1015] on Problems of Hydrogen and the Rare Gases 551 



Air and air nitrogen were the materials examined, as stated in the 

 lirst column. As oxygen is absent in the second case, some idea of 

 the effect of its presence can thus l)e obtained. The second column 

 gives the partial pressure of the residue of rare gases uncondensed at 

 2i»° Abs. The next column shows the reduced pressure in the 

 apparatus when the exhaust pump was connected, and continued in 

 action for over 10 minutes, the uncondensable residue being thereby 

 removed down to the limit which the pump can maintain under the 

 circumstances. The exhaust was then closed off, and the condensed 

 material allowed to evaporate off into the apparatus, by lowering the 

 liquid hydrogen vessel away from the bulb H until everything was at 

 room temperature. The bulb was then again cooled by cautiously 

 and slowly raising the vessel of liquid hydrogen to its former 

 position ; the steady pressure then obtained is recorded in the last 

 column. This value is seen to be higher than that left by the pump 

 exhaust. By repeating the operation a state can be reached when 

 only a very small increase is registered, as shown when the air 

 nitrogen was examined. 



This means that even at the low pressure maintained by the pump 

 some helium or neon or hydrogen has been held bound, or dissolved 

 in the solid condensed nitrogen and oxygen ; so that the value 

 obtained by simple condensation to 20^ Abs. is low by at least this 

 amount, though no doubt the amount of occlusion at the higher 

 initial observed pressure of the uncondensable is even greater than 

 that at the lower pressure of the pump ; the nature of this occluded 

 material remains to be seen. Having regard to the relative volatility 

 of hydrogen, neon and helium, one might expect that hydrogen 

 would most likely be present in any gas' remaining occluded under 

 high exhaust in solid nitrogen and oxygen at 20° Abs. Experiment, 

 however, seems to suggest that the very opposite is the case. To 

 examine this question"^ mixtures of hydrogen with nitrogen and 

 carbonic oxide respectively were employed ; the table on the following 

 page gives the results. 



" The first column shows the gases used in the experiments. In 

 the second column is given the partial pressure of each constituent 

 after admission to bulb H and the McLeod gauge of Plate 1. 

 Following this is recorded the actual pressure registered on cooling 

 the bulb H in liquid hydrogen ; and in the next column the 

 apparent occlusion before "^exhausting, being the difference between 

 the partial pressure of the hydrogen in the case of the mixtures shown 

 in column 2 and the reading of column 3, which is the pressure 

 observed. At this point, following the former procedure, the 

 molecular pump exhaust was connected for 15 minutes ; then, as 

 before, the condensed material remaining was evaporated off into the 

 gauge by removing the liquid hydrogen. But before re-cooling to 

 determine any uncondensable residue due to occlusion at the high 

 exhaust, which would be a completion of the previous experiment, a 



