94 Bulletin de la Société Entomologique d'Egypte 



enclosing sheallis. But our lingers are too clumsy and 

 our tools loo coarse to deal with so delicate a creature 

 and the specimens liberated in this manner are nearly 

 always ciipples. 



Further, oi)servalion shows Ihal, in order to moult 

 successlully the Empusa has only a short time at its 

 disposal namely that during which the new cuticle is 

 still malleable. As Fabhe has shown, it is owing to 

 this condition that the spines of the femora and tibiae 

 can be depressed at the right time. After that, in sj)ite 

 of all attemjits, llie legs remain caught within the sheaths. 



As to the causes ol failure, the (irsi to be noted is 

 tlie debili ly resulting from breeding in captivity. Whe- 

 ther its lavourite food is missing or whether, l)eing 

 dei)rived of Ihe morning dew, it cannot drink suffi- 

 ciently, a specimen bred in a cage is nearly always weak. 

 H is in vain tha', under these circumstances tbe Insect 

 tries to retard its moult as long as possible (1)- 

 Dénutrition continues and, as a consecjuence, atony of 

 the muscles, diminulion of tbe blood, lowering of pres- 

 sure, perhaps abnormal diying of the e})idermis (2). 



When llie critical moment arrives, lher]mi)usa tries 

 hard to moult, but, doomed before hand, at best pro- 

 duces a cripple. 



Another cause of failure is the mutilation of one of 



I 



1. Two larvae o[ Einpaan taken to Paris on June 1st. 

 191() and fed only on liouse flies postponed llieir moult till 

 Se])tcniber. Both died after failing to free llicmselves. 



2. Certain facts observed in Lepidoptera allow one to 

 suppose that in the case of other insects equally, a liquid 

 destined to facilitate nioillting is secreted between the old 

 and tlie new cuticle. 



