164 THE ENTOMOLOGIST. 



spotted form (males and females), the single exception remaining still 

 in pupa, which it was presumed would appear during the coming May 

 in tlie vernal dress. In this latter case he had reared a second hrood 

 of larvae from eggs laid by some of the Jul}' females, all of which were 

 now in the chrysalis state. Mr. Cole added : — ' May not the above 

 be considered a case of "season-dimorphism" analogous to that 

 occurring in Pieris, Amschnia, Selenia, &c., as investigated by Dr. 

 Weismann, a slow process of development during the winter being 

 necessary for the May form (which may be considered the type), 

 whilst if the development of the pupa is hastened by the heat (and 

 light?) of summer, the smaller and less perfect individuals are the 

 result ? Referring to the similar case of Selenia illustraria, Dr. Knaggs 

 (Ent. Mo. Mag. iii. 238) remarks as follows : — " It is pretty well 

 known that in the natural sequence S. illustraria reproduces itself in 

 the form of S. delunaria, and vice versa. But what I assert is, that 

 whenever (whether at large, owing to exceptionally hot or long 

 summer seasons, or in captivity from warmth, assisted perhaps by 

 what Mr. Crewe has happily termed ' feeding up quickly ') the 

 completion of the pupal stage is accelerated, then S. delunaria 

 produces delunaria, not illustraria. Further, it is my belief 

 that the converse will be found to hold good, viz., that should the 

 completion of the pupal stage be retarded either by cold seasons or 

 climates in a state of nature, or artificially by aid of an ice-well, 

 S. illustraria, not S. delunaria, would be found to result from 

 S. illustraria." And again (loc. cit , p. iJ56) he puts it thus: — "If 

 1. = illustraria, D. = delunaria, and — = winter ; then if there 

 be but one brood in the year the sequence will be I. — I. — /., and 

 so on ; if two broods, I. D. — I. D. — /. D., and so on ; if three 

 broods, /. D. D. — I. D. D., and so on." I have not yet tried the 

 effect of artificial retardation on the pupse of Ephyra, but intend to 

 do so when opportunity offers. My experiment shows that the effect 

 of natural retardation over the winter months is to produce the type 

 whatever may be the form of the parents ; and that such natural 

 retardation does usually (? always) occur in double-brooded species I 

 believe to be true from my experience in breeding various insects. 

 Remembering that the summer broods of season-dimorphic species 

 are smaller, and apparently vitally weaker tlian the spring ones, and 

 that it is from the former that the latter are usually descended, may 

 we not assume that the provision by which some few of the direct 

 offspring of the spring forms are preserved through the winter in the 

 pupal state, and so are enabled to pair with the offspring of the 

 summer form, is of advantage to the species, in affording a "cross" 

 between individuals which have developed under very different 

 conditions? A similar benefit may be derived in the conunonly 

 observed case of individutil pupte of single-brooded moths (e.g., Erio- 

 gaster and many Nutodontida) rcinaiuing two, three, or more years in 

 that stage, and then eventually making their appearance at the 

 proper season with the ordinary flight of the species. As bearing on 

 tlie above suggestion, I may refer to what occurs in those single- 



