ACCIDENTS TO GESTATION 389 



fifteen feet from the high water mark ; Marshall also found large numbers 

 on the sandy shore of the mainland. From these observations it will be 

 seen that the breeding grounds of palpalis may be very extensive, and 

 although they .only relate to the shores of a large body of water, there can 

 be very little doubt that the fly must deposit its larvae in the sand on the 

 banks of rivers and small streams where it is said to be so 'prevalent ; it 

 is more than extraordinary that information on this point is so deficient. 

 The breeding grounds of palpalis and probably of all the Glossininae 

 recall those of the genus Stomoxys ; the latter often select the sand banks 

 of streams and rivers for oviposition, the damp sand being best suited 

 to the habits of the larvae. 



Much more information is yet required as to exactly where and at 

 what time of the day palpalis deposits its larvae, especially in the 

 temporary gites of Roubaud. The subject is of the utmost importance 

 in the fight against Sleeping Sickness. 



The duration of the pupal state is from thirty-two to thirty-five days ; it 

 can be reduced experimentally by an increased temperature. The pupae 

 possess greater powers of resistance to unfavourable conditions than the 

 imagines, but cannot resist a temperature of 30 C. to 35 C. They, do 

 not survive prolonged immersion in water or damp earth. 



It has been determined by observations on captive flies that the average 

 duration of the life of a female Glosslna palpalis is about three months, 



during which period she gives birth to from eight 



, , . i r r j Number of larvae 



to ten larvae, at intervals of from nine to ten days, 



produced 



once the reproductive functions have become esta- 

 blished. The period of development of the first larva is a long one, 

 and its birth may not take place for as long as three weeks or a month, 

 on account of the undeveloped state of the nourishing glands, which are 

 not then able to provide a sufficient supply of food. Once the function 

 is established, however, the reproduction proceeds very regularly, pro- 

 vided the food supply is copious and constant, and that the external con- 

 ditions remain favourable. The different seasons do not appear to 

 affect the production of larvae to any considerable extent. 



The pregnancy of Glossina palpalis, in captivity at least, does not 

 invariably proceed to a successful issue. In some cases, when the 



conditions are unfavourable, the larva is expelled 



. ,, r ... ,. Abortion and intra- 



prematurely, and before it is in a condition to pupate. uterine nympho8is 



This is especially liable to occur with flies captured 



and brought to the laboratory in an advanced condition of pregnancy ; 



it is probably due to their endeavours to escape from their unaccustomed 



