82 FLUIDITY AND. PLASTICITY
divergences from it seem to be more clearly marked out in the
case of viscosity than, with other properties, such as refractive
index." Thorpe and Rodger (1897) say, "The observations
described in this paper afford additional evidence of the fact indi-
cated by Wijkander and supported by Linebarger, that the vis-
cosity of a mixture of miscible and chemically indifferent liquids
is rarely, if ever, under all conditions, a linear function of the
composition. It seldom happens that the liquid in a mixture
preserves the particular viscosity it posesses in the unmixed
condition. To judge from the instances heretofore studied, the
viscosity of the mixture is, as a rule, uniformly lower than the
mixture law would indicate, but no simple relation can yet be
traced between the viscosity of a mixture and that of its constit-
uents." Thorpe and Bodger were so struck by the absence of
linearity in the viscosity curves, that they thought that an ex-
planation was needed for the fact that the viscosity curves of
some mixtures measured by Linebarger (1896) are indeed linear.
"The observed viscosities in general are less than, those calculated
by the mixture rule, except, possibly, in the case of mixtures of
benzene and chloroform and mixtures of carbon disulfide with
benzene, toluene, ether, and acetic ether, where, possibly, the
temperature of observation (25°) was too near the boiling-point
of the carbon disulfide to make any specific influence, which that
liquid might exert at lower temperatures, perceptible/'
Lees (1900) showed what are the necessary assumptions in
regard to the nature of flow in mixtures, so that the viscosities
should be additive, but by making a careful study of existing
data, he found little justification for these assumptions. Simi-
larly Lees tried the assumptions that fluidities or logarithmic
viscosities are the characteristic additive property, but he was
unable to obtain a satisfactory verification of either from the
experimental results.
The question before us seems to "be: "Is viscosity or fluidity
or some function of one of them the characteristic additive prop-
erty?" The answer to this question is imperative before we
can intelligently discuss the relation of viscosity to other proper-
ties. This statement requires no proof in view of the statements
which we have quoted to show that in some cases the viscosity
concentration curve is linear according to assumption, but in the