44 riPTERA OF NORTH AMERICA. [PART IV. 



TABLE FOR DETERMINING THE GENERA.' 



TTwo longitudinal veins between the fifth vein and the posterior 

 , ! margin. 2 



I Only one longitudinal vein between the fifth vein and the posterior 

 L margin ; Tab. II, fig. 19 and 20 (Sect. VIII. Ptychopterina). 45 

 Last joint of the palpi shorter or not much longer than the two pre- 

 ceding joints taken together ; the auxiliary vein usually ends 

 in the costa, and is connected with the first longitudinal vein 

 by a cross-vein. 3 



Last joint of the palpi very long, whiplash-shaped, much longer than 

 the three preceding joints taken together ; the auxiliary vein ends 

 in the first longitudinal vein ; no cross-vein between it and either 

 of the two veins running alongside of it (Tipulida longipalpi). 

 o ( A single submarginal cell ;2 Tab. I, fig. 1-13. 4 



\ Two submarginal cells ;» Tab. I, fig. 14-20, and Tab. II, fig. 1-18. 6 



' This table contains all the known European and North American 

 genera of the eight first sections of the TipuUdie; the table for the follow- 

 ing sections (^Tip. longipalpi) will be appended to the volume treating 

 of them. In using dichotomical tables it should always be remembered 

 that to construct them in such a way as to meet all cases, to include all 

 the anomalous structures, is impossible, and if it were possible, it would 

 be only through the use of anatomical characters, which would defeat the. 

 object in view, the facility of determination. Thus, if, in order to accommo- 

 date Chionea, we had abstained from the use of any character connected 

 with the wings, we would perhaps have rendered the table more precise, 

 but certainly less useful. As it is, Chionea, although wingless, is placed 

 among the genera provided with two submarginal cells, where it belongs. 

 These imperfections of the dichotomical tables occur especially in those 

 portions of them which refer to the larger divisions ; as soon as the genera 

 and species are reached, more precision can be expected, although even 

 there it can never be absolute. 



2 Toxorrhina has none at all. 



' Cladolipes has only a single submarginal cell, although it belongs to 

 this division. 



