201 



through the night rested on the trunks of an avenue of rain trees- 

 [Eiiterolobiuin saina/i, Prain.) 



In order to obtain eggs for study, a specially constructed build- 

 ing, covered with brass gauze and having a cement Hoor was 

 erected, being sufficiently large to contain four to six carabaos. 

 In one portion of this a concrete water-tank tilled with growing- 

 plants was provided. Two carabaos were placed in this with 

 thousands of flies which were gatliered while resting on the rain 

 trees. In a short time females were observed feeding on the 

 hosts provided and several were found ovipositing in various 

 places about the enclosure. The author observed that the eggs 

 were invariably laid during the early afternoon and never later 

 than 2 o'clock (50 cases). They were laid in compact masses, 

 usually surrounding such objects as projecting splinters of wood, 

 fibres of jute sacking, fine brass wire, etc. When deposited on a 

 flat surface, leaves were selected on two occasions. In all other 

 cases the eggs were deposited on woodwork, etc., and invariably 

 on the shady portions. The method of oviposition is described at 

 length, and the flies were observed to lay with clock-like precision 

 at the rate of 10 eggs per minute ; in three cases the whole process 

 occupied 40 to 45 minutes. The eggs of this species of Tahamis 

 are laid with very little cementing material, which is waterproof 

 and was found to be insoluble in various strengths of alcohol and 

 xylene. The numbers laid in a mass vary greatly. In four 

 masses counted, there were respectively 270, 340, 417 and 425. 

 Hatching takes place in between three and five days, the period 

 being influenced by temperature and moisture. 



Immediately after emergence the young larvae hide themselves. 

 They were fed on various kinds of food and throve on minute 

 Crustacea, larvae of Stomod'ij.s and of mosquitos, and on young- 

 angleworms. Blowfly larvae were not satisfactory, unless pre- 

 viously killed, as they were capable of killing or injuring even 

 well-grown Tahanus larvae. Feeding is not a continuous opera- 

 tion and there is a long rest after a sufficient meal. At two days 

 old a Tahanus larva is capable of devouring 2i well-grown 

 Stovioicys larvae in 25 minutes. The method of feeding is care- 

 fully described, and the author says that the Tahanvs larva when 

 full-grown does not wait for its food, as in the young stage, but 

 actively pursues its prey. Cannibalism occurs, especially in the^ 

 presence of food which all cannot reach at the same time, and in 

 one experiment 85 per cent, of a large number of larvae were 

 destroyed in this way. 



The larvae can remain submerged in water for several minutes 

 at a time without apparent discomfort. They are capable of 

 swimming, but if placed in deep water the movements of the body 

 are apparently purposeless. The young larva is well adapted to 

 an aquatic life and was observed to have no difficulty in keeping- 

 afloat in water containing mosquito larvae and feeding upon 

 them, being quite capable of killing one fully five times its size. 

 The more developed larvae do not appear to be able to adapt them- 

 selves to life in water so well as the young ones. The process of 

 moulting is described at considerable length and also the full- 

 grown larva and pupa. 



