287 



(KCiipics an incrcdihlv sliort time, generally not more than five or six 

 seconds. 



Characters of the hnacjincs 



In this pa[)er I have divided the Chiroiiomida: into three subfam- 

 ilies, viz., Ccratopogonincc, Tanypincc, and Chironoviincc. In the Cera- 

 topogonina: there are, as already indicated imder the two previous 

 heads, two distinct groups. The known species of Forciponiyia and of 

 Ccratopogon, sens, stric. of authors, have terrestrial larva?, or larvic 

 which are nf)t truly aquatic, and which have many distinct spines or 

 bristles on the body. In imagines of this group there are generally 

 distinct hairs on the wings, and in practically all cases the apex of the 

 wing, at least, bears microscopic hairs. The empodium is always dis- 

 tinct, and generally large. The second group has, as far as is known, 

 larvae of a snakelike form, which are entirely aquatic, and wholly bare 

 except at the apex of the abdomen, where four pairs of long hairs are 

 generallv present. Imagines of this second group have the wings bare, 

 or, in CuUcoidcs, with microscopic hairs, and the cmpodia indistin- 

 guishable or very small. In all species of Ccratopogonincc examined 

 bv the writer the mouth parts are well developed, and have chitinized 

 piercing parts, whereas in Tanypincc and CJiirononiincc the mouth 

 parts are very poorly developed and not fitted for piercing. 



The species of the group Tanypincc may be distinguished in the 

 larval stage from those of CJiirononiincc by the structure of the head, 

 which is dealt with at length in a subsequent part of this paper. In the 

 imagines of this group the characters which most readily separate its 

 species from those of Chirononiincc are the 15-jointed antennae in both 

 sexes and the presence, near the middle of the wing, of a cross vein 

 which connects the cubitus with the media. One section of this sub- 

 family has the w'ings with surface hairs; the other section has the 

 wings bare. The insufiiciency of material in hand does not permit my 

 forming an opinion as to whether this difference in the imagines is sup- 

 ported by corresponding differences in the larvae. The imagines in 

 CJiirononiincc are distinguished from those of Tanypincc by the ab- 

 sence of the cross vein between the media and the cul)itus and by the 

 8-jointed antenna; of the females, and from those in Ccratopogonincc 

 by their more slender and elongated legs and by the structure of the 

 antenna and thorax. These distinctions are indicated in the generic 

 key presented in this paper. The only deviation from the above rule 

 is in the genus Diamcsa, which has the medio-cubital cross-vein pres- 

 ent on the wing, but the larva is essentially of the chironomid type, and 

 the adult female has only 8 antennal joints. 



