Observations were first made on the heat emission from 30 milligrams of radium bromide. The difference in pressure observed on the manometer was standardized by placing a small coil of wire of known resistance in the place of the radium. The strength of the current through the wire was adjusted to give the same difference of pressure on the manometer. In this way it was found that the heat emission per gram of radium bromide corresponded to 65 gram-calories per hour. Taking the atomic weight of radium as 225, this is equivalent to a rate of emission of heat from one gram of metallic radium of 110 gram-calories per hour. The emanation from the 30 milligrams of radium bromide was then removed by heating the radium (section 215). By passing the emanation through a small glass tube immersed in liquid air, the emanation was condensed. The tube was sealed off while the emanation was still condensed in the tube. In this way the emanation was concentrated in a small glass tube about 4 cms. long. The heating effects of the “de-emanated” radium and of the emanation tube were then determined at intervals. It was found that, after removal of the emanation, the heating effect of the radium decayed in the course of a few hours to a minimum, corresponding to about 25 per cent. of the original heat emission, and then gradually increased again, reaching its original value after about a month’s interval. The heating effect of the emanation tube was found to increase for the first few hours after separation to a maximum, and then to decay regularly with the time according to an exponential law, falling to half its maximum value in about four days. The actual heat emission of the emanation tube was determined by sending a current through a coil of wire occupying the same length and position as the emanation tube. The variation with time of the heating effect from 30 milligrams of radium and the emanation from it is shown in Fig. 99. [Illustration: Fig. 99.]