THE VARIATION OF LATITUDE.—In the Astronomical Journal, No. 19 (November 14), Prof. S. C. Chandler gives the eighth of the important series of articles that he has been contributing on the variation of latitude. The special part of the subject which is referred to deals with the direction of the rotation of the pole and is accompanied by an explicit demonstration which includes all the data bearing upon it. Owing to the insufficient extent of series of observations in widely different longitudes to furnish independent values of the constants for both terms of the variation, Prof. Chandler has thought well to combine short series made in nearly the same longitudes, and so deduced fourteen determinations of the numerical equations for the latitude variation. Reducing the values so obtained to a common epoch, he found that the values of the observed Julian date when the latitude would be a minimum, or when the pole of figure would pass the meridian of the respective stations by virtue of the fourteen months revolution alone, and of the sun's longitude on the observed date when the same phase would occur by virtue of the annual term alone, both decreased from Pulkowa towards Madison showing that the direction of the rotations in both the elements was fröm west to east. In the latter part of the article Prof. Chandler refers not only to our knowledge of the general law of latitude variation, but to the accuracy of the necessary constants which afford us a means of predicting the immediate future course. The minimum of the curve of April, 1893, will be followed by an interval of nearly two years, and will be marked by very slight fluctuations, so that from the maximum of October, 1893, to that of August, 1895, or from minimum April, 1893, to that near the beginning of 1895, “there will apparently be but a single decidedly marked period of, say 20–22 months,” the total range amounting to 0″.10 as against 0″.56 which prevailed in 1889 and 1892. In May, 1896, the same dimensions as in 1889 will be again attained, and the variation from that time forward to 1898 it will be in full play with the range of 0″.5 or 0″.6, a period of nearly 390 days which prevailed between 1889 and 1892. In § 2 of the article Prof. Chandler adds a few words as to the reality of these movements of the earth's axis as against the motions being “merely misinterpretations of the observed phenomena” or an illusory effect of instrumental error due to the influence of temperature. Those of our readers who are still sceptical on the subject will learn that the observed law of latitude variation includes two terms, one with a period of fourteen months, and another with twelve months, making the phases come in very different relations to conditions of temperature dependent on season, an argument greatly against that brought forward by temperature-variation believers.