1910.] Sickness Studied by Precise Enumerative Methods. 



415 



very large and their graph is quite irregular. They abandoned counts in 

 measured quantities of blood as " unreliable." Thomas and Breinl* showed 

 that in three cases of Sleeping Sickness the numbers of trypanosomes found 

 in " fresh cover-slip preparations " varied irregularly from time to time. 

 Koch, Beck, and Kleine (1909) remark on the irregularity of the appearance 

 of T. gambicnse in African natives, and state the parasites are present 

 for two to five days and absent for two to three weeks. Salvin-Moore and 

 Breinlf show a graph with two undulations and a final premortal rise in two 

 heavily infected rats, and give a detailed description of corresponding 

 changes in the parasites. Apparently, hitherto, only irregular variations in 

 the numbers of the parasites seem to have been recognised ; probably the 

 large error due to inadequate methods of counting has disguised the regular 

 periodicity of the variation shown by more exact counts in the 11 successive 

 undulations observed in our case. 



We should add that our methods enable us to detect parasites when they 

 are in numbers so small that their detection by the ordinary methods would 

 be exceedingly laborious. Hence if our case had been studied by the 

 ordinary methods, probably only the crests of the rises would have been 

 visible in the chart, and it would have been said that the parasites had 

 disappeared in the intervals. 



* ' Memoirs of the Liverpool School of Tropical Medicine,' vol. 16, 1905. 

 t 'Annals of Tropical Medicine, Liverpool,' vol. 1, No. 3, 1907. 



