222 



Cholera in Relation to Certain Physical Phenomena. 



[ 1/ l-TI 



Comparisons of the phenomena of individual years fail to show such close corre- 

 spondence between conditions of atmospheric humidity and prevalence as is indicated 

 by the present data. It must, however, be borne in mind, that in order to institute 

 accurate comparisons for brief periods, such as individual months, it would be necessary 

 to know the actual distribution of the cholera prevalence throughout their course, and 



A combined Relative Humidity and 

 Temperature curve [12 years]. 



Total monthly Cholera Deaths [12 

 years]. 



ilHiiiiili 





 nil lllgll 



Diagram 4. — Relations of Cholera-prevalence for 12 years to contemporaneous Humidity 



and Temperature. 



to compare the data with those of the actual humidity coincident with the prevalence 

 at different times, the mere monthly averages of the two phenomena being in this 

 case evidently capable of giving rise to very incorrect conclusions. Moreover, the 

 degree of atmospheric humidity cannot be supposed to act directly in producing pre- 

 valence ; it can only act by increasing predisposition or b}'' favouring the devel opment, 

 diffusion, or preservation of the agent producing the disease, so that any influence 

 which it possesses may be neutralised by the action of other conditions. 



(d) Rainfall. 

 TABLE XXIV. 



Average monthly Rainfall (47 years) compared with Gholera-jjrevalence. 



In this diagram, as in Diagram 3, the line representing the rainfall has been 

 inverted, in order to show the somewhat reverse relation occupied by the rainfall and 

 cholera-prevalence. Beginning with the question of average rainfall, we find, first, a 

 period of months in which the rainfall does not amount to one inch, then three months 



