BY R. J. TILLYARD. 89 



zontal channel under C, we found two in quick succession, 

 both males and fairly well-grown. I was surprised to find 

 them so different in appearance from what I had imagined 

 when I examined the exuviae. They are very flabby, and 

 appear ill-nourished. In colour, and to the touch, they are 

 rather like the fleshy white grubs of Scarabeid beetles so 

 commonly found when gardening. The abdomen is fleshy- 

 white all over, except along the lateral edges of the segments 

 and round the anal extremity, where the dark stain of the 

 ooze seemed to have become more or less permanent. The 

 thorax is also tleshy-white, but not quite so flabby, with a 

 distinct orange band passing on each side from the wing-base 

 to the mesocoxa. The head is whitish behind, but much 

 darker on the vertex and front, and especially on the labrum, 

 which is a hard plate, nearly black. The eyes are quite black 

 in front. The labium is pale glaucous-greyish, with the 

 lateral lobes and hook brownish. Wing-cases pale greyish. 

 Femora whitish, darker on the inner side; tibiae and tarsi 

 fairly hard and dark. From the fact that only the front of 

 the head and parts of the legs are hard, the rest of the insect 

 being soft and flabby, it is clear that the insect usually in- 

 habits the soft mud, and uses its head and legs to push itself 

 along, and to scoop out new channels and passages in the 

 soft ooze. 



We now continued our investigation by slicing out the 

 clump A, in a direction perpendicular to the plane of the 

 diagram. In this way we dug out a space of six feet by three 

 feet, including the whole of C and the adjoining side of the 

 two clumps A and B. At the further end of C, we dug out 

 one clump of sedge-roots containing a distinctly marked 

 passage, from which we secured a fine larva, nearly full-fed. 

 This was the only larva we found actually in a channel, but 

 in most cases it was impossible to dig without stirring up the 

 mud, and disturbing everything completely. The whole 

 space was dug out to a depth of nearly three feet, and the 

 mud and roots carefully dredged with the net. In all, we 



