1 day 0·027 „ 9 „ 0·0041 „ 3 „ 0·011 „ 11 „ 0·0020 „ 4 „ 0·0095 „ 12 „ 0·0011 „ 6 „ 0·0063 „ 28 „ 0·0004 „ The volume contracted with the time, and was very small after a month’s interval, but the minute bubble of the emanation still retained its luminosity to the last. The tube became deep purple in colour, which rendered readings difficult except with a strong light. There was a sudden decrease in the first day, which may have been due to the mercury sticking in the capillary tube. The experiments were repeated with another capillary tube and the volume of gas observed at normal pressure was 0·0254 c. mm. The gas obtained was found to obey Boyle’s law within the limit of experimental error over a considerable range of pressure. But, unlike in the first experiment, the gas did not contract but expanded rapidly during the first few hours, and then more slowly, finally reaching a volume after 23 days of 0·262 c. mm. or about 10 times the initial volume. The measurements were complicated by the appearance of bubbles of gas in the top of the mercury column. The differences observed in these two experiments are difficult to account for. We shall see, later, that the emanation always produces helium, and, in the first experiment, the decrease of the volume to zero indicates that the helium was buried or absorbed in the walls of the tube. In the second case, probably owing to some difference in the glass of the capillary tube, the helium may have been released. This suggestion is confirmed by the observation that the volume of gas, after the experiment ended, gave a brilliant spectrum of helium. We shall see later that there is considerable evidence that the α particles expelled from radio-active substances consist of helium atoms. Since the particles are projected with great velocity, they will first be buried in the walls of the tube, and then may gradually diffuse out into the gas again under conditions probably depending on the kind of glass employed. Since α particles are projected from the emanation and also from two of the rapidly changing products which arise from it, the volume of helium should, on this view, be three times the initial volume of the emanation. If the helium produced escaped from the walls of the tube into the gas, the apparent volume of the gas in the capillary should increase to three times the initial volume in a month’s interval, for during that time the emanation itself has been transformed into a solid type of matter deposited on the walls of the tube.