POSSIBILITIES AND COSTS. 63 



fully with pieces of fresh fish. Allowing for five broods 

 of mosquitoes in a summer, and that one hibernated female 

 may be responsible for at least ten thousand larvae by the 

 time the Odonats appear, it will be seen that the one 

 brooded enemy will need to be produced on an enormous 

 scale. Liberated in great quantities, as they would be if 

 bred in sufficient numbers, the dragon flies might migrate 

 to less crowded localities. 



The question of transporting the young larvse from the 

 breeding tanks to the mosquito infested jjonds is to be 

 considered ; although they are tough and can stand jos- 

 tling, only a few can be carried in one receptacle. Twenty 

 put in one jar would be found to be an inextricable kick- 

 ing mass of cannibals after a mile's transportation. 



We have foiled to find the natural enemy of the mos- 

 quito to be the dragon fly, and have intentionally re- 

 served to the last that remedy which to us seems to far 

 outrank all others of which we have any knowledge or 

 have been able to devise, viz., the oil treatment. 



The United States Department of Entomology and the 

 various State Reports, as well as numerous economic ento- 

 mologists abroad, have long recommended the use of petro- 

 leum in some form for the extermination of plant lice and 

 many other noxious insects. Petroleum emulsion, sprayed 

 petroleum, the naphtha compounds, and others from the 

 same source, are prompt and deadly insecticides. With 

 this in mind we early began a series of tests with common 

 illuniinatins: oil on Culicid larvee under all circumstances. 

 The narration of one series of experiments, typical of all, 

 will illustrate the efficacy of this treatment. Into a shal- 

 low pool of wtiter with an area of ten square inches, five 

 pupse, two grown larvae, and about sixty others in various 



