Dealing with the very important question whether the two streams are actually intermingled in space, Eddington finds them nearly at the same mean distance and thoroughly intermingled, and there is no possible hypothesis of Drifts I and II passing one behind the other in the same line of sight. A third drift, to which all the Orion stars belong, is under investigation, together with comprehensive analysis of the drifts according to the spectral type of all the stars included. The farther research on star-streaming is pushed, the more it becomes evident that a third stream, called Drift O, is necessary, especially to include B-type stars. The farther we recede from the sun, the more this drift is in evidence. At the average distances of B-type stars, the observed motions are almost completely represented by Drift O alone. Halm of Cape Town concludes from recent investigations that the double-drift phenomena (Drifts I and II) is of a distinctly _local_ character, and concerns chiefly the stars in the vicinity of the solar system; while stars at the greatest distances from the sun belong preeminently to Drift O. The 60-inch reflector on Mount Wilson gathers sufficient light so that the spectra of very faint stars can be photographed, and a discussion of velocities derived in this manner has shown that Kapteyn's two star streams extend into space much farther than it was possible to trace them with the nearer stars. Star-streaming, then, may be a phenomenon of the widest significance in reference to the entire universe. As to the fundamental causes for the two opposite and nearly equal star streams, it is early perhaps to even theorize upon the subject. Eddington, however, finds a possible explanation in the spiral nebulæ, which are so numerous as to indicate the certainty of an almost universal law compelling matter to flow in these forms. Why it does so, we cannot be said to know; but obviously matter is either flowing into the nucleus from the branches of the spiral, or it is flowing out from the nucleus into the branches. Which of the two directions does not matter, because in either case there would be currents of matter in opposite directions at the points where the arms merge in the central aggregation. The currents continue through the center, because the stars do not interfere with one another's paths. As Eddington concludes: "There then we have an explanation of the prevalence of motions to and fro in a particular straight line; it is the line from which the spiral branches start out. The two star streams and the double-branched spirals arise from the same cause." CHAPTER LVII THE GALAXY OR MILKY WAY