PRESIDENTIAL ADDRESS. 667 
one homozygous for the character, and another heterozygous, and a third 
homozygous for the absence of the character—and that, further, these types of 
individuals occur in the proportion of 1:2:1. Needless to say, the prediction 
is susceptible of verification by experimental breeding from the heterozygote. 
These are Mendelian commonplaces with which I should have hesitated to 
occupy our time were it not for the fact that I desire to emphasise the epoch- 
making nature of Mendel’s method. The magic wrought by genius is potent 
because it is simple. The rules of Mendelian method are simple. If it be urged 
that I have broken my promise and strayed from method to doctrine I would ask 
which of the simple propositions I have stated may be demurred to by any student 
of biology? 
The supreme importance of Mendel’s contribution to science consists in this : 
that instead of mixing anything with anything ‘in the gruel thick and slab’ of a 
witches’ cauldron, he has taught us to cast the horoscope of Fate by the 
method of genetical analysis of individual characters. Thus the first part of the 
Mendelian restatement of the old problem of Heredity reads: Investigate 
one by one the modes of inheritance of the several characters of an individual. 
Choose for this purpose organisms which are as tar as possible alike in all 
respects except for the character under investigation. Carry the experiment to 
its conclusion, even to the third or fourth generation. If uncertain results are 
obtained, ascertain before discarding the method whether the uncertainty may 
not be due to the interference of other characters not to be suspected a priori 
of exercising an influence upon the expression of the character under investigation. 
Who, for example, would suspect a morphological character lke thickness 
of stem of exercising an influence on the time of flowering of a plant? Yet 
such is the case with the pea (Pisum sativum), and there is evidence that when 
this disturbing influence is removed inheritance of time of flowering follows 
Mendelian rules. : 
The second part of the restatement of the problem of genetics may be expressed 
as follows: Only by the use of individuals of proved constitution with respect 
to a given character may the effect of external conditions on organisms be deter- 
mined. The study of variation must be preceded by Mendelian analysis and 
synthesis. Let me illustrate this theme by an example. 
The species Primula sinensis, the Chinese primrose, has given rise to many 
distinct varieties. Among these varieties are some with white flowers and 
others with magenta, blue, red, or other coloured flowers. Each of these 
varieties may be obtained of florists in a pure strain—that is to say, in a strain 
which breeds true to flower-character. For our immediate purpose we will group 
these varicties into white and coloured forms. 
It has been shown, however, that this apparently natural mode of grouping 
is inadequate to give a correct idea of the genetic constitutions of these races. 
It would seem self-evident that the white races differ from the coloured races 
by the lack of flower-pigment; yet Mendelian analysis demonstrates that there 
are more subtle differences between the different races. These differences 
become apparent when true-breeding white and coloured plants are crossed with 
one another; for it is then discovered that two types of white-flowered plants 
exist, and it is only by their fruits—their offspring—that ye may know them. 
Thus if certain white-flowered races are chosen for the experiment, the result 
of crossing white and colour is a coloured F, generation. If certain other white 
races are used and mated with the coloured form the offspring of the cross all 
bear white flowers. The different genetical behaviours of these heterozygous first 
generations give the clue to the difference between the two forms of white used 
as parents. In the former case—that in which the first (F,) generation consists 
of coloured offspring—the second (F,) generation, raised by self-fertilising F, 
individuals or by crossing them with one another, consists of coloured : white in 
the proportion of 3:1. 
Whence we conclude that the white used in this experiment owes its 
character of whiteness to lack of the pigment-producing factor which is present 
in the coloured parent race. This conclusion is confirmed by the genetical 
behaviour of the whites of the F, generation. Such extracted whites breed true 
to flower-character—that is, give rise to white-flowered offspring only. White- 
flowered races which behave in this manner are termed recessive whites. 
