estimating the average distance of certain classes of stars.

The parallaxes of which we have heretofore spoken consist in the
change in the direction of a star produced by the swing of the
earth from one side of its orbit to the other. But we have already
remarked that our solar system, with the earth as one of its
bodies, has been journeying straightforward through space during
all historic times. It follows, therefore, that we are continually
changing the position from which we view the stars, and that, if
the latter were at rest, we could, by measuring the apparent speed
with which they are moving in the opposite direction from that of
the earth, determine their distance. But since every star has its
own motion, it is impossible, in any one case, to determine how
much of the apparent motion is due to the star itself, and how
much to the motion of the solar system through space. Yet, by
taking general averages among groups of stars, most of which are
probably near each other, it is possible to estimate the average
distance by this method. When an attempt is made to apply it, so
as to obtain a definite result, the astronomer finds that the data
now available for the purpose are very deficient. The proper
motion of a star can be determined only by comparing its observed
position in the heavens at two widely separate epochs.
Observations of sufficient precision for this purpose were
commenced about 1750 at the Greenwich Observatory, by Bradley,
then Astronomer Royal of England. But out of 3000 stars which he
determined, only a few are available for the purpose. Even since
his time, the determinations made by each generation of
astronomers have not been sufficiently complete and systematic to
furnish the material for anything like a precise determination of
the proper motions of stars. To determine a single position of any