, AQUATIC INSECTS IN THE ADIRONDACKS 581 



ments are much the same, but the individual tubercles are in general 

 a Httle more distinct in T. p i c t i p e s ; the brush of hooked bristles on 

 the dorsal locomotor tubercles is a Httle shorter (this is omitted altogether 

 from Dufour's figure of the larva of T. ferruginea); there are four 

 fairly distinct lateral tubercles each side in the median ring of the seg- 

 ment ; the subterminal ring of lesser tubercles is less distinct than in 

 S. fuscipennis, except on the ventral side where it is more distinct, 

 and has the pair of little tubercles beside the midventral line fused more 

 or less completely into one. 



The respiratory float at the end of the body (pi. 2>^, fig. 4) is similar 

 to that of S. fuscipennis, but is a little less copiously fringed, and 

 the fringe is a little shorter, reaching but about two thirds of the distance 

 to the margin of the disk. The fleshy processes and tubercles bordering 

 the disk are very different. The ventral, submedian pair of processes are 

 broad and blunt. The lateral processes have a broad obtuse base sup- 

 porting a slender second joint. The tubercles forming the dorsal border 

 of the disk, excepting the external one each side, are nearly obsolete. 

 By the characters mentioned in this paragraph the larva of this species 

 (possibly, of this genus) may readily be distinguished from S e p e d o n — 

 at least, from Sepedon fuscipennis. 



Puparium. (PI. 14, fig. n, 12) Length 6 mm; diameter 2 mm. 



Color reddish brown, more reddish on the obsolescent larval tubercles, 

 and on the upper side of the beak into which the body is contracted at 

 the front end. There is no black arcuate mark at the base of the beak 

 on the dorsal side. 



Body more slender than in Sepedon, less convex below, less 

 flattened above, though of much the same general appearance. The tail- 

 like prominence at the posterior end bears at its apex the unaltered 

 larval float, and at its sides the scarcely distinguishable remains of the 

 processes and tubercles which surrounded the caudal disk of the larva. 

 This ^' tail " is bent upward at an angle of about 45° with the axis of the 

 body. By this lesser degree of angulation, as well as by the general 

 slenderness of form, this species is readily distinguished from S. fusci- 

 pennis in the pupal stage. 



Roederiodes juncta Coquillet 



Plate 15, fig. 5-8 



The reader is referred to page 586 for the technical description of the 

 imago of this species. It was found in a rapids in Little Clear creek just 

 below the railroad embankment. Larvae and pupae were found first, the 

 former crawling among the pupa cases of S i m u 1 i u m , the latter 

 usually resting within an abandoned S i m u 1 i u m pupa case. After 

 these had been bred, the imagos were found, clinging in companies to 

 the under side of pieces of boards which rested just above the level of 

 the water. They were not seen flying, except when disturbed, and then 

 they took flight slowly and flew poorly. A piece of board might be lifted" 

 with a score of the flies sticking to it within an area of a few square 

 inches, and most of them could be gathered into a cyanid bottle before 



