The author places the family Rhyacophilidae lowest in the group of 

 families with campodeiform larvae, because its primary chaetotaxy is 

 almost unchanged compared with that of the 1st- stage, and its secondary 

 chaetotaxy is little developed; he also stresses the high specialization of 

 certain organs (anal legs). 



The author places the Hydroptilidae next to the Rhyacophilidae (including 

 Glossosomatinae) in spite of the special character of the full-grown larvae of 

 136 this family because the primary chaetotaxy (though highly specialized) 

 persists during the entire larval life of Hydroptilidae, and because of the 

 relatively poor development of the secondary chaetotaxy. The morphological 

 characters of the free -living larvae of the first 4 stages of this family, 

 which do not build cases and which have deep constrictions between the 

 abdominal segments, resemble the typical campodeiform larvae of other 

 families. 



The author separates the Philopotamidae as an independent, very 

 primitive family including the genera Philopotamus, Wormaldia and 

 Chimarrha, on the basis of the morphology of the larvae; the primary 

 chaetotaxy predominates in these genera; they have characteristic labrum, 

 claws and structure of the anal legs in all 3 genera. 



Siltala considers the composition of the family Polycentropodidae 

 (according to Ulmer, 1906) as not sufficiently uniform according to the 

 morphology of the larva; it is not correct in his opinion to unite the 

 Psychomyiinae and Polycentropodiinae in one family, Polycentropodidae; 

 he considers the Psychomyiinae, because of the prevalence of the primary 

 chaetotaxy of the larva, as a more primitive group; this subfamily differs 

 markedly from the Polycentropodiinae with their rich secondary chaetotaxy 

 in the morphology of certain organs (even in the first stage). 



Siltala places the family Hydropsychidae highest in the group of families 

 with campodeiform larvae because of their characteristic structures, and 

 their rich and highly differentiated secondary chaetotaxy; the full-grown 

 larvae of this family differ markedly from the lst-stage larvae. 



Siltala places the family Phryganeidae lowest in the group with cruciform 

 larvae; he considers this family as distinct from the Limnophilidae, and 

 describes a number of primitive characters in the organization of the larva 

 of Phryganeidae; he describes a certain resemblance to campodeiform 

 larvae in a number of structures, e. g., the slightly elongate, not completely 

 hypognathous head, the flattened abdomen with its deep constrictions between 

 the segments, the little differentiated primary chaetotaxy and the little 

 developed secondary chaetotaxy. 



Siltala noticed the close relationship between the larvae of Leptoceridae 

 and other eruciform larvae; he negated the close relationship between this 

 family and other families with campodeiform larvae. He placed the family 

 Sericostomatidae highest in the group with eruciform larvae. Among the 

 subfamilies of Sericostomatidae he placed the subfamily Lepidostomatiinae; 

 lowest (according to the morphology of the larvae). He noticed a certain 

 resemblance between the larvae of this subfamily and those of the subfamily 

 Goerinae on the one hand and family Limnophilidae on the other hand. 



Ulmer ("Genera Insectorum," 1907:11—20) considered the Psychomyiidae 

 as an independent family. He placed this rather primitive group between 

 the Philopotamidae and Polycentropodidae (Ulmer, 1907:19-20, 191—254).* 



* In the key to the families and also in the text, the family Polycentropodidae is given number 9, the 



Psychomyiidae number 10, and the Philopotamidae number 11 (Ulmer, 1907:19— 20); the numbers of the first 

 2 families are different on page 11, apparently because of a misprint. 



128 



