white dent cobs in the next row), we get 73 per cent., or 
55*5 out of 76 as the proportion of white grains which were 
homozygotes in the previous generation, the expectation 
being 50 out of 76. The remaining grains were heterozy¬ 
gotes which showed the white character. In view of the 
small number of plants examined the result is in reasonably 
close agreement with Mendelian expectation. 
The offspring of blue grains pollinated by the recessive 
showed a similar proportion of blue grains to that in the 
previous generation. Altogether 4,034 blue and 10,300 
white grains were counted, or 28T per cent, of the former. 
We must suppose that exactly the same phenomenon is here 
repeated. 
