174 J. W. SCOTT MACFIE AND A. INGRAM. 



as the main factor. That this is the sole factor is unhkely, as ah-eady pointed out in 

 this paper wath regard to those larvae wliich are reared in domestic pots ; for if it were, 

 it would be difficult to account for the relatively gi'eater frequency of C. duttoni 

 amongst the domestic mosquitos of Lagos as compared with those of Accra, or to 

 explain the presence of C. fatigans in considerable numbers amongst the domestic 

 mosquitos of Accra and its absence from the collections made at Lagos. Again 

 C. tigripes y&r.fusca, the larva of which is carnivorous and is not particular as to the 

 species of larvae it preys upon, occurs in the proportion of 5'3 per cent, amongst the 

 domestic mosquitos of Lagos, whereas its proportion is only 0"23 per cent, amongst 

 similar mosquitos at Accra. 



If the food supply of the larva is the preponderating factor in determining the 

 distribution of the adult mos^iuito and if, as seems probable, the growth of the minute 

 organisms, chiefly of a vegetable nature, which constitute the food of the larva are 

 just as dependent upon varying conditions of atmosphere and the composition of the 

 medium in Avhich they grow as are the higher plants of the country, then it should be 

 possible to indicate roughly in a table, which gives a return of the mosquitos taken 

 in different districts that vary in their physical conditions, a corresponding variation 

 in the distribution of the mosquitos. Table M!L, which has been drawn up from the 

 " Distribution of Mosquitos in West Africa " and from our own experience of the 

 prevalence of certain species in different districts^ is of interest in this connexion. 



It has been shown by one of us* that certain mosquito larvae are very tolerant 

 of salt in the medium in which they live. As there is a considerable amount of 

 chlorides in Accra water, its presence may have some bearing on the anomaly in the 

 distribution of C. duttoni and C. fatigans in this district. C. fatigans is capable of 

 breeding in brackish water, while C. duttoni, according to Graham.f is intolerant of 

 even a moderate amount of chlorides. Mosquitos such as 0. irritans and C. thalassius, 

 which breed by preference in brackish water, are not found far from the coast. 



Graham J has pointed out that larvae of A. costal is failed to increase in size when 

 placed in clear tank-water, but that growth proceeded upon the addition of small 

 quantities of human urine (the addition of which may, however, have transformed 

 the clear tank- water into a culture medium) which contains sodium chloride, and that 

 the water of Accra and Sekondi, where A. costalis is the dominant Anopheline, is 

 slightly brackish. A. costalis certainly appears to be more common on the coast than 

 it is inland, where its place as a carrier of malaria is taken by A.funestus {of Stephens 

 and Christophers). 



Some mosquitos appear to flourish best in an atmosphere which is humid, e.g. 

 Eretmopodites, Toxorhynchites, and Taeniorhynchus. Others such as C. ager var. 

 elhiopicus, C. consimilis, and C. annulioris, the larvae of which are generally found 

 embedded in flimsy algae, do not appear to be found on the coast. Anopheles sqiiamosus 

 and A. rvflpes also seems to. prefer inland stations, as does Stegomyia sugens. 

 Stegomyiafasciata on the other hand is a lover of the coast, but as this is a domestic 

 mosquito, the explanation of its wider distribution near the sea is probably that given 

 by Boyce,§ who states that S. fasciata follows " the tin invasion which extends up 

 from the coast to the ijiterior villages," the discarded tins favouring the development 

 of this particular species. 



* Bull. Ent. Kes., vi, p. 225. f Bull. Ent. Res., ii, p. 132. 



X Report upon Entoinolooical Observations made in Southern and Central Ashanti. 

 1907. p. 14. Colonial Ollice, Miscellaneous No. 230. 

 § Bull. Ent. Res., i, p. 230. 



