— 443 — 
But if a « forcing » temperature is so applied to the larva, the 
pupal period is greatly shortened (1). 
STAGES IN LARVAL LIFE WHEN TEMPERATURE EFFECTIVE. 
Experiments in large numbers have been made by me with the 
view of ascertaining at what stage in larval growth the low and 
the high temperatures respectively should be applied to be 
effective, larva being transferred from one temperature to the 
other in different stages, in the Ist, 2nd, 3rd, 4th, and sth 
instars, both with A. /evana and S. bilunaria. Without troubling 
you with details, I may state that experiments made upon many 
hundreds show conclusively that, when the larva are subjected to 
a « spring » temperature during the early part of their lives, the 
proportion of those, whose pupal period is absolutely the short one 
(that is the one proper to the pure summer phase), is much larger 
than in those whose young larva were at the « cool summer » or 
« full summer » temperature; and when the usual duration of that 
short summer phase period has been exceeded by both, its 
duration has in general been considerably shorter in the former 
case, where the first part of larval life was at « spring » temperature, 
than where the young larva were at the « cool summer » or « full 
summer » temperature. 
I have not been able to say, however, that there has been any 
transfer of summer phase S. bilunaria larva from the lower to the 
higher temperature in the ji/th instar, the reason being that I am 
nearly sure that under these circumstances there was no fifth 
instar. My time for observation, owing to other engagements, was 
usually limited to an hour or two in the early morning or late 
afternoon, and I found it impossible to note when the larval 
skins were changed, but certainly what I had thought, during the 
erowth at the lower temperature, to be their fourth instar proved 
(1) This may explain why in countries like Scotland, where the summer 
temperature is low, our S. dz/unaria is « single brooded » i. e. has but one, the 
winter phase, and why in other countries, in very hot summers, an additional 
summer phase brood is often interpolated in seasonally dimorphic species, 
making three generations within the year. 
