XVl] REPRODUCTION 283 



had a directly injurious action on the fly. Further observa- 

 tions on this question are necessary in order to explain the 

 disappearance of G. morsitans from certain parts of Africa in 

 recent times. 



Reproduction. Most of the observations on the breeding 

 habits of G. morsitans have been made during the past two 

 or three years and are still very incomplete. Unlike G. longi- 

 palpis, morsitans readily copulates in captivity and in conse- 

 quence the flies can be raised in the laboratory without much 

 difficulty. 



Fischer found that on Lake Victoria, bred flies began to 

 drop larvae about the twentieth day after hatching. Kinghorn, 

 in Northern Rhodesia, states that 14 to 15 days is the usual 

 gestation period under laboratory conditions (temperature 

 58-5° to 77'8° F,), but much irregularity was displayed by the 

 females, and after the first larva had been born, many of them 

 did not produce a second one for a considerable time. In 

 Upper Dahomey, Roubaud found that the interval between 

 successive deposits of larvae, at a temperature of about 32° C, 

 was usually eight to nine days. 



Newstead found the pupae in three separate spots in the 

 forest, about one and a half miles from the banks of the 

 Shire River, at a place lying about 18 miles due north of 

 Liwonde, Nyasaland. The pupae were found buried in soil at 

 the foot of various trees. In Rhodesia, Jack has found the 

 pupae in several localities, e.g. under a clump of Mubula trees ; 

 on a high ant-heap ; under the exposed roots of a Baobab ; 

 at the base of a Mopani tree, etc. The author writes : " In all 

 places except two, where pupae were found, the soil was either 

 sandy and easily worked, often rich in humus, or covered with 

 leaves which afforded an easily penetrable shelter. In two 

 instances cases were taken from hard soil, in one instance one 

 and a half inches below the surface, but the chitin which forms 

 the case is an enduring substance in a dry situation, so that the 

 age of these cases is difficult to judge. The soil may have been 

 soft when the larvae entered or they may have penetrated along 

 a crack. It is in the highest degree improbable that a larva 

 could penetrate one and a half inches of hard baked ant-heap. 



