LATITUDE, VARIATION OF. For many years it has been suspected that terrestrial latitudes might be subject to small changes, and that these might possibly affect the results, of ordinary astronomical observations. But in spite of all efforts to detect with certainty the existence of such changes. it was not until the year 1888 that Kfistner proved beyond a doubt that latitudes vary by observable amounts. His observations were made in Berlin, and he found that the •ati tude of that place was less by two-tenths of a second of arc in the spring of 1885 than it had been in the spring of 1884. His result has been abundantly- confirmed by subsequent observers.
If we imagine two straight lines drawn from the centre of the earth, one to the pole. and the other to a given observatory, then the angle be tween these two lines is called the co-latitude of the observatory. The latitude, in the ordi nary geographical or astronomical sense, is ob tained simply, by subtracting this co-latitude from 90°. It is evident that the co-latitude (and therefore also the latitude) will remain constant for any given observatory if the pole maintains an immovable position on the earth. Now. if we disregard, as we may, moderate irregularities in the earth's surface, we can take its figure to be a slightly flattened globe or sphere. The shortest possible line through the centre, and limited by the surface at each end, may he called the axis of figure of the earth. The points where this axis meets the surface are the of fig ure. But the earth has still another axis. viz. the axis of rotation. About this axis the planet revolves once in twenty-four hours, giving rise to all the diurnal phenomena of astronomy. Con stancy of latitude would imply the relative fixity of these two axes. That each shall maintain con tinuously exactly the same position with respect to the other is the one necessary and snffieient condition for perfectly invariable terrestrial lati tudes. If the axis of figure he subject to a slow revolution about the axis of rotation. there will be a corresponding variation of astronomically determined latitudes. The maximum amount of the variation will be the same as the small angle between the two axes, and its period will be equal to the time required for the rotation of the one axis about the other. At the end of every such period, the latitudes of all places on the earth should return to their original values.
Up to the publication of the work of Kiistner, in ISRS, fundamental astronomy had adopted invariability of latitude as a fact practically established. All the results of astronomical ob servations made prior to that date must there fore he subject to so much error as might be prodneed by assuming a constancy of latitudes in the discussion of the observations. Now, the aberration constant (see ABERRATION OF LIGHT) is particularly liable to error from this source. It is evident, therefore, that a redetermination of this quantity was imperatively needed, and that the necessary observations must he arranged in such a manner as to take account of the effect of latitude variation. Careful study of the problem
brought out the fact that the two quantities in volved are entangled in such a way that no avail able method of research could prove satisfactory, unless it had for its object the Alin') itmwous de termination of both aberration and variation of latitude. It seemed best to employ what is known as the 'zenith-telescope method,' and the best modification of this was suggested by Kiist ner himself. The observations, when made, must 1)e continued more or less through the entire night and must be made on every clear night for a period of fourteen months, if it be desired to determine the aberration. For the latitude va riation, the observations must of course be eon tinned for a term of years—indeed, they must be kept up as long as we wish to trace the polar motion. It will he seen that the problem is one of great dillieulty. testing to the utmost the pa tience rind endurance of the astronomer. 'Alore over, as in the ease of the solar parallax, the precision can be enhanced by making a simul taneous series of observations at more than one observatory. If the participating observatories are situated upon the same parallel of latitude, or very near it. the results obtained will enjoy a further increase of precision. This was first suggested by Fergola of Naples. The advan tage consists in the possibility of observing just the same stars at all the observing stations, so that the differences of latitude of the stations are determined independently of any knowledge of the exact positions of the stars on the sky. This is most important, for the positions of the stars are never known with absolute precision, being themselves but the results of fallible hu man observation. Moreover, the polar motion can be deduced from the latitude differences of the observatories just as well as from the actual lati tudes. If, for instance, the pole happens to he moving at a given time toward an American observatory, it will be moving away from ob servatories in .Japan or the Philippine Islands. So that, if we can but measure, from time to time, the latitude differences of observatories properly situated. we can get an accurate and complete idea of the actual motions of the pole. The International Ceodetie Association, which includes all the civilized nations, has now under taken the systematic observation of latitude ation. Four stations have been established upon the same parallel of latitude. Two are in the United States, one in Japan, and one in Sicily. Two private observatories participate volun tarily. Systematie observations were begun about the end of 1899.
The accepted constant of aberration must also he rerarded as subject to slight correction. so long as the latitude problem remains unsettled, especially as the most recent aberration results exhibit rather large diseordanees among them selves. Consult Chandler's articles in the As tronomical Journal (Cambridge. Mass.), and Al brerlit's articles published in the reports of the International Ceodetie Association, and also, in abridged form, in the Astronomische Yachrichten Cermany). See PARALLAX.