770 Prof. F. Horton and Miss. A. C. Davies on 



metastable atoms should result in the reversion of these 

 abnormal atoms to the normal state with the emission of 

 radiation, which would ionize the impurity and act pkoto- 

 electrically on the electrodes, giving a current which it 

 should be possible to detect. Whether it is thought that the 

 admission of impurity to an accumulation of He 2 molecules 

 would split up the molecules into metastable atoms and cause 

 these to revert to the normal condition in the same way is 

 not clear. 



Experiments were made to test if the admission of im- 

 purity to pure helium, which had been bombarded by 

 electrons with energies intermediate to 20*4 volts and 

 21 - 2 volts, caused the production of radiation. For tins 

 purpose, the stream of helium through the apparatus was 

 stopped and a constant pressure of gas was used, but in 

 order to ensure that the purity of the helium was equal to 

 that of the helium used by Franck and Knipping, the 

 experiments were made after the carbon purifying tubes on 

 either side of the apparatus had been immersed in liquid 

 air for several days. The potential difference Y : was then 

 adjusted to 20'9 volts so that 20'4 volts electron collisions 

 with helium atoms were occurring. The electrometer current 

 obtained at this value of Y 1 was 4"17 x 10 -u amp., and 

 the bombardment of the pure helium was continued for 

 20 minutes. At the end of this time, the current was again 

 measured and was found to be identical with that obtained 

 at first. The current heating the lower filament, which was 

 supplying the electrons, was then cut off, and the earthing 

 key of the electrometer was raised, one pair of quadrants 

 being thus insulated. The liquid air was then removed from 

 the purifying U-tube attached to the side tube, so as to allow 

 some impurity to enter the apparatus, and the indications 

 of the electrometer were carefully noted. No measurable 

 current was obtained. In order to be quite sure that by this 

 time impurity had entered the apparatus, the filament was 

 reheated and it was found that the electrometer current was 

 3856 X 10 -u amp. — much larger than before. There is thus 

 no doubt that after the removal of the liquid air, impurities 

 from the carbon had mixed with the helium. The experiment 

 therefore shows, either that no accumulation of abnormal 

 atoms was produced as a result of the continued bombard- 

 ment of pure helium by electrons with energies slightly in 

 excess of 20*4 volts, or that such abnormal atoms were not 

 affected by the presence of impurity in the way supposed by 

 Franck and Knipping. 



