Recent Advances in the Study of Heredity. 245 
The question to which I now propose to direct your attention 
is a Mendelian one in the proper sense of that term. It will have 
been gathered from the description of the phenomena which has 
been given, that the characters which behave in a Mendelian fashion 
in inheritance are associated together in pairs, of which one 
member is dominant and the other recessive. Such characters, 
which behave as independent units in inheritance in the manner 
described above, are termed unit characters ; and also, in view of 
the fact that every such character has its counterpart, which is 
either dominant or recessive to it,— allelomorphs. The adjectival 
form of this term is useful ; because we can express the fact that 
round is one of the two members of a pair of characters, of which 
the other is wrinkled, by the statement that round is allelomorphic 
to wrinkled. 
We have so far confined our attention to crosses in which the 
two individuals mated differed in respect of a single pair of characters 
only. Now let us proceed to the consideration of one in which the 
two individuals mated differ in respect of two pairs of characters ; 
for instance—not to leave Pisum —a cross between a yellow wrinkled 
and a green round pea. The actual result is given in the following 
scheme. 
YW x GR 
I I I I 
F 2 9YR 3YW 3GR 1GW 
A cross involving the same two pairs of characters might 
have been made by crossing a YR by GW ; and the result in F 1 
and Fo would have been the same. But it is better to make the 
cross in the way represented in the above table because the fact is 
brought out that the matter of dominance is absolutely independent 
of the individuality of the parent which bears the dominant 
character. The result of a cross between a YR and a GW does 
not tell us that the dominance of the two characters Y and R is not 
an attribute of the individual which manifests them ; but the result 
of a cross between a YW and a GR excludes the possibility of such 
an interpretation. The result in the two cases is, of course, the 
same ; but in the former case both dominant characters are borne 
by one parent; whilst in the latter each of the two parents bears 
one of them. 
The result in P t and P 2 of the cross we are considering follows 
from the knowledge which already we possess of the mode of 
