It is necessary to adjust the quantity of light from the lamp in the field, so that the wires are sufficiently visible while the stars are not put out by too much illumination; for the majority of stars a red glass before the lamp is best. This gives a field of view which renders the wires visible without masking the stars, but a green or blue light is sometimes very serviceable. A shaded lamp should be used for reading the circles on the micrometer, so as not to injure the sensitiveness of the eye by diffused light in the observatory. A lamp fixed to the telescope, having its light reflected on the circles, but otherwise covered up, is a great advantage over the hand-lamp. In very faint stars, which are masked by a light in the field sufficient to see the wires, the wires can be illuminated in the same manner as in the transit, but there is this disadvantage—the fine wires appear much thickened by irradiation, so that distances, especially of close stars, become difficult to take. * * * * * We come now to the differential observations made with the equatorial. Let us explain what is meant. Suppose it is desired to determine with the utmost accuracy the position of a new comet in the sky. If we take an ordinary equatorial, or an extraordinary equatorial (excepting probably the fine equatorial at Greenwich), and try to determine its place by means of the circles, its distance from the meridian giving its right ascension and its distance from the equator giving its declination, we shall be several seconds out, on account of want of rigidity of its parts; but if we do it by means of such an instrument as the transit circle at Greenwich, we wait till the comet is exactly on the meridian, and determine its position in the way already described. As a matter of fact, however, the transit circle is not the instrument usually used for this purpose, but the equatorial. We do not however just bring the comet or other object into the middle of the field and then read off the circles, but we differentiate from the positions of known stars; so that all that has to be done in order to get as perfect a place for the comet as can be got for it by waiting till it comes to the meridian—which perhaps it will do in the day-time, when it will not be visible at all—is to determine its distance in right ascension and declination from a known star, by means of a micrometer. Of course one will choose the brightest part of the comet and a well-known star, the place of which has been determined either by its appearance in one of the catalogues, or by special transit observations made in that behalf. We then by the position micrometer determine its angle of position and distance from the known star at a time carefully noted, or we measure the difference in right ascension and the difference in declination. [Illustration: FIG. 163.—Ring Micrometer. ]