SURFACES OF DISCONTINUITY 583 



gave much the same value for the area per molecule of adsorbed 

 benzene on the mercury surface. Experiments were carried 

 out with ethyl alcohol, propyl chloride, and ethyl bromide, 

 showing that, as in the previous cases, the adsorption of these 

 substances appears to be within certain limits a reversible 

 phenomenon. Iredale expresses surprise that these substances, 

 "which are more definitely polar than benzene and, especially 

 in the case of the alkyl halides, possess an atom or group more 

 likely to form a definite finking at the mercury surface, should 

 have no more marked effect on the surface tension than benzene 

 itself." 



SI . The Experiments of Micheli, Oliphant, and Cassel 



Subsequently Micheli at Donnan's suggestion {Phil. Mag., 3, 

 895 (1927)) took up the same problem. He examined the va- 

 pors of benzene, hexane, heptane, pentane and octane, all in a 

 high state of purity, at a water-vapor interface using the drop- 

 weight method. It was found that if <r is plotted against the 

 partial pressure of the vapor (in the vapor-air mixture) the result 

 is a straight line; hence F = kP, where A; is a constant. From a 

 knowledge oi k, Fg the amount adsorbed when the partial pres- 

 sure is equal to the maximum vapor pressure at the temperature 

 of adsorption could be determined. His comment on the re- 

 sults is as follows: "The fact that a linear relationship holds 

 between the partial pressure of the vapor and F right up to 

 the value F„ and also that this quantity changes so rapidly with 

 increasing temperature, indicate clearly that a limiting condi- 

 tion, such as would obtain if a closely-packed adsorbed layer 

 were formed, had not been reached." He also shows from his 

 calculations that the values for the area occupied by one mole- 

 cule of benzene is larger than Adam's value for a closely packed 

 layer of certain benzene derivatives on a water surface. "In 

 this case, clearly an unsaturated layer is formed." His pro- 

 portionality factor decreases as the temperature at which the 

 experiments are performed is raised. This indicates that ad- 

 sorption is accompanied by an evolution of heat, but we shall 

 postpone the discussion of this matter until we reach the com- 

 mentary on the subsection dealing with thermal effects. 



