MANDUCA ATROPOS. 431 



greened on the back at the segmental divisions, and there are 

 two very rough pear-shaped spots between the thorax and 

 abdomen. The colour is rich mahogany-brown, clouded in places 

 with a darker brown (Hellins). The age and sex of the pupae 

 have much to do with their weight, and the sex also influences 

 size in many cases (see Meldola, Annals and Magazine of Nat. 

 History, ser. 4, vol. xii., 1873). Fryer gives the weight of some 

 pupae of M. atropos collected in 1900 as follows : — 



14 collected in September — Greatest weight, 11-945 . . Least weight, 7706 grammes 



Mean weight, 9-004 ,, 

 7 ,, ,, October — Greatest weight, 12-515 .. Least weight, 8-387 ,, 



Mean weight, 10-340 ,, 



Mean weight of 2 r pupae 9472 ,, 



Russell notes (Ent. Rec., xii., p. 345) that, of five pupae that he 

 measured, two were 2^in. and three 2Jin. in length. Mitchell records 

 (Ent., xxiv., p. 76) an instance of a pupa in which, after the larval 

 skin had been successfully shed, the fluid contents of the pupa were 

 discharged through the wings which assumed a most abnormal 

 globular form. Glenny states (Ent. Rec, xiii., p. 157) that "it is quite 

 easy to distinguish during the day the pupae from which imagines will 

 emerge at night, since the wings dry away from the wing-cases, and 

 the yellow and blue-black markings of the body are quite easily 

 discernible." 



Pupal habits. — The pupal stage varies from a few weeks to 

 many months, those pupae that are formed in July and August 

 may emerge in from 6-8 weeks, or they may go over until May-July 

 of the following year. Sharp observes {Ent., xxix., p. 327) that the 

 pupa of this species is excessively active with its abdominal segments. 

 The abdominal segments 6-8 of the pupa are very deeply impressed 

 and ridged on each side, and this condition is correlated with the 

 capacity of movement,* for, though this sculpture looks quite shapeless, 

 yet, when the pupa is fully bent laterally, the largest ridge, close to 

 the stigma, exactly fits to the side of the preceding segment. (See also 

 Chapman, anted p. 428.) Morres also says that a healthy pupa of M. 

 atropos is exceedingly lively, so that, unless one is careful in handling 

 them, they will sometimes nearly wriggle out of the hand. It is, he 

 says, remarkable how different is the movement of the pupa if 

 it has lain long in the ground. After some time, say a couple 

 of months, they are often quite quiescent, and scarcely move at 

 all, sometimes being so stiff as to lead one to suppose that 

 they are dead. As to the time in which the pupal stage is reached in 

 Britain, i.e., the period at which pupae may be found, refeience 

 must be made anted pp. 408-413. Burrows writes (Ent. Rec., xiii., 

 p. 156): "The earliest date — spread over a series of years — on 

 which pupae have been brought to me, is September 2nd, 1894, at 

 Rainham, and the latest on October 20th, 1900, at Mucking, 

 a period really coincident with the potato-digging. The pupae, 

 when in health, are remarkably active ; on receiving their morning 



*A remarkable statement occurs (Ent. Rec, vi., p. 58) that the pupa has 

 no free segments and so cannot work its way through the earth. The pupa, of 

 course, has the normal number of free segments for Sphingid pupae, but, in common 

 with most other obtect pupae, does not leave the puparium on the emergence of the 

 moth. 



