INSTINCT OF INSECTS. 47 
“Si seulement ils élevoient une fois des cibanes quar- 
rées,” (says Bonnet when speaking as to what faculty 
the works of the beaver are to be referred, ) “ mais ce sont 
éternellement des cAbanes rondes ou ovales? :”—and so 
we might say of the phenomena in question :—Show us 
but one instance of bees having substituted mud or mor- 
tar for mitys, pissoceros, or propolis, or wooden props 
for waxen ties, and there could be no doubt of their 
being here guided by reason. But since no such instance 
is on record; since they are still confined to the same 
limits—however surprising the range of these limits—as 
they were two thousand years ago; and since the bees 
emerged from their pupze but a few hours before, will 
set themselves as adroitly to work and pursue their opera- 
tions as scientifically as their brethren, who can boast 
the experience of a long life of twelve months duration ;— 
we must still regard these actions as variations of instinct. 
In the second place, no degree of reason that we can 
with any share of probability attribute to bees, could be 
competent to the performance of labours so complicated. 
as those we have been considering, and which, if the re- 
sult of reason, would involve the most extensive and va- 
ried knowledge in the agents. Suppose a man to have 
attained by long practice the art of modelling wax into a 
-congeries of uniform hexagonal cells, with pyramidal 
bottoms composed each of three rhombs, resembling the 
cells of workers among bees. Let him now be set to 
make a congeries of similar but larger cells (answering 
to the male cells), and unite these with the former by 
other hexagonal cells, so that there should be no disrup- 
~ tion in the continuity or regularity of the whole assem- 
4 (uvres, ix. 159. 
VOL. Il, 2k 
