94 
A Study of Trypanosome Strains 
(b) The next comparison I make is between the Mzimba (Donkey) Strain 
taken through rats and the above wild-game strains. I have added the data for 
the Mzimba Sti'ain to the last table (p. 93) : it is given by Sir David Bruce and 
others in a paper on the Mzimba Strain*. I compare the Reedbuck and the 
Mzimba (Donkey) strains first. We find: 
;3^;== 53-37, P = -000,05. 
Thus only once in 20,000 trials would a divergence as great as this arise, if the 
two strains were samples from the same population. 
The results of comparing the Mzimba strain with Waterbuck and Hartebeeste (1) 
give respectively 
X'=IU-2S, P = < -000,000,1, 
and 171-00, P = < -000,000,1. 
These give for practical purposes impossibility of a common source, thus still 
further demonstrating that the marked feature of the wild-game and Mzimba 
strains is divergence, not sameness. 
Sir David Bruce and his colleagues writef : " The trypanosome of the Mzimba 
strain is the same species as that occurring in the wild-game inhabiting the 
Proclaimed Area, Nyasaland." In an earlier paper a diagram j is given of the 
frequency distribution of 3600 trypanosomes of Human strain taken from the rat 
alone. These are drawn from four native cases of sleeping sickness in Nyasaland 
and from one European case from Portuguese East Africa. As the individual 
cases for the rats alone are not given, they have had to be read off the per- 
centage diagram, but the frequencies must be very nearly correct. This Human 
strain may be compared with the T. vhodesiense, the T. brucei, the Mzimba 
(Donkey) strain and a strain obtained from a native woman suffering from 
" Kaodzera," the so-called sleeping sickness of Nyasaland. The frequencies of 
these five strains are given in the following table. I first compare the trypano- 
somes of Nyasaland given as (b) above with T. brucei and T. rhodesiense, for this 
is the comparison made by the authors themselves§. 
Taking the trypanosomes of Nyasaland (b) and the T. brucei as figured in 
percentage curves by Bruce and others, we have 
;^- = 72-17, P < -000,000,1, 
or it is impossible to ascribe any degree of sameness to these two strains. We 
now compare the Nyasaland strain (6) with T. rhodesiense, and find 
= 69-95, P = -000,01; 
* R. S. Proc. Vol. 87, B, p. 31, 1913. 
t E. S. Proc. Vol. 87, B, p. 34. 
+ R. S. Proc. Vol. 86, B, p. 301. 
§ E. S. Proc. Vol. 85, B, pp. 431 and 433. 
