1890.] 263 



inch, and then somewhat abruptly, indicating, I believe, that the first 

 moult has been accomplished, enters the pith chamber, in which it 

 afterwards keeps. Occasionally the first portion runs backwards, but 

 it is invariably changed to a forward direction as soon as the pith 

 chamber is invaded. I have been unable to find any remains of the 

 egg, but I feel sure it is deposited at the spot where the mine is found 

 to commence ; the insect, so far as can be judged by the examination 

 of a dry specimen, being provided with a cutting ovipositor, quite 

 capable of penetrating the bark of a twig. Probably, the site of the 

 egg, which, when the larva is full grown, is usually removed by two 

 internodes from the end of the shoot, was at the time of laying, the 

 actual end itself, and, therefore, tender enough for the ovipositor to 

 pierce. This would explain, too, why the larva occasionally mines 

 backwards at starting. 



I imagine that on these occasions the tissues at the time of hatching 

 are still very immature, and that, instead of eating away forwards and 

 jeopardising the terminal vitality of the shoot, the larva wisely works 

 in the opposite direction. That it has a marvellous knowledge of the 

 geography of its surroundings is shown by the ease with which it finds 

 the leaf stalk ; there is no experimental boring here and there to hit 

 it, but the larva enters at once without any hesitation, taking no doubt 

 as its guide the fibro-vascular bundle given oif to the leaf. It might be 

 thought perhaps that the little creature was sometimes just a trifle 

 puzzled, because it may at times overshoot the mark a bit and have to 

 retrace its steps, but I am persuaded that it knows very well what it is 

 about, and passes by the leaf stalk in the first instance only for the 

 reason that it is not yet quite mature enough for entering it. 



The larva lias neither thoracic nor anal legs, and the four pairs of ventral ones 

 are very small and ill-developed. The thoracic segments are broad, the ventral more 

 slender and tapering towards the anal extremity. The head is small and flat, re- 

 tracted under the 2nd segment, and has the posterior lobes produced. In the 

 penultimate skin the colour is whitish, with the intestinal contents red. The head 

 shining black. A T-shaped blackish mark (the cross line posterior) on the back of 

 segment two, and on the under-side a pair of blackish longitudinal lines. There are 

 three anal plates, grey with black margins ; one is placed on the flap, the others 

 latero-ventrally beneath it. In the last skin, colour white. Head pale brown. 

 Segment two entirely grey on both sides ; a faint grey spot (plate) on the back of 

 three, and a still fainter one on the venter ; a small ventral plate on four, better 

 recognised by its texture than by its colour. The three anal plates very pale brown 

 ■with red margins. Compared with resplendellum in the last skin, that larva has the 

 head dark grey ; the 2nd segment blackish on both sides ; and distinct black plates 

 on the dorsum of three and on the venter of three and four. The thi-ec anal plates 

 grey with black margins. 



