Aphilotrix radicis. 29 



line in front of the scutellum ; further the metathorax/ 

 and an irregular blotch on the first segment of the 

 abdomen, are also dark, as well as the bases of the coxae 

 and tibiae of the hind legs, and the claws. The thorax 

 is closely covered with silky pubescence : the antennae 

 are variously coloured, but the basal four joints are 

 reddish brown, and the apex dark. 



Experimental breeding. After the flies quit the 

 galls they usually rest a few days before beginning to 

 lay their eggs. When I first made experiments with 

 this species in 1875 I expected them to select the roots 

 or the lower part of the trunk to lay their eggs in, but 

 I soon discovered that they always creep up the trunk 

 in search of the buds. After they have carefully 

 examined the buds with their antennae they begin to 

 prick them. They do this just as it is done by 

 a Neuroterus, but the fly takes up its position nearer 

 to the base of the bud. It drives its terebra under 

 one of the bud scales until it reaches the foot of the 

 bud-axis. Thence it directs the channel not towards 

 the centre of the bud, in which lie the rudimentary 

 leaves, but below this point, and the tip of the terebra 

 thus enters tissues from which later shoots are de- 

 veloped. Some eggs may, it is true, come to lie upon 

 the base of the rudimentary leaves, but the greater part 

 will be found lower down, and there are usually no 

 galls found on the leaves themselves. 



When the buds that have been pricked begin to 

 shoot, a long interval takes place before there is any 

 sign of gall formation. The earliest symptoms recog- 

 nizable are a delay in the development of the bud, 



