112 
ARCHITECTURE OF BEES. 
the direction of their combs, when I approximated a 
surface too smooth to admit of their clustering on it. 
They always sought the wooden sides. I thus com- 
pelled them to curve the combs in the strangest shapes, 
by placing a pane of glass at a certain distance from 
their edges. These results indicate a degree of in- 
stinct truly wonderful. They denote even more than 
instinct ; for glass is not a substance against which 
bees can be warned by nature. In trees, their natu- 
ral abode, there is nothing that resembles it, or with 
the same polish. The most singular part of their 
proceeding is changing the direction of the work, be- 
fore arriving at the surface of the glass, and while yet 
at a distance suitable for doing so. Do they antici- 
pate the inconvenience which would attend any other 
mode of building ? No less curious is the plan adopted 
by the bees for producing an angle in the combs ; 
the wonted fashion of the work, and the dimensions 
of the colls, must be altered. Therefore, the cells 
on the upper or convex side of the comb are enlarged ; 
they are constructed of three or four times the width 
of those on the opposite surface. How can so many 
insects, occupied at once on the edges of the combs, 
concur in giving them a common curvature from one 
extremity to the other ? How do they resolve on 
establishing cells so small on one side, while dimen- 
sions so enlarged are bestowed on those of the other ? 
And is it not still more singular that they have the 
art of making a correspondence between cells of such 
reciprocal discrepance ? The bottom being common 
to both, the tubes alone assume a taper form. Per- 
