an electrometer. An insulating key was arranged so that any one

of the electrodes E, F, H could be rapidly connected with one pair of quadrants of the electrometer, the other two being always connected with earth. The current observed in the testing cylinder vessel was due entirely to the ions produced by the emanation carried into the vessel by the current of air. On substituting a uranium compound for the thorium, not the slightest current was observed. After a constant flow has passed for about 10 minutes, the current due to the emanation reaches a constant value.

The variation of the ionization current with the voltage is similar to that observed for the gas ionized by the radiations from the active bodies. The current at first increases with the voltage, but finally reaches a saturation value.

141. Duration of the activity of the emanation. The emanation rapidly loses its activity with time. This is very readily shown with the apparatus of Fig. 51. The current is found to diminish progressively along the cylinder, and the variation from electrode to electrode depends on the velocity of the flow of air.

If the velocity of the air current is known, the decay of activity of the emanation with time can be deduced. If the flow of air is stopped, and the openings of the cylinder closed, the current steadily diminishes with time. The following numbers illustrate the variation with time of the saturation current, due to the emanation in a closed vessel. The observations were taken successively, and as rapidly as possible after the current of air was stopped.

Curve A, Fig. 52, shows the relation existing between the current through the gas and the time. The current just before the flow of air was stopped is taken as unity. The current through