102 FLUIDITY AND PLASTICITY

the mixture has a critical-solution temperature. The nearest
approximation is in the case of isobutyric acid, <p2o° = 76.0, and
water, <p2o° = 99.8, examined by Friedlandcr. As can be seen
from Fig. 39 taken from the work of Drapier, the irregularity is
very slight. The calculated deviation is

(36)

a-ni + orjz
which for a 50 per cent mixture corresponds to 27.6. The irregu-
larity is greatest in the case of hexane, <p2o° = 314.0, and nitro-
benzene, <?2o° = 50.1, where the fluidities are also the most
unequal. The calculated deviation is in this case 95.6 for a
50 per cent mixture. The deviations actually read from the
curves for isobutyric acid and water and hexane and nitro-
benzene are of the order of 11 and 25 respectively. That these
numbers are so much smaller than the calculated values may be
easily accounted for on the supposition that the drops are not all
sufficiently large to fill the cross-section of the capillary, and
hence the viscosities are not strictly additive.
SUSPENSIONS
According to the view that viscosities are always additive,
the viscosity of all suspensions should be infinite. On the con-
trary, as already stated, Friedlander found that colophonium
suspended in water had practically no influence on the viscosity
of water. Similarly Bose measured the viscosity of suspensions
of finely-divided quartz, whose viscosity may be taken as infinite,
in bromoform and water, and he found that the viscosity of the
medium was but little changed. Had they measured the visco-
sities at increasing concentrations of the solid, they would have
undoubtedly found that the viscosity was altered and in a
perfectly definite manner. As already indicated, page 55 and
Fig. 19, the fluidity curves of such suspensions are normally
linear. Since, however, suspensions and emulsions are closely
allied, it is important to inquire why viscosities are not additive
in suspensions as well as in the emulsions already considered.
In suspensions, we have practically the same conditions as in
emulsions in which the drops are so small that they do not nearly
fill a cross-section of the capillary tube. In this case the viscosi-
ties are not strictly additive.