414 Transactions of the Society. 



experience, I associate with the food of the predaceous insect ; 

 this observation is, however, contradicted by that on the next pre- 

 paration in the point of colour, the granules being white and 

 sinning, and reflecting the light in the manner of starch. An 

 almost identical appearance is seen on a preparation of E. bruni- 

 pennis Mg. 



A female of E. livida L. affords valuable data, as it contains 

 digested and undigested food ; the hairs and scales of a gnat 

 (Culex), and a joint (probably of the palpus of a male) are quite 

 recognisable, and the digested food has a reddish brown appear- 

 ance of a finely granular texture ; where the hairs and scales are 

 thickest, the black colour is also present, combined with a more 

 digested portion. 



Dolichopodid.e. — The black stain in most of my preparations 

 is very strongly marked. Psilopa wiedemanni Fin. $ shows 

 pollen and spores of fungi. Another preparation has minute rods, 

 also probably fungoid. The contents of the stomach contradict 

 the character of the mouth, which has sharp teeth on the labrum, 

 and is, on that character, raptorial, though the tracheae of the 

 labium are singular, and unlike that part in all other insects of 

 the family. 



Gymnopternus assimilis Staeg. $ has pollen in the stomach. 

 The labium of these insects is longer than is usually found in the 

 Dolichopodidae, and they seem to have more of the characters of 

 flower feeders than is the rule in the species of this family. Their 

 mouth-parts appear to be intermediate between the ordinary and 

 a specialised type. There is one species — Orthochile rdgrocosrulea 

 Ltr. — that has quite a long labium, obviously modified to enable 

 the insect to reach the nectary of flowers. 



Medetcrus truncortim Mg. $ has, like E. livida, hair and scales 

 of Culex inside. 



Another species contains a greenish mass, which, however, 

 seems to be mixed with albumen. 



Two preparations of Campsicnemus curvipes Fin., male and 

 female, show a somewhat similar appearance to E. livida in the 

 digested part. 



Lonchopterid^e. — L.flavicauda Mg. $ is quite full of trans- 

 parent, structureless filaments, long, cone-shaped, and tapering, 

 which I have not met with in other insects, and are probably the 

 mycelium of some fungus or mould ; the other specimens of the 

 same insect only show the black granule. 



Platypezid^e and Pipunculid^e. — I cannot arrive at definite 

 conclusions as to what I find in my preparations of these 

 insects. 



Syrphid^e. — S. balteatus Deg. and S. ribesii L., like most of my 

 preparations of the insects of this family, show an immense mass 

 of undigested pollen. 



