52 DR. W. M. GRAHAM — 



precipitation the water becomes clear, the motile alg?e become stationary, 

 their transparent cytoplasm dissolves in the water, and the chloroplasts 

 (chlorophyll corpuscles) fall to the bottom of the vessel. Then in the 

 absence of their natural food, the larvre in the jar become cannibalistic and 

 destroy one another. In lesser concentration salt appears to iidiibit the 

 growth of very young larvae, probably by diminishing the supply of food, 

 but the development of fully-grown larvse appears to be hastened in a 

 hypertonic medium, and they p:iss into and through the pupal stage with 

 unusual rapidity. From this it is evident that there are alternative methods 

 available for the destruction of mosquito larvae: 



(1) Methods intended to destroy the larvae. 



(2) Methods intended to destroy the food of the larvae. 



Very little attention has been paid to the second method, though it is 

 apparently worthy of further study. Unfortunately, any such investigation 

 demands an exact knowledge of West African fresh-water algse, a subject 

 upon which very little information is at present available. 



Whenever female mosquitoes full of eggs were caught, they were im- 

 prisoned in one of the glass jars, and in several cases egg-rafts were laid 

 upon the water after a few days' captivity. Under such conditions, no eggs 

 were laid by Mansonia females, and in most cases eggs so laid did not develop 

 into larvae. 



It remains to describe the district in which the larvae were found, and 

 to explain some of the local conditions. 



The country in the immediate vicinity of the Research Institute is flat 

 grass-land covered with orchard-bush, with a few tall trees. There are 

 several large w^ater-holes close to the Institute, an extensive swamp about a 

 mile to the east, and numerous borrow-pits and puddles along the course of 

 the railway-line. The nearest water-hole is typical. In the dry season it 

 measures about 30 X 40 yards, but during the rains it overflows its banks and 

 extends over an area of 200 or 300 yards. The water-hole is used by the 

 natives for bathing in and for washing clothes, but not as a drinking- supply. 

 It contains cat-fish, but I have not observed any very small fish in it. A 

 large herd of oxen is watered there twice daily, and the water at the edges is 

 fouled with excrementitious matter, by the cattle. 



In April the larva and pupa of Pyretophoriis costalis were found in large 

 numbers in the little bays along the edge of the water-hole. With the onset 

 of the rains and the formation of road-puddles and the filling of borrow-pits, 

 the numbers in the water-hole decreased, but were never wholly absent. 

 They were always found where the water had been fouled by cattle, from 

 March to November. 



The larvae of Myzorliynclws mauriliamts and of Cellia pliaroensis were also 

 found here, Imt only during the months of June, July and August. The 

 swamp about a mile behind the Institute was examined, but though the pools 



