the Biology of some British Neuroptera. 521 



bubble of gas, and this is of such size as to render the 

 larva of the same specific gravity as the water. For a 

 time it drifts aimlessly, but if the water be disturbed, it 

 swims by jerkily straightening the abdomen two or three 

 times, at the same time using the legs. This swimming is 

 not very effective and after three strokes the larva rests 

 for a time, to make another attempt after an interval. 

 What does the larva do under natural conditions ? I have 

 tried it in the presence of a freshwater sponge, Ephydatia 

 flnviatilis. It floats about until a current of water from 

 the sponge is felt, then it becomes active and strikes out 

 towards the sponge. Little progress is made, hoAvever, 

 and I am quite sure that in nature the larva drifts rather 

 than swims into contact with the sponge. Having now 

 gained a foothold, the larva does not hesitate to probe 

 into the sponge body with its sucking spears, and soon the 

 gut is seen to contain food. Thus the true food of the 

 larva is from the sponge itself, and does not consist of 

 insect larvae, etc., living within the sponge. 



The newly hatched larva (Plate XXXVIII, fig. 3) is 

 about -5 mm. long, exclusive of jaws. The head is dark 

 blackish and rounded, and is the broadest part of the body 

 at first. The body is almost as broad as the head in the 

 thoracic region, but tapers gradually to the tip of the 

 abdomen, where there are a number of long black setae. 

 Apart from the matter of size and general proportions, the 

 following dift'erences are noticeable from the full-fed larva. 



Antennae are five- jointed, the terminal joint being small 

 and narrow, terminated by a bristle. The jaws are short 

 and fairly stout, being only about twice the length of the 

 head. The ])roximal half is thicker, becoming narrower 

 distally. Palpi are absent throughout the whole of larval 

 life. The thorax bears two blackish sclerites to each 

 segment, those of the prothorax closely approximating in 

 the middle line ( ? fused). Internally, but visible through 

 the chitin body walls may be seen the bubble of gas in the 

 anterior part of the alimentary canal and also some 

 yellowish-white fat body Ijang under the hypodermis. The 

 only other patcli of this is at the apex of the abdomen ; 

 otherwise the body is transparent. The legs have only 

 one tarsal claw. The tergites of the abdomen are less 

 distinct than in the later larva. I have been unable as 

 yet to find any trace of abdominal gills in the first-instar 

 larva. 



