249 



importance of certain structural details of the imagines is not suf- 

 ficiently realized as yet, and as we gradually accumulate data upon the 

 larval and pupal stages I am confident that certain groups at present 

 considered as genera in the family will be elevated to subfamily, and 

 some even to family, rank. I have taken upon myself the responsi- 

 bility of separating some of the so-called subfamilies from Myceto- 

 philidae — a step which I consider justified because of » distinctions 

 which are evident in the adult insects, and also in the known larvae and 

 pupae. 



FAMILY CHARACTERS 



Larva. — Head complete, not very heavily chitinized, usually con- 

 spicuously different from the remainder of body in color; antennae 

 usually short, appearing in many genera as pale rounded areas on each 

 side of head near anterior margin ; mandibles toothed ; maxillae well 

 developed, their inner surfaces usually conspicuously dentate; max- 

 illary palpi developed or very slightly so ; labium not in the form 

 of a flat plate; median dorsal sclerite of head tapering to a point 

 posteriorly. Ventral surface of thoracic and abdominal segments 

 sometimes with a transverse band of black locomotor spinules, 

 which are occasionally very conspicuous ; prothoracic and first 7 ab- 

 dominal segments each with lateral spiracles; apex of abdomen sim- 

 ple ; anal spiracles terminal. 



Pupa. — Head unarmed ; antennae curving well over upper mar- 

 gins of eyes, forming a semicircle and ending about middle of wings 

 along their costal margins. Thorax conspicuously elevated, the an- 

 terior margin almost vertical, the dorsum appearing in lateral view 

 almost globose; spiracles not elevated; legs elongated; entire fore legs 

 visible ; mid coxae visible only in part ; apices of all legs extending 

 beyond apices of wings, those of the hind pair sometimes reaching 

 apex of abdomen. Abdomen unarmed ; 7 pairs of spiracles present, 

 the basal pair usually hidden by wings. 



Imago. — As limited in this paper this family contains only those 

 genera that belong to the following subfamilies of Johannsen's paper 

 on Mycetophilidae : Diadocidiinae, Sciophilinae, and Mycetophilinae. 

 T consider that these three groups really constitute separate families, 

 but in the absence of larval and pupal material as criteria I prefer to 

 leave matters at present as they are, hoping at some future time to 

 elucidate further their relation, or that some other student of the order 

 mav nnd time to do so. 



