In this work the calculations proceed directly from the actual observations, without any presumptions as to the result. All of the original observations have been included in the calculation, without any omissions and without the assignment of weights. No corrections of any kind have been applied to the observed quantities. This procedure has been adopted as the only safe one in the first search for a hitherto unidentified effect. The present results strikingly illustrate the correctness of this method, as it now appears that the forty-six years of delay in finding the effect of the orbital motion of the earth in the ether-drift observations has been due to the efforts to verify certain predictions of the so-called classical theories and to the influence of traditional points of view.
Absolute Motion of the Solar System and The Earths Orbital Motion Determined
Northern apex of the solar motion rejected
As already explained, the interferometer determines the line in which the motion of the earth with respect to the ether takes place but does not determine the direction of motion in this line. The results of the observations given in Tables I and II indicate either an apex located near the north pole of the ecliptic or one diametrically opposite, near the south pole of the ecliptic. The choice between the two possible directions of motion is determined by the consistency of the results in satisfying the original observations taken as a whole and in connection with known phenomena. The studies of the proper motions and of the motions in the line of sight of stars in our own cluster have shown that the solar system is moving with respect to the nearby stars towards an apex located in the constellation Hercules, about 42 from the northern one of the two apexes indicated by the interferometer observations, the velocity of this motion being about nineteen kilometers per second. This circumstance seemed confirmatory of an absolute motion towards the north and the northern apex was chosen for further study of the problem.
Upon the completion of the observations for three epochs at Mount Wilson, corresponding to April 1, August 1, and September 15, 1925, a study of the results was made upon the pre
sumption of a northern apex. Various trial solutions were checked with the parallelogram apparatus, Fig. 18, and finally by a partial least squares solution, for the determination of the velocity of the cosmic motion. The effects which should be characteristic of the several epochs because of the varying direction of the orbital motion could not be identified in the corresponding curves of observation, indicating that the orbital component is probably much smaller than the cosmic component. The curves for the three epochs were simply averaged and it was found that when plotted in relation to local civil time, the curves are in such phase relations that they nearly neutralize each other; the average effect for the three epochs thus plotted is very small and unsystematic. The curves of observation were then plotted with respect to sidereal time and a very striking consistency of their principal characteristics was shown to exist, not only among the three curves for azimuth and those for magnitude, but, what was more impressive, there was a consistency between the two sets of curves, as though they were related to a common cause. The average of the curves, on sidereal time, showed conclusively that the observed effect is dependent upon sidereal time and is independent of diurnal and seasonal changes of temperature and other terrestrial causes and that it is a cosmical phenomenon. The results of this study were presented as the address of the President of the American Physical Society at the meeting in Kansas City, on December 29,
1925.13 The conclusion stated that there is a positive, systematic ether-drift effect, corresponding to a constant relative motion of the earth and the ether, which at Mount Wilson has an apparent velocity of ten kilometers per second; and that the variations in the direction and magnitude of indicated motion are exactly such as would be produced by a constant motion of the solar system in space towards an apex, near the north pole of the ecliptic, having a right ascension of 171/2 hours and a declination of +65. On the hypothesis of the Stokes ether concept, that the ether is partially entrained by matter moving through it, it was suggested that the observed velocity of ten kilometers per second might be only a fraction of the absolute velocity;
13 D. C. Miller, Science 63, 433 (1926).
and further, assuming that the earths orbital velocity of thirty kilometers per second, similarly reduced, would be so small as to be near the limit of perceptibility in these observations, it was suggested that the actual velocity of the cosmical motion might be two hundred kilometers or more, per second. It was also reported that, for some unexpected reason, all the azimuths were displaced to the westward.
A fourth series of observations was made at Mount Wilson, corresponding to the epoch, February 8, 1926, and a reexamination of all the observations for the four epochs was made, as before, on the presumption of an apex near the north pole of the ecliptic and, because no consistent effect of the orbital motion could be found, the four series of observations were simply averaged for the determination of the cosmical motion of the solar system. The results of this elaborate study were reported at the Pasadena Ether-Drift Conference of February 4 and 5,
1927,14 indicating that there is a constant cosmical motion, of the same general characteristics as were reported at Kansas City, the apex having a right ascension of 17 hours and a declination of +68. It was understood at this time that the process of averaging the observations for the four epochs eliminated the orbital effect, since two of the positions of the earth in its orbit were nearly diametrically opposite the other two, as is shown in Fig. 24. It was announced at the Pasadena Conference that the orbital effect, if it existed, was certainly small and though the search so far had failed to demonstrate its influence, yet the writer was confident that it would be found and that further study and observation would be carried on for this purpose. It may be added that an adequate analysis and calculation of the observations of the four epochs, upon any one set of assumed conditions, requires the time of an expert computer for perhaps a full year. This and other considerations, such as the making of further observations in Cleveland in 1927 and 1929, have delayed the plan for the restudy of the Mount Wilson observations until the autumn of 1932.
As explained in the next section, the new study of the problem, based upon the presumption of a
14 D. C. Miller, Astrophys. J. 68, 341 (1928); Contrib. Mt. Wilson Obs. No. 373, 12 (1928).
solar motion directed to the southward, has given consistent results for both the cosmic motion of the solar system and for the orbital motion of the earth. For this reason the northern apex of the solar motion is rejected in favor of the southern apex.
Southern apex of the solar motion adopted
Beginning in the autumn of 1932, a reanalysis of the ether-drift problem, and a recalculation of the observations made at Mount Wilson in 1925 and 1926 have now been completed. By adopting the alternative possibility that the motion of the solar system is in the cosmic line previously determined but is in the opposite direction, being directed to the apex near the south pole of the ecliptic, a wholly consistent solution has been obtained. This gives for the first time a quantitative determination of the absolute motion of the solar system and a positive detection of the effect of the orbital motion of the earth, by means of the ether-drift interferometer.
The apexes derived from the observations for the four epochs, determined by the right ascensions and declinations given in Tables I and II, are shown on the chart, Fig. 23, which represents the south circumpolar region of the celestial sphere. The observed apexes derived from the azimuth curves are indicated by squares
Fig. 23. Chart of the observed apexes of the resultant absolute motion of the earth.
