ere is another principle, the prodigious influence of which, at
least with regard to bees, has hitherto been unknown, that is the
sentiment of aversion which all females continually feel against each
other, a sentiment whose existence is so fully demonstrated by my
experiments, and which explains many important facts in the theory of
swarms.

_PREGNY, 10. September 1791._

LETTER XII.

_ADDITIONAL OBSERVATIONS ON QUEENS THAT LAY ONLY THE EGGS OF DRONES, AND
ON THOSE DEPRIVED OF THE ANTENNAE._

In relating my first observations on queens that lay male eggs alone, I
have proved that they lay them in cells of all dimensions indifferently,
and even in royal cells. It is also proved that the same treatment is
given to male worms hatched from eggs laid in the royal cells, as if
they were actually to be transformed to queens; and I have added, that
in this instance the instinct of the workers appeared defective. It is
indeed most singular, that bees which know the worms of males so well
when the eggs are laid in small cells, and never fail to give them a
convex covering when about to transform to nymphs, should no longer
recognise the same species of worms when the eggs are laid in royal
cells, and treat them exactly as if they should change to queens. This
irregularity depends on something I cannot comprehend.

In revising what is said on this subject, I observe still wanting an
interesting experiment to complete the history of queens that lay only
the eggs of drones. I had to investigate whether these females could
themselves distinguish that the eggs they deposit in the royal cells
would not produce queens. I have already observed that they do not
endeavour to destroy these cells when close, and I thence concluded,
that in general the presence of royal cells in their hive does not
inspire them with the same aversion to females whose fecundation has
been retarded; but to ascertain the fact more correctly, it was
essential to examine how the presence of a cell containing a royal nymph
would affect a queen that had never laid any other than the eggs of
drones.

This experiment was easy; and I put it in practice on the fourth of
September, in a hive some time deprived of its queen. The bees had not
failed to construct several royal cells for replacing their females. I
chose this opportunity for supplying them with a queen, whose
fecundation had been retarded to the twenty-eighth day, and which laid
none but th