LONDON. Physical Society.—Meeting held in University College, March 22.—Prof. S. P. Thompson, president, in the chair.—A paper on the expansion of silica was read by Prof. Callendar. The extreme smallness of the thermal expansion of silica (fused quartz) renders the determination of its coefficient of expansion more difficult than is the case with many substances. The author has made experiments upon a rod of pure silica 40 cms. long and 1 mm. diameter. This rod was enclosed in a platinum tube about 3 mms. diameter, which could be raised to various temperatures by the passage of a suitable electric current. Both the rod and the tube were firmly fixed at one end, and the positions of the other ends were accurately observed by a micrometer microscope reading to a thousandth of a mm. The expansion of the tube, in conjunction with a knowledge of its coefficient of expansion, served as a means of determining the temperature of the tube, and, therefore, of the rod. The increase in length, the original length and the range of temperature of the silica being known, the coefficient of expansion can be at once calculated. In some previous experiments the author has investigated the distribution of temperature along a heated platinum rod subjected to cooling at the ends. These experiments prove that the error due to end effect, in the case of the silica rod, can be neglected. The expansion of silica up to 1000° C. is regular, and is about one-seventeenth that of platinum. Between 1000° C. and 1400° C. silica expands more quickly than below 10000 C., and if left at any temperature for a considerable time continues to slowly increase in length. If a curve be plotted having temperatures as abscissae, and increases in length as ordinates, a straight line will represent the expansion of silica up to 1000° C. Above 1000° C. the curve bends upwards, and upon cooling it returns along a path above the ascending curve, so that the final length of the bar is greater than the original length when the lower temperature is reached. The determination of the coefficient of expansion at these high temperatures was made by means of a variable zero, that is by using for the length of the rod that obtained by suddenly cooling from the higher to the lower temperature. At 1400° C. the properties of silica alter and the expansion is replaced by a contraction. On cooling from above 1400° C. to ordinary temperatures there is first an expansion and then a contraction. This property was illustrated by Prof. Callendar, the small changes in length of the rod being magnified by a lever and shown upon a screen by an optical arrangement. The critical point at which contraction occurs on heating has been found by Le Chetalier at about 8oo° C. His experiments were made by a differential methcf, using porcelain as a standard substance. As the expansion of porcelain is uncertain, the author thinks it probable that the effect noticed may be due to irregularities in the expansion of porcelain rather than in that of silica. Mr. Boys expressed his interest in the experiments and asked if the small coefficient of expansion of slate had ever been measured. The small expansion of silica would make it a useful suspension for pendulums on account ot the small compensation necessary. Its perfect elastic properties might be made use of in hair springs for chronometers. Prof. Threlfall said that he had tried to measure the expansion of silica between 0° C. and 70° C. by weighing rods in distilled water, but the method was not accurate. He had made experiments similar in principle to the author's, using temperatures from 0° C. to 100° C. The devitrification of silica is troublesome, and he thought that the rate of devitrification in presence of air increased with the temperature. Dr. Donnan thought that the irregularities in the expansion of silica pointed to a complex composition. Mr. Porter (Eton) asked if the effect ot fused quartz on polarised light had been investigated, and if this effect altered after heating to 1400° C. Mr. Boys said that quartz rods formed by fusion depolarised light. The chairman said that he had noticed the effect spoken of by Mr. Boys, due to strain, but he had been unable to detect any rotatory power. Prof. Callendar, in reply to Mr. Lupton, said that the expansion of quartz crystals was much larger than that of fused silica.—The spectroscopic apparatus of University College was then exhibited by Dr. Bily.—The Society then adjourned until April 26.