152 SUMiMER DIARRHCEA 



" If the house-fly is the transmitting agent in summer diarrhoea, 

 the following conditions should be fulfilled : 



(i) {a) "There should be evidence that the house-fly carries 

 bacteria under the ordinary summer conditions ; (I?) house-flies 

 should be present in sufficient numbers in houses invaded b}' 

 fatal diarrhoea. 



(2) "There should be a close correspondence between the 

 aggregate number of house-flies in houses and the aggregate 

 number of deaths from diarrhoea week by week. 



(3) " The life-history of the house-fly should explain any 

 discrepancy between the observed number of flies and the 

 observed number of deaths. 



(4) " The minority of breast-fed children not apparently 

 accessible to infection should receive explanation. 



(5) "There should be a closer correspondence of diarrhoeal 

 fatality with the number of flies than with any other varying 

 seasonal fact. 



(6) " Any other closely corresponding seasonal fact should 

 be capable of interpretation in terms of the number of house- 

 flies. 



(7) " Any variation from district to district in the annual 

 curve of deaths should be accompanied by a similar variation in 

 the curve of flies. 



(8) " No other available hypotheses must be capable of 

 explaining the course of summer diarrhoea." 



In regard to i (a) it has been shown in previous chapters 

 that flies carry numerous bacteria under ordinary summer con- 

 ditions. — I (d) and 2. — In order to estimate the number of flies 

 in houses from 1904 onwards, Niven provided certain reliable 

 householders with bell-glass traps, which were baited with beer. 

 Careful and continuous daily counts were made. At each station 

 the numbers caught per week varied, and it was only when the 

 numbers caught at all the stations were added together that 

 coherent and graduated curves could be obtained. Niven records 

 for five years the number of deaths week by week, the number of 

 flies captured, the mean atmospheric temperature and the mean 

 temperatures at depths of i ft. and 4 ft. Charts 3 and 4 giving 

 the numbers of flies captured and the diarrhoea deaths in 1904 



