1388 REPORTS ON THE STATE OF SCIENCE.—1918. 
indirect, but it indicates an electrical effect independent of the 
chemical nature of the ion. Whether the conclusion is justified that 
it implies the production of a definite salt seems doubtful. The 
difference between the osmotic pressures of equimolar solutions of 
sodium salts with uni-and bivalent anions must not be forgotten. 
A further difficulty is presented by the fact that calcium salts were 
not found to form ionizable compounds, whereas in the experiments 
of Pfeiffer and Modelski (1912), the “compounds” of amino-acids 
with calcium chloride were some of the easiest to prepare. If the 
contrast is due to the colloidal nature of gelatin, it is an additional 
reason to doubt the adequacy of explanation on pure electro-chemical 
lines. Why does Loeb speak of proteins as “so-called colloids” ? 
If a salt of a protein with a strong base or strong acid is exposed 
-to an electric field between electrodes, the colloidal ion is deposited 
at one pole or the other according to whether it is anion or cation 
and the protein is naturally said to have a charge of the sign opposite 
to that of the pole at which it is deposited. But the sign of the 
charge on such a surface as that of paper can be changed by electro- 
lytes, even by weak acids and by neutral salts. In such cases, it 
seems that something other than actual salt formation must be the 
cause. Whatever may be the way in which the electric charge on | 
the surface of insoluble particles or colloids is produced, whether by 
electrolytic dissociation at the surface, one ion being insoluble and 
fixed (see Hardy, 1910), or by other causes more allied to the 
phenomena of static electricity (Lewis, 1909), as appears to be the 
case with drops of petroleum in water, it is clear that the surface 
energy due to this charge would be reduced if ions of the opposite 
sign were deposiled on it. The second law of thermodynamics 
would allow us to predict this fact, which has heen called “ electrical 
adsorption.” It isshown in a striking way by the opposite effect of 
bivalent ions on the adsorption of colloidal dye ions by paper, when 
these have a different sign of charge. Some data on this point will 
be found ina paper by myself (1906). I was inclined to attribute the 
effect to a reversal in sign of the charge on the colloidal particles, but 
it is more satisfactorily explained as exerted on the paper. While, 
however, it is comparatively easy to see why the charge should 
be reduced to zero, it is not quite so easy to see why it should be 
replaced by a charge of the opposite sign. as Perrin, (1904, 1905), 
Mines (1912) and others have shown to be the case. Freundlich 
(1905, p. 245) has suggested that certain ions may be adsorbed in the 
mechanical way to a much greater extent than certain others with an 
opposite charge. The result would be a deposition on the surface in 
amount greater than merely necessary to neutralize the existing 
charge and sufficient to confer a charge of the opposite sign. (See 
especially the experiments of Freundlich and Schucht, 1913, and of 
Ishizaka, 1913.) ; 
We have, then, in addition to the chemical compounds of certain 
amphoteric electrolytes with strong acids and bases, as association of 
electrolytes with the surfaces of colloids in general, a relationship 
which appears to be of the naturejof adsorption. Thisis indicated by 
the law which expresses the proportion of electrolyte held by the 
colloid to that in the external phase. The law is expressed by the 
