COLLECTING AND IDENTIFYING LARVAE 267 



among the stems and leaves of the water plants and can only be 

 captured one at a time. In this case it is best to pull out most of 

 the plants and, when the larvae have collected together after an interval, 

 to dip for them in the usual way. 



Larvae may be collected from wells by using a simple and efficient form 

 of apparatus devised by Bentley, the construction and use of which he 

 describes as follows : ' A loop about fifteen to eighteen inches in diameter 

 ' is made of very thick brass or galvanized iron wire, and upon this frame 

 ' a shallow net of very thin mull-mull or muslin is stretched. The net is 

 ' suspended by three short brass chains about two feet long, which are 

 ' attached to the frame below and to a small ring above. To the ring 

 ' a rope is tied, and the net can be lowered to any required depth. When 

 ' a well has to be examined the net is lowered to about a foot beneath the 

 ' surface of the water, the weight of the frame causing it to sink. It is 

 ' gently moved below the surface, and pulled up at a point some distance 

 ' from where it first entered the water. On examining the net, larvae, 

 ' if plentiful, are easily seen and can be secured by washing them off into 

 ' a bowl of clean water '. Bentley also recommends washing the net in 

 a large-sized w r hite enamel frying pan. 



In the laboratory the different kinds of larvae should be separated out, 

 care being taken that predaceous species, such as those of Culex concolor, 



and dragon fly larvae, are removed to separate trays. 



_, , r . . Identification of 



The more important characters or taxonomic value, larvae 



such as the structure of the antenna and syphon tube, 

 the position of the palmate hairs and the number and structure of the 

 leaflets, can be studied in the living larva ; for finer details it is best to 

 make cleared preparations. With a little practice the worker will learn 

 to recognize the larvae of the various species which are present in his 

 neighbourhood. The several species of mosquito larvae which may be 

 found breeding in the same collection are best identified by transferring 

 them singly, with the aid either of a glass pipette or a small spoon, to a 

 slide and placing a coverslip over them. In this way they are immobi- 

 lized, though in no way damaged, and can be examined with a low power 

 objective ; in the case of larger larvae it may be necessary to place them 

 in a hollow glass slide. 



If a few imagines are required merely for purposes of identification 

 larva may be kept in small trays, but it will soon be found that they 

 do not thrive in these receptacles. It is difficult to obtain water 

 containing suitable food and to keep it fresh. 



In order to obtain the best results in rearing mosquito larvae they 



