152 (July. 



display are the leaves occupied by apicella and intimella, provided only 

 the season is favourable to the production of rich autumnal colouring. 

 To see, as I have more than once, an aspen tree with nearly every leaf 

 of a pure yellow, save for a bold splash of vivid green striking across 

 from stalk to margin, is an extraordinary sight, and one that can 

 scarcely fail to arrest attention. Equally remarkable are the leaves of 

 the Bedford willow {Salix Russelliana) when intimella is present, for 

 one lateral half will be yellow or even brown, while the other half 

 remains green. In both cases too the death of the larva whilst in the 

 stalk in no way lessens the effect on the blade, and the simplest, indeed 

 the only plausible, explanation I can see is, that some substance is 

 produced, which being absorbed by the vascular bundles, among which 

 the creature is burrowing, gets distributed to the parts of the leaf they 

 supply, where it is taken up and appropriated by the cells. 



But this curious power is not restricted to the Nepticulcs. It is 

 equally noticeable in the Lithocolletes, yet with a distinct difference as 

 regards one point. In the former the patch faded more or less in- 

 sensibly into the surrounding area ; in the latter it is sharply defined 

 and coterminous with the limits of the mine, in fact, it is neither more 

 nor less than the mine itself. The first thing the Lithocolletis larva 

 does is to separate the cuticle of the leaf, until an area sufiicient for 

 its future needs is mapped out. It then spins strands of silk across 

 this separated cuticle, and presently by the natural contraction of the 

 silk the sides of the mine are drawn together, and at the same time, 

 aiding the operation, the larva nibbles all round the edge of the mine, 

 picking out the parenchyma but leaving the veinlets untouched. After- 

 wards it continues to feed upon the parenchyma from the commence- 

 ment made at the edge, leaving the central parts to the last, and never 

 at any time touching the network of veinlets. The virtue of the 

 process lies, as it did in the JS'epticula, in the first stage of the work, 

 namely, the separation of the cuticle, for the larva, having effected 

 this, may come to gi'ief, and yet the patch it had so cunningly marked 

 out for its future use will remain green and fresh as if nothing were 

 amiss, whilst the rest of the leaf may long since have gone through 

 the whole series of autumnal changes. 



Striking as the contrast is between the restriction of the process 

 within definite limits in the one case, and the absence of any such 

 limitation in the other, it can scarcely point to any real difference in 

 the principle itself, but rather suggests a want of agreement in some 

 detail of proceedure. Now, the Nepticula larva, though it may not 

 eat down the veinlets to the same level as the parenchyma, does, never- 



