bours, and Bessel's labours had for their aim the
reconstruction, on an amended and uniform plan, of the entire science of
observation.

A knowledge of the places of the stars is the foundation of
astronomy.[62] Their configuration lends to the skies their distinctive
features, and marks out the shifting tracks of more mobile objects with
relatively fixed, and generally unvarying points of light. A more
detailed and accurate acquaintance with the stellar multitude, regarded
from a purely uranographical point of view, has accordingly formed at
all times a primary object of celestial science, and was, during the
last century, cultivated with a zeal and success by which all previous
efforts were dwarfed into insignificance. In Lalande's _Histoire
Celeste_, published in 1801, the places of no less than 47,390 stars
were given, but in the rough, as it were, and consequently needing
laborious processes of calculation to render them available for exact
purposes. Piazzi set an example of improved methods of observation,
resulting in the publication, in 1803 and 1814, of two catalogues of
about 7,600 stars--the second being a revision and enlargement of the
first--which for their time were models of what such works should
be.[63] Stephen Groombridge at Blackheath was similarly and most
beneficially active. But something more was needed than the diligence of
individual observers. A systematic reform was called for; and it was
this which Bessel undertook and carried through.

Direct observation furnishes only what has been called the "raw
material" of the positions of the heavenly bodies.[64] A number of
highly complex corrections have to be applied before their _mean_ can be
disengaged from their _apparent_ places on the sphere. Of these, the
most considerable and familiar is atmospheric refraction, by which
objects seem to stand higher in the sky than they in reality do, the
effect being evanescent at the zenith, and attaining, by gradations
varying with conditions of pressure and temperature, a maximum at the
horizon. Moreover, the points to which measurements are referred are
themselves in motion, either continually in one direction, or
periodically to and fro. The _precession_ of the equinoxes is slowly
progressive, or rather retrogressive; the _nutation_ of the pole
oscillatory in a period of about eighteen years. Added to which, the
non-instantaneous transmission of light, combined with the movement of
the earth in its or