Benj. D. Walsh on Insects inhabiting Willow Galls. 223 



On the Insects, COLEOPTEROUS, HYMEITOPTEROUS and DIPTEROUS, in- 

 habiting the Galls of certain species of Willow. — Part 2d and last. 



BY BENJ. D. WALSH, M. A. 



DIPTERA.— Supplement. 

 GALLMAKERS.— Genus CECIDOMYIA, Subgenus CECLDOMYIA. 



No. 3. Gall S. strobiliscus Walsh. — I described this gall from 

 a single dried specimen found by Mr. Bebb on Salix rostrata in North 

 Illinois. I have since found very numerous specimens of what for 

 the present I regard as the same gall on S. discolor near Rock Island, 

 111. Of 23 gathered March 23d one was undistinguishable from the 

 S. rostrata gall ; the rest had the tips of the external leaves (except at 

 the tip of the gall) not angulated, but more or less rounded with a 

 subobsolete midrib outside which terminated in a minute tooth or 

 beak. In other respects they did not differ, and especially in the 

 veins on the inside of the leaves being obsolete or subobsolete. The 

 general outline of this gall was ovate lanceolate, rarely ovate ; length — re- 

 jecting one stunted specimen, which however contained a larva — 1.05 

 — 1.G5 inch, diameter .57 — .72 inch. The stunted specimen was not 

 porrect, but deflected at an angle with the axis of the twig, and I sub- 

 sequently found a few others varying in the same way. In one gall 

 I met with 2 or 3 of the same rchelimum eggs which occur so copiously 

 in S. strobiloides 0. S., and May 26th I bred several Orchelimum 

 larvae from these galls. 



The larva and pupa, as well as the pupal integument, are undis- 

 tinguishable from those of S. strobiloides 0. S., but the cocoon is 

 shorter, being only lg — 2 times as long as the larva, instead of 2z — 

 3 times as long : 2 larvae and 2 pupae examined April 9. 



Imago. Cecidomyia s. strobiliscus n. sp. — Differs from Gee. 

 8. strobiloides,W alsh, only in the % antenna) being 23 — 24-joiuted, (not 

 21 — 22-jointed,) with 1 or 3 of the terminal joints sessile and the right 

 aud left antenna varying iu the same S in the number of joints; and 

 in the origin of the anterior branch of the 3rd longitudinal wing-vein 

 being usually pretty distinct. Hence it can scarcely be separated from 

 Gee. s. rhodoides Walsh, though the galls are quite different. One % . 

 eleven $ , bred April 30 — May 8. 



No. 4. Gall S. unaplialioides Walsh. — I found a single specimen 



