218 HYMENOPTEEA. 



lioiiej'-j'ellow costa, and a dusky stigma, edged with honey- 

 yellow. The male difi'ers a little in having black coxae. Mr. 

 Walsh states that the larva is a pale grass-green Avorm, half 

 an inch long, Avith a black head, which becomes green after 

 the last moult, but with a lateral brown stripe meeting with 

 the opposite one on the top of the head, where it is more or 

 less confluent ; and a central brown-black spot on its face. 

 It appears the last of June and early in July, and a second 

 brood in August. They spin their cocoons on the bushes on 

 which they feed, and the fly appears in two or three weeks, the 

 specimens reared by him fl3'ing on the 26th of August. P. 

 sycopJianta Walsh is an "• inquiline," or guest gall-saw-flj'', 

 .inhabiting a Cecidomyian gall on a willow. 



The genus Exmra comprises several gall-making species. It 

 differs from the preceding genus in the second, instead of the 

 first, submarginal cell haA'ing two recurrent venules. Mr. 

 Walsh has raised E. orbitalis Norton (E. genuina Walsh) from 

 galls found on Salix humilis. This gall is a bud which is 

 found enlai'ged two or three times its natural size, before it 

 unfolds in spring. The larva is twenty-footed, is from .13 to 

 .19 of an inch long, of a greenish white color, and the 

 head is dusky. It bores out of its gall in autumn, descending 

 an inch into the ground, where it spins a thin, silken, whitish 

 cocoon. The gall of E. saUcis-ovum Walsh is found on Salix 

 cordata. The female is shining yellow, Avhile the ground color 

 of the male is greenish white. The gall of this species is an 

 oval roundish, sessile, one-chambered, green or brownish swell- 

 ing, .30 to .50 of an inch long, placed lengthAvise on the side of 

 small twigs. The larva is pale j^ellowish, and the fl}' appears 

 in April. The fly is, according to Walsh, " absolutel}' undistin- 

 guishable by any reliable character from the guest gall-saw-fl3-, 

 Euura perturlans Walsh," Avhich inhabits dipterous galls made 

 by Cecidomyian flies on the willow and grape (Walsh) . If these 

 two "'species" do not differ from each other, either in the larva 

 or adult state, "by any reliable characters," then one must 

 question whether the variation in habits is sufficient to separate 

 them as species, and Avhether E. salicis-OA'um does not, some- 

 times, instead of forming a new gall, laj^ its eggs in a gall read}- 

 made by a dipterous gall-fly. We have seen that Odynerus 



