Chironomidae 393 



in feeding. I have found evidence of a selection of milk in preference to 

 algae; a fact which agrees with the observations of Branch (1923) 

 in regard to C. cristatus [see p. 395 J . Except for milk, algae are 

 preferred to other materials. Obviously the larval habitat determines 

 the variety of food, but where algae are present in sufficient quantities 

 they comprise the bulk of the diet. 



Three methods of feeding midge larvae have been tried experimentally. 

 In all cases the actual procedure in rearing the larvae has been the same, 

 namely that of treating stagnant concrete ponds or wooden troughs or 

 tanks with different kinds of fertilizers and stocking them with eggs. 

 The fertilizer, besides attracting the adults for the purpose of egg laying, 

 produces a rich culture of microscopic algae. In ponds fertilized with 

 soybean meal, Chlamydomonas and Chlorella were the two principal 

 algal forms; in those fertilized with sheep manure and superphosphate, 

 diatoms and desmids predominated. 



Sheep manure and superphosphate, sheep manure and soybean meal, 

 and soybean meal alone have been used as fertilizing agents. Where 

 the soybean meal was used by itself or in combination with sheep 

 manure, it was first poured into a good-sized pan, and enough water 

 added to form a thin soup. The pan was then set out in the open and 

 the mixture allowed to ferment for at least 24 hours before it was poured 

 into the pond. This procedure prevented over-pollution of the water 

 unless an excessive amount of fertilizer was used. After the ponds were 

 fertilized, they were allowed to stand until a culture of larvae was 

 produced. 



Controlled spawning. In this method of propagation all operations 

 were performed within a screen wire enclosure. Under such conditions 

 it was possible to control the spawning of the adults and thus insure that 

 the basins in which the larvae were reared were sufficiently stocked 

 with eggs. 



The equipment used for testing the practicability of this method con- 

 sisted of a screen wire cage 8% x 1 1 x 12% ft., enclosing three of a series 

 of six concrete ponds, 5 x 8 ft. with a water volume of 18 cu. ft. Water 

 was supplied to each pond from a raceway controlled by slash boards. 

 Each pond was equipped with a separate intake and drain. By plugging 

 the drains the ponds could be kept stagnant, and at the same time a 

 constant level maintained. 



One pond was used to supply adults. It was stocked with grown 

 larvae at the beginning of the season and a few were added about 20 days 

 later. After that the eggs deposited in the pond kept it well stocked. 

 The other two ponds were used for rearing larvae. Since the number 

 of eggs deposited in these ponds was always more than sufficient for 

 stocking them, and as natural enemies were excluded, it would seem that 



