148 CONCEK>Nli\G GLOSSINA MOKSITANS. 



advisable, if not essential. Any preconceived notion as to the association of 

 fly "with big game should be set aside. This study should be continued over 

 a year at leasts so as to include wet and dry seasons and intermediate periods. 

 The area should be accurately plotted on a large-scale map and the result 

 checked at regular intervals to determine the seasonal variations. No such 

 study has been made of a morsitans area. The distribution of wild animals 

 (not only " big game '') which might serve as food should be similarly 

 plotted. 



Breeding-grounds. — It is probable that the localisation, both general and 

 special, of morsitans depends on the requirements of the pupa, and that,, 

 when the breeding-grounds are found and studied, much that is puzzling in 

 its topical distribution will be explained. The natural breeding-grounds^ 

 have never been discovered. Their site Avill be determined chiefly by 

 conditions of shade and soil. They will occur in places sheltered from grass 

 fires. When the breeding-grounds have been discovered, it may be possible 

 to rid large areas from fly by the removal of certain shrubs. The relation of 

 breeding to season must be carefully studied. It has been pointed out that 

 tsetse are often found in sandy places. Snnd is the ideal soil for the 

 protection of the pupse. 



Diet. — Observations whether morsitans drinks water or feeds on anything 

 but blood are needed. 



Seasonal variations. — On what do the seasonal variations in numbers 

 depend, and what is their exact relation to meteorological conditions ? Is 

 the diminution of numbers during the dry season due to scarcity of animals 

 on which to feed, or to the attacks of birds or other enemies, or to abeyance 

 of breeding, or to delayed pupation so that few flies hatch out ? In 

 prolonged drought it is possible that all the adult flies die off, and that the 

 continued existence of the species is insured by the hatching out of old pupae 

 at the beginning of the rains. If larvae or pupae can be obtained, the effect 

 on development of varying degrees of dryness and moisture should be 

 experimentally tested. 



Enemies. — What enemies has the fly, either in the pupal or mature 

 stage ? 



Presence on one hank only of a river. — Tsetse sometimes are found on one 

 bank of a river but not on the other. What is the reason ? Is there in such 

 cases a diff*erence of soil, which favours perhaps the deposit of larvae on one 

 side but not on the other? 



Sex proportion. — On what does the usual predominance of males depend? 

 Does it, as with palpalis, vary with locality ? Does it depend on the position 

 of the breeding-grounds ? Is it a real excess, or only that males allow 

 themselves to be caught more readily than females ? 



Animals specially/ attacked. — It has been frequently noticed that tsetse 

 flies will often settle in some numbers on an animal that has just been shot. 



