sustaining effect will be much the same, whatever the figure of
the outline of the superficies may be, and a circle perhaps
affords the best resistance of any. Take, for example, a circle
of twenty square feet (as possessed by the pelican) loaded with
as many pounds. This, as just stated, will limit the rate of
perpendicular descent to 1,320 feet per minute. But instead of
a circle sixty-one inches in diameter, if the area is bounded by
a parallelogram ten feet long by two feet broad, and whilst at
perfect freedom to descend perpendicularly, let a force be
applied exactly in a horizontal direction, so as to carry it
edgeways, with the long side foremost, at a forward speed of
thirty miles per hour--just double that of its passive descent:
the rate of fall under these conditions will be decreased most
remarkably, probably to less than one-fifteenth part, or
eighty-eight feet per minute, or one mile per hour.'

And again: 'It has before been shown how utterly inadequate the
mere perpendicular impulse of a plane is found to be in
supporting a weight, when there is no horizontal motion at the
time. There is no material weight of air to be acted upon, and
it yields to the slightest force, however great the velocity of
impulse may be. On the other hand, suppose that a large bird,
in full flight, can make forty miles per hour, or 3,520 feet per
minute, and performs one stroke per second. Now, during every
fractional portion of that stroke, the wing is acting upon and
obtaining an impulse from a fresh and undisturbed body of air;
and if the vibration of the wing is limited to an arc of two
feet, this by no means represents the small force of action that
would be obtained when in a stationary position, for the impulse
is secured upon a stratum of fifty-eight feet in length of air