1912.] T. gambiense and its Relation to G. palpalis. 531 



objection to the compass method is that the last fraction has often to be 

 estimated. This last criticism is not important in the present instance, since, 

 for example, the point 19 on the squared paper simply implies that the 

 length of the trypanosome falls between 18"50 /j, and 19 - 50 /x. The minute 

 fraction of a micron need not cause undue anxiety when it is remembered to 

 what rigorous treatment a soft protoplasmic organism has been subjected in 

 the obviously energetic process of fixing and staining. 



Moreover, it is the relative and not the absolute lengths that are significant 

 in the present connection, and, further, in no case do the important considera- 

 tions turn on very slight differences of size. All possible care has been taken 

 with the drawings, and it is hoped that the inevitable coefficient of human 

 error will be compensated for and evened out by the relatively large numbers 

 • of individuals involved. 



By comparing the daily curves the distribution of the different types of 

 trypauosomes in relation to the endogenous cycle can be observed with con- 

 siderable precision. Cages of newly hatched laboratory-bred flies were fed 

 daily when possible upon the selected monkey, and the nature of the trypano- 

 some infection at the time of feeding was further studied in relation to the 

 number of infected flies produced. Each cage was fed only once on the 

 infected monkey, and every possible care was taken to see that the flies had 

 actually fed. To obviate individual idiosyncrasies in the blood of clean 

 monkeys used to nourish the flies during the experiment, the monkeys were 

 pooled, and every cage was fed on each of the group in turn during the first 

 15 days. The nature of the food does not seem to affect the production of 

 infected flies after the first days. 



Owing to some other experiments being carried on at the same time in the 

 laboratory, most of the cages were actually tested as regards infectivity, and 

 produced typical infections in the test animals in every case in which plus 

 flies were present. 



IV. Endogenous Cycle. 



Before considering the experiments in detail, it is best to give a brief 

 account of the general results obtained. 



T. gambiense varies in length from 10 microns to 34 microns. The 

 minimum and maximum lengths are rarely reached, the bulk of the forms 

 varying between 15 fx, and 32 p. There is really no sharp distinction into 

 separated types, though the range of variation is relatively very wide. The 

 trypanosomes may, nevertheless, be divided up readily enough into short, 

 long, and intermediate individuals, but the intermediate forms insensibly 

 join up the short and the long specimens. There is a continued transition 

 from one type to another, but the duration of the different states varies 



