2. The declination circle must read 0° when the telescope is at right angles to the polar axis.
3. The polar axis must be placed in the meridian.
4. The optic axis of the telescope, or line of collimation must be at right angles to the declination axis, so that it describes a great circle on moving about that axis.
5. The declination axis must be at right angles to the polar axis, in order that the telescope shall describe true meridians about that axis.
6. The hour circle must read 0h. 0m. 0sec. when the telescope is in the meridian.
When these are correctly made the line of collimation will, on being turned about the declination axis, describe great circles through the pole, or meridians, and when moved about the polar axis, true parallels of declination; and the circles will give the true readings of the apparent declination, and hour angles from the meridian.
To make these adjustments, the telescope is set up by means of a compass and protractor, or otherwise in an approximately correct position, the declination circle put so as to read nearly 90° when the telescope points to the pole, and the hour circle reading 0h. 0m. 0sec. when the telescope is pointing south.
First, then, to find the error in altitude of the polar axis.
Take any star from the Nautical Almanac of known declination on or near the meridian, and put an eyepiece with cross wires in it in the telescope, and bring the star to the centre of the field as shown by the wires. Then read the declination circle, note the reading down and correct it for atmospheric refraction, according to the altitude[[18]] of the star by the table given in the Nautical Almanac, turn the telescope on the polar axis round half a circle so that the telescope comes on the other side of the pier. The telescope is then moved on its declination axis until the same star is brought to the centre of the field, and the circle read as before and corrected. The mean of the two readings is then found, and this is the declination of the star as measured from the equator of the instrument, and its difference from the true declination given by the almanac is the error of the instrumental equator and of course, also of the pole at right angles to it.
It is obvious that if the declination circle were already adjusted to zero, when the telescope was pointing to the equator of the instrument, one observation of declination would determine the error in question; and it is to eliminate the index error of the circle, as it is called, that the two observations are taken in such a manner that the index error increases one reading just as much as it decreases the other, so that the mean is the true instrumental declination.