32 TRANSACTIONS, NATURAL HISTORY SOCIETY OF GLASGOW. 



instances of this kind is found in Periclistus 

 canince, which inhabits the gall of Rhodite 

 centifolice, increasing its size by nearly one-half, 

 and forming numerous cells inside, while it would 

 have been a roomy shell tenanted by one larva if 

 the proper tenant had been left undisturbed. Cer- 

 tain galls, too, can be detached from the plant 

 when half-grown, and, if supplied with moisture, 

 will grow as well as in their natural position. 



It must, however, be mentioned that the above 

 view — of the action of the larva as being the most 

 potent agent in causing the growth of galls — 

 applies only to the Cynipidcp. In the Tenthredini- 

 dce the larva certainly has nothing to do ^vith the 

 formation of the gall, which is fully formed before 

 the larva quits the egg. Here the only agencies 

 at work are the irritation made by the saw-like 

 ovipositor, and by the egg, ^vhich increases in size 

 considerably after being placed in the plant. It is 

 true that there is a liquid deposited in the wound 

 along with the eggs, but there is not the slightest 

 doubt that its function is to close the wound. It is 

 laid by all Tenthredhndce along with the eggs, and 

 in some cases it apparently prevents the wound 

 from closing in on the eggs and crushing them. 



In connection with the Tenfhredmidce the question 

 naturally arises how one species raises galls, and 

 another, which can hardly be distinguished from it, 

 does not do so, both depositing their eggs moreover 

 in the leaves. I believe the answer is that the 

 former lays its eggs in the cambium-layer and the 

 other does not. 



The poison, or, as it might be termed, the infec- 

 tion theory, as we have seen, accounted for the 

 forms of the galls by the specific nature of the 

 secretion. How can we explain, on the larval 

 irritation theory, for each species always having 

 a particular gall of its own? Many gall-flies, 

 which give origin to most diverse galls, are not 

 distinguishable, and all of them in the same genus 



