This important investigation was very greatly facilitated by the general catalogue of 6,188 stars well distributed over the entire sky, the work of the late Professor Boss. It was published by the Carnegie Institution of Washington, and includes all stars down to the sixth magnitude. Boss was very critical in the matter of stellar positions and proper motions and his work is the most accurate at present available. Excluding stars of the Orion type and the known members of moving clusters, Kapteyn's investigation was based on 5,322 stars, which he divided into seventeen regions of the sky, each northern region having an antipodal one in the southern hemisphere. Mathematical analysis of these regions showed them all in substantial agreement, with one exception, and enabled Kapteyn to draw the conclusion that the stars of one stream, called Drift I, move with a speed of thirty-two kilometers per second, while those of the other, Drift II, travel with a speed of eighteen kilometers per second. Their directions are not, like those of east and west bound ships, 180 degrees from each other, but are inclined at an angle of 100 degrees. Drift I embraces about three-fifths of the stars, and Drift II the remaining two-fifths. Quite as remarkable as the drifts themselves is the fact that the relative motion of the two is very closely parallel to the plane of the Milky Way. This epochal research has very great significance in all investigations of stellar motions, and it has been verified in various ways, particularly by the Astronomer Royal, Sir Frank Dyson, who limited the stars under consideration to 1,924 in number, but all having very large proper motions. In this way the two streams are even more characteristically marked. But radial velocity determinations afford the ultimate and most satisfactory test, and Campbell has this investigation in hand, classifying the stars in their streaming according to the type. Type A stars are so far found to be confirmatory. Turning to the question of physical differences between the stars of the two streams, Eddington inquires into the average magnitude of the stars in both drifts, and their spectral type. Also whether they are distributed at the same distance from the sun, and in the same proportion in all parts of the sky. His conclusion is that there is no important difference in the magnitudes of the stars constituting the two drifts. Regarding their spectra, stars of early and late types are found in both streams, with a somewhat higher proportion of late types among the stars of Drift II than those of Drift I. Campbell and Moore of the Lick Observatory have investigated seventy-three planetary nebulæ which exhibit the phenomena of star-streaming, and have motions which are characteristic of the stars.