116 THE ROYAL SOCIETY OF CANADA 



a lowering of the pressure proportional to n from that given in the 

 above equation. Also, since the pressure of the gas is mounting nearly- 

 proportional to n, there is extra solubility on this account, and this 

 causes another lowering of the pressure, also roughly proportional to n. 



.*. Final pressure = kn — ln^ 

 which tells us that the curve between pressure and n is a parabola 

 with its axis vertical. The apex is reached when the introduction of 

 more gas does not raise the pressure. This is borne out generally 

 by the shapes of the curves in Figs. 2 and 3. 



If instead of writing 



^ = kn-ln2 



we put the lowering of the pressure (due to the increased solubility 



under the greater pressure) proportional to p, we get 



p = k.n — \np 



whence 



kn 

 p= ■ 



1 + ln 

 which gives a straight line between nip and n. 



The Solubility of the Nitrogen in the Condensate 

 If now the liquid air tube is withdrawn from the condenser G, 

 and the latter allowed to warm up, one would expect the mercury in 

 the gauge to drop gradually down to the level of the mercury in the 

 dish M and then to blow out if the tap O is not turned. This, indeed, 

 often happens, but also very often a curious phenomenon occurs. 

 The mercury drops gradually to a certain point, then halts for a while 

 and then suddenly jumps up several centimetres, keeping again steady 

 for a short time after which it gradually descends. It may do this 

 more than once before the mercury drops to the dish level. Some- 

 times the upward kick is accompanied by rapid oscillations. The 

 kick may extend many centimetres. It should be remembered that 

 an upward kick means decrease of pressure. This phenomenon has 

 been observed over and over again with many gases. 



A possible explanation is that as the temperature rises the methane 

 vapour pressure and nitrogen gas pressure increase, and suddenly some 

 of the nitrogen is forced into solution and leaves the space above. I may 

 say the condensate is sometimes solid when the condenser is in liquid 

 air and in many cases the upward kick in the gauge occurs at the 

 instant the solid melts. 



Suppose now that the tube has warmed up so that the mercury is 

 down low. Now if the liquid air be applied again the contents of G 

 cool and the mercury rises. It rises steadily for some time, then, in 



