382 
Faraday Society, February 19.—Sir Robert Robert- 
son, president, in the chair.—A. W. Porter and J. J. 
Hedges: The law of distribution of particles in 
colloidal suspensions with special reference to Perrin’s 
investigations. Pt. ii. The behaviour of particles 
specifically lighter than the medium has been ex- 
amined in regard to distribution with height, using 
for the purpose emulsions of paraffin in water. The 
change of concentration occurs only at the bottom 
of the containing vessel. There is an increase of 
concentration with height reckoned from the bottom. 
A type of curve is suggested which fits closely the 
experimental results.—D. B. Macleod: On a relation 
between surface tension and density. The empirical 
relation y/(p;—p,)*=C, where y is the surface tension 
at any temperature, p,; and p, the densities of the 
liquid and the vapour at the same temperature and 
C is a constant for each liquid, fits’ the experimental 
figures with remarkable accuracy for temperatures 
ranging from the melting-point to the critical tempera- 
ture—D. B. Macleod: (1) On a relation between 
the viscosity of a liquid and its coefficient of expansion. 
If % be the volume of the free space in I c.cm. of 
a liquid at o° C. and 1-4) the volume occupied by 
the molecules, it is assumed that at a temperature 
t° C. the volume of the free space is %, +at+f+7@ 
—the volume of the molecules remaining constant. 
The viscosity of liquids is expressed as a function 
of the free space, thus 7.%;=C. For normal liquids 
A is nearly unity. For associated liquids it has a 
higher value. The values obtained for the free space 
for various liquids at their boiling-points are practically 
constant and of the order required by Van der Waal’s 
theory. An expression is given for the viscosity 
of liquids at different temperatures and pressures. 
(2) On the viscosity of liquid mixtures showing 
maxima. The viscosity of liquid mixtures is a function 
of the free space of the constituents and of the mixture. 
In the case of liquid mixtures showing a maximum, 
the increase of viscosity is due mainly to the increase 
of density, which in turn is due to the chemical 
affinity between the constituents. It is probable 
that complexes which are formed further reduce the 
free space and consequently increase the viscosity.— 
F. H. Jeffery : Electrolysis with an aluminium anode, 
the anolyte being (1) solutions of sodium nitrite, 
(2) solutions of potassium oxalate. With solutions 
of sodium nitrite probably the primary product of 
reaction at the anode is aluminium nitrite which is 
hydrolysed rapidly to hydrated aluminium oxide 
and nitrous acid, the latter giving rise to nitric oxide 
and nitric acid. With solutions of potassium oxalate 
the product of reaction at the anode is a com- 
plex anion derived from aluminium. The salt 
K,{Al(C,O,),} . 3H,O can be derived from the anolytes 
after electrolysis. It is probable that the salt is a 
true complex salt comparable with potassium chromi- 
oxalate, and if this be true, the alumini-oxalate 
complex can be represented in three dimensions 
just as Werner represented the chromioxalate. The 
isolation of a complex salt from an anolyte dots 
not imply necessarily that the constitution of the 
anionic part of this salt is identical with that of the 
complex anion present in the anolyte after electrolysis. 
—Maurice Cook: Crystal growth in cadmium. Evi- 
dence has been obtained that unworked crystals can 
grow under certain conditions. The usual methods of 
preparing metallic specimens for microscopic examina- 
tion are useless, since the specimen cannot be regarded 
as unworked after it has been sawn off the original, 
ground, and polished. In these experiments the 
metal is cooled in such a way as to be free from the 
stresses usually set up during solidification. The 
results obtained indicate that during annealing 
considerable crystal growth has taken place. Irregu- 
NO. 2785, VOL. 111] 
NATURE 


[Marcr 17, 1923 
larity in the shape of the grains is probably a factor 
greatly facilitating crystal growth.—S. D. Muzaffar: 
Electric potential of antimony-lead alloys. Measure- 
ments of the electric potential of the antimony-lead 
alloys were made by means of a quadrant electro- 
meter against a calomel electrode in N KOH, 
N Pb(NO,),, and tartar emetic with tartaric acid 
solutions. The results reveal an identity of potential 
up to 98 per cent. antimony with that of lead, which 
show the formation of no solid solution and no 
chemical compound between the two metals. 
Royal Microscopical Society, February 21.—Prof. 
F. J. Cheshire, president, in the chair—Sir W. M. 
Bayliss: Colloids and staining. The histologist is 
concerned with the staining of particles, large or 
small, sometimes present in the living cell, some- 
times formed by fixing agents. The process is a 
complex one; but, as would be expected from the 
heterogeneous nature of the systems concerned, 
adsorption is the chief factor, especially in its electrical 
aspect ; chemical combination seems to be of less 
importance. Thus, surfaces with a positive charge 
take negative (‘‘ acidic ’’) dyes, those with negative 
charge take ‘‘ basic ’’ or positive dyes. The degree 
depends on the magnitude of the charge, as shown 
by the effect of electrolytes, alcohol, heat, isoelectric 
point, etc. The removal of the amino groups from 
proteins has no effect on the process. Adsorption 
can be distinguished from chemical combination in 
certain cases, such as silk dyed with the acid of 
Congo-red. The fixation of stains by heat is difficult 
to explain. The action of mordants is also obscure ; 
chemical combination as “‘ lakes”’ is only a partial 
explanation, since these are stated to be resistant 
to acids. Differentiation appears usually to be a 
process of colloidal dispersion of the “lake.” In a 
few cases, as the staining of fat by Soudan III., parti- 
tion in accordance with solubility is the main factor. 
A. Mallock: The resolving power and definition of 
optical instruments. Resolving power is taken as 
indicating the least distance (angular or linear) at 
which two points can be seen as separate in the 
field of the instrument; definition is the ratio of 
that area of the field over which the resolving power 
is maintained to the whole area, or, shortly, the 
dimensions of the least objects appreciable and the 
range over which the appreciability extends. Optical 
images are formed when and where a number of paths 
from one point to another have the same optical 
length, in which case either point may be considered 
as the image of the other. By optical length is 
meant the length measured in wave-lengths in the 
medium through which the path proceeds. The 
constancy of this length causes all the waves emanat- 
ing from one of the points to arrive at the other im 
the same phase, and this condition may be used to 
determine the form of the reflecting or refracting 
surfaces required to make one point the image of 
another. Resolving power depends on the rapidity 
with which the length of the optical path varies as 
the distance from the geometrical focus is increased : 
the more rapid the variation the greater is the con- 
centration of the light and the smaller the luminous 
area which forms the image of a point. For telescopes 
where the angular aperture of the lens is small the 
variation is proportional to the diameter of the object 
glass, and a perfect lens one inch in diameter shoul 
have a resolving power of 4in. of arc. For microscopes 
where the angular aperture of the lens is-large the 
least appreciable distance is about \/2 or 1/100,000 in. 
with ordinary light. Test plates for microscope 
objectives consist of groups of fine lines ruled on 
films of anilin colour, the thickness of which is only 
a small fraction of a wave-length of light. 

