A HUNT FOR THE ATRIL SVANGLK (JALT,. 247 



irregularl}' dislocated, go back nearl}' to their ])r()2)('i- i)laccs, when the 

 soft tissues of the gall shrivel u}*. 



In a few eases, the gall occupies one side only of the hud, and a few 

 leaves or catkins develop on the other. The great majority of the galls 

 are one-celled, but a very large number are double, and a ver^^ occasional 

 one contains three or more cells ; I incline to think that the latter, at 

 any rate, are due to more than one oviposition taking place in the same 

 bud. 



The next step is to obtain the summer gall, l)ut here Cameron is 

 not so clear a guide as one might wish, and there appears to be still 

 room for some doubt as to what the summer (agamic) gall may be. 

 This would corres])ond to the spangle or winter gall of the more tyjiical 

 species of Ncuroterus, from which (ijyn'liuHS clearly differs considerably 

 in habit. 



On May 12th, a search on <iprilinus trees showed some male catkins 

 Avith certain anthers looking a trifle larger and a trifle greener than 

 normal, and an examination showed the ujiper part of the filament, or 

 l^erhaps rather the dissepiment between the antliers, to be thickened 

 and swollen, slightly separating the bases of the anthers ; these anthers 

 Avere also less easily broken off than those supposed to be normal and 

 healthy, some of which had already dehisced ; still the difference Avas 

 so slight that I did not feel sure that I Avas not deceiving m3'self Avith 

 some ordinary A-ariation in the anthers. On May 14th, the swollen 

 stamens presented many examj^les in Avhich the bases of the anther 

 cells Avere Avidely separated, and the Avhole stamen of about tAvice the 

 normal bvilk : the normal stamens were dehisced, sliriA-elled and 

 withered ; Avhilst, on the trees examined, a large proportion of the galled 

 anthers Avere also dead, but in a different way from the healthy stamen, 

 being brown and fleshy looking, rather as if they Avere rotting instead of 

 shrinking into a dr}^ husk. 



On the 16th, these galls of scldcchtendali Avere fully developed as re- 

 gards size, and Avere beginning to fall. XotAvithstanding that the ther- 

 mometer stood at the remarkable figure, for May, of 75° in the shade, I 

 was surprised by this remarkably rapid dcA-elopment. I A'isited mA* 

 two most accessible trees on Avhich aprilinus had been freely i)resent, in 

 order to see how the galls Avere progressing, and Avas surprised to find 

 them well dcA-eloped and dropping off Avithout more than a slight 

 touch. They Avere, hoAvever, rather scarce ; it Avould have been difficult 

 to get a few score of them, as a great man}' of them appeared not to 

 have properly developed and the catkins had dropped off except Avherc 

 kept on by 6aroar«m and qiKidrilhwdtus. I tlierefore visited the (ijrr/lintdi 

 tree at once and found KchJccJiieii(htli abundant ; that is, the catkins 

 freel}' remained on the tree and ;ill had several galls — many of them a 

 dozen or tAvo. They were already but lightly attached and easil}' broke 

 off in gathering. Though I describe them as abundant, of cour&e only 

 a fcAv low branches Avere accessil)le and from these I made a fair bag 

 for Mr. Bignell ; if the tree Avas as Avell-tenanted throughout, there Avere 

 no doubt a good many hundred thousand galls upon it. Why this tree 

 was so Avell tenanted, Avhilst my trees Avere largely failures, I don't 

 knoAv. It is strong evidence in favour of this being the summer gall 

 of apyiliitus, that only these two trees exliibited any schlechtniddli, and 

 that this one Avas the most fully supplied Avith both forms, Avhilst trees 

 that had shown no ajn-diuns Avere free from ^rhUrlitcnddll. 



