5^ IJ. S. BUREAU OF FISHERIES. 



grown intact Culex jnpiens larvse. In the others no.mosquito remains 

 were found among tlie cliironomid larvae which formed the chief 

 item of food of the larger and Entomostraca of the smaller fishes. 

 On August 27, 1 larva per dipper in the open waters and 20-}- in the 

 checks were found. 



Both of these areas (D 1 and J 3) as well as others were examined 

 in August, 1920. Mosquito breeding about the lake shore generally 

 at that time was at a low ebb. It was evident, however, that the 

 work of the summer of 1919, supplemented by further clearing up 

 of the shore lines while the water level was lowered, had iiad a lasting 

 effect. Small sunfishes and other fishes were plentiful, and careful 

 search in the opened areas disclosed absolutely no mosquito larvae. 

 In sample check pools and other spots that remained closed to the 

 fishes counts averaged for 10 samples each 2, 2^, 3|, 6J, 18, and 20, 

 most of the larv^ being Culex pipien.s but some Aedes sylvesf^^s. 



The above recorded data, together with many others of a similar 

 character, should remove any doubt concerning the place occupied 

 by the common sunfish in natural societies in which they are associ- 

 ated with breeding mosquitoes. It is unquestionably an important 

 agent antagonistic to the multiplication of the latter, tlie aquatic 

 stages of which serve as food. Under the most favorable circum- 

 stances this influence may amount to total suppression, in which, 

 however, other fishes usually play a part. It is seldom that such fav- 

 orable conditions exist throughout the entire extent of a pond or other 

 body of water or continue through the entire season. Usually some 

 mosquitoes will be produced in any body of water offering otherwise 

 favorable conditions, and it is evident that in many cases the density 

 of the immature mosquito population varies positively with the limi- 

 tations placed upon the activity of sunfishes in feeding upon them. 



These limiting conditions are numerous and often delicate and com- 

 plex. The principal ones disclosed by the data recorded in this 

 paper are (1) the presence of barriers which more or less full}^ pre- 

 vent the passage of the fishes and (2) an abundant supply of other 

 easily accessible food. Under the former head the most important 

 are dense growths of plants and a great variety of other mechanical 

 obstructions which cut off areas of water often of very small extent 

 and i:)rovide shelter and concealment where the mosquitoes may 

 breed nearly unmolested. Bat the barriers may be physiological, 

 such as result from changes in acidity,^ temperature, or contamina- 

 tion of the v\-ater. The influence of the food supply is complex and 

 bears a close relation to the density of the fish population. If other 

 food is abundant and the number of mosquito larvae small, the latter 

 occup3'' a correspondingly unimportant place in the diet; but if for 

 any reason the total available food supply be reduced or the supply 

 of mosquito larva? available to the fishes be increased, the ])orcentage 

 of the latter found in the food correspondingly rises. The influence 

 of the fishes on moscpiito breeding is therefore one of C()m])ensatory 

 regulation which operates most vigorously when mosquitoes are 

 breeding in large nunil)ers and other equally accessible food is scarce 

 and which falls as the supply of mosquitoes is diminished until a 

 point of equilibrium for the existing conditions is reached. How 

 "to bring the mosquito element in this balance to the lowest possible 



^ H ion concentration may be an important factor, which of the autiior's stiulpiits 

 has agreed to investigate. 



