radium emanation present in the large tank, special experiments were made with a smaller tank in which a known quantity of the radium emanation was introduced by employing a solution of pure radium bromide of known concentration. A central wire was made the negative electrode as before, and, after removal, it was wound on the frame and its activity tested. In this way it was found that the amount of radium emanation present in the large tank, in order to produce the excited activity observed, must have been equal to the equilibrium or maximum amount to be obtained from 9·5 × 10^{-9} grams of pure radium bromide. The volume of the large tank was 17 cubic metres, so that the amount of emanation present per cubic metre was equivalent to that liberated from 5·6 × 10^{-10} grams of radium bromide in radio-active equilibrium.
If the amount of the emanation in the tank is taken as the average amount existing in the outside air, the amount of radium emanation present per cubic kilometre of the air is equivalent to that supplied by 0·56 grams of radium bromide.
For the purpose of calculation, suppose the emanation is uniformly distributed over the land portion of the earth (1/4 of the total surface), and to extend to an average height of 5 kilometres. The air over the sea is not taken into account as its radio-activity has not been examined. The total amount of emanation present in the atmosphere under these conditions corresponds to that supplied by about 400 tons of radium bromide. In order to maintain this amount of emanation in the atmosphere, it must be supplied at a constant rate from the earth's surface. Since the greater amount of the emanation probably escapes into the air by transpiration and diffusion through the soil, the emanation cannot reach the surface except from a very thin layer of the earth. The probable thickness of this layer can be estimated if it is assumed that the present loss of heat from the earth is supplied from the radio-active matter contained in it. We have seen (section 271) that, on this hypothesis, there must be an amount of active matter in the earth corresponding to about 300 million tons of radium. If this is supposed to be uniformly distributed, a thickness of layer of about 13 metres will suffice to maintain the calculated amount of emanation in the atmosphere. This thickness of layer is about the order of magnitude to be expected from general considerations.
