of the two emanations to be made under comparable conditions. The emanation, mixed with a small amount of the gas to be used, was introduced into the cool spiral, which had been exhausted previously by means of a mercury pump. The amount of emanation remaining uncondensed after definite intervals was rapidly removed by means of the pump, and was carried with a constant auxiliary stream of gas into the testing vessel.
Tested in this way, it was found that the volatilization point of the radium emanation was very nearly the same as that obtained by the blowing method, viz. -150° C. With thorium, on the other hand, the condensation started at about -120° C., and, as in the blowing method, continued over a range of about 30° C. The proportion of the emanation condensed at any temperature was found to depend on a variety of conditions, although the point at which condensation commenced, viz. -120° C., was about the same in each case. It depended on the pressure and nature of the gas, on the concentration of the emanation, and on the time for which it was left in the spiral. For a given temperature a greater proportion of the emanation was condensed, the lower the pressure and the longer the time it was left in the spiral. Under the same conditions, the emanation was condensed more rapidly in hydrogen than in oxygen.
168. Thus there is no doubt that the thorium emanation
begins to condense at a temperature higher than that at which
the radium emanation condenses. The explanation of the peculiar
behaviour of the thorium emanation is clear when the small
number of emanation particles present in the gas are taken into
consideration. It has been shown that both emanations give
out only α rays. It is probable that the α particles from the
two emanations are similar in character and produce about the
same number of ions in their passage through the gas. The
number of ions produced by each α particle before its energy
is dissipated is probably about 70,000. (See section 252.)
Now, in the experiment, the electrometer readily measured a current of 10^{-3} electrostatic units. Taking the charge on an ion as 3·4 × 10^{-10} electrostatic units, this corresponds to a production in the testing vessel of about 3 × 10^6 ions per sec., which would be
