the knoll and four concrete piers were formed to support the iron mercury tank at a suitable height. This was enclosed in a light housing, Fig. 12, twenty feet square and about twelve feet high at the ridge of the roof. The sides of the house were enclosed with sheets of corrugated
Fig. 12. Interferometer house on “Ether Rocks," Mount Wilson.
iron, except that at a height of from four to seven feet above the floor, on all sides there were continuous “windows” of white canvas cloth. The canvas was attached to a series of frames so that the windows could be opened on all sides at the level of the interferometer, for a width of three feet. In the south end was a small door with iron and canvas panels to match the sides of the house. A rough flooring was placed a little above the rock; on this floor a smooth circular track was constructed on which the observer could walk comfortably while following the interferometer as it turns slowly on its axis. This house was purposely constructed with wide cracks at the various joints in the sides and floor and under the eaves, so that there should be a very free circulation of air to secure equalization of temperature with the outside air. The opening of the windows on all sides greatly facilitated this condition. In order to secure sufficient darkness for the observation of the fringes in the daytime, sheets of thin black paper were placed over the canvas windows and over such holes and cracks as admitted too much light. Electric current was supplied to the house and several fixed and portable lamps were available. Common and
precision thermometers were hung on each side of the house and were read at the beginning and end of each set of observations. A barograph and a thermograph were carried at all times on the interferometer itself. An anemometer was attached to the roof of the house. A copy of the Mount Wilson Observatory meteorological records was also secured for the duration of the observations. These general arrangements apply to all subsequent experiments.
Observations were begun on April 8 and continued till April 21, 1921, by means of the apparatus and methods employed by Morley and Miller in 1904 and 1905, with certain modifications and developments in details. The first observations of sixty-seven sets consisting of 350 turns gave a positive effect such as would be produced by a real ether-drift, corresponding to a relative motion of the earth and ether of about ten kilometers per second. Before announcing such a result, it seemed necessary to study every possible cause which might produce a displacement of fringes similar to that caused by ether-drift; among the causes suggested were radiant heat, centrifugal and gyrostatic action, irregular gravitation effects, yielding of the foundation, magnetic polarization and magnetostriction. In order to test the first, the metal parts of the interferometer were completely covered with cork about one inch thick, and fifty sets of observations consisting of 273 turns were made; there was a periodic displacement of the fringes, as in the first observations, showing that radiant heat is not the cause of the observed effect.
Observations of December, 1921. Concrete interferometer
In the summer of 1921, the steel frame of the interferometer was dismounted and a base of one piece of concrete, Fig. 13, reinforced with brass, was cast in place on the mercury float. All the metal parts which were supported on the concrete base were made of aluminum of brass. The entire apparatus was free from magnetic effects and the possible effects due to heat were much reduced. In December, 1921, forty-two sets of observations, consisting of 422 turns, were made with the nonmagnetic interferometer. These show a positive effect as of an ether drift, which is
Fig. 13. Interferometer with base of concrete, 1921.
entirely consistent with the observations of April, 1921.
Many variations of incidental conditions were tried at this epoch. Observations were made with the centering pin tight in its socket and then loose; with rotation of the interferometer clockwise and counterclockwise; with a rapid rotation of one turn in 40 seconds and a slow rotation of one turn in 85 seconds; with a heavy weight added first to the telescope arm of the main frame and then to the lamp arm; with the float extremely out of level because loaded first in one quadrant and then in the next quadrant; with the recording assistant walking round in different quadrants and standing in different portions of the house, near to and far from the apparatus.
The results of the observations were not affected by any of these changes.
It was demonstrated that the use of the concrete base did not change the effect observed with the steel base either in magnitude or azimuth. The concrete base was less affected than the steel by change of dimensions due to changes of temperature; but this slight advantage was counterbalanced by the fact that it accommodated itself more slowly to a change of temperature. In spite of the fact that the concrete was considerably heavier than the steel parts which it displaced, it was much less rigid. Tests showed that a weight of 30 grams placed on the end of the arm of the interferometer would produce a displacement of the fringes of one fringe width, while nearly ten times as much weight is required to produce the same effect with the steel base. The concrete base was abandoned and the original steel base has been used in all subsequent observations.
Laboratory Tests of the Interferometer, Cleveland, 1922-1924
The entire apparatus was returned to the laboratory at Cleveland; during the years 1922 and 1923 many trials were made under various conditions which could be controlled and with many modifications in the details of the apparatus. An arrangement of mirrors and prisms was made so that the source of light could be placed outside of the observing room, Fig. 14, the
Fig. 14. The interferometer in the laboratory, 1923.
