48 U. S. BUEEAU OF FISHERIES. 



potting shed) and ovipositing. In the pond they made no headway, 

 except within the fine wire check pen and in a gutter, also screened 

 off from the fislies. 



By the first week in July these conditions had greatly changed. 

 The vegetation had become much more dense, and masses of Con- 

 ferva? had collected along the shore. On July 8 mosquito larva?, 

 mostl}^ in early stages, were found almost everywhere near the shore 

 and in places where the vegetation was heaviest were as numerous 

 as within the check pens. But even large larvse were found in places 

 fully accessible to the fishes. At this time the water was swarming 

 with Daphnia, and great numbers of plant lice were falling from the 

 leaves of water lilies and other plants. In the bottom mud and on 

 submerged plants chironomid larvee of several kinds were in very 

 great numbers. Examination of stomachs of sunfishes (Table 7, 

 No. 18T08d, p. 42) at this time showed them to be packed full of this 

 abundant and nutritious food to the neglect of mosquito larvae. 

 During July the numbers of entomostracans and plant lice gradually 

 diminished, but the chironomid larvse remained plentiful and formed 

 the chief food of the sunfishes. 



About the middle of August a reduction was apparent in these 

 also. On August 13 mosquito larvse were again detected in the 

 stomachs (Table 7, No. 18813a, c), the contents of which became 

 much more varied, indicative of a necessity of covering a wider range 

 in foraging. On August 28 mosquito breeding had been nearly 

 abolished in the pond generally but continued with undiminished ac- 

 tivity in the check pens, while some tubs of water nearby were swarm- 

 ing with larvse. The stomachs of sunfishes (No. 18820d) examined at 

 this time (Aug. 20) contained many young larvae of mosquitoes; 

 one 2-inch specimen had eaten little else. Later, as the rate of breed- 

 ing declined, they again disappeared from the stomachs. Until mos- 

 quito breeding ceased about the middle of October conditions re- 

 mained the same. Breeding was negligible, the number of larvse 

 averaging about 1 in each 4 samples taken in all parts of the pond, 

 while in the check pens the average fell graduall}^ from 13 to 2.3 per 

 sample on October 11. when the last count was made. 



In 1919 the first oviposition of Culex pijneTis in this pond was 

 observed on May 2. The first Culex larvse were found in stomachs 

 of fishes taken on May 16 (Table 7, No. 19516a). On May 19 light 

 breeding was found in the check pens only. On May 24 the general 

 average in the pond accessilile to sunfish was three-fifths larva per 

 sample. Most of these were small Culex jn]ri^ns, but an occasional 

 Anopheles larva, mostly of large size, was found. In the coarse 

 screen pen the average was nearly 1, and in the fine screen pen 

 betAveen 7 and 8 per dipper, some being nearly full-groAvn Culex. 

 In tlie gutter the rate Avas about the same as the last and a few 

 Culex pupse and Anopheles larvse were found. 



On June 5 the pond generally yielded one-fourth larva per sample. 

 This rose to 4 to 6 in the coarse screen pen and the densest vegeta- 

 tion, to 20 (mostly first and second stage Culex) in the fine screen 

 pen and to 8 to 15 Culex and 1 to 6 Anopheles in the gutter. Many 

 of the latter were full grown or nearly so. The stomachs of fishes 

 (Table 7, Nos. 19524c, d, and 19611c, e) taken during this period 

 frequently yielded mosquito larvse. On June 18 these breeding av- 



