INTRODUCTION. 
87 
from food till their task is completed, and if they 
do not, they may frequently increase the suppleness 
of the leaf, although we do not allow them credit 
for being sufi&ciently sagacious to gnaw its fibres for 
that express purpose. 
The proceedings of different caterpillars, in forming 
dwellings of this kind, vary considerably ; but the 
mechanical means which they employ are in most 
cases similar to the above. Some roll the leaves 
from the point doivn the mid-rib, others from the 
side towards the centre. Many compose their 
habitations by rolling two leaves together, while 
others draw the two sides towards each other and 
convert the whole leaf into a capacious tube. A 
few, less laborious than their fellows, select the long 
and nearly parallel leaves which terminate the 
yoimg shoots of the vrillow, and very expeditiously 
adapt them to their own purposes, simply by wind- 
ing round them a number of silken cords. Such 
kinds show a wonderful degree of instinctive fore- 
sight, in carefully gnawing the bud at the extremity 
of the shoot, which is enclosed in the centre of the 
packet, and thereby preventing it from sprouting, as 
its doing so would have the effect of disarranging 
the whole tenement, and probably rupturing the 
bands by which it was held together. A small 
number take up their abode in the centre of um- 
belliferous herbs. It is well known, that in many 
of these plants the little umbels gradually become 
longer as they recede from the centre, and that 
consequently when they approximate to each other, 
