comparisons up to the present century having been made either by the naked eye, or with the assistance of an opera-glass only. About the middle of this century photometers specially adapted for comparing the light of the stars were first used by German astronomers. Zollner invented a photometer, consisting of a telescope in which the light from a kerosene-lamp, admitted through a very small hole and presenting the appearance of a star, is compared with the real star under observation. C. S. Pierce, of the United States Coast Survey, used such a one in the construction of a photometric catalogue of 494 stars, published in Vol. IX of the "Annals of Harvard College Observatory." His description of his difficulties with this very imperfectly contrived and still more imperfectly constructed instrument would be amusing, if it did not excite regret that so accurate an observer and excellent a mathematician should have been weighted with an instrument so poorly adapted to the work.
Up to within a few years, only about 500 stars had been the subject of photometric observation, and for some time it has been regarded as highly desirable that systematic comparisons should be made of the light of all stars visible to the naked eye.
In Vol. XIV of the "Annals of the Observatory of Harvard College" we have the records of the most extensive and complete photometric observations ever undertaken. The great attention which Professor Pickering, the director of the observatory, has given to astro-photometry, and the large experience he has had with photometers, both of his own and others' construction, peculiarly fitted him for such a task.
Part I of the "Annals" opens with a description of the meridian photometer devised by Professor Pickering. Having ascertained by experiment that any change of position on the part of the observer had an injurious effect upon the observations, he constructed his instrument in the form of a broken transit, in which, the line of sight being always horizontal, stars at all altitudes could be observed without moving the head. Experience with other photometers had also satisfied him that no artificial light could be a proper standard of comparison for the light of a star. A real star was therefore chosen, and the fact that the pole-star is always visible, and its light, on account of its very slight changes of altitude, a constant quantity, directed him in his choice of it as the standard of comparison for all stars. To make sure that it was otherwise suitable, a large number of observations were made of the pole-star, to ascertain if its light was subject to any periodical variation.
A reliable instrument and perfect standard having been thus obtained, and many preliminaries settled which it is impossible to touch upon here, observations were begun in October, 1879 and continued till September, 1882. During this period seven hundred series of observations were made, including 94,476 separate comparisons, the result being that every star not fainter than the sixth magnitude, between the north pole and thirty degrees of south declination, was compared from three to fourteen times with the pole-star. The whole number of stars thus compared is 4,260.
The space to which this review is necessarily restricted renders it impossible to give even a passing notice to the immense amount of work expended upon the subject of astro-photometry as recorded in these "Annals." An idea can be formed from the statement that there are in all ninety-one tables, some of which occupy several pages. The "General Catalogue," constituting Table XXVII, alone occupies 211 pages, each line having twenty-six columns, one of which has fourteen sub-headings.
Part II of the "Annals," the publication of which has been delayed till the present year, is largely devoted to a discussion of the work of those astronomers, from Ptolemy to the present day, whose estimates of the relative magnitudes of the stars Professor Pickering has chosen for comparison with his own. A very complete list of all known or suspected variable stars is also given, with copious notes as to dates and observers. There is also a chapter on the distribution of the stars.
Professor Pickering does not close his work with the advancement of any theories of his own. But, for whatever purpose an exact determination of the relative magnitudes of the stars may be desired, either for the discovery of variable stars, or to as-
