The Orbits of Four Spectroscopic Binaries - The Total Solar Eclipse of January 24, 1925 W. E. Harper J. Patterson PAGE 1 4 C. A. Chant 10 J. F. Frédette 13 H. B. Collier 17 From Scientific American 19 A Short Method of Obtaining the Azimuth of Polaris Valuable Investigations on Mars Review of Publications Notes and Queries Meetings of the Society Upon request, made previous to publication of article, contributors will be supplied free either with fifty copies of the issue containing the article, or with the sheets containing the article stitched in a printed cover. If separate reprints are desired the cost will be as follows: Covers, extra, 50 copies, $1.50; 100 copies, $2.00; 200 copies, $2.50. Business correspondence, remittances, etc., should be addressed to H. W. Barker, Treasurer of the Society, 198 College Street, Toronto. Communications regarding Library matters should be addressed to the Librarian. $2.00 per annum. Single numbers, 25 cents. Subscription to the JOURNAL is included in membership fee. THE By W. E. HARPER "HE following summary of the investigation of four spectroscopic binaries may be of interest to readers of the JOURNAL. One has a period of about three days while the other three are long. The data of the stars themselves are given in the table immediately following whilst the orbital elements are collected at the end of the paper. This star, which is 5 Tauri, was announced a binary by the Lick observers from observations made during the years 1897 to 1907. Other velocities were obtained by Küstner, by the Mount Wilson observers and by the writer while at Ottawa. These early observations have naturally helped to define the period but apart from that they have not been used. Twenty-six single-prism spectra made at Victoria during the years 1919 to 1924 and covering two cycles of the long period of 960 days have formed the basis for the determination of the elements. The plates were measured on the comparator against Arcturus as a standard and the resulting probable error of a plate is ±0.9 km. per sec. Boss 1190 This star, known as zeta Aurigae, was also announced a binary from the Lick Observatory in 1908. Miss Maury at Harvard had previously classified the spectrum as composite suggesting that it consisted of a K-spectrum and one of B-type. Our own plates bear out this view and in general the spectrum has a "washed out" appearance due to the overlapping. However, one plate taken at the time when the K-type star was in its orbit nearest the sun has a spectrum markedly different in that this "washed out" appearance is absent, the absorption lines being particularly intense. The probability of an eclipse in an orbit where the period is 973 days is necessarily small but some explanation is demanded by the change in the spectrum. The plates were likewise measured on the spectrocomparator against Arcturus and the probable error of ±1.0 km. per sec. was satisfactorily low. The star will be kept under observation here. Boss 1452 Twenty-five spectrograms obtained with the single-prism were made the basis of a determination of the orbital elements. The measures depend for the most part on four lines, the two hydrogen lines X4340 and X4101, the magnesium X4481 and the calcium 13933. True, there are numerous other metallic lines but in general they are too faint and elusive to measure excepting occasionally X4549 and X4233. The measures were made upon the micrometer engine with a resulting probable error of a plate of ±3.1 km. per sec. Boss 5192 Twenty-seven single-prism spectra of this star measured on the comparator against Arcturus as a standard give fairly reliable |