TIPULIDJ3. 269 



flattened lobes ; these brunches are divergent, aud sometimes 

 curled up in dried specimens. The short palpi (bi-articulate, 

 according to Mr. Curtis) are inserted about the middle of the 

 proboscis to the anterior angles of the rostrum." 



The mouth parts, it is probable, do not attain in this family any 

 higher complexity than that herein described, in fact it seems 

 probable that in many genera the organ is of much more simple 

 form. Meigen speaks of a pair of horny linear pointed maxillae 

 in Glochina, a non-Oriental genus which I have had no oppor- 

 tunity of examining. 



The palpi. These organs are, in the large majority of genera, 

 four-jointed, and afford useful characters for subdividing the 

 family. The following table indicates how these organs differ 

 in the various subfamilies. 



The four palpal joints all considerably elongated . . PTYCHOPTERIX^E. 

 The first three palpal joints moderately elongate 

 only, the 4th always at least as long as the 2nd 

 and 3rd joints together. 



The 4th palpal joint very long, whiplash-like, 

 generally as long as or longer than the three 



preceding joints taken together TIPULINJE. 



The 4th palpal joint of about the length of the 

 2nd and 3rd taken together, or only a little 

 longer than this, generally not whiplash-like LIMNOBIIXJE. 



In the TIPULIN^E the palpi are generally of more or less uni- 

 form thickness throughout, the last joint being often rather more 

 slender ; but in the LIMNOBIINJE this comparative uniformity is 

 not perceived, the 2nd joint frequently being much stouter alto- 

 gether than the rest, the whole palpus generally being incurved, 

 or bent approximately into a circle, often making it difficult to 

 decipher the limits of each joint. The longer and whiplash-like 

 (generally curved or twisted) nature of the 4th palpal joint in the 

 TIPULIN.E is easily recognisable after a little practice. 



The antennae are always elongate, often exceptionally so, reaching, 

 if bent backwards, to about the root of the wings. The number 

 of the joints varies, the two basal joints, forming the scape, being 

 invariably differentiated from the remainder, which are called col- 

 lectively the flagellum, these generally being all alike except that 

 the first or basal one is sometimes slightly different, although never 

 to such an extent as the scapal joints. The terminal joint may 

 be reduced almost to a short style and is often easily overlooked. 

 In the subfamily PTYCHOPTERIN^E the antenna? are 16-jointed 

 (sometimes expressed 2 + 14). In the TIPULIX^E they are 13- 

 jointed ; whilst in the LIMNOBIIN^E they are 14-jointed in the 

 LiMNOBina and 16-jointed in the CYLINDBOTOMINI, EBIOPTERINI, 

 AMALOPINI (normally) and LIMNOPHILINI. The ASISOMEBINI 

 form a somewhat abnormal section, with antenna? of from 6 to 

 10 joints. Exceptions occur in most of the sections : Toxorhina 



