FURTHER SUGGESTIVE CRITICISMS 263 



curve is that the same fly population is accompanied by a rise in 

 the number of cases in early August and by a foil in early 

 September. If the time relations of fly prevalence and diarrhoea 

 cases is to be regarded as something more than an interesting 

 coincidence a satisfactory explanation must be found for these 

 facts. He rightly states that the number of flies is dependent 

 upon the accumulated effect of temperature because a considerable 

 period of warm weather, weeks or months, is required to produce 

 an abundance of flies from the few which come out of hibernation. 

 The dependence of the epidemic upon this factor is not so obvious 

 in the absence of exact knowledge as to the etiology of diarrhoea. 

 But assuming an infective agent of a bacterial nature with which 

 food might become contaminated he shows that the dose of in- 

 fection would not be dependent upon the accumulated effect of 

 temperature during the previous three months. It is just con- 

 ceivable that the virulence of the organism might be enhanced 

 by the continued influence of warm weather. Describing the 

 excellent a priori reasons for supposing that flies could transmit 

 the infective agent of diarrhoea he says : " Anyone familiar with 

 the domestic menage of the average working man on a hot summer 

 day, with the baby sick with diarrhoea and other small children 

 to care for, must realize that the opportunities afforded for fly 

 transmission are adequate enough." 



After discussing the data relating to the question to which 

 I have previously referred, namely, the decline of the epidemic 

 while the number of flies is still considerable, he groups the two 

 factors which may contribute in a varying degree to the decline 

 of the epidemic. First, a fall in temperature, diminishing (a) the 

 activity and number of the supposed transmitters, and (6) the dose 

 of infection which the child ingests owing to the effect upon the 

 rate of multiplication of the infective agent. Second, the exhaus- 

 tion of the more susceptible individuals, as pointed out by Niven 

 (I.e.), and also by Peters (1910) in describing an epidemic in 

 Mansfield, where, under similar conditions, in one section of the 

 town the epidemic was nearly finished whilst in another the 

 aforementioned factors appears to predominate. 



In commenting upon the evidence of Nash and Niven, which 

 I have already given, upon the relation between the number of 



