. There
had been a spontaneous decision, one half in favour of the left, one
half in favour of the right.

The arrangement presents another merit, one superior to that of
symmetry: it has the merit of corresponding with the minimum expenditure
of force. To admit of the exit of the whole series, if the string
consists of n cells, there are originally n partitions to be perforated.
There might even be one more, owing to a complication which I disregard.
There are, I say, at least n partitions to be perforated. Whether each
Osmia pierces her own, or whether the same Osmia pierces several, thus
relieving her neighbours, does not matter to us: the sum-total of the
force expended by the string of Bees will be in proportion to the number
of those partitions, in whatever manner the exit be effected.

But there is another task which we must take seriously into
consideration, because it is often more troublesome than the boring of
the partition: I mean the work of clearing a road through the wreckage.
Let us suppose the partitions pierced and the several chambers blocked
by the resulting rubbish and by that rubbish only, since the horizontal
position precludes any mixing of the contents of different chambers. To
open a passage for itself through these rubbish-heaps, each insect
will have the smallest effort to make if it passes through the smallest
possible number of cells, in short, if it makes for the opening nearest
to it. These smallest individual efforts amount, in the aggregate, to
the smallest total effort. Therefore, by proceeding as they did in my
experiment, the Osmiae effect their exit with the least expenditure of
energy. It is curious to see an insect apply the 'principle of least
action,' so often postulated in mechanics.

An arrangement which satisfies this principle, which conforms to the law
of symmetry and which possesses but one chance in 512, is certainly no
fortuitous result. It is determined by a cause; and, as this cause
acts invariably, the same arrangement must be reproduced if I renew the
experiment. I renewed it, therefore, in the years that followed, with as
many appliances as I could find bramble-stumps; and, at each new test, I
saw once more what I had seen with such interest on the first occasion.
If the number be even--and my column at that time consisted usually
of ten--one half goes out on the right, the other on the left. If the
number be odd--eleven, for instance--the Osmia in the middle goes