198 DK. JAMES J. SIMPSON. 



With regard to G. morsitans the following data are available. Bruce in his Report 

 of the Sleeping Sickness Commission of the Royal Society states that out of 500 flies 

 examined 288 or 57*6 per cent, contained mammalian blood, while only three con- 

 tained nucleated corpuscles, and two contained a mixture of both. Of 30 female 

 flies examined he states that 13, or 43*3 per cent, contained mammalian blood. In 

 another experiment he notes that 99 per cent, contained mammalian blood and only 

 1 per cent, avian. 



Lloyd states that out of 67 flies examined 24 per cent, contained mammalian blood, 

 and 7.4 non-mammalian.* 



Eminson in 1914 found that out of 500 flies examined there was blood in 43 ; in 41 

 there was mammalian blood, and two contained non-mammalian. 



This would tend to show that G. morsitans is much more dependent on mammalian 

 blood than is G. jpaljpalis. I had very little opportunity of studying the latter species 

 in large numbers, but several thousands of G. tachinoides were examined microscopi- 

 cally for stomach contents. In none of these could I detect anything of a vegetable 

 character. At Yapi 1,289 G. tachinoides were captured containing blood. The 

 percentage of mammalian and non-mammalian blood in the different batches examined 

 varied considerably, but taking the observations as a whole the result was as follows : 

 Mammalian 67*3 per cent., non-mammalian 32*7 per cent. 



Other observations at different places gave the following figures : — 



Flies. Mammalian blood. Non-mammalian blood. 



375 59 % 41 % 



287 68-5% 31-5% 



329 74-3% 26-7% 



It must be remembered that out of those taken with mammalian blood a certain 

 number must be deducted, as undoubtedly many of them had bitten the collectors. 

 This fairly exhaustive experiment goes to show that non-mammalian blood forms a 

 distinct food supply for G. tachinoides, and the point is raised : What is the source of 

 this food supply ? In my animal house at Yapi I kept various types of reptiles, 

 including the puff adder (Bitis arietans), black cobra (Naia nigricollis), the yellow- 

 spotted monitor (Varanus niloticus), and the common crocodile (Crocodilus niloticus)> 

 and have repeatedly seen G. tachinoides gorge itself on each and all of these species. 

 This is especially the case after these animals have had a heavy meal and are very 

 somnolent. On one occasion at Prang on the River Pru I came across a large sleeping 

 puff adder on which one G. palpalis and two G. tachinoides were feeding. 



Observations on birds were very different. Very few G. tachinoides managed to get 

 a full meal from any of the species kept. They never fed on fowls unless these were 

 tied down. This would distinctly point to reptilian rather than avian fauna as the 

 source of food in the case of those found with nucleated corpuscles ; and reptiles are 

 extremely common all along the river banks where G. tachinoides abound. 



The source of the mammalian blood is not so easy of solution. Various species of 

 small and large game come to the river in the mornings and evenings to drink, and of 



* [In Northern Rhodesia, Kinghorn, Yorke and Lloyd record (Ann. Trop. Med. Paras., 7 

 vii, 1913, p. 282) that out of 82 flies containing recognisable blood, 12 (or 14-6 per cent.) 

 contained nucleated red cells. — Ed.] 



