474 L. Rogers — 'Relationship of the water-supply, water-logging [No. 4, 



there must be several hundred tanks in the five square miles of Manik- 

 tolla alone, the chances of being able to destroy these larvee appears to be 

 very remote. Further observations are being made on the seasonal 

 distribution of these larvse and the amount of fever, but it may be 

 mentioned that they nearly disappeared from the tanks after the first 

 burst of the rains, and remained absent during a break which followed, 

 although fever now began to be prevalent, so that up to the present the 

 number of the Anopheles has been in inverse proportion to the amount 

 of fever. Possibly the tank forms are different from those of the rainy 

 season in the small pools, but I have not yet been able to settle this 

 point. The differences will be only microscopical, so that this would 

 not lessen the practical difficulties in lessening malaria in Bengal by 

 destroying the mosquitos, the only possible way of partially affecting 

 which would appear to lie in the time-honoured method of extensive 

 drainage in order to lessen the number of suitable breeding-grounds for 

 the mosquitos. 



The great difficulty of destroying the Anopheles larvse in Bengal 

 enhances the importance of the influence of a filtered water-supply in 

 reducing so materially the amount of fever, which has been shown to 

 be the case in portions of this tract of country, while the much greater 

 liability of the drinkers of tank water to malaria suggests that the 

 disease may commonly be obtained by drinking infected water, as has 

 for centuries been considered to be the case. Such a mode of infection 

 may be easily reconciled with the mosquito theory if we allow that 

 these insects, in addition to directly inoculating the disease, may also 

 take the parasite back to water, perhaps by means of the black spores 

 described by Ross, in which they may survive for a limited time only, 

 so that the infection has frequently to be renewed by the mosquitos. 

 This is a point which can only be settled by investigation, which I hope 

 shortly to be able to undertake. 



Lastly, an examination of charts showing the monthly rainfall and 

 fever-rate in this tract of country revealed the fact there is no constant 

 relationship between either the amount or monthly distribution of the 

 rainfall of different years and the amount of malarial fever. A more 

 detailed examination, however, showed that there is a relationship between 

 the daily distribution of the rain and the fever ; those years in which 

 the rainfall is very irregularly distributed with frequent and prolonged 

 breaks, being those in which malarial fevers are most prevalent. This 

 point is also being more closely studied, in conjunction with the 

 observations on the variations in the distribution of the Anopheles. 



