XVI] BIONOMICS 265 



Minchin thought that the fish-eating birds of this region 

 might furnish a constant supply of food and this view is sup- 

 ported by the fact that in captivity the flies readily feed on 

 fowls and moreover thrive on this diet. G. palpalis has been 

 observed to feed on lizards, chameleons, snakes, frogs and mud- 

 fish, in addition to birds and mammals, but the latter constitute 

 the most important hosts. There is little doubt that the fly 

 prefers man to any other host and in captivity it thrives better 

 when fed on human blood than on that of experimental 

 animals. 



As the fly feeds only during the daytime, it is evident that 

 the various larger animals, such as antelopes, buffaloes, etc., 

 that visit the water only at night time, or in the early morning 

 must be excluded as a probable source of food, and it is generally 

 agreed that palpalis is not dependent on the presence of big 

 game. 



Roubaud found that at 26 C. under ordinary conditions 

 the female flies, if permitted to gorge themselves, fed once 

 every three days, whereas the males would feed after two days, 

 even when they had not digested their previous food. If the 

 temperature was raised to 28 C. the females could be induced 

 to feed more often, and at this temperature digestion was 

 practically complete after 48 hours. In captivity the most 

 favourable temperature for the flies was found to be about 

 28 C., but this is near the limit ; an alternation of 26 C. 

 by day and - 30 C. by night was well borne, but a con- 

 tinuous temperature of 30 C. was rapidly fatal. The degree 

 of humidity has a marked effect on the ability of the fly to 

 resist starvation, for at 33 C. in a saturated atmosphere insects 

 were found to live six times longer than in one of moderate 

 humidity, and twelve times longer than in a perfectly dry 

 atmosphere. Similar results were obtained in the case of flies 

 kept at 26 C., for in a saturated atmosphere a male and female 

 were observed to live 10 and 13 days respectively, whereas 

 in a dry atmosphere most of the flies were dead after one 

 day and only one, a male, lived two days. 



It follows, therefore, that dryness of the air accelerates 

 nutritive changes and causes the fly to feed more frequently ; 



