134 Mr. F. Merrifield's systematic temperature 



species belonging rather to the warmer parts of Europe, 

 and alniar'ia, as belonging more to the northern parts. 

 This year, instead of mixing in equal proportions the 

 eggs of several parents, so as to have a better chance of 

 obtaining hereditary varieties, I limited myself to the 

 eggs of a single pair, thus providing more exact means 

 of comparison. 



(1) Preliminary cooling and icing experiments with 

 illustraria. — Those who are acquainted with Prof. Weis- 

 mann's studies in heredity as translated and added to 

 by Prof. Meldola, will remember that both in his experi- 

 ments and in those of Mr. Edwards as recorded in that 

 work, icing was generally not very efficacious unless 

 applied to the pupa in an early stage, and that when so 

 applied it was often fatal. Mr. Poulton's discoveries 

 with regard to the pupfe and pupating larvae of some of 

 the butterflies showed that the effect of the surrounding 

 circumstances was in them produced only at a still 

 earlier stage, viz., on the larva shortly before pupation. 

 Though there seemed no great reason to suppose that 

 the same principle would apply to pupte protected by 

 enclosure within leaves, it appeared desirable to test the 

 matter, and to give the icing every opportunity for 

 operating. My first experiments were designed to ascer- 

 tain whether cold could be borne in the earlier part of 

 the pupating and pupal stages, and, as regards pupte 

 which would ordinarily pass the pupal stage in summer, 

 what degree of cold they would bear as fully-formed 

 and hardened pupae. 



(i) Icing larvcejust spun up. — A first experiment with 

 three larvae of illustraria which had recently spun up, 

 and were cooled for 16 days, showed that they had all 

 jDupated at that low temperature, and I then determined 

 to try if such larvae would pupate at the " icing" tem- 

 perature of 33^. Accordingly (Table III.) I placed 8 in 

 ice just after they had spun up. After 29 days they 

 looked healthy and had spun a few threads, but had not 

 turned; I therefore moved them to the " cooling" tem- 

 perature, and then in about 6 days 4 of them pupated. 

 These were afterwards iced as pupae for 42 days ; one 

 died and the other three came out as perfect moths. The 

 other 4 were after 8 days' cooling exposed to the ordinary 

 temperature and dry air of the room — about 70° at this 

 time — and all died in a day or two without pupating. 



