This anomalous behaviour was found to be due to the presence of dust particles in the air of the vessel, in which the bodies were made radio-active. These particles of dust, when shut up in the presence of the emanation, become radio-active. When a negatively charged rod is introduced into the vessel, a part of the radio-active dust is concentrated on the rod and its activity is added to the normal activity produced on the wire. After the air in the vessel has been left undisturbed for an interval sufficiently long to allow each of the particles of dust to reach a state of radio-active equilibrium, on the application of an electric field, all the positively charged dust particles will at once be carried to the negative electrode. The activity of the electrode at once commences to decay, since the decay of the activity of the dust particles on the wire quite masks the initial rise of the normal activity produced on the wire. Part of the radio-active dust is also carried to the anode, and the proportion increases with the length of time during which the air has been undisturbed. The greatest amount obtained on the anode was about 60% of that on the cathode. These anomalous effects were found to disappear if the air was made dust-free by passing through a plug of glass wool, or by application for some time of a strong electric field. =182. Decay of excited activity from radium.= The excited activity produced on bodies by exposure to the radium emanation decays much more rapidly than the thorium excited activity. For short times of exposure[275] to the emanation the decay curve is very irregular. This is shown in Fig. 66. It was found that the intensity of the radiation measured by the α rays decreased rapidly for the first 10 minutes after removal, but about 15 minutes after removal reached a value which remained nearly constant for an interval of about 20 minutes. It then decayed to zero, finally following an exponential law, the intensity falling to half value in about 28 minutes. With longer times of exposure, the irregularities in the curve are not so marked. [Illustration: Fig. 66.] Miss Brooks has recently determined the decay curves of the excited activity of radium for different times of exposure, measured by the α rays. The results are shown in Fig. 67, where the initial ordinates represent the activity communicated to the body from different times of exposure to a constant supply of emanation. It will be observed that in all cases there is a sudden initial drop of activity, which becomes less marked with increasing time of exposure. The activity, several hours after removal, decreases exponentially in all cases, falling to half value in about 28 minutes.