The light of the pole star, which was taken for the standard or unit of measurement, was reflected by a prism into one object glass, and that of the star to be measured into the other. The light of the brighter star was then reduced to exactly that of the fainter by the turning of a screw having a register attached. The indication of the register gave the measure, which was confirmed by repeated observations. Telescopes mounted in the ordinary way continued to be used in other branches of photometric work. [Illustration: HARVARD OBSERVING STATION AT WILLOWS, CAL., JANUARY 1, 1889.] The photometric survey of the sixth magnitude and brighter stars was completed Aug. 25, 1881. In 1882 a new and more powerful meridian instrument was constructed and a photometric survey of a list of about 21,000 stars, from the sixth to the ninth magnitude, was entered upon. This work was finished Sept. 29, 1888, and soon afterwards the instrument, with others, was sent to Peru in charge of Mr. S. I. Bailey of the observatory corps, where, May 11, 1889, a corresponding survey of the stars, from the first to the ninth magnitude, inclusive, between 30° south and the southern pole, was begun. Thus the facts relating to all the stars in the sky of these classifications will be embodied in the final result. The record, which will comprise several volumes, one or more of which have already been published, will have an identity throughout as respects the method, the instrument, and the unit of measurement. It will be authoritative as a text book or series of text books, and will enhance the value for reference, and comparison of various records of the light of stars, both those of modern and ancient date. The successful working of the two meridian photometers led to the construction of one still more powerful, having an aperture of 12 inches. The first was of 1½ inch aperture, and the second of four inches. The three differ somewhat in mechanism, but are the same in principle. The 12-inch is called by distinction the “horizontal telescope.” It will be available in case a photometric survey of stars of fainter magnitudes shall be undertaken, but its use is not limited to photometry. In 1879, a photometer was devised for measuring the light of nebulæ, thus applying to these objects and to stars the same unit and scale. In 1881, photometric observations of certain bright parts of the moon, were made for the Selenographical Society of England, the particular parts being selected by that society. It thus was shown that the lunar scale of light in common use may be closely expressed in terms of stellar magnitude, each degree of the lunar scale answering to six-tenths of a magnitude. Photometry has been very extensively applied at Harvard in study of variable stars.