Distillation of Liquid Air. 



519 



In the experiments on the composition of the vapour, the 

 method of working was very simple. The thermometer-bulb 

 was immersed in the liquid air contained in a vacuum- 

 jacketed vessel, the top of which was loosely stoppered with 

 cotton-wool. Samples of the gas evolved were taken from 

 time to time as the air boiled away, by means of a tube 

 passing through the cotton-wool, and the thermometer 

 reading was noted at each time. The samples of gas were 

 afterwards analysed by means of burning phosphorus, and 

 the results plotted as percentages of oxygen, against the 

 corresponding absolute temperatures. Several series were 

 thus obtained, and the ends of the curve, that is to say 

 where the mixture contained great excess of nitrogen or 

 oxygen, were obtained by mixing liquid oxygen or nitrogen, 

 as the case might be, with the liquid air. The curve drawn 

 Fij 













l r> ' 













St 























90 

 88 





















^7 

















^ 



^ 

























6 

 3 86 



Ci 



s 



J 84 





































A- 



























Absolute 



i PS 

























A 



















80 





/* 























/ 



















76 



/ 





^if 

















l^*^ 





















.76 





















10 



£0 



30 



70 



80 



90 



40 50 60 



Oxygen percentage, 

 through the points so obtained was completed by the deter- 

 mination of the boiling-points of nitrogen and oxygen, and it 

 was found that the extrapolated curve met these points abso- 

 lutely. The smoothed curve is shown on fig. 2 (curve A). 



100 



