204 DR. JAMES J. SIMPSON. 



During the dry season, when there was no water in the stream, only 3 per cent, of 

 those liberated at 2 miles, 8 per cent, of those liberated at 3 miles, and 6 per cent, of 

 those liberated at 4 miles were again caught, and that at no great distance from the 

 road. None of those liberated at 5 miles were again seen. 



In the wet season, however, when water was abundant in the stream, 10 per cent, 

 of those liberated at 2 miles, 13 per cent, of those liberated at 3 miles, 9 per cent, of 

 those liberated at 4 miles, and 6 per cent, of those liberated at 5 miles, were again 

 captured. 



Before being liberated, these tsetse were all fed as far as possible, and the cages 

 containing them were taken out at night and left open. It was found that by 2 a.m. 

 all the tsetse had escaped. Of course it must be remembered that there is considerable 

 traffic on this road, and many of these tsetse may have followed the natives for some 

 distance. 



However, from these experiments the following deductions may be drawn : — 



(1) The greatest distance covered by a single tsetse was four miles. 



(2) Apart from this none returned over two miles. 



(3) All seemed to return to the water, as none were caught at a greater distance 

 from the river than where they were liberated. 



(4) If there was any intermediate water between where the tsetse were liberated and 

 the large river, the tsetse discovered this and were making their way down stream to 

 the large river. 



(5) Where there was shade on the bank of a stream but no water, the tsetse 

 congregated there, but did not attempt to proceed towards the large river. 



(6) A thorough examination of the open bush on each side of the road revealed 

 not a single tsetse at a greater distance than 200 yards. 



(7) There was practically no discrepancy in the proportion of the sexes recaptured ; 

 if anything, there was an excess of males, but the numbers were too small to oe of 

 value. 



It must be remembered that these numbers may not — in fact, cannot — actually 

 represent the number of tsetse which really returned. The percentages must therefore 

 be taken not as absolute, but as proportionate. 



I then tried a similar experiment in the open bush, but found that this was practi- 

 cally impossible owing to the various directions in which the tsetse might return to 

 the sinuous river, and the impracticability of patrolling the whole area. 



On another occasion I liberated 1,000 tsetse at distances varying from one to ten 

 miles, along the bank of the river. It was found impossible to collate the returns, 

 but the greatest distance covered by an individual tsetse was seven miles, while 

 numerous flights of 2-4 miles were observed. It is more than probable, however, 

 that these followed the numerous canoes which constantly ply on the river. I have 

 already shown in my various reports the great attraction moving objects have for 

 tsetse. 



Of a large number of tsetse liberated at Yapi in September and October 1916, 

 some were taken as late as December 1916, while one was captured 15 miles down the 

 river in January 1916. This is evidently a clear case of a tsetse following a canoe. 

 While tsetse liberated on one bank have been captured on the other, I have not been 

 able to ascertain the total distance covered in one distinct flight. 



