mentioned the subjects of precession, nutation, and aberration. The star's places most sensibly change from circumstances unconnected with the parallax. We must know these accurately beforehand; and knowing these accurately beforehand, we proceed as follows.
We observe the star with the mural circle while the earth is in the position E'. We apply the corrections for precession, nutation, aberration, and refraction; and we shall know what the corrected position of that star should be half a year hence, as observed when the earth is at E'". Now, suppose we go to this second state of things, and when the earth is at E"', we observe the meridional zenith distance of the star, we correct it for refraction, precession, nutation, and aberration. Now, do these two corrected zenith-distances agree? Are the stars (after all these corrections are applied) seen exactly in the same direction when the earth is at E' and when it is at E'"? All calculations for these accidental causes of disturbance being affected, the result is this:—for the vast majority of stars we do not discover any sensible difference; the difference is, at any rate, exceedingly small; the stars are so far off that, for the vast majority of them, we can see no difference in the directions of the line E'S and the line E'"S. There are some stars, however, that are not at so great a distance, so that the inclination of these lines to each other can be ascertained; but the angle is exceedingly small, and is measured with much difficulty. In the southern hemisphere, there is the bright star of the Centaur, (Alpha Centauri,) for which it would seem that the inclination of the two lines from the opposite sides of the earth's orbit to the star, is an angle of two seconds and no more.
