244 NEW YORK STATE MUSEUM 



mind in using this key, that I have seen but few species save those 

 mentioned in my list or figured in the plates, and other species may 

 exhibit characters different from those I attribute to the genus. 

 Notwithstanding this, the things found in the species I have seen 

 should be suggestive, and should add something to the final work- 

 ing out of the system for this group. It is certain that a thorough- 

 going study of the antennae, the mouth parts, the legs, or the 

 appendages of the abdomen would likewise yield valuable results, 

 also, the study of the immature stages, for which I am endeavoring 

 to gather material for future use. 



Key to the North American genera of crane Hies 

 (Based on venational 'characters) 

 a Rs four branched, the posterior fork more deeply branched or, if but 

 three branched, it is R' and R^ that are fused. A single anal 



vein PTYCHOPTERIDAE 



b Rs four branched, M three branched cord deeply indented by first fork 



of M; vSca and cross vein w present ; Idioplasta only. .TAN YDERINAE 



bb Rs three branched; M^ two branched; cord not indented by the median 



fork, Sc2 and cross vein m absent PTYCHOPTERINAE 



c M two branched Pt3'choptera 



cc M one branched Bittacomorpha 



aa Rs never four branched ; when three branched it is Ri and R^ that have 

 fused (for a few apparent exceptions, all of which have the forks of 

 nearly equal length, see p. ante). Two anal veins present (three in 



the fossil Cladoneura) TIPULIDAE 



b Sci wanting: R2 directed forward and reduced, or wanting: basal de- 

 flection of Cwi never extensively fused with Ms TIPULINAE" 



bb Sci usually present and joining the costa beyond Sc-z, which often 

 recedes toward the wing base : sometimes both tips are wanting. 

 Cui usually extensively fused with Ms at its basal deflection, the cross 



vein in m-cu being eliminated thereby LIMNOBIINAE 



c Ri, Rt. and Rz extensively fused together from the tips backward in 

 a long, straight vein ; M skewed forward at its first fork, its base 

 being in line with the tip of Cu. Tips of Sc often rudimentary. 

 Cross vein m never atrophied: Cylindrotommi 



d Ml and Mo separate at tips Cylindrotoma 



dd Ml and Mo fused to the wing margin 

 e Cross vein m present 



/ Antennal joints subcylindric Liogma 



ff Antennal joints subglobular Triogma" 



ee Cross vein m eliminated by fusion of adjacent veins. Phalacrocera 

 cc Ri, R2 and Rs not all fused together into a long, straight tip 

 d Rs three branched 



ol am not sufficiently acquainted with the genera of this family to attempt a venational 

 key for them. I know only those genera figured herewith and Tipula and Pachyrhina , 

 Antennal characters seem to have furnished. the basis forjmost of the genera, 



b Triogma I have not seen. 



