Nielsen did not deal with the phylogeny of the order as a whole, because 

 he thought that the morphological material available was inadequate. He 

 used his data for the clarification of problems of phylogeny of certain 

 groups and their systematic position. However, he restricted himself to the 

 morphology and biology of the early stages of development. Hence, 

 Nielsen's methods were the exact opposite of those of Schmid (an interesting 

 phenomenon of specialization!). 



Entomologists have been of the opinion for a long time that the family 

 Rhyacophilidae is primitive; Nielsen confirmed this by the following 

 characters of the full-grown larvae (Nielsen, 1942:178—179); l) the head 

 capsule of some species of Rhyacophila resembles that of Rhaphidiidae; 



2) there is a simple articulation between the pleural sclerite and the coxa; 



3) the arrangement of the gills is primitive; 4) the anal legs (like those of 

 Hydroptilidae) are intermediate between the long legs of Philopotamidae and 

 Polycentropodidae and the short legs of Integripalpia. 



Nielsen wrote a large work on the morphology and biology of the larvae 

 and pupae of Hydroptilidae (Nielsen, 1948); according to the larvae, he found 

 that this family (which had been considered as homogeneous) consists of 

 2 different complexes. On the basis of a small amount of European material, 

 he separated these complexes as 2 new subfamilies: Hydroptilinae (genera 

 Hydroptila, Agraylea and Oxyethira) and Orthotrichinae (genera 

 Orthotrichia and It hy t r i c h i a). We consider these subfamilies as 

 groups of genera. The family Hydroptilidae is closely related to the 

 Rhyacophilidae in some characters, according to Nielsen; according to 

 other characters to the Glossosomatidae. 



In his "Trichopterologische Notizen" (Nielsen, 1943—1944), which deals 

 with the phylogeny of a number of groups, Nielsen revised the classification 

 of the Limnophilidae; he separated the new subfamily Ecclisopteryginae 

 on the basis of the morphology of the larvae and pupae (Nielsen, 1943); it 

 consists of the genera Ecclisopteryx and Drusus. The validity of 

 the subfamily is supported by the following characters: l) the characteristic 

 form of the head (which differs from that of Limnophilinae) with a dorsal 

 148 ridge and secondary setae; 2) the mandibles (like those of Apatania and 

 unlike those of Limnophilinae) are without denticles, have a sharp line along 

 the distal margin and are adapted for scraping; 3) the pronotum has a 

 characteristic form, with a transverse carina; 4) the chaetotaxy of the legs is 

 characteristic; there are numerous secondary setae on the femora, and the 

 claw bears a long basal seta like the Apataniinae; 5) there are dorsolateral 

 processes on the 8th abdominal segment; 6) hfead and mandibles of the pupa 

 have a characteristic form; 7) the female lays eggs under water as 

 Apatania. 



Nielsen considers the subfamily Ecclisopteryginae as related to the 

 Apataniinae. 



In his recent work on the family Limnophilidae, Schmid accepts the 

 subfamily of Nielsen, and places 9 genera in it; he wrote a monograph on this 

 group ("La sous-famille des Drusinae"); larvae of this subfamily are known 

 for the genera Drusus and Ecclisopteryx (Schmid, 1956). Contrary 

 to the rules of nomenclature, Schmid renamed the subfamily after its largest 

 and most ancient genus (Drusus), which contains 33 species of the 

 42 species of the whole subfamily; in his phylogenetic scheme, Schmid shows 



138 



