Time by clock. Hour circle reading. 11h. 23m. 52s. 11h. 55m. 30s. Telescope east. 11h. 31m. 55s. 24h. 8m. 24s. Telescope west. ———— ———— ———— ———— ———— —————— 8m. 3s. 12h. 12m. 54s. 8m. 3s. ———— —————— 2) 4m. 51s. ———— —————— Collimation error at 2m. 25·5s. dec. 46´·5 angle between object glass and declination axis acute. If this error is not corrected, it must be added when the telescope is on the east side of the pier, and subtracted when on the west.[19] 5th Adjustment.—Place a striding level on the pivots of the declination axis and bring the bubble to zero by turning the polar axis; read off the hour circle and note it; then reverse the declination axis east and west and replace the level; bring the bubble to zero and again read the circle. The readings should show the axis to be turned through half a circle, and the difference shows the error. If the second reading minus the first be more than half a circle or 12 hours, it shows that the pivot at the east at the first observation is too high, and therefore in bringing the declination axis level, the first reading of the hour circle is diminished from its proper amount and increased on the axis being reversed. To adjust the error, find half the difference of circle readings and apply it, with the proper sign, to each of the two circle readings, which will then differ by exactly twelve hours; bring the circles to read one of the corrected readings and alter the declination axis until the bubble of the level comes to zero. If the pivots of the declination axis are not exposed, so that the level can be applied, the following method must be adopted:—Fasten a small level on any part of the declination axis or its belongings, say on the top of the counterpoise weight; bring the axis apparently horizontal and the bubble to zero; turn the telescope on the declination axis, so that by the turning of the counterpoise the level comes below it; if then the bubble is at zero, the axis of the level is parallel to the declination axis, and both are horizontal, and if not it is clear that neither of these conditions holds; therefore bring the bubble to zero by the two motions of the level with reference to the counterpoise and the motion of the declination axis on the polar axis, so that the error is equally corrected between them; repeat the proceeding until the level is parallel with the axis, when it will show when the axis is horizontal as well as the striding level. For example:—