INQ UIR V INTO MILK S UP PL V 379 



tion of the cases as a whole. Sometimes this may be done by a 

 " spot-map," on which each case of the disease is marked. By such 

 a simple plan groups of cases, if such groups exist, at once become 

 evident. Or lists of the customers of the suspected dairy can 

 generally be obtained and compared with a list of the persons 

 suffering from the disease. Charts of one kind or another have 

 proved very serviceable in tracing out a community of the milk 

 supply. One of the most effective and graphic of these methods 

 was devised by Dr Davies, the Medical Officer of the City of 

 Bristol. In 1897 there occurred in Bristol an epidemic of enteric 

 fever attacking more than 240 persons. This was traced by Dr 

 Davies to three milk-rounds of polluted milk coming in to Clifton 

 from the country. The facts which he was able to discover and 

 the graphic manner in which he presented these facts are so full 

 of instruction that we may quote his words at some length. Not 

 only is the "graphic method" recorrtmended by Dr Davies one 

 very useful for adoption, but the method of inquiry and the kind 

 of fact elicited, as revealed in the following paragraphs, provide an 

 admirable illustration of how to investigate milk-borne disease. 

 Speaking of the sources of the infected milk, he writes : — ^ 



" I. The milk from the farm X, situate in the county of Somer- 

 set, was consigned to a distributor A whose man B met it on its 

 entry into Clifton, over the Suspension Bridge, and distributed it 

 directly from the churn to the customers without its going to 

 any local dairy. Any milk not used on the round was, however, 

 returned to the branch dairy and sold to casual customers. Before 

 the 26th September, when the big schools reopened, the supply 

 was sufficient to supply two districts. A, the lower level, and B, 

 the higher level : but after this date the supply was confined to 

 the high level district {see diagram). Some of the cases seem to 

 have developed on the low level round at the beginning of October, 

 while the incidence on the high level round was more marked in 

 the second week of October. 



"On the low-level round : — 



Batio of attacks 

 to houses. 

 27 houses were supplied . . . .) 



12 were attacked /44-4 per cent. 



31 cases resulted. 



" On the high-level round : — 



29 houses were supplied . . . .1 ^ 



19 were attacked |&S-5 per cent. 



83 cases resulted. 



^ Trans, of Epidemiological Society of LoTidon, 1898, vol. xvii., p. 82. 



