258 THE RELATION OF FLIES TO SUMMER DIARRHOEA 



necessary defects, they show a degree of correspondence which 

 creates a high degree of probability that flies are the transmitting 

 agents in summer diarrhoea. In all the curves it will be seen that 

 deaths diminish more rapidly than do flies in the middle part of the 

 decline. For this there are two causes. In years of high diarrhoea 

 incidence the more susceptible and exposed infants have been 

 killed off or rendered immune. In every year towards the close 

 of the fly season the flies are attacked by Empusa muscae, and are 

 hindered by cold from leaving the house, so that they cease to act 

 as transmitting agents." 



For a full discussion of the careful observations made by Niven 

 in his exhaustive study of the epidemiology of this disease in 

 Manchester, the reader is referred to the original paper. In 

 summarising the results of his analysis Niven states, after de- 

 claring that summer diarrhoea is an infectious illness : " The 

 health of infants prior to attack — in other words, the social 

 condition — has much to do with the fatality. The summer wave 

 is not due to dust, nor is it conditioned by any growth of bacteria 

 in or on the soil. There is nothing to support the view that the 

 infective organisms are of animal origin, and the connection 

 between privy middens and diarrhoea goes far to prove the 

 contrary. The disease becomes more fatal only after house-flies 

 have been prevalent for some time, and its fatality rises as their 

 numbers increase and falls as they fall. The correspondence of 

 diarrhoeal fatality is closer with the number of flies in circulation 

 than with any other fact. The next closest connection is with 

 the readings of the four-foot thermometer, with which, however, 

 diarrhoeal fatality can have no direct relation. Flies and the 

 readings of the four-foot thermometer are both functions of air 

 and surface temperatures and of rainfall. Certain facts in the 

 life-history of the fly throw light on discrepancies arising in the 

 decline of flies and cases. The close correspondence between flies 

 and cases of fatal diarrhoea receives a general support from the 

 diarrhoea history of sanitary sub-divisions of the Manchester 

 district. The few facts available for the study of the correspond- 

 ence of flies and fatal cases in different sub-divisions, in the course 

 of the same year, also lend support to this view. No other 

 explanation even approximately fits the case." 



