THE NATURAL HISTORY OF BOMBAY MALARIA. 407 



very like those seen in the larvas of N. stephensi but 

 usually a little clearer and longer. The external frontal 

 hairs are also often slightly branched or frayed but some- 

 times appear almost plain. Examine the fans or palmate 

 hairs and note that they are large and prominent. A pair 

 is present on the thorax. The ends of each leaflet is 

 broad, broken-looking and slightly pigmented as in N. 

 jamesi and the spine or filament is of medium length. 



Breeding Places of N. stephensi. 



As I have already pointed out the commonest dangerous anophe- 

 les in Bombay is N. stephensi and if we wish to reduce the 

 number of this species and so lessen the chance of the spreading 

 of malaria, we must know a good deal more about the habits of 

 this mosquito and the places in which it breeds so that we can 

 attack it with some hope of success. 



Last year shortly after the • commencement of the present en- 

 quiry, it became apparent that the breeding places usually selected 

 by N. stephensi could be roughly divided into two classes : — 



(1) Those that were permanent. 



(2) Those that were temporary. 



The permanent breeding places take the form of wells, tanks, 

 iron cisterns, fountains and garden tubs. The temporary breeding 

 places usually consist of surface pools of rain-water that collect in 

 the monsoon, tin-pots, hollows in machinery, kutcha and pucca 

 drains containing clean rain-water, &c. In regard to these two 

 types of breeding places, it is possible to lay down certain definite 

 rules : — 



(1) Larvas of N. stephensi are practically never found in 



really foul or dirty water. 



(2) Temporary breeding places only appear to become in- 



fected with larvas of A', stephensi when permanent 



breeding places exist in their vicinity. 



The point regarding dirty water is of importance because a great 



many people imagine that dirty puddles in the streets, of foul 



water in gully and drain traps is a source of malaria. This is not 



the case, because, though some kinds of mosquitoes will breed in 



