1864.] 609 



the place where it worked was discolored and brown. On August 19, 

 from about half a gallon of galls, which had been gathered only 5 or 6 

 days before, there came out about a dozen larvae, apparently with the 

 intention of going under ground, and after some considerable search I 

 discovered one inside the substance of the stem of a jrall. No others 

 came out subsequently, so flir as I observed, from that large mass of 

 galls ; and if many of them had come out they could scarcely have 

 escaped notice, for there was nothing in the breeding-jar but the galls 

 themselves, and no earth at the bottom of it. Those that came out 

 were .05 inch long, rather elongate, and with the head more porrect 

 and pointed than is usual, except in the larvae of C. s. siliqna and C. 

 s. nodulus. The breast-bone was rather indistinct, but seemed to be 

 about twice as long as its basal width and tapered to I the basal width 

 at tip. Repeatedly at other times during the season I had cut into 

 these galls, both at home and in the field, and always fiiiled to find Ce- 

 cidomyidous larva? in them, though I sometimes found that of an in- 

 quilinous Lepidopteron. It is possible that the above larvae might also 

 have been inquilinous; but if so, where were the Cecidomijin that really 

 made the galls? I am persuaded that the gall is really Cecidomyidous, 

 because, l.s^, I obtained from them the same Lepidopterous imago that 

 I obtained in great numbers from the Cecidomyidous gall ^S'. hrassi- 

 coides n. sp.;* 2;^/, I noticed on them in considerable numbers, and both 

 in the larva and imago states, the same Heteropterous insect — Antho- 

 coris psf'tidorhinche Fitch — that occurs also in great numbers on the 

 Cecidomyidous gall *S'. brassicoides n. sp., and in less numbers on the 

 Cecidomyidous galls S. rhodoides n. sp. and S. strobiloidcs 0. S. ; 3/y/, 

 As already stated (p 551) I believe that all Willow Galls are either the 

 work of Glall-gnats or of Saw-flies, and the larvae of Saw-flies being com- 

 paratively large and conspicuous, if ^S'. senigvia was a Tenthredinidous 

 gall I must have found some Tenthredinidous larvse in it, so often as I 

 dug into it; whence by the method of exhaustion it follows that it must 

 be a Cecidomyidous gall. — As on June 19 I noticed on these galls the 

 larva of a lai'ge Thrips, and a few others subsequently in the imago 



*The species here referred to belongs, according to Dr. Clemens, to Tortri- 

 cid<E, but has not as yet been described by him. It is remarkable for varying 

 in the most surprising manner, and I sent Dr. Clemens a very large series of 

 all the variations. 



