278 J. W. SCOTT MACFIE. 



fasciala and Culexjatk/mi^, but the eoninionest mosquitos in the houses are Mmisoni- 

 oides <ifricanus and M. unifonnis. The peculiar habits of the larvae of Mansonioiiles 

 and theii' association with tlie water-weed, Pistio straliotcs, make it possible to locate 

 their breeding places fairly accurately-, and it can confidently be affirmed that at 

 Accra these mosquitos must often fly more than a mile. It is useless to expect, or 

 even to hope, for a total abolition of mosquitos by oiling in these circumsta;ices. 



Another feature at Accra that militates against the success of " oiling " is the 

 presence of lagoons of such a size and in such situations that it is imjiracticable to deal 

 with them by this method. In the brackish water of these lagoons C. thalassins, 

 a species often found in the houses in Accra, breeds freely. The larvae, it is true, 

 seem to be limited to the shallo\\ margins of the lagoons, but this does not render 

 them vulnerable to " oiling,"* and as will be mentioned later, this species is able to 

 live submerged for a very long time without having direct access to the air. 



The Mode of Action of Kerosene as a Larvicide. 



Many methods of applymg the oil have been recommended, and the amount neces- 

 sary per square metre of surface has been estimated at widely differing figures. For 

 small pools Boss (1911) advocates " painting " the oil over the surface with a bunch 

 of rag tied to the end of a stick ; but such details, he says, are unimportant, the essential 

 point being, as he has stated elsewhere (19(,i(.)), that "' the film must spread all over the 

 surface of the water, and must last for at least half an hour." In each locahty the 

 best method of application and the amount of oil required per square metre must be 

 determined by experiment. The presence of weeds must also be considered, as they 

 interfere Mith the action of the oil, probably by breaking the film. In laboratorv 

 experiments, however, the mere presence of organic matter in the water is found to 

 diminish the action of crude kerosene. This is illustrated by the experiment sum- 

 marised in Table I, in which young larvae of Stegonujia fasciata were used, namelv 

 larvae less than 24 hours old and still in the first phase of their development. The 

 larvae were kept in identical vessels in (1) distilled water, (2) tap water, and (3) a 

 natural ^\ fasciata medium, all being fluids in which control larvae lived for days. 

 The kerosene was applied to the surface in a proportion equal to 110 cc. per square 

 metre, and as will be seen from the table, its action was much more pronounced in the 

 pure medium. In the media containing much organic matter, indeed, some of the 

 larvae escaped its action and were still alive after 57 hours, when the experiment was 

 ended as it was thought that lack of food might be becoming an appreciable factor. 

 There was still a continuous film of oil over the surfaces of the media when the experi- 

 ment was discontinued. AVhy this should be so I am unable to explaiji, for. as will 

 be shown later, solution of the oil does not appear to play any part in its larvicidal 

 action, aiui the fact is all the more remarkable because in other circumstances the 

 ])resence of organic matter tends to diminish the period during which submerged 

 larvae can survive. But perhaps it may have been due to the tendency shown by the 

 larvae of S. fasciata.. and indeed of most if not all mosquito larvae, to lie at times in an 

 apparently inert condition on the nnid or debris at the bottom. In this " resting'" 

 state their oxygen absorption may be but slight, and if they remain below long enough, 

 they may escape the more pronounced action of kerosene and on eventually coming 

 to the surface mav have to encounter onlv a thin film of the oil. 



