328 AMERICAN PSYCHODIDyE. 



LARVA. 



The young Itirvse make their appearance on the sides of the jar 

 just above the surface of the water. Here they remain for some 

 time, feeding upon the decaying vegetable matter found in the sur- 

 face film. The number of moults cast was not determined, but 

 specimens from 2-3 mm. long were found moulting. The skin 

 splits open along the top of the head and thorax, after wiiich the 

 larva crawls bodily forward, leaving the moult behind. In one 

 instance under observation the moult refused to slip, and in the 

 struggle to free itself from it the larva severed one of the main 

 tracheae near the posterior end, and died within a few minutes. 



About one day after the larvae hatch they become quite active, 

 descending into the water, where, apparently urged by an unquench- 

 able hunger, they scramble over and gulp in with their fang-shaped 

 mandibles all tliat their spacious alimentary canal will hold, after 

 which they return to their "airy" perch, so to speak. This opera- 

 tion is repeated from time to time, the older larvse spending most of 

 the time in the water. 



There are no ventral sucking discs whatever. They stick to the 

 glass simply by means of a film of water and perhaps a slime-like 

 secretion, which continuously surrounds them and often makes it 

 very difficult to study the more minute details. 



The anal breathing tube is somewhat similar to that of mosquito 

 larvse, although it is relatively much stronger. The two main 

 trachese run side by side and open separately at the end of the tube. 

 The external openings are armed with strong, lanceolate cilia, 

 mounted upon three small, retractile processes. The two main 

 trachese may be traced forward to the thoracic region, where each 

 will be seen to have a side branch, which terminates as a short, 

 black nipple on the dorsal surface of the prothoracic segment. 

 While these thoracic air nipples are present in the younger larvse, 

 they are not so pronounced as in the older. They are slightly pro- 

 tractile in the older larvse. The writer noticed in one living adult 

 larva just before pupating that these nipples had already taken on 

 the elongated structure of the thoracic breathing tube of the pupse, 

 but he has been unable to find examples of it in his preserved mate- 

 rial. In larvse that are " most at home " in water just deep enough 

 to cover their bodies, it can well be seen how these nipples can be 

 used to advantage in respiration, but in these larvse which have 



