56 
tissues in the pupa, so that the normal pigment (if the brighter coloured 
pigment be considered normal) has not sufficient time to mature ; or 
it may be that the pupa has not sufficient material at disposal to form 
the pigment matter. I take it that the second of these possibilities 
is the correct one. It seems to me that this change of colour being 
due to physiological causes, does not in any way affect the specific 
characters of the species, and what is true of this is true of all 
other Lepidoptera that vary under different climatic conditions. 
When we turn to Merrifield’s experiments, where, as far as possible, 
all other factors have been eliminated, we are much struck with the 
fact that the exposure of the pupa to a difference of temperature, does, 
in some species, produce a marked difference in the imago. This 
difference, however, is not by any means identical, even with species 
of the same brood, yet we may safely say that a general change in the 
same direction takes place, under given conditions, in the indi¬ 
viduals of the same species, some individuals showing much, others 
comparatively little, change. The more extreme results, however, 
of these experiments have not been the production of useless specific 
characters, but have resulted in the production of atavic forms, often 
quite unlike any species now in existence. That these modifications 
take place only at the critical point of pigment formation, Mr. Merri- 
field has, I think, abundantly proved, and that the changes are due to 
the effect of the high or low temperature on the energy of the pupa at 
the time of the formation of the pigment is, I think, certain. I am, 
therefore, inclined to consider that climate does not, except as (1) acting 
prejudicially (or the reverse) on insects through their food (making it 
stunted and less nutritious, and vice versa), and thus leaving them 
with an insufficiency (or excess) of energy for the production of an 
imago of normal size and pigmentation, (2) causing less (or more) 
rapid metabolism of the pupal tissues during the period at which 
the imaginal tissues are being perfected, really affect any species of 
Lepidoptera at all. Certainly it does not do so in the direction of 
producing real structural modifications. 
THE EFFECT OF FOOD IN PRODUCTION OF SPECIFIC CHARACTERS. 
When I wrote Melanism and Melanochroism in British Lepidoptera , 
I stated, I believe, that food could not produce changes in the colours 
of insects. I have long since retired from this untenable position, 
although I cannot give, off-hand, a single illustration from nature, 
where food, without doubt, causes a change in a species. Probably 
the stunted moorland forms of Hijpsipetes sordidata and Cidaria russata, 
with which Mr. Porritt has made us acquainted, are due to the in¬ 
nutritions nature of their food. I am also aware that ill-fed larvse 
produce either dwarfed or ill-pigmented imagines, the larva* evidently 
not being strong enough, nor having reserve-material and energy 
enough, to produce the normal coloration. All variations brought 
about by food, too, appear to act in this direction, i.e., by increasing 
or decreasing the amount of material available for wing formation and 
pigment, and, thus, variations produced by this means are those of 
size or colour. In no way can these be considered as hereditary, 
since they change with the substitution of a different food-plant, with 
each brood, or oven part of a brood, if it be separated. Hence these 
differences can in no way be considered as specific characters. 
