488 DIPTERA CHAP. 



after day, changing its position frequently, until it has completely 

 emptied the pupa of its contents, nothing being left but the skin. 

 Although this is accomplished without any wound being inflicted, 

 so effectual is the process that all the Chalicodoma is gradually 

 absorbed. The time requisite for completely emptying the victim 

 is from twelve to fifteen days ; at the end of this time the 

 A n t/ir ax-larva, is full grown, and the question arises, how is it to 

 escape from the cell of solid masonry in which it is imprisoned ? 

 It entered this cell as a tiny, slender worm through a minute 

 orifice or crack, but it has now much increased in size, and exit 

 for a creature of its organisation is not possible. For some 

 months it remains a quiescent larva in the cell of the Chali- 

 codoma, but in the spring of the succeeding year it undergoes 

 another metamorphosis, and appears as _a pupa provided with a 

 formidable apparatus for breaking down the masonry by which 

 it is imprisoned. The head is large and covered in front with 

 six hard spines, to be used in striking and piercing the masonry, 

 while the other extremity of the body bears some curious horns, 

 the middle segments being armed with rigid hairs directed back- 

 wards, and thus facilitating movement in a forward direction and 

 preventing slipping backwards. The pupa is strongly curved, 

 and fixes itself by the aid of the posterior spines; then, unbend- 

 ing itself, it strikes with the armour of the other extremity 

 against the opposing wall, which is thus destroyed piecemeal 

 until a gallery of exit is formed ; when this is completed the 

 pupa -skin bursts and the perfect fly emerges, leaving the 

 pupa-case still fixed in the gallery. Thus this species appears 

 in four consecutive forms in addition to the egg each of 

 which is highly specialised for the purposes of existence in that 

 stage. 



The habits of our British Bomlnjlius major have been partially 

 observed by Dr. Chapman, 1 and exhibit a close analogy with 

 those of Anthrax trifasciata. The bee-larva that served as food 

 was in this case Andrena labialis, and the egg was deposited 

 by the fly, when hovering, by jerking it against the bank in 

 which the nest of the bee was placed. 



It has recently been discovered that the larvae of various 

 species of Bombyliidae are of great service by devouring the 

 eggs of locusts. Eiley found that the egg-cases of Caloptcnus 



1 Ent. Mag. xiv. 1878, p. 196. 



