ON THE OSMYLIN.E OF JAPAN. 



49I 



FAMILY HEMEROBIIDAE. 



Subfamily Osmylinas. 



Certain authors have separated from the Hemerobiidae such genera 

 as Osmyhts, Dilar, etc., as representing distinct families, but I 

 agree with Banks 15 in thinking that it would be more natural to keep 

 up this family in the broad sense, dividing it into four subfamilies as 

 follows: 1) The Dilarinae, characterized by the peculiar ovipositor of 

 the female and the pectinate antennae of the male. 2) The Psycho- 

 psinae, characterized by the union of subcosta, radius and radial sector. 

 3) The Osmylinae, in which the subcosta and radius are united near 

 tip of wing. 4) The Hemerobiinae, to include all the rest of the 

 family. 



Of these subfamilies, the Dilarinae seems to be well differentiated 

 from all the others on account of the male antennae being of a 

 structure somewhat resembling that in the Sialidae, and of the 

 ovipositor resembling that of the Rhaphididae. The remaining three 

 subfamilies are very nearly allied with one another ; especially close 

 seems to be the relation between the Osmylinae and the Hemerobiinae, 

 and it may even become necessary in the future to unite these two 

 subfamilies into one. Here I should mention that the fact of the union 

 of the subcosta and radius, taken alone by itself, does not seem to be 

 always sufficiently adequate to serve as distinctive criterion between 

 the Osmylinae and the Hemerobiinae ; for, in a new Japanese Osmyline 

 form {Sisyra japonica n. sp.) I have found the union of the said veins 

 only imperfectly carried out. Further, it may be pointed out that 

 Sisyrella nikkoana (Navâs), a form generally placed under the Heme- 

 robiinae, 25 so closely approaches Sisyra in several respects that it may 



1) Synopses and Descriptions of Exotic Neuroptera. Trans. Amer. Ent. Soc, xxxix 

 (I9I3)- 



2) Navâs, Hémérobides neuveaux du Japon. Rev. Russ. d'Ent., No. 4, p. 397-98 (1910), 

 — Banks, I.e., p. 218. 



