466 Mr. R. M‘Lachlan’s Observations on 
were yellowish, when on Centaurea nigra reddish, when on Matri- 
caria whitish, &c.; afterwards I placed them all on Senecio Jacobea, 
they being then nearly full grown. I did not find that they showed 
a tendency to become yellowish ; and this proved to my mind, first, 
that it was necessary for the larva to have fed on the one kind of 
flower from the egg in order to acquire this power of mimicry, and 
secondly, that the colour of the larva could not be caused by the 
food showing through the somewhat transparent integuments. 
Hence, it appears that the colours of Lepidopterous larvee are in 
a great measure adapted to save them from being carried off by 
birds and other enemies, and it is reasonable to suppose that in 
those cases, where the colours do not ‘assimilate with, or are 
directly opposed to, those of the food-plant, some other circum- 
stances may exist, rendering such mimicry unnecessary. The 
larvee of most internal feeders, which are not greatly exposed to 
external enemies, show little variation, either in particular species 
or as a whole. 
I conclude, therefore, that food has an immediate though in- 
direct effect in producing variation in the larva, but that in the 
imago it possesses this attribute in a very small degree. In 
the latter we must look to other and more subtle causes. That 
mimicry does not exist in the imago to the same extent as in the 
larva appears evident, but I cannot believe that Nature is ever 
aimlessly prodigal, and, no doubt, the causes of variability in the 
imago-state are as potent as in the larva-state, but at present they 
are, for the most part, beyond our comprehension. 
With respect to range of variation I will say but little. Tt 
appears to me that ordinarily varieties have a tendency to revert 
to what we consider as the type, but that under certain cir- 
cumstances, not only will they not so revert, but that the diver- 
gence will gradually become wider, until eventually they develope 
into what is considered as a species. 1 do not say that I am pre- 
pared to accept the ‘development theory ” to the full extent to 
which some would apply it; but that it is a reasonable way of 
accounting for phenomena, which otherwise cannot be satisfac- 
torily settled, must, I think, be evident to all who endeavour to 
rid themselves of hereditary prejudices. ‘The acceptation, partial 
or entire, of this theory is not so disastrous as some would appear 
to consider it. The most inveterate describer of new species 
need not fear that the darling object of his existence is useless 
and aimless if well done, for the process of development is of 
necessity so immeasurably slow, that to all intents and purposes, 
a faithful description of a new “ form” or “species” is as useful 
