PISHES FOR CONTROL OF MOSQUITOES. 41 



toad eggs, larval, pupal and adult insects, amphipods, crayfishes, 

 leeches, snails, sponges, and algae and other plants, and continues in 

 this summary : 



The food of the pumpkiu seed was made up of insects (22.1 per cent), large 

 Crustacea (10.3 per cent), snails (25.8 per cent), plants (25.5 per cent), and 

 other things. Forbes and Richardson (1908) found that more than half the 

 food of the fish they examined was niollusks; the rest was amphipods, isopods, 

 and insects. Hankinson (1908) reported midge larvae, may-fly nymphs, cray- 

 fishes, amphipods, snails, leeches, and caddis-fly larvje. Iteighard (1915) found 

 snails, insect larvae, and Chara. Insects formed the chief food of those fish 

 examined by Baker (1916). 



In the course of this investigation 360 stomachs of the common 

 sunfish were examined. All but 30 of these were of young or small 

 fishes, as it soon became evident that these and not the adults fre- 

 quented the extreme margins of ponds and similar situations where 

 mosquitoes are likely to oviposit. Table 7 (p. 42) records the sum- 

 marized stomach contents of 224 examples of fishes measuring from 

 18 to 97 mm. long, the great majority being under 80 mm. The grand 

 summary gives as food: Mosquitoes in all stages, 9 per cent; chirono- 

 mid larvae and pupee with some other Diptera, 33 per cent ; all other 

 insects, 14 per cent ; crustaceans, chiefly entomostracans of all kinds, 

 18 per cent; mollusks, 7 per cent; vertebrates, 4 per cent; all other 

 animal matter, 6 per cent; algae and other plants, 5 per cent; and silt 

 and debris, 4 per cent. The great diversity and richness of the 

 dietary are only partially apparent, however, and it would seem that 

 anything that is edible and obtainable is eaten. 



The principal contents of individual stomachs or of uniform lots 

 of stomachs differed greatly. In many cases they were chiefly or solely 

 chironomid larvae, in others Entomostraca, sometimes purely a single 

 species of Cyclops or Daphnia, in a few cases dragonfly or mayfly 

 nymphs or the larvae and pupae of culicine mosquitoes, or amphipods, 

 isopods, tadpoles, or annelids, and in one case goldfish eggs. Adult 

 fishes were found with the stomachs packed with grasshoppers or 

 17-year cicadas or dragonfly nymphs. There is great variation with 

 age, season, and locality. The young conspicuously subsist more on 

 entomostracans and small insect larvae, the adults on the larger in- 

 sects, crustaceans, snails, and leeches. In the spring mayflies and 

 lamellicorn beetles will be prominent, in the fall grasshoppers and 

 crickets. In brooks caddis-fly, stone-fly, and crane-fly larvae with 

 crayfishes will often dominate; in ponds, chironomid larvae; and in 

 sluggish plant-grown creeks and rivers, snails. Propinquity and 

 abundance appear to be of even more importance than choice in de- 

 termining what is eaten, though the latter is undoubtedly a factor 

 when not overcome by the stronger stimulus of nearness and bigness. 



In correspondence with the abundance, ubiquity, and importance 

 for the investigation of this species observations were made over a 

 wide extent of territory and under a variety of conditions. The 

 principal experiments were made near Philadelphia and in Palisades 

 Park. 



In the first place abundant confirmation was found of the state- 

 ment made by Seal and Smith that "this is undoubtedly the most 

 useful species of sunfish as a destroyer of mosquito larvae." Smith 

 again states (1904, p. 108) that it "keeps the ditches and streams in 

 cranberry bogs free from wrigglers." It may be added with reser- 



