562 RICE 



ABT. L 



and Davies (J. Am. Chem. Soc, 49, 2230, (1927)). Brief 

 accounts of it will be found in the books by Adam and Rideal. 

 The substances examined were aqueous solutions of p-toluidine, 

 of amyl alcohol and of camphor. The method used for deter- 

 mining r was the bubble method much improved as to accuracy 

 over previous investigations, an accuracy of a few per cent 

 being claimed. If this is so, there is no doubt that these experi- 

 ments have left the whole matter in some doubt. Hitherto, 

 it had been regarded as very satisfactory that an agreement 

 in order of magnitude between calculated and observed values 

 had been reached, in view of the manifest difficulty of the 

 measurement of the adsorbed amounts. If the claim to high 

 accuracy made by McBain and his co-workers is justified, this 

 state of satisfaction is hardly possible any longer. The general 

 idea of the method is that bubbles of very pure nitrogen satu- 

 rated with the vapors of the solution are passed up a long 

 inclined tube of large diameter containing the solution. The 

 slope of the tube is adjusted so that the time occupied by the 

 bubbles in passing to the top end of the tube is amply sufficient 

 to insure that the surface of each bubble has attained the full 

 adsorption concentration corresponding to the bulk concentra- 

 tion of the solution, the tube being so large that the adsorption 

 does not appreciably lower this bulk concentration. At the 

 top of the inchne the bubbles rise into a vertical tube so narrow 

 that each bubble fills its diameter. Each bubble in the vertical 

 tube rapidly overtakes its predecessors and draining is so rapid 

 that within a few inches there is a continuous column of cylindri- 

 cal bubbles in contact with one another. At the height at 

 which draining is found to be sufficiently complete the narrow 

 tube is curved over and down. The films break in the down- 

 ward portion of the tube and collapse to a liquid which is caught 

 and analyzed. For a full account of the very stringent pre- 

 cautions taken to insure accuracy the literature should be 

 consulted. It must be admitted that little was left undone in 

 that direction. Perhaps the only possible source of trouble 

 has been indicated by Harkins (Colloid Symposium Mono- 

 graph, 6, 36, (1928)). As bubbles pass along the tube, they 

 oscillate in shape; this involves an oscillation in the extent of the 



