the vessel. A glass vessel with a removable base was employed and the vessel was lined with the material to be examined. The following table shows the relative results obtained. The amount of leakage observed is expressed in terms of the number of scale divisions of the eye-piece passed over per hour by the gold-leaf:
| Material of lining of vessel | Leakage in scale divisions per hour |
|---|---|
| Tinfoil | 3·3 |
| „ another sample | 2·3 |
| Glass coated with phosphoric acid | 1·3 |
| Silver chemically deposited on glass | 1·6 |
| Zinc | 1·2 |
| Lead | 2·2 |
| Copper (clean) | 2·3 |
| „ (oxidized) | 1·7 |
| Platinum (various samples) | 2·0, 2·9, 3·9 |
| Aluminium | 1·4 |
There are thus marked differences in the leakage observed for different materials and also considerable differences in different samples of the same metal. For example, one specimen of platinum caused nearly twice the leakage of another sample from a different stock.
McLennan and Burton, on the other hand, measured by means of a sensitive electrometer the ionization current produced in the air in a closed iron cylinder 25 cms. in diameter and 130 cms. in length, in which an insulated central electrode was placed. The open cylinder was first exposed for some time at the open window of the laboratory. It was then removed, the top and bottom closed, and the saturation current through the gas determined as soon as possible. In all cases it was observed that the current diminished for two or three hours to a minimum and then very slowly increased again. In one experiment, for example, the initial current observed corresponded to 30 on an arbitrary scale. In the course of four hours the current fell to a minimum of 6·6, and 44 hours later had risen to a practical maximum of 24. The initial decrease observed is probably due to a radio-activity of the enclosed air or walls of the vessel, which decayed rapidly with the time. The decay of the excited activity produced on the interior surface of the cylinder when exposed to the air was probably responsible for a part of the decrease observed. McLennan ascribes the increase of current with time to a radio-active emanation which is given off from the cylinder, and ionizes the enclosed air. On placing linings of lead, tin, and zinc in the iron cylinder, considerable differences were observed both for the minimum current and also for the final maximum. Lead gave about twice the current due to zinc, while tin gave an intermediate value. These results are similar in character to those obtained by Strutt.
McLennan and Burton also investigated the effect of diminution of pressure on the current. The cylinder was filled with air to a pressure of 7 atmospheres, and allowed to stand until the current reached a constant value. The air was then allowed to escape and the pressure reduced to 44 mms. of mercury. The current was found to vary approximately as the pressure over the whole range. These results are not in agreement with the results of Patterson already described, nor with some later experiments of Strutt. McLennan’s results however point to the conclusion that the ionization was mainly due to an emanation emitted from the metal.