13 



numbers of cases of enteric fever (commencing) and the diar- 

 rhoea deaths for each week of the years from 1891 to 1908. 

 Whilst the latter show well-marked rises during some 

 10-12 weeks of the summer, the seasonal prevalence of the 

 former is not so striking. Speaking generally, the typhoid 

 season in Manchester is considerably later than the diar- 

 rhoea season, and the cases of enteric fever reach a maximum 

 in late autumn at a time when most of the flies have 

 already disappeared. The census of flies commenced in 

 1904, and coincides for the years 1904-1909 fairly well in 

 time with the diarrhoea deaths, if these are antedated 

 10 days to allow for the average period ensuing between 

 commencement of the disease and death. If we assume 

 that the same close time relation existed throughout, the 

 observations taken as a whole seem to me to lend little 

 support to the view that typhoid fever in Manchester is to 

 any material extent dependent upon transmission by flies. 



The time relation of fly prevalence and cases of typhoid 

 fever was also studied in Washington during 1908 by 

 Eosenau, Lumsden, and Kartle (1909) in collaboration with 

 Howard, entomologist to the United States Department 

 of Agriculture. In Washington in 1908 the flies (musca 

 domestica) appeared in considerable numbers in April and 

 reached a maximum in July. During August and September 

 they steadily diminished to about one-third and rose again 

 in October. The relation of the two curves was not such as to 

 suggest that flies are an important factor in the dissemination 

 of typhoid fever in Washington. In a subsequent report on 

 typhoid fever in Washington, Lennoden and Anderson (1911), 

 from an analysis of the incidence of typhoid fever upon the 

 population using privies or yard closets, discovered that the 

 greater prevalence of the disease among these persons during 

 the summer coincided with the fly season. 



Fig. 6 shows the time relation of fly prevalence to notifica- 

 tions of enteric fever for London during 1908. The data are 

 taken from Dr. W. H. Hamer's (1909) observations giving 

 the mean number of flies caught each day in 141 situations 

 grouped around nine centres. The observations being upon 

 a larger scale the sampling was presumably more representa- 

 tive of the fly population than the records in Manchester. I 

 have antedated the records of the number of cases of enteric 

 fever by three weeks, 7 days for average lapse of time prior 

 to notification, and 14 days for average incubation period 

 (Klinger, 1909). It will be seen that the number of cases of 

 enteric fever rises steadily during the fly season, but that the 

 rise continues independent of the fact that flies are rapidly 

 diminishing, and reaches a maximum at a time when they 

 have retired for the winter. 



As regards seasonal incidence of typhoid fever 1908 is 

 fairly typical of what has occurred in London during the 

 last 20 years. As the weather becomes warmer the number 

 of infections with typhoid fever increases and is not cut 



