Correlation of Structure and Host- Relation i8i 



■ ■ ■ Pemphigus spirothecEe, (?) Siphonophora aveiis:, (Ind.) Aphis 

 pruni, (la.) Lecanium, 3 spp., (Cal., Fla., Mo., Neb.) Pulviuaria 

 innumerabilis, (Mo.) Dactylopius destructor, (Fla.) Kertnes, 

 3 spp., (Cal., N. Y., Mo.) Aspidiotus corticalis, (Fla.) Diaspis 

 rosae, (D. C, Mo., Cal.) 



From this condensed statement certain interesting facts 

 plainly appear. Rhopus, Holcothorax, Aphycus, Blastothrix, 

 Psilophrys, L,eptomastix, Chiloneurus, Comys, Cerchysius, 

 Rileya, Cerapterocerus, Habrolepis, Phsenodiscus, Ericydnus 

 and Baeocharis, or 15 out of the 23 genera of which we know 

 the habits, are parasitic upon bark lice exclusively. Copido- 

 soma, Pentacnemus and Dinocarsis are parasitic exclusively 

 upon lepidopterous larvae. Bothriothorax is parasitic upon 

 dipterous larvae only. Honialotylus is parasitic exclusively 

 upon coleopterous larvae of the families Coccinellidse and 

 Chrysomelidse. Isodromus is parasitic exclusively upon 

 Chrysopa larvae, issuing from their cocoons. Tanaostigma is 

 parasitic upon the larvae of seed-inhabiting weevils. 



Thus far there has been absolute uniformity in host relation 

 within generic bounds in so far that the host insects of each 

 particular genus are closely related and of the same general 

 type. There is one genus remaining, however, which is a 

 biological complex and, from the uniformity which has existed 

 among other members of the group, the natural inference that 

 it is also a morphological complex would be justified. Close 

 study of the classificatory characters bears out this assertion. 

 Encyrtus is one of those unwieldy genera of varying limit, 

 found in nearly every large family of insects, in which many 

 species have been lumped, frequently for insufficient reasons, 

 and really for want of a better place to put them. Up to a 

 certain stage in the classification of the group, authors have 

 not felt justified in separating the species generically, since 

 their characters have not seemed as important as those which 

 have been considered of generic value, while the subgenus is 

 an element of convenience or confusion, according as you may 

 view it, which has not as yet been adopted to any extent in 

 entomology. Encyrtus is such a genus. Coming at the end 

 of a synoptic table, by a process of elimination the refuse has 



