the radio-element. The term "conservation of radio-activity" is thus a convenient expression of the facts known at the present time. It is quite possible, however, that further experiments at very high or very low temperatures may show that the radio-activity does vary.
Although no difference has been observed in the radio-activity of uranium over an interval of five years, it has been shown (section 261) that on theoretical grounds the radio-activity of a given quantity of a radio-element should decrease with the time. The change will, however, be so slow in uranium, that probably millions of years must elapse before a measurable change can take place, while the total radio-activity of a given quantity of matter left to itself should thus decrease, but it ought to be constant for a constant mass of the radio-element. It has already been pointed out (section 238) that the activity of radium, measured by the α and β rays, will probably increase for several hundred years after its separation. This is due to the appearance of fresh products in the radium. Ultimately, however, the activity must decrease according to an exponential law with the time, falling to half value in about 1300 years.
The conservation of radio-activity applies not only to the radiations taken as a whole, but also to each specific type of radiation. If the emanation is removed from a radium compound, the amount of β radiation of the radium at once commences to decrease, but this is compensated by the appearance of β rays in the radiations from the vessel in which the separated emanation is stored. At any time the sum total of the β radiations from the radium and the emanation vessel is always the same as that from the radium compound before the emanation was removed.
Similar results have also been found to hold for the γ rays. This was tested by the writer in the following way. The emanation from some solid radium bromide was released by heat, and condensed in a small glass tube which was then sealed off. The radium so treated, and the emanation tube, were placed together under an electroscope, with a screen of lead 1 cm. thick interposed in order to let through only the γ rays. The experiments were continued over three weeks, but the sum total of the γ rays from the radium and the emanation tube was, over the
