1 68 AFRICAN NATURE NOTES CHAP. 



to the Zambesi, a few miles to the south of Daka, to 

 the " fly "-infested forest, which lay across the same 

 road a little to the north of Pandamatenka. The 

 country between these two " fly "-infested areas was 

 exactly similar in its vegetation, its altitude above 

 the sea, and in every other particular, as far as 

 one could see, to the " fly " belts which bounded 

 it to the north and south, and during the rainy 

 season buffaloes must have wandered through the 

 intermediate country, as well as through the two 

 " fly " belts. That tse-tse flies used to be found 

 in the greatest numbers along the wooded banks 

 of rivers such as the Chobi, the Zambesi and 

 many of its tributaries, was owing, I think, to the 

 fact that buffaloes had become excessively plentiful 

 in the same districts, not because the near neigh- 

 bourhood of water was necessary to them ; for many 

 "fly" belts, e.g. those extending across the road to 

 the Zambesi, were absolutely destitute of water 

 during several months of every year. 



The tenacity with which tse-tse flies cling to 

 certain tracts of country, or even narrow belts of 

 forest, is wonderful, but they sometimes move 

 beyond their usual limits nevertheless. About 

 forty years ago, a waggon track was made by 

 elephant hunters from Matabeleland to Hartley 

 Hills in Mashunaland. Not more than ten miles 

 to the north of the points where this waggon road 

 cut the Umzweswe and Umfuli rivers, the country 

 was always frequented by buffaloes and infested 

 with tse-tse flies. In my own experience, I have 

 often known large herds of buffaloes to come south 

 along the Umfuli river up to and beyond the 

 waggon road. I hunted and shot them there on 

 horseback for the last time in 1885, and used my 

 waggon and oxen to bring the meat and skins to 

 my camp ; and as my cattle did not suffer in any 

 way, there could not have been any " flies " about. 



