A 
usually associated with great difference in mass (1), and probably 
in the number of larval instars. 
2. It is in the larval stage that the life-habits and mass of the 
two phases are, as a rule, determined, but in the pupal stage the 
facies may be very materially affected, and even transformed. 
3. The larva, when first hatched, can develop into either the 
- winter or the summer phase, according to the conditions to which 
they are subjected, temperature being the governing if not the 
only one. 
4. The effect of temperature is different according to the diffe- 
rent periods of life in which the larva is subjected to it. 
5. The operation of temperatures on larva is by no means in 
direct proportion to their degree. It depends in great measure on 
the period of larval life in which they are applied. 
6. In S. dilunaria a forcing temperature 75°-80° F. (24°-27° C.) or 
one that is considerably higher than is usual in nature, applied 
during the whole of larval life causes large size, and a short pupal 
period. 
7. A mean «spring» temperature, applied during the whole 
of larval life, shortens in this species the pupal period more than 
does a mean temperature several degrees higher. 
8. Applied during the earlier part of larval life only, a « spring » 
temperature in S. bi/unaria and some other Heterocera promotes 
smallness of mass; and, in these species, also in A. levana, short 
pupal period; and probably in S. bilunaria it reduces the number 
of instars from 5 to 4. A summer temperature so applied promotes 
in S. dilunaria large size and also, unless several degrees above 
60° F. (15°-16° C.), in both long pupal period. 
(1) This difference in mass has been observed on by Prof. PouLTON 
(« Trans. Ent. Soc. Lond. », 1902, p. 415) and Dr. Dixey (« Ib. ». 1903, p. 159) 
and by others as indicating that in some cases of seasonal dimorphism condi- 
tions antecedent to the pupal stage must have been the cause. 
