42 THE FEVER CENSUS 



plague or cholera ; but this is not much use, for it 

 is difficult to say what a man died of by merely 

 looking at his skin. The Arab knows only one cause 

 of death ; it is, " God willed it that the heart should 

 stop." In India one may see in the death register 

 " Plenty much belly pains." But these are not very 

 scientific causes of death. Even suppose a death is 

 certified as fever, it is rarely stated which fever. 

 Some, but not very much, knowledge can be obtained 

 from the death registers ; even the little gained is 

 rarely reliable. 



The health officer should apply to the doctors. 

 If there is a local medical association or institute he 

 may utilise it for the purpose. Or he may write a 

 circular letter to all the medical men practising in 

 the town, inviting them to inform him of such cases 

 of fever as come to their knowledge ; but it is neces- 

 sary to explain to them the reason for the demand. 

 Experience shows that it is better not to make fever 

 a compulsorily notifiable disease. It is very tire- 

 some for any medical practitioner to be constantly 

 filling up and posting notification forms ; in many 

 countries there is no compulsory notification. Then 

 a medical man may notify a case as fever, and it 

 turns out to be appendicitis, and he is obliged to 

 correct his diagnosis ; this gives him extra trouble, 

 and it vitiates the returns. It is better to ask them 

 to send in a biennial list of all cases of " fever " 

 other than typhoid, typhus, Malta fever, or tubercu- 

 losis, to the health office. This gives less trouble to 

 the practitioner than regular notification. He can 



