t 20 miles an hour, is the mystery to be explained.
It is not accounted for, quantitatively, by any of the theories which
have been advanced, and it is the one performance which has led some
observers to claim that it was done through 'aspiration.' i, e., that
a bird acted upon by a current, actually drew forward into that current
against its exact direction of motion."

Buzzards Soar in Dead Calm.

A still greater mystery was propounded by the few observers who asserted
that they had seen buzzards soaring in a dead calm, maintaining their
elevation and their speed. Among these observers was Mr. E. C. Huffaker,
at one time assistant experimenter for Professor Langley. The writer
believed and said then that he must in some way have been mistaken, yet,
to satisfy himself, he paid several visits to Mr. Huffaker, in Eastern
Tennessee and took along his anemometer. He saw quite a number of
buzzards sailing at a height of 75 to 100 feet in breezes measuring 5
or 6 miles an hour at the surface of the ground, and once he saw one
buzzard soaring apparently in a dead calm.

The writer was fairly baffled. The bird was not simply gliding,
utilizing gravity or acquired momentum, he was actually circling
horizontally in defiance of physics and mathematics. It took two years
and a whole series of further observations to bring those two sciences
into accord with the facts.

Results of Close Observations.

Curiously enough the key to the performance of circling in a light wind
or a dead calm was not found through the usual way of gathering human
knowledge, i. e., through observations and experiment. These had failed
because I did not know what to look for. The mystery was, in fact,
solved by an eclectic process of conjecture and computation, but once
these computations indicated what observations should be made, the
results gave at once the reasons for the circling of the birds, for
their then observed attitude, and for the necessity of an independent
initial sustaining speed before soaring began. Both Mr. Huffaker and
myself verified the data many times and I made the computations.

These observations disclosed several facts:

1st.--That winds blowing five to seventeen miles per hour frequently had
rising trends of 10 degrees to 15 degrees, and that upon occasions when
there seemed to be absolutely no wind, there was often nevertheless a
local rising of the air estimated at a rate of four to eight miles or
more per hour. This w