184 
POPULAR SCIENCE REVIEW. 
invalidates the conclusion that the direction of flow is independent of 
chemical composition and molecular structure of the liquid. It would 
appear, however, that this direction depends ‘ much more frequently upon 
the direction of the electric current than upon the internal architecture of 
the liquid.’ » 
A paper complementary to this, by the same distinguished physicist, con- 
siders the Electric currents caused by liquid diffusion and Osmose. It states 
that other observers have already obtained electric currents by the contact 
of two liquids — Nobili the first. His arrangements consisted of a series of 
four glass cups, each containing an electrolyte, the liquids in the two ter- 
minal vessels being similar, with platinum electrodes ; a second and third 
hind of liquid being in the two intermediate cups. The solutions were con- 
nected by liquids similar to those in the cups contained in glass syphons, 
with turned-up capillary ends. Fechner, Wild, L. Schmidt, and Wullrer, all 
attacked the same problem in various ways. To ascertain whether difference 
of facility in diffusion, caused by the action of gravity upon two portions of 
of solution of different degrees of concentration and of specific gravity would 
produce a current, twenty-five small glass tumblers were taken, each alter- 
nate one containing liquid of a different degree of concentration. Pctassic 
nitrate was the salt employed. Platinum wire electrodes completed the 
circuit. The liquids in the vessels were connected by means of inverted bent 
tubes half an inch in diameter, alternately filled with the two solutions. 
Where the liquids came in contact at the end of each of the tubes a wet 
septum of parchment paper was secured. On interposing a galvanometer a 
feeble current was obtained, not due to differences in the electrodes. Its 
direction was downwards, through the septa and surfaces of contact. With 
sulphate of copper and sulphuric acid, a stronger current of similar direction 
was obtained. Four other experiments, somewhat varied in their conditions, 
produced a much stronger current, in a reverse direction, from the strong 
solution upwards through the diaphragm into the weak one. The difference 
of osmose, through a porous partition, was opposite in direction to that pro- 
duced by difference of diffusion. 
The phenomenon of Regelation , originally discovered by Faraday in 1850, 
and extended in its physical bearings to explain glacier motion by Tyndall, 
has been recently made the subject of a memoir in the Annales de Chemie et 
de Physique, by Mons. W. Spring, Professor of the University of Liege. He 
adverts to the more recent researches of Helmholtz, Bottomley, and es- 
pecially of Tresca, on the viscosity of solid bodies, but is of opinion that we 
have not yet attained a satisfactory solution of the question. James 
Thomson and Clausius have given mathematical expression to the physical 
laws involved in the phenomenon, the general cases of which have been 
verified experimentally by Bunsen. But two substances still form excep- 
tions to the rule — water and bismuth. Their specific volume being greater 
in the solid than in the liquid state, pressure lowers, instead of raising their 
point of fusion. That of ice lowers 0°*0075 for each additional atmosphere 
of pressure. Mousson was able to melt ice completely, without the addition 
of any external heat, by a pressure of 13,000 atmospheres. Mons. Spring 
considers that the binding together of solid bodies under pressure is a much 
more general phenomenon, comparable to the liquefaction of gases. He has 
