REPORT ON MOSQUITOES. .39 



maintain themselves, for in that case there would be a normal 

 ovarian development. It seems rather as if the inability to re- 

 produce resulted in a restless desire to wander. Just why so 

 large a percentage of the females should be barren I do not 

 know, nor can I see how this may be in any way of use to the 

 species. It can scarcely be due to any lack of males, because 

 when the species is bred in the laboratory both sexes develop in 

 about equal numbers ; the males usually a day or two before the 

 females. So also on the marshes the grass about breeding pools 

 may be full of males before a single female has developed. Nor 

 is it possible that I have examined only fresh specimens in which 

 eggs have not yet had time to develop. Mr. Viereck's observa- 

 tions indicate that sollicitans may oviposit a week after develop- 

 ment to the adult stage, and my collections have extended over 

 a period of more than a month from the same swarm. Further- 

 more, it is a fact that even in marsh collections only a small per- 

 centage of the females show developing ovaries. It is a fair 

 statement from examinations made that in sollicitans not over 

 25 per cent, of the females are able to reproduce their kind. As 

 to cant at or ^ the percentage of barren females seems not nearly so 

 great. 



Prof. F. M. Webster, of Urbana, 111., recently published con- 

 cerning the flights of the Buffalo gnat, and records among other 

 things that he found all the migratory forms barren. As the mos- 

 quito and the Buffalo gnat are members of the same order and 

 not too distantly related this record forms an interesting supple- 

 ment to my observations and indicates that among the flies there 

 is some correlation between a migratory habit and the inability 

 to reproduce. 



Meanwhile it is certain that the migratory species dominate a 

 large section of New Jersey, and make their control a State 

 rather than a local problem. 



The question has been asked whether, assuming that New 

 Jersey actually controlled her own output, she would not suffer 

 from swarms bred beyond her jurisdiction. The answer to that 

 is yes, as to one point only. The marshes on Staten Island, 

 opposite the Elizabeth River, are as bad as anything within the 

 jurisdiction of our State, and until these marshes are treated, 

 the city of Elizabeth will never be free from mosquitoes. 

 Rahwa}^ is threatened to- a less extent; but at no other point is 

 there danger of any incoming supply from points beyond the 

 State line. 



Delaware Bay is too broad for ordinary mosquito flights over 

 water, and their tendency is not in that direction as a rule. The 

 marsh area along the coast of Delaware is narrow and as the 



