442 Osten Sacken: on the cliaraders of the three divisions of 



froni tliem in tlic absence of the thoracic suture, of tlic „sensitive 

 liairs" on the antennae, in the sniall nuniber of the abdominal Seg- 

 ments (scven, according to the authors; I cannot count them on niy 

 dry speeimcns) and in the rudimcntary condition of the seventh vein. 

 The thorax resembles that of CMronomus in its sliape. The larva 

 has the most rcmarkable reseniblance to that of Anoplceles (com- 

 pare the tigurcs of Mcinert); the pupa has the convolutcd legs 

 forming with thorax and wings a more or less uniform mass, a struc- 

 ture characteristic of the pupa of the Oalicidae and Chironomidae. 

 Dixa therefore must be placed, as a separate family, between the 

 Tipididae and the group Ctdicidae + Chironomidae + Psychodidae. 

 (Compare the Additions.) About 27 spccies of this genus arc known 

 to exist: 15 in Europe, 8 in North-America, one in China, and three 

 in New South-Wales. (The latter are not described yct, but merely 

 mentioned by Mr. Skuse in the Trans. Austral. Ass. Adv. Sei. l>5lK>, 

 p, 350). Four species were found by Loe\Y in amber. 



The larvae of the Nemocera vera. 



The Nernocera vera can be subdivided into two natural groups, 

 in conformity not only to the structure of their imagos, but also of 

 their larvae (compare above, p. 428). 



I. The Mycetophilidae and Cecidomyidae have peripneustic 

 larvae, the former with eight, the latter with nine pairs of spiraclesi) 

 The larvae are strictly terrestrial and have none of the numerous 

 adaptions for aquatic life which distinguish the next group. 



In my: Characters of the larvae of Mycetophilidae (Proc. Ent, 

 Sog, Phil. 1862; also reprinted separately, with additions, in Heidel- 

 berg 1886) I have shown the perfect unity of type, prevailing among 

 the larvae of the different genera of this family. There is one excep- 

 tion to this rule however, to which I have alluded in my paper, but 

 which deserves a more detailed notice than it received at that time. 

 It is the larva of Mycetohia pallipes, which is not peripneustic, likc 

 the larvae of the other Mycetophilidae, but amphipneustic; it shows 



i) Stannius, as early as 1831, has noticed this relationship bet- 

 ween the two families, which he connects with a character common to 

 both, the faculty of spinning a cocoon for the pupa. 



Stannius, Observationes etc. 1831, p, VIII: 



Imo etiam contigit ut in Cecidomyiarum larvis Stigmata observavcrini 

 lateralia, quod memoratu dignissimuni, quum haecce simili modo ac Ti- 

 pularlum fungicolarnm larvae, antequam in nyrapharum statura abeant, 

 contextum sibi parent sericeum. Singularis igitur nexus huic respirandi 

 rationi cum tela conficiendi facultate interesse videtur! 



My attention to this passage was drawn by Westwood, Introd. 

 II, p. 519, footnote. 



