ROUGH METHOD OF CALCULATION 49 



head of population per year, while in the latter it is 

 nearly two shillings. This difference is largely due to 

 the many regularly irrigated gardens at Ismailia, 

 which have to be dealt with every week. But the 

 comparative size of the two towns is also an im- 

 portant factor in producing the difference in the cost 

 of the two campaigns. It may be taken as a general 

 rule that the larger the town, the less, relatively, is 

 the cost per head of population per year ; and at the 

 same time, the more scattered the houses, the greater 

 becomes the annual cost of the work. Thus Port 

 Said is a compactly built town of 50,000 inhabitants, 

 whose houses are closely packed together, while 

 Ismailia is an openly built village of 10,000 people 

 only. 



In dry summer climates, such as that of the 

 Mediterranean littoral, it may be accepted that a cam- 

 paign against domestic mosquitos will cost 1,000 a 

 year for each 50,000 inhabitants in towns having less 

 than 250,000 population. In towns having less than 

 25,000 inhabitants the cost will be relatively greater, 

 while in those over 250,000 it will be relatively much 

 less ; but the scale is a geometrical one, and cannot 

 be estimated in arithmetical proportion. However, 

 a municipal rate of sixpence per head of population 

 per year should be sufficient for the upkeep of a mos- 

 quito campaign in towns and cities having a popula- 

 tion of more than 25,000 inhabitants ; but this rate 

 may have to be increased owing to summer rains or 

 floods, where there is no drainage or outflow for the 

 water over an impermeable soil, or if there is much 

 4 



