An experimental attempt was made to produce a fly-free corridor through a belt of savannah woodland containing the tsetse fly G. swynnertoni.

An area two miles wide and four miles long was treated with a coarse spray of a 4·6 per cent, w/v solution of DDT in Shell Diesoline. The dosage per application was 0·5 gallons per acre, and seven applications were made, at intervals of approximately two weeks, so that the treatment covered two pupal periods.

The fly density had fallen to a very low level by the end of the experiment, and the area remained virtually free from flies for the subsequent two months. An examination of the data suggests, however, that the fly population was maintained largely by immigrant flies, and was certainly subject to wide variations, and it seems certain that the effect of the applications would have been considerably less upon a stable, self-supporting population.

The drop spectrum of the ground deposit had a mass median diameter of 0·35 mm., and the recovery of insecticide in the area was approximately 60 per cent. Leeward and under sides of obstacles did not receive a dose although in some cases dosages were obtained on apparently leeward sides, probably because of local reversals of wind direction.