The nematine larva, after its final molt, 1 generally enters the ground 
to pupate, spinning a double or single silken cocoon more or less incor- 
porated with particles of earth exteriorly. In the case of the species 
having several broods annually, the cocoons, at least of the summer 
generation, are frequently constructed above ground, either among the 
dry leaves and rubbish at the base of the host plant, or on the twigs, or 
in crevices of the bark of the latter. Some of the gall species pupate 
in their galls, but many of them abandon their galls to undergo their 
transformations in rotten wood, in the pith of plants, in deserted galls, 
or in the earth. 
Species living exposed on the leaves will also sometimes enter deserted 
galls, either to transform or to hibernate. 2 
Range of species and economic importance. — Some few species are known 
to be widely distributed, and this is particularly true of the larch saw- 
fly, which occurs throughout Canada and the ^Northern States, and also 
in Europe. Whether this species (Lygwonematus crichsonii ITartig) 
can be called an introduced species or net is a question. Its wide dis- 
tribution throughout the Northern States would seem to indicate that it 
has, perhaps for many centuries, occurred on both continents. The 
gooseberry and currant sawflies, however ' (Pteronus ribesii Scop, and 
Pristiphora appendiculata Hartig), are undoubted cases of importation. 
Tbe economic importance of the group is well illustrated by the species 
just mentioned, the last two being among the most serious enemies of 
several small fruits, and the first threatening the almost total destruction 
of the larch forests in many districts. Other examples of very destruc- 
tive species are the willow sawfly {Pteornus ventral is Say), the wheat 
sawfly (Pachyncmatns extensicornis Xort.), the Western pear sawfly 
(Gymnonychus calif ornicits n. sp.), and the cranberry sawfly (Pristiphora 
idiota Kort.). 
Difficulties arising from confusion of species and loss of types. — The 
classification of this natural and distinctly differentiated subfamily has 
been, until quite recently, in a very experimental and unsatisfactory 
condition, and this is particularly the case with the genus Kcmatus, 
which, cumbersome from the number of species referred to it, has been 
invariably a stumbling block to every student of the Tenthredinidie. 
Following the lead of the earlier European writers on the group, Amer- 
ican describers of species in the old genus Ncmatus have based their 
characterizations almost solely on mere differences in coloration, with 
such references to structural features as are of little value or of generic 
rather than specific importance. The failure to note the variations in 
the structure of different parts of the insect has led to the most 
l See> "The Final Molting of Tenthredinid Larva','' Proc. Ent. Soc. Wash., vol. n, 
p. 115. 
2 Seo "Hibernation of Nematids and its bearing on Inqnilinous Species/' Proc. 
Ent. Soc. Wash., vol. m, p. 263. 
