TRYPANOSOMES IN TSETSE-FLIES AND OTHER DIPTERA. 209 
trypanosomes from an infected to a clean animal by means 
of its proboscis. 
To prove direct transmission, however, does not disprove 
the existence of an indirect or cyclical method. It is 
logically impossible to prove a universal negative since any 
such proposition is invalidated by a single instance to the 
contrary. We can only say that our attempts to obtain 
evidence of a cyclical infection gave negative results. Such 
attempts were of two kinds: first, experiments to show 
periodicity in the effectiveness of the flies, such as is known 
to occur in the case of malaria; secondly, observations on the 
fate of T. gambiense when taken up by the fly. By 
experiment we obtained infection up to, but never after, 
forty-eight hours from the time that the fly fled on the 
infected animal. By observation, as already stated at length 
in this memoir, it was found that T. gambiense at first 
multiplied and became differentiated in the gut of Glossina 
palpalis or other Diptera, and also increased continually in 
size, but sooner or later died out and could never be found in 
G. palpalis on or after the fifth, sometimes not on the fourth 
day after infection. In all forty-two tsetse-flies have been 
examined by me at ninety-six hours or later after feeding 
on infected animals, with invariably negative results, although 
all the flies from the same batches were found to contain 
T. gambiense more or less abundantly when dissected at 
earlier periods. 
Very different results from those obtained with sleeping 
sickness are furnished by the recorded experiments and 
observations with regard to nagana, the tsetse-fly disease 
of animals caused by T. brucii. If we take first of all 
~Bruce’s experiments on the transmission (8) we are 
struck at once by the much smaller number of flies used to 
produce a result. Where hundreds were required for an 
infection with sleeping sickness, tens or even units were suffi- 
cient for an infection with nagana. ‘Thus in one experiment 
(228, loc. cit., p. 5) eight flies, fed four times on a healthy 
dog immediately after feeding on an infected one, produced 
