Another series of experiments was made, in which an active aluminium disc was placed in an exhausted tube, and exposed to the cathode ray discharge. Under these conditions, a part of the activity of the disc was removed. When the disc was made the anode, the loss of activity was usually 20 to 60 per cent. for half-an-hour’s exposure. If the disc was made the cathode, the loss was much greater, amounting to about 90 per cent. in 10 minutes. Part of the active matter removed from the disc was collected on a second disc placed near it. This second disc on removal lost its activity at a far more rapid rate than the normal. The rate of decay on the first disc was also altered, the activity sometimes even increasing after removal. These results indicate that, in this case, the apparent volatility of the products is reversed. Thorium B is driven off from the disc more readily than thorium A. The rates of decay obtained under different conditions were satisfactorily explained by supposing that the surfaces of the discs after exposure to the discharge were coated with different proportions of thorium A and B. The escape of thorium B from the disc under the influence of the discharge seems rather to be the result of an action similar to the well-known “sputtering” of electrodes than to a direct influence of temperature. The results obtained by von Lerch[304] on the electrolysis of a solution of the active deposit also admit of a similar interpretation. Products were obtained on the electrodes of different rates of decay, losing half their activity in times varying from about 1 hour to 5 hours. This variation is due to the admixture of the two products in different proportions. The evidence, as a whole, thus strongly supports the conclusion that the active deposit from thorium undergoes two successive transformations as follows: (1) A “rayless” change for which λ₁ = 1·75 × 10⁻⁵, _i.e._, in which half the matter is transformed in 11 hours; (2) A second change giving rise to α, β and γ rays, for which λ₂ = 2·08 × 10⁻⁴, _i.e._, in which half the matter is transformed in 55 minutes[305]. It is, at first sight, a somewhat unexpected result that the final rate of decay of the active deposit from thorium gives the rate of change not of the last product itself, but of the preceding product, which does not give rise to rays at all. A similar peculiarity is observed in the decay of the excited activity of actinium, which is discussed in section 212. For a long exposure in the presence of a constant supply of thorium emanation, the equation expressing the variation of activity with time is found from equation (8), section 198, $$ \frac {I_t} {I₀} = \frac {Q} {Q₀} = \frac {λ_2} {λ_2 - λ_1} e^{–λ_1 t} − \frac {λ_1} {λ_1 − λ_2} e^{–λ_2 t} $$ $$ = \frac {λ_2 e^{–λ_1 t}} {λ_2 - λ_1} (1 − \cdot083 e^{−1\cdot90 × 10^{−4} t} ) $$ .