Ielworthy] hot springs OF BANFF, ALTA. 33 



softening point of the glass. The unabsorbed gas continued its 

 passage over heated copper oxide, soda lime and phosphorus pent- 

 oxide, and was then pumped back into the burette by means of a 

 Sprengel pump. The residue from several hundred cubic centi- 

 meters was collected in this way and continually repassed over the 

 calcium till no further reduction in volume was observed. 



Finally the gas was allowed to enter the Plucker tube included 

 in the system and the spectrum of the gas was observed. 



As a result of several experiments 1-25 % of unabsorbable gas 

 was found to be present in the sample. The gas was argon together 

 with a trace of helium.. Having no liquid air it was impossible to 

 isolate helium or neon. 



The origin of the gases is probably dissolved air in the water, 

 which percolating through the ground eventually rises through thous- 

 ands of feet of limestone to form the springs. In its passage under- 

 ground the oxygen in the w^ater is used up in oxidizing sulphides while 

 the inactive nitrogen remains. This is borne out by the argon con- 

 tent. Argon is more soluble than nitrogen and therefore increased 

 proportion in the dissolved gases of rain water would be expected. 

 100 c.c. of atmospheric nitrogen contains 1-18% of argon,^ while 

 100 c.c. of nitrogen from the Banff gases contains about 1 -28%. 



lA. M. Kellas Proc. Roy. Soc. Lond. 59-66. 

 Th. Schloesing Comp. Rend. 121-605. 



Sec. III. Sig. 3 



