38 U. S. BUREAU OF FISHERIES. 



seine yielded also roach, catfish, common sunfish, and ye41ow perch, 

 all of small size and in small numbers. Whirligig beetles (Oyrinus) 

 were abundant in the quiet sheltered waters. Close to the shore, 

 among the vegetation, egg boats of C'ulex pipieTis and a few newly 

 hatched larvae, both of this species and of Aedes sylvestris, were found 

 in considerable numbers on June 14 and subsequent dates, but not 

 a single advanced larva or pupa was found in the lake at this point. 

 Eggs and young larvae of mosquitoes were found in 30 per cent of the 

 stomachs of blue-spotted sunfish taken here on July 15 (Table 5, No. 

 2345, p. 37) — strong evidence of its importance as a factor in pre- 

 venting the maturation of mosquitoes in these plant-choked waters. 



The pools cut off from the lake waters and inaccessible to fishes 

 and in which no fishes could be found were, on the other hand, 

 sources of prolific breeding. In these and especially in the rock 

 pools not only were egg boats and young larvae found in great num- 

 l3ers, but there were all stages of larvee as well as pupae of both 

 Culex pipiens and Aedes sylvestris, together with many cast pupal 

 skins, showing that development had been completed. In one of 

 these pools (fig. 11), measuring 6 by 2 feet and separated from the 

 lake by a barrier less than a foot wide, formed partly by a rock 

 and partly by the root of a tree, and in which Utricularia was 

 growing in abundance, 12 blue-spotted sunfishes were placed on 

 July 15. At that time culicine breeding was so dense that the 

 larvae could not be accurately counted in the dipper, but every 

 sample contained 50 and more larvae of all sizes, together with 

 pupae of both species, besides frequent egg boats of Culex pipiens. 

 Ten days later, on July 25, the rate of breeding in this pool had 

 been reduced to an average of 6, and on August 7 to 4.6 per dipper 

 for 20 samples, while neighboring similar pools in which no fishes 

 had been placed showed no diminution in the rate. During the in- 

 terval exceptionally heavy and frequent rains had raised the level 

 of water in the lake, and it is possible that the barrier had been 

 slightly submerged, though there was no evidence of this. Owing 

 to the irregular contour of the bottom and sides of the pool it was 

 found impossible to secure any of the sunfishes for stomach ex- 

 aminations. 



To overcome the latter difficulty, a mosquito screen box, open 

 above, was placed in a neighboring similar pool (M 3), where the 

 rate of breeding had been maintained, care being taken that condi- 

 tions within and without the screen both as to plants and the density 

 of mosquito breeding were exactly similar. On August 7, after 

 allowing the screen to remain undisturbed for a week, 10 small 

 blue-spotted sunfishes were placed therein. On this date the number 

 of mosquito larvae (mostly half grown) both within and without 

 the screen averaged 30-j- per sample. On August 11 the average 

 within the screen was 8.1, outside it was 22. The stomachs of two 

 (Table 5, No. 2372) of the fish taken from the screen on this date 

 contained, besides a large number of Cyclops and Daphnia (about 

 60 per cent) and plants, consisting of Lemna, filamentous algae, and 

 diatoms (about 12 per cent), no less than 10 larvae, 1 pupa, and 

 several disintegrated egg boats of Culex pipiens and 6 larvae of 

 Aedes sylvestris. 



