304 AGRICULTURAL EXPERIMENT STATION. 



the usual double tuft, each having one very long- hair on the 

 dorsal part of the apical margin. The barred area on the ven- 

 tral part, with ten to twelve tufts of live or six hairs each. The 

 anal gills are moderatel_y long-, without obvious tracheae. 



Habils of fJic Bcirly Stages. 



Practically all our knowledge of this species comes from Mr. 

 Brakeley's observations, though in 1904 Mr. Grossbeck took 

 lai-va; in the Great Piece Meadow, May loth, from which a male 

 adult was obtained May 13th, as against a dozen canadensis 

 which emerged on the same date. Mr. Brehme, in collections 

 of larvae and pupae made at Arlington, May 9th, also had one 

 male of this species, which emerged May 13th. This indicates 

 a somewhat general distribution in the State, but a rare occur- 

 rence at any point. No' one save Mr. Brakeley has taken the 

 adults in New Jersey, and all w'ho have taken the larva have 

 found it associated with that of canadensis. 



The captures of Messrs. Grosbeck and Brehme were purely, 

 accidental, while, after I once recognized the larva as distinct, 

 Mr. Brakeley's takings were intelligent, the species being recog- 

 nized as soon as attention was called to it. As a matter of fact, 

 indeed, Mr. Brakeley himself recognized the larva as unlike the 

 general run of those that he was taking. 



The larva first came under my obser^'ation in the early spring 

 of 1902, a single specimen being in material sent in from Laha- 

 way by Mr. Miller Emley, under instructions from Mr. Brakeley. 

 It was not bred and escaped attention until it was too late to 

 search for other specimens. At that time I believed that I had the 

 larva of perfurbans, then the onlv species known to me from 

 Lahaway of which I did not have the early stages. 



In 1903 Mr. Brakeley sent in the first larvae, March 23d. as 

 extra large canadensis. Being advised of the difference, he found 

 them afterwards in the larger bodies of water covering the bogs 

 and reser\'oirs and in pools of consideraljle size nearby, always 

 associated with canadensis and always comparatively rare. They 

 do not hug the edge of the pool so generally as do^ some of the 

 other species, but favor tufts of grass, rushes or vines several 

 feet from shore. Larvae only w^ere collected until April 14th, at 

 wdiich time the first pupa was taken. Thereafter larvae and pupae 

 were taken until ^^^ell along in May, the latest actual date at 

 which either larvae or pupae occurred being May 13th. No anrifer 

 larvae have been collected with the earliest canadensis, and from 



