st be _at least_ as
great as that of steel.[888]

Ratification from an unexpected quarter has lately been brought to this
conclusion. The question of a possible mobility in the earth's axis of
rotation has often been mooted. Now at last it has received an
affirmative reply. Dr. Kuestner detected, in his observations of 1884-85,
effects apparently springing from a minute variation in the latitude of
Berlin. The matter having been brought before the International Geodetic
Association in 1888, special observations were set on foot at Berlin,
Potsdam, Prague, and Strasbourg, the upshot of which was to bring
plainly to view synchronous, and seemingly periodic fluctuations of
latitude to the extent of half a second of arc. The reality of these was
verified by an expedition to Honolulu in 1891-92, the variations there
corresponding inversely to those simultaneously determined in
Europe.[889] Their character was completely defined by Mr. S. C.
Chandler's discussion in October, 1891.[890] He showed that they could
be explained by supposing the pole of the earth to describe a circle
with a radius of thirty feet in a period of fourteen months.
Confirmation of this hypothesis was found by Dr. B. A. Gould in the
Cordoba observations,[891] and it was provided with a physical basis
through the able co-operation of Professor Newcomb.[892] The earth,
owing to its ellipsoidal shape, should, apart from disturbance, rotate
upon its "axis of figure," or shortest diameter; since thus alone can
the centrifugal forces generated by its spinning balance each other.
Temporary causes, however, such as heavy falls of snow or rain limited
to one continental area, the shifting of ice-masses, even the movements
of winds, may render the globe slightly lop-sided, and thus oblige it to
forsake its normal axis, and rotate on one somewhat divergent from it.
This "instantaneous axis" (for it is incessantly changing) must, by
mathematical theory, revolve round the axis of figure in a period of 306
days. Provided, that is to say, the earth were a perfectly rigid body.
But it is far from being so; it yields sensibly to every strain put upon
it; and this yielding tends to protract the time of circulation of the
displaced pole. The length of its period, then, serves as a kind of
measure of the plasticity of the globe; which, according to Newcomb's
and S. S. Hough's independent calculations,[893] seems to be a little
less than that of steel. In an earth compacted