Observed dec. of Capella 43° 50´ 30˝ Telescope west. 47° 0´ 0˝ Telescope east. ——— ——— ——— 2) 90° 50´ 30˝ ——— ——— ——— 45° 25´ 15˝ 47° 0´ 0˝ Error due to refraction 0° 0´ 7˝ 43° 50´ 30˝ ——— ——— ——— ——— ——— ——— Instrumental declination 45° 25´ 8˝ 2) 3° 9´ 30˝ True declination 45° 52´ 0˝ ——— ——— ——— ——— ——— ——— Index error 1° 34´ 45˝ 26´ 52˝ This indicates that the pole of the instrument is pointing below the true pole, and index error 1° 34´ 45˝. Observed declination of Pollux 6h. west 28° 19´ 18˝ Refraction 0° 0´ 46˝ ——— ——— ——— 28° 18´ 3˝ True declination 28° 20´ 10˝ ——— ——— ——— 0° 1´ 38˝ This shows the pole to be 1´ 38˝ east of true pole. 4th Adjustment.—For the estimation and correction of the third error, that of collimation, an equatorial star is brought to the centre of the field of the telescope, the time by a clock noted, and the hour circle read. The polar axis is then turned through half a circle, and the star observed with the telescope on the opposite side (say the west) of the pier, the time noted, and the hour circle read. Subtract the first reading from the second (plus twenty-four hours if necessary) and subtract the time elapsed between them, and the result should be exactly twelve hours, and half the difference between it and twelve hours is the error in question. If it is more than twelve hours the angle between the object end of the telescope and the declination axis is acute, and if less then it is obtuse. This error can then be corrected by the proper screws. A little consideration will show, that if the angle between the object end of the telescope and the declination axis be acute, and the telescope is on the east side of the pier, and pointing to a star, say on the meridian, the hour circle will not read so much as it would do if the line perpendicular to the declination axis were pointing to the meridian. When the telescope is on the wrest side of the pier, the circle will read higher for the same reason, and therefore the difference between the angle through which the hour circle is moved and 180° is equal to double the angle between the line perpendicular to the declination axis and the collimation axis of the telescope; allowance being made for the star’s motion. For example γ Virginis, Dec. 0° 46´·5.