The effect of temperature on gravitative attraction

Abstract

Previous work by one of us with a torsion balance of the Boys-Cavendish type had appeared to indicate that the so-called Newtonian Constant of Gravitation increases slightly as the attracting bodies are heated. The attempt has now been made to carry the work to a higher order of accuracy with a view to confirming or disproving that result. The clearing up of this point has become of greater importance on account of the recent development in the theories of gravitation. The same delicate torsion balance was used as in the former experiments. This consisted of a fine quartz fibre carrying a light beam, from which two silver balls were suspended by fibres of unequal length. It was contained in a vacuum tube, and the observations were made by means of a beam of light proceeding from an illuminated scale through a window in the vacuum tube, and reflected by a small plane mirror attached to the torsion fibre. Two lead spheres, each of 100 lbs. weight, were placed near the silver spheres, so that by their attraction first to one side and then to the other, a torsional deflection might be obtained from which to calculate G. These deflections were not determined directly on account of the length of time required for the torsion system to come to rest, but were calculated from observations of successive turning points of the image of the scale.