( Iviii ) 
the theory of natural selection into the subject of insect 
colouration have not yet been exhausted. From the obser- 
vation that the species which are mimicked are generally 
gaudily coloured and take no special means to hide them- 
selves, it is but a step to the well-known theory of warning 
colours propounded by Wallace in 1867. That theory, in 
itself the outcome of a question raised by Darwin in connec- 
tion with his theory of sexual selection, stimulated the 
experiments of the late Jenner Weir and of A. G. Butler, 
the striking observations of Thomas Belt in Nicaragua, the 
detailed researches of Weismann into the origin and meaning 
of the colours of caterpillars, and the later systematic series 
of experiments conducted by Poulton. Yet another example 
I will permit myself to make use of because it is one in 
which I have some personal interest. In considering the 
subject of adaptive colouration as explained by Bates and 
Wallace, a difficulty occurred in the case of species which 
are of variable colouring ; I ventured to suggest, as far back 
as 1873, that this kind of colouring would be explicable by 
natural selection, if we supposed that this agency could 
confer a power of adaptability on the individual. At that 
time no mechanism could be conceived of by which such 
individual adaptability could be acquired, excepting the direct 
assimilation of the colouring-matter of food-plants in the 
case of caterpillars or other vegetable feeders. This, of 
course, carried with it the implication that natural selection 
could work on physiological processes if they were of use, 
just as well as upon any external morphological character. 
Stimulated by this hypothesis, other cases of variable 
colouring were sought for and found. The subject was later 
taken up by Professor Poulton, who, for many years, con- 
ducted experiments and obtained results which are now 
familiar to all naturalists. The original speculation, that 
variable colouring was the result of an individual adapta- 
bility due to natural selection, implies that this faculty is of 
bionomic value. I am not now concerned with the validity 
or otherwise of this assumption ; that is an issue on which 
opinion appears to be divided ; although, I have no 
doubt in my own mind on the point, it is not necessary 
