MANDUCA ATROPOS. 457 



Chichester, he had 10 pupae in August and September, 1877, 

 and that, without forcing, one emerged in November, another matured 

 at the same time, but died without disclosing the imago, that, by the 

 following April (1878), two pupae had died and one moth was 

 found dead in the cocoon, the others being then taken into a greenhouse 

 when imagines emerged as follows: June 18th J, June 23rd 

 cf , June 24th ?, June 25th $, June 28th $, June 29th $, 

 and July 5th $. Livett's note (En/., xix., p. 126) on forcing 

 pupae, suggests that his would naturally have gone over the 

 winter, for, although forced from October, the moths emerged 

 December 19th, January 7th, February 1st, and March 21st, 

 whilst one other pupa died on March 14th just as the imago 

 was ready for emergence, further, in 1896, the same observer 

 notes that he similarly forced pupae from September and obtained 

 imagines on November 4th, December 2nd-7th, 1896, and January 

 1st and 10th, 1896. Cambridge also states (Enf., xxxiv., p. 227) 

 that, of two pupae obtained from larvae in August, 1900, and 

 treated similarly (kept in room heated by ordinary fireplace), one 

 emerged in October, 1900, and the other on July nth, 1901, at 

 Bloxworth; whilst, in 1896, Newnham obtained from 6 pupae, 5 imagines 

 in October and one on April 17th, 1897, tne pupae being kept in 

 the house and the moss covering them slightly damped. Again, 

 comparing the records from the Marseilles districts with our own records, 

 we find that Siepi notes that, at Marseilles, his earliest capture of the 

 imago wild, has been on June 7th, his latest on September 24th, 

 his earliest bred example appearing on June 6th (from a larva that 

 pupated on October 20th the preceding year), whilst he also bred an 

 imago on August 23rd that pupated at the end of July. Of forced pupae 

 (kept at a dry heat of 20 C.) he has had them emerge as early as 

 November 14th and as late as April 14th. Powell states {in litt.) 

 that the larvae are found at Hyeres in October and November, 

 but that the late examples do not mature ; from two larvae taken 

 nearly fullfed on October 31st and November 1st, 1902, two healthy 

 pupae, however, resulted, which were kept out-of-doors all winter, 

 exposed to' all weathers and all temperatures, and from these two ima- 

 gines emerged, one on May 30th, 1903, the other on June 3rd, yet, he 

 says, one rarely sees the imago here at large nor has he ever seen 

 the larvae before September and has taken them quite young 

 in October, so that they can hardly be the progeny of May- 

 June emergences, although he has never himself taken, nor heard 

 of anyone else taking, summer-feeding larvae, but at Entrevaux, 

 from July 13th, 1903, onwards, he obtained 7 fullgrown larvae (4 of 

 which were ichneumoned) the 3 others pupating and producing imagines 

 (1 $ , 2 2 s) in August of the same year. Yet, across the Mediterranean, 

 Blackmore found larvae maturing all through the winter and 

 spring of 1869-70, in Morocco, and spring imagines emerging in 

 due course in March- April. But 170 years ago, the species 

 appears to have occurred in the Paris district much as it occurs 

 with us now, for Reaumur records (Memoires, ii., p. 295) fullfed larvae, 

 from Nainvilliers on July 16th, 1734, that pupated between July 17th 

 and 24th, the pupae producing imagines that did not lay eggs, in 

 Paris, between September 5th and October 29th, and adds that, in 

 nature, the imagines are disclosed about the end of September 



