3 92
DISCUSSION
As to the average displacement of the azimuth to the west (50°), it seems to be very hard to explain. Fortunately, however, it also changes about this position periodically with sidereal time. Otherwise, one could hardly escape the suspicion that the whole effect might be due to some cause in the laboratory.
Now some words about Piccard’s experiment: I saw the fringes in his laboratory, and they were extremely nice indeed. As a matter of fact, Piccard considers his experiments only preliminary ones, which he hopes still to improve considerably. He worked the first time under very unfavorable conditions, as the night of his first ascent was unusually warm. I might mention, just for general interest, that such observations as Piccard’s are very exhausting. Those of Mr. Miller are, too, of course. Piccard told me that he did not notice any physiological effect in the turning balloon due to centrifugal force; but motions in the vertical direction, nodding the head, for instance, were very painful, because of the effect of Corioli’s force.
Professor A. A. Michelson: There are one or two questions which I should like to ask. Did Mr. Miller put his results together with the intention of finding an orbital effect (effect due to the motion of the earth in its orbit around the sun) ?
Professor D. C. Miller: Certainly. It was for this purpose that the observations were made at four epochs, approximately at intervals of three months; so that the direction of the orbital component of motion changes about 90° from epoch to epoch. The observations for each epoch have been reduced to determine the actual resultant motion for that epoch. The apex of the motion indicated by all the observations is near the pole of the ecliptic, and hence the orbital motion would manifest itself in a change in the position of this apex from epoch to epoch; that is, it would produce a sort of annual aberration in the apex. A comparison of the results for the four epochs fails to show conclusive evidence of this effect. I hope, however, that when several sets of observations for each epoch are available, the effect of the orbital motion may be evident. The positive effect actually obtained corresponds to a relative motion of the earth and ether of about 10 km/sec., with a probable error of I km/sec. It follows that the effect of the orbital motion on the observed resultant velocity must therefore be less than \ km/sec.
CONFERENCE ON MICHELSON-MORLEY EXPERIMENT 393
Michelson: What is the probable error for this | km/sec.?
Miller: This \ km/sec. is itself the probable error in the measurement of the magnitude of the effect, as determined from the calculations. Since no effect has yet been detected which can positively be attributed to the orbital motion one can only say that such an effect, if present, must be less than \ km/sec.
Michelson: Excuse me if I insist on this point. This estimation of the probable error is based on an interpretation of the experiments which does not intend to find the effect of the motion of the earth at all. Can you not find the probable error in discussing the observations from the point of view of finding the orbital motion?
Miller: I have not calculated the error from such a point of view.
Michelson: It would, however, be possible to do so. I should really like to see such calculations carried out.
Had I known earlier of the beautiful and ingenious apparatus of Mr. Kennedy, I probably should not have undertaken my experiments now going on in the same form. In any case, the problem in question must be investigated further. Even a more precise repetition of experiments with older devices already used will be of great value for the reliability of the results. We have now to find out definitely what actually is the truth, without going at it with any prejudice.
I am happy in respect to Mr. Kennedy’s experiment that I had the idea of this device, too. I also had intended to use the photometric comparison of the field produced by light which is reflected from a divided mirror, the two half-surfaces being at a distance of a fraction of a wave-length. But it did not occur to me that the separation could be made so nicely by sputtering. I intended to take the layer off by acids. The apparatus is indeed so beautiful that I should like to work with a similar device, in case Mr. Kennedy has no objections.
As far as Mr. Piccard’s remarks are concerned (see Lorentz) I must say that every beginner thinks himself lucky if he is able to observe a shift of 1/20 of a fringe. It should be mentioned, however, that with some practice shifts of 1/100 can be measured, and that in very favorable cases even a shift of 1/1000 of a fringe may be ob-