240 NEW YORK STATE MUSEUM 
Classification of the Tipulidae or crane flies 
It has been one of the great pleasures of my brief study of the 
Tipulidae to note the broad and catholic spirit in which Baron Osten 
Sacken studied them and dealt with their classification. -I do not 
always trust the characters he used, but he did not use them slav- 
ishly. He was always searching for further light, always open to 
conviction. I am encouraged to offer a few further notes on his 
several sections or tribes, by the following invitation contained in 
his monograph [p. 25]: “ The more characters peculiar to each 
one of the sections we accumulate, the stronger we render the basis 
on which the classification is established, and the easier the solution 
we prepare for all future doubtful cases. In this respect a great 
deal yet remains to be done.” 
The primary division of crane flies into two families, based 
originally and mainly on the profound differences in the larvae, finds 
its venational justification in the distinct behavior of the radial 
sector and the median vein, as illustrated in figures 14 and 15, and 
in the absence of a second anal vein in the Ptychopteridae and its 
presence in Tipulidae. 
Ptychopteridae 
The Tanyderinae are distinguished by the possession of the full 
‘ complement of branches of the radial vein. They are in this respect 
the most generalized of Diptera. Idioplasta is our only representa- 
tive of the group, which, like many other archaic groups, finds its 
other representatives in the antipodes (Chili and Australian region), 
and in fossil remains. | 
The Ptychopterinae have the radial sector reduced to three 
branches but with R* and R® remaining separate however, and they 
are further distinguished by the absence of Sc®, by a better de- 
veloped cord, by the brevity of the base of the radial sector, and by 
the sinuosity of Cu?— all marks of specialization. 
Tipulidae 
The Tipulinae are distinguished from other Tipulidae by the loss 
of Sc!, the skewing of R? forward, carrying the cross vein r often 
into a longitudinal position, and the slight tendency toward fusion 
of Cu! with M®. Within this group the tribe Dolichopezinae ap- 
pears to be marked off by a tendency of the first fork of media to 
progress outward beyond the cross vein m-cu. I have seen too few 
representatives of the other tribes of this subfamily, but they are 
based on antennal characters, for which I have found no vena- 
tional counterparts. 
