aluminium. As the occluded emanation was slowly released or lost its activity, the activity of the metal fell to a limiting value. The absorption of the radium emanation by lead, paraffin, and caoutchouc has been noticed by Curie and Danne (section 182).
The residual activity on the plates comprised both α and β rays, the latter being present, in all cases, in a very unusual proportion. The equality of the activity and the identity of the radiation emitted from each plate show that the residual activity is due to changes of some form of matter deposited on the plates, and that it cannot be ascribed to an action of the intense radiations; for if such were the case, it would be expected that the activity produced on the different plates would vary not only in quantity, but also in quality. This result is confirmed by the observation that the active matter can be removed from a platinum plate by solution in sulphuric acid, and has other distinctive chemical and physical properties.
The variation with time of the residual activity measured by the α rays will first be considered. A platinum plate was exposed in the presence of the radium emanation for seven days. The amount of emanation initially present was equal to that obtained from about 3 milligrams of pure radium bromide. The plate immediately after removal gave a saturation-current, measured between parallel plates by a galvanometer, of 1·5 × 10^{-7} ampere. Some hours after removal, the activity decayed according to an exponential law with the time, falling to half value in 28 minutes. Three days after removal the active plate gave a saturation-current, measured by an electrometer, of 5 × 10^{-13} ampere; i.e. 1/300,000 of the initial activity. The activity was observed to increase steadily with the time. The results are shown in Fig. 93, where the time is reckoned from the middle of the time of exposure to the emanation.
The curve is initially nearly a straight line passing through the origin. The activity increases with the time for the interval of eight months over which the observations have extended. The latter portions of the curve, however, fall below the tangent to the curve drawn through the origin, showing that the activity is not increasing proportionately with the time.
The active deposit, obtained in a different manner, has been
