linear relation as required by the theory of Hicks; this relation is shown in Fig. 30.
The Entrained Ether Hypothesis
In order to account for the results here presented, it seems necessary to accept the reality of a modified Lorentz-FitzGerald contraction, or to postulate a viscous or dragged ether. In commenting upon the preliminary report of this work presented to the National Academy of Sciences in April, 1925, Dr. L. Silberstein said: “From the point of view of an ether theory, this set of results, as well as all others previously discovered, is easily explicable by means of the Stokes ether concept, as modified by Planck and Lorentz, and discussed by the writer (Silberstein) in the Philosophical Magazine.”16 The theory of Stokes may be described by means of the following sentences selected from Sir Joseph Larmor’s treatise on Aether and Matter, pages 10, 13, 35 and 36:
As Sir George Stokes was not disposed to admit that the aether could pass freely through the interstices of material bodies in the manner required by Fresnel’s views, and as any other theory of its motion which could be consistent with the fact of astronomical aberration required irrotational flow, an explanation of the limitation to that flow had, he considered, to be found. This chain of argument, that motion of bodies disturbs the aether, that aberration requires the disturbance to be differently irrotational, that this can only be explained by the dispersion of incipient rotational disturbance by transverse waves, and further that radiation itself involves transverse undulation, he regards as mutually consistent and self supporting, and therefore, as forming distinct evidence in favor of this view of the constitution of the aether. . . . The question then arises how far this explanation will extend to the case in which the aether is entrained by the matter that is moving through it.
There are systematic differences in the so-called constant of aberration and in standard star places as determined at different observatories, which might be explained on the hypothesis of a variation in ether drift due to differences in the local coefficient of drag. The drag at any given station may depend more or less upon altitude, local contour and the distribution of large masses of land such as mountain ranges. The ether-drift experiments have never been
16 L. Silberstein, Phil. Mag. [6] 39, 161 (1920).
made at sea-level, nor, in fact, at any place except Mount Wilson, with sufficient completeness to give accurate measures of the effects. The evidence now indicates that the drift at Mount Wilson does not differ greatly in magnitude from that at Cleveland and that at sea-level it would probably have about the same value.
The reduction of the indicated velocity of two hundred or more kilometers per second to the observed value of ten kilometers per second may be explained on the theory of the Lorentz-FitzGerald contraction without assuming a drag of the ether. This contraction may or may not depend upon the physical properties of the solid and it may or may not be exactly proportional to the square of the relative velocities of the earth and the ether. A very slight departure of the contraction from the amount calculated by Lorentz would account for the observed effect. Sir Oliver Lodge in his autobiography says: “I still cling to the idea that the FitzGerald contraction is a reality which must be taken into consideration in any physical contemplation of the universe.”17
One is compelled, therefore, to consider whether there are possible readjustments of the theories of the ether that will account for the reduction in the observed velocities of absolute motion and for the displaced azimuths. The difficulties presented by these anomalies are certainly not greater than those existing in many other fields of experimental research.
Other Recent Ether-Drift Experiments
Since the announcement of the evidence of absolute motion of the solar system made at Kansas City in 1925, several other experimenters have performed ether-drift experiments with interferometers of various designs and under various conditions, leading to results which are generally considered to be at variance with the conclusions of this paper. Brief reference to these experiments will be made but without extended analysis.
Dr. Roy J. Kennedy, at Pasadena, used an interferometer with an optical device of original design, giving great sensitivity.18 The length of
17 O. J. Lodge, Past Years, 206 (1932).
18 R. J. Kennedy, Proc. Nat. Acad. Sci. 12, 621 (1926); Astrophys. J. 68, 367 (1928).
the light path, to the end mirror, represented by D in the formula previously given, was 200 centimeters. The apparatus was in a sealed metal case filled with helium. The conclusion was that any indicated ether-drift must be less than 2.5 kilometers per second; this limiting value was later reduced by Illingworth to 1 kilometer per second.
Professor A. Piccard and E. Stahel, of Brussels, thinking that the height above the earths surface might influence the ether-drift effect, placed an interferometer in a balloon which ascended to an altitude of 2500 meters.19 The balloon was rotated about a vertical axis by means of a propeller. The interferometer had a light path in which D was 280 centimeters; it had a self-recording device and a thermostatic control; it was enclosed in a metal case which was evacuated. The indicated velocity of ether-drift might have been as large as 7 kilometers per second, which was the limit of precision. This interferometer was later taken to the summit of the Rigi in Switzerland, at an altitude of 1800 meters, where the observations showed an upper limit to the possible ether-drift of 1.5 kilometers per second.20
The late Professor Michelson, together with F. G. Pease and F. Pearson, used an interferometer mounted in the laboratory of the Mount Wilson Observatory in Pasadena, having a light path, D, equal to 1616 centimeters, which was later increased to 2592 centimeters. The readings were made in the vertical axis of the interferometer, the observer being located in the room above the apparatus. “The results gave no displacement as great as one-fiftieth of that to be expected on the supposition of an effect due to a motion of the solar system of three hundred kilometers per second.”21
Professor Georg Joos, working at Jena, used an interferometer mounted on a quartz base suspended in an evacuated metal housing and provided with photographic registration. The interferometer had a light path, D, equal to 2099
19 A. Piccard and E. Stahel, Comptes Rendus 183, 420
(1926); Naturwiss. 14, 935 (1926).
20 A. Piccard and E. Stahel, Comptes Rendus 185, 1198
(1927); Naturwiss. 16, 25 (1928).
21 A. A. Michelson, F. G. Pease and F. Pearson, Nature
123, 88 (1929); J. Opt. Soc. Am. 18, 181 (1929).
centimeters. The results indicated that any existing ether drift could not exceed 1 kilometer per second.22
In three of the four experiments, the interferometers have been enclosed in heavy, sealed metal housings and also have been located in basement rooms in the interior of heavy buildings and below the level of the ground; in the experiment of Piccard and Stahel, a metal vacuum chamber alone was used and in the experiment of Michelson, Pease and Pearson, the interferometer was in the constant temperature vault but did not have a vacuum case. If the question of an entrained ether is involved in the investigation, it would seem that such massive and opaque shielding is not justifiable. The experiment is designed to detect a very minute effect on the velocity of light, to be impressed upon the light through the ether itself, and it would seem to be essential that there should be the least possible obstruction between the free ether and the light path in the interferometer. It is planned to make a direct study of this factor of the problem.
In none of these other experiments have the observations been of such extent and of such continuity as to determine the exact nature of the diurnal and seasonal variations.
While the interferometer used by Kennedy is more sensitive than that of ordinary type, it is doubtful whether the precision of the result equals that obtained from the very large number of readings made under all conditions of temperature and season with the interferometer of the usual type which is much less sensitive to disturbing causes.
The limitations of the direct-reading method have been recognized but it has been adopted because of its simplicity and because it permits the accumulation of a large number of readings in the shortest time. It is believed that any lack of precision in making a single reading is fully compensated by the large number of readings and by the use of an interferometer of longer light-path and therefore of greater initial sensitivity. The interferometer used in the experiments here reported has a light-path, D, equal to 3203 centimeters.
22 G. Joos, Ann. d. Physik [5] 7, 385 (1930).
