Manchester Memoirs, Vol. xlvi. (1902), No. 8. 3 



The pressure required for driving the gas through 

 this apparatus is represented by the difference of height 

 between the working level in DD and X ; this is equal to 

 the resistance in the coil added to the difference of level 

 between X and Y, allowance being made for the large 

 percentage of air bubbles in the coil and downtake Y. 



The internal diameter of the coiled tube should be 

 about >8 inch; if made smaller, the capacity of the 

 apparatus is much reduced and the resistance increased. 

 \i the diameter is made larger than one quarter inch, 

 bubbles are not formed, the fluid is not carried over into 

 DD, and the apparatus is then, in fact, reduced to a 

 Winkler worm of inefficient design. The essential 

 principle of my apparatus is that the fluid must be carried 

 over with the gas. 



Of the several models shewn, one of the more satis- 

 factory ones had a coil 30 inches long and yi inch internal 

 diameter, with a difference of level between X and Y of 

 1)4^ inch. A stream of air was completely broken up 

 into bubbles even when as much as seven litres of air 

 were passed through per hour. Under these conditions 

 the estimated duration of contact of bubbles and fluid 

 was 2^/^ seconds. 



A larger model, having a coil four feet long and 

 Yj^ inch bore, worked best when about 50 litres per hour were 

 passed through it, contact lasting for about 3 seconds. A 

 much larger volume of air, driven by laboratory bellows 

 and charged with hydrochloric vapour, has been passed 

 through this apparatus, the absorption being complete as 

 shewn by the nitrate of silver test, even although the 

 bubbles were very large and the absorbing fluid was very 

 weak. 



