118 THE ROYAL SOCIETY OF CANADA 



The issuing gas was not inflammable, indicating the presence of nitro- 

 gen. 



On December 19, 1917, the observation was repeated with gas 

 from the Landrum Well at Petrolia, Texas. Here, after passing five units 

 of gas into G (see Fig. 3), I was in the act of pumping over the un- 

 condensed gas to the next part of the apparatus, when the pressure 

 having been lowered to 18 cm. suddenly rose to 28 cm. Later on in 

 the same evening I observed a similar sudden elevation of pressure, 

 from 8 cm. to 23 cm. In this case the condensate was a white solid. 



The earliest instance noted, however, of these pressure kicks 

 was on May 9, 1916, when working on the gas from Medicine Hat. 

 The liquid air having been removed, the gauge was observed to halt 

 for some time as the condensate melted and when the pressure had 

 risen to 35 cm., the mercury column in the gauge oscillated rapidly 

 up and down several times, indicating instability of the liquid and 

 gaseous phases within the condenser. 



The next notable instance occurred on June 9, 1916, with Bow 

 Island Gas. Gas had been passed into G (see Fig. 2, BI (1) ) until the 

 final pressure was 34 cm. The following observations were then 

 made: 



Removed L,* momentarily. The deposit was found to be a white 

 solid. 



Replaced L. Pumped off all the gas. Got the pressure down to 

 4, to 2-8 and finally to 2-5 cm. 



Removed L, momentarily. The deposit was drier and whiter 

 than before. 



Replaced L. 



Removed L. The gauge fell. When the pressure was 17 cm. 

 there was an upward kick. Then pressure gradually rose to 19 cm. 

 and another kick occurred. The condensate was all liquid at a pressure 

 of 21 cm. 



Replaced L. The gauge rose and became steady at 10-7 cm. 

 The increase of 10-7 above 6-5 (the vapour pressure of methane) is 

 evidently due to gas which has escaped from the liquid. 



Readings with Medicine Hat Gas, June 13, 1916. After 10 col- 

 lectors had been passed in, the final pressure was 15-6 cm. (see curve 

 in Fig. 1). 



Removed L. White condensate. Icy streaks on walls of G. 

 Replaced L, quickly. 



Exhausted by air pump until pressure fell to 3 cm. 

 * L signifies the liquid-air container. 



