V. H. Blackman. 
hybrids are so definite and comparatively simple that there is 
probably some simple cause at work leading the hybrids des¬ 
cribed above to produce regularly three classes of offspring in the 
strict numerical proportion of 1 : 2 : 1 . From a study of the 
hybrids with a single pair of differentiating characters just des¬ 
cribed and from that of the more complex cases where the hybrids 
possess two or more pairs of differentiating characters Mendel was 
led to the conclusion that in respect of the pairs of characters 
which he investigated, the gametes* (sexual cells) of the Pea 
hybrids transmit only one or other of the characters, but not both; 
in other words, in relation to these characters the gametes them¬ 
selves are pure, not hybrid, in nature. This hypothesis, assuming 
that the various kinds of male and female gametes are produced on 
the average in equal numbers, fully explains the definite and 
regular behaviour of the hybrids and their offspring. 
If A and a be used as above then there are four possible com¬ 
binations of the gametes of the hybrid Aa when self-fertilised. 
$ A x A $ 
2 A x 
? a x A $ 
? A X A $ 
The first combination will give rise to a pure dominant form, and 
the last to a pure recessive form ; the second and third combi¬ 
nations will produce similar hybrid forms. If the two kinds of 
gametes occur in equal numbers it is clear that the chances are 
equal for the occurrence of any of the four combinations, and as 
two of the combinations produce similar hybrid forms there will be 
two hybrid forms for every pure dominant and pure recessive form, 
and A -f- 2 Aa + A will express exactly the relative proportion of 
their occurrence. Further, all the first three combinations will 
produce forms with the dominant character and only the last 
recessive, so that the dominant forms will bear to the recessive a 
proportion of 3 to 1 . 
The hypothesis that the gametes of the hybrids transmit only 
one character of the pairs in question explains then so completely 
the striking numerical results obtained by Mendel in his experi¬ 
ments that there is a very strong prima facie case for believing that 
it is the correct explanation. The theory is also supported by the 
behaviour of the offspring of the hybrids in the more comple'x cases 
where several pairs of differentiating characters are combined 
together. In the case where the hybrids unite two pairs of differen¬ 
tiating characters (A a and B b) it is clear that the number of 
*Mendel uses the terms egg and pollen cells. 
