44 AGRICULTURAL EXPERIMENT STATION. 



the end of July. At New Brunswick,, where the conditions were 

 under my own observation, cant at or arriving May 12, yet Hn- 

 gered August ist. 



The term Hngered is advisedly used because before the middle 

 of July the city was practically free ; but on the outskirts and in 

 the gardens, where, as in my own, there were excellent hiding 

 places where the specimens found shelter during the day and 

 invaded the porches at night, the supply seemed undiminished. 

 So in Newark it was only from the gardens that any number of 

 specimens came, while the town itself, which had been covered 

 in May, was practically free in early July ; before the end of the 

 month the Jersey City and Elizabeth marshes had sent in a new 

 supply. 



It is fair, under conditions as observed, to give the migratory 

 forms inland, an active period of nearly or quite three months. 

 The importance of dealing with the early mosquito broods is 

 especially emphasized by these observations, for when a place 

 is once thoroughly seeded down, it takes a long time to work 

 out the pest. 



Newark's experience proves one side of this proposition; the 

 first brood was allowed to develop, but though the drainage work 

 cleaned up the second and later broods, specimens of the first 

 brood lingered until after August ist. The experience of Mon- 

 mouth Beach is on the other side; the work done there got ahead 

 of the first brood and, as a result, there were practically no mo- 

 squitoes at all, while the other shore communities were suffering. 



In 1903 the second brood, developing in June and on the wing 

 in early July, supplemented the left-overs of the May migration 

 and gave the appearance of a "double infection" — as it really 

 was — which lasted until September. 



Anopheles has not been as closely observed because this is a 

 continuous breeder; larvse of all sizes being constantly found in 

 the same pools. It is more than probable, however, that the 

 life period is not less than that of the species of Culex, and one 

 month for the individual is certainly a conservative estimate. It 

 is probably longer; but even one month is amply sufficient to 

 provide for the development of the malarial parasite and the 

 transference of the disease. 



BREEDING PIvACES. 



It is a popular belief, which has been already referred to, that 

 mosquitoes develop in the grasses and shrubbery in which they 

 are so frequently found; but that is a mistake, for no species 



