way as the tangential component, with the solar movement; and since complete knowledge of it, in a sufficient number of cases, is rapidly becoming accessible, while knowledge of tangential velocity must for a long time remain partial or uncertain, the advantage of replacing the discussion of proper motions by that of motions in line of sight is obvious and immediate.  And the admirable work carried on at Potsdam during the last three years will soon afford the means of doing so in the first, if only a preliminary investigation of the solar translation based upon measurements of photographed stellar spectra.

The difficulties, then, caused either by inaccuracies in star catalogues or by ignorance of star distances may be overcome; but there is a third, impossible at present to be surmounted, and not without misgiving to be passed by.  All inquiries upon the subject of the advance of our system through space start with an hypothesis most unlikely to be true.  The method uniformly adopted in them—­and no other is available—­is to treat the inherent motions of the stars (their so-called motus peculiares) as pursued indifferently in all directions.  The steady drift extricable from them by rules founded upon the science of probabilities is presumed to be solar motion visually transferred to them in proportions varying with their remoteness in space, and their situations on the sphere.  If this presumption be in any degree baseless, the result of the inquiry is pro tanto falsified.  Unless the deviations from the parallactic line of the stellar motions balance one another on the whole, their discussion may easily be as fruitless as that of observations tainted with systematic errors.  It is scarcely, however, doubtful that law, and not chance, governs the sidereal revolutions.  The point open to question is whether the workings of law may not be so exceedingly intricate as to produce a grand sum total of results which, from the geometrical side, may justifiably be regarded as casual.

The search for evidence of a general plan in the wanderings of the stars over the face of the sky has so far proved fruitless.  Local concert can be traced, but no widely diffused preference for one direction over any other makes itself definitely felt.  Some regard, nevertheless, must be paid by them to the plane of the Milky Way; since it is altogether incredible that the actual construction of the heavens is without dependence upon the method of their revolutions.

The apparent anomaly vanishes upon the consideration of the profundities of space and time in which the fundamental design of the sidereal universe lies buried.  Its composition out of an indefinite number of partial systems is more than probable; but the inconceivable leisureliness with which their mutual relations develop renders the harmony of those relations inappreciable by short-lived terrestrial denizens.  “Proper motions,” if this be so, are of a subordinate