570 
CHEMISTRY: W. D. HARKINS 
The data obtained in this laboratory indicate that the adhesional 
work between a mercury surface and the surface of an organic liquid is 
always greater, at least for all substances investigated, than that between 
the same organic liquid and water, and also greater than that between 
the organic liquid and itself. This last might be more properly de- 
scribed by the term cohesional surface work. A second point of interest 
is that for about half the organic substances investigated, the difference 
between the adhesional work against mercury and that against water, 
is nearly constant, with a value between 80 and 92 ergs. This is true 
for such Hquids as the paraffin hydrocarbons, benzene, toluene, the 
xylenes, carbon tetrachloride, chloroform, and nitrobenzene, so the work 
is by no means entirely specific. On the other hand the adhesional work 
toward water is specifically high in the case of the alcohols, water and 
ether, while that toward mercury is very high in the case of the com- 
pounds of sulphur, iodine, and bromine, and also oleic acid, which 
indicates that the specific effects are very marked. 
At the interface between organic compounds and water, groups which 
contain oxygen or nitrogen, double or triple bonds, orient toward the 
water; at a mercury surface the sulphur, bromine, or iodine, should turn 
toward the mercury. 
In order to illustrate these relations a few of the data obtained in the 
laboratory are presented in table 1 . The experimental work on mercury 
was done under my direction by Dr. E. H. Grafton and Warren W. Ewing, 
while that on water was selected from an extensive set of data obtained 
by Drs. F. E. Brown, G. L. Clark, E. C. H. Davies, Mr. L. E. Roberts, 
and Mr. Y. C. Cheng. 
The well known rule of Antonow^ states that the interfacial tension 
between two liquids is equal to the surface tension of the liquid which 
has the higher surface tension minus that which has the lower surface 
tension. Where this rule does not hold for an interface organic liquid- 
water, the discrepancy is explained by Antonow as due to the fact that 
the interface is between two solutions and not two pure liquids, so the 
surface tensions used should be those of the solutions or phases, instead 
of the pure liquids. The Antonow rule sprang from the idea that an 
interface at the two individual surfaces still exists, but that one acts in a 
negative sense on the other. This rule is in direct discordance with the 
theory developed independently by Langmuir and by Harkins, that the 
molecules in surfaces and interfaces are oriented. This theory need not 
be discussed here, since the above table clearly proves the Antonow rule 
to be incorrect for immiscible liquids. Thus his rule if it were applica- 
