668 TRANSACTIONS OF SECTION K. 
In the second case—that in which the F, generation consists of white-flowered 
olfspring—the IF’, generation, from selfed or intercrossed F, plants, consists of 
three white: one coloured. The coloured offspring breed true. Of the three 
whites one breeds true to whiteness and the other two give rise, like the white F, 
generation, to three white: one colour. White races which thus impose their 
whiteness on the offspring of their union with a coloured race are known as 
dominant whites. Mendelians account for the genetical behaviour of 
dominant whites by assuming that they carry the character for colour and also 
a character for colour-inhibition. This hypothesis is amply justified by genetical 
results. Nevertheless it is an hypothesis which is novel to biology. It pro- 
pounds a series of questions to the physiologist and biochemist, and in 60 doing 
exemplifies the fruitfulness of Mendelism. We shall see immediately whether the 
biochemist is able to take up this Mendelian challenge and what answer he can 
ive to it. 
‘ At present, however, we are concerned to show by an example the necessity 
of prefacing the study of variation by Mendelian analysis. It was stated just 
now that the cross, dominant white by colour, results in a white F,. That 
statement requires amplification. Grown under normal conditions the 
F, individuals bear pure white flowers; but if grown in somewhat higher tempera- 
tures the flowers develop a distinct though’ pale flush of colour. It is easy to 
show that the factor for colour is unaffected by the changed conditions, for 
the flushed F, individuals yield offspring of the same kind and in the same 
proportions as those produced by white F, plants. 
It is fairly evident that the flushing is produced by the destructive action 
of heat on the inhibitor. In pre-Mendelian times this response to temperature 
would have been added without more ado as yet another ornament to dress the 
window of that old curiosity shop which is stocked with miscellaneous and 
heterogeneous articles all ticketed with the label ‘ variation.’ 
But in the light of Mendelism we may see in this effect of temperature the 
result of the casting-vote of circumstance on a heterozygous constitution. We 
may recall instances—as, for example, those provided by the well-known experi- 
ments on the effects of high temperatures on insect larve—which seem to show 
that environmental agencies may single out not only characters but also factors 
for attack. Thus we may begin to cohere in series the hitherto sundered and 
scattered phenomena of variation. 
It is not yet possible to say how much of variation is to be put down to 
the interplay of characters, or, rather, to the differential effects of external 
conditions on characters which tend to balance one another; but this at least 
may be said—that the old and worn controversy on acquired characters was 
so much waste of words, because the problem purporting to be discussed had 
never been defined. Like the half of human quarrels, it was a quarrel about 
words. 
It is stated in the books that the formation of peloric (regular) flowers may 
be induced by uniform illumination. Was the material used in the research 
homozygous or heterozygous? Does uniform illumination just prevent the 
unpaired factor from inducing normal growth? If so, what is the effect on the 
homozygous normal? These are examples of questions which suggest themselves 
at every turn, and they will abide the answer of experiment. The time is 
approaching when it will be possible to test the validity of the hypothesis on 
which the super-hypothesis of natural selection rests apparently secure from 
verification or disproof. 
That hypothesis maintains that everything is in a state of flux; that variation 
occurs at all times and affects all parts. This may be true of multiple mongrels; 
of organisms which are heterozygous for many characters. On the other hand, 
nothing is more surprising than the stability of forms which are pure-bred for 
a fair number of characters, and it is at all events a suggestion not to be 
rejected summarily that plants pure bred for a considerable number of characters 
may exhibit a constancy and stability not usually associated with our ideas of 
living things. 
In any case, it is open to the biologist to provide himself with suitable 
material wherewith to study the range and scope of variation and to investigate 
the conditions under which the organism discards old characters and regresses 
fain a 
