378 C. F. M. SWYNNERTON. 



completely by leaf-fali or simultaneous burning. It would be difficult to ascertain 

 their value until one bad already successfully applied tbe local measures. 



There are other difficulties also to be faced in dealing with the individual centres. 

 Their probable large number in a given area — ^though in the area visited they 

 occupied only a small number of the places that appeared suited to them — gives rise 

 to difficulties (1) in relation to finding them all, (2) in relation to the number of 

 white workers that it might be necessary to employ when simultaneity seemed 

 important and time short, as in work that might be planned against the emerged 

 flies. Here the period when emergence is already rapid, but the fly not yet greatly 

 scattered, might be of special importance. There is also the fact that the game 

 must first be banished from each centre before the ffies will present themselves for 

 treatment in anything approaching their full numbers. 



Broader lines of work are undoubtedly preferable if any should be foimd effective^ 

 and it will be most interesting to see the result of the Southern Rhodesian Govern- 

 ment's present measures against morsitans, planned, as I understand they are, 

 on game removal and large measures of bush destruction. 



To discuss one or two of the points, nevertheless, it may be said that the great 

 reinforcement of the visible fly sometimes called out by the presence of my cattle 

 showed that ordinary collecting by hand is a very inadequate method, especially 

 with game about. Collecting round cattle would be far better, but their length 

 of life under continual attack is a point to be considered in relation to the financial 

 feasibility of the scheme. It might well be prolonged by antimony injections, 

 but would otherwise (to judge from my experience) be a matter of a very few wrecks 

 indeed. Better still would it be if the flies coming to the cattle could be automati- 

 cally poisoned. The effect of frequent arsenical spraying, although it seems quite 

 unlikely to be useful, might be tested experimentally, for it has not been my experience 

 that a tsetse keeps its proboscis buried to the bulb all the time it is feeding, and 

 a possibility I have referred to elsewhere (p. 341) might also be worth a trial. 



If the breeding concentration seen by me at the Kanyezi centre should be typica], 

 it suggests that in areas where such concentrations are the rule it might be possible 

 to devise useful measures against puparia. Artificial breeding places, unless 

 scattered nearly as freely as the present fallen logs, though with a regularity that 

 would make them more easily found, would not, I think, be useful ; for my indirect 

 evidence all pointed to the probability that the heavily pregnant female does not 

 fly many yards from where she last fed. 



The present situation at Gundoda's vleis suggests that the placing of Kafir villages 

 at the fly-centres might break these up through the effect on the game, but with 

 so many apparently eligible sites vacant the concentrations would presumably 

 reform elsewhere. This measure alone would only be useful if it were possible to 

 plant a population at such places through the area generally. 



So far as G. brevipalpis is concerned^, the observations spoken of on p. 351 would 

 require to be greatly added to before we could assume that this fly's permanent 

 boundary could be put back merely by the splitting up of the bush on its western 

 margin. The point is worth looking into further. In any case the game would 

 require to be destroyed or excluded in order that the fly might not be carried back 

 in the rains. 



