184 
POPULAR SCIENCE REVIEW. 
While the analysis agrees with that of a barytic andesine, the mineral 
shows by its optical properties and the angle of its cleavage planes, that it is 
not that ; in fact, it agrees with none of the felspars which have yet been 
examined. (Jahrbuch fur Mineralogie, 1879, 591.) 
PHYSICS. 
The Thermo-electric behaviour of Aqueous Solutions with Mercurial Elec- 
trodes, has been communicated to the Royal Society by Mr. G. Gore. He 
used for this purpose an apparatus consisting of two thin glass basins, 
containing each a layer of mercury about one centim. deep, cohered by a 
layer three centim. deep, of the aqueous solution to be examined. A bent 
glass tube of the shape of an inverted U with a vertical canal at the highest 
point was filled with the liquid, and inverted over the mercury ; two stout 
glass tubes with a platinum wire sealed into their lower ends, one filled with 
mercury, were placed in the mercury of the basins before adding the liquid. 
Terminals of platinum touched the mercury, but not the liquid. Thermo- 
meters were immersed in either pool of mercury. One basin was supported 
on wire gauze with a Bunsen’s burner beneath it, the other on a shelf at the 
ordinary temperature. The mercury was carefully purified. A galvanometer 
was used to test the completeness of the circuit. One basin was then heated, 
and the deflection of the galvanometer noted at various temperatures. 
Solutions were chosen which had little chemical action on mercury, and 
which were free from any visible film or dulness over the heated metal. 
The galvanometer was astatic with two coils, each of 50 ohms resistance, 
connected together. A table was obtained with the solution at the top, in 
w hich the hot mercury w r as most positive at 180° Fahr., and that at the 
bottom in w r hich it w r as most negative. Far at the head stands phosphate of 
ammonium, with a deflection of + 28° followed by carbonate of sodium with 
+ 19*0. At the other extremity of the scale stands cyanide of potassium, 
which sinks from — 14°'0 in a solution of 5 grs. to — 50 o, 0 in one of 100 grs. 
In order to ascertain whether this order agreed with that of a series arranged 
according to the different degrees of electromotive force of the various 
couples, two similar basins w r ere charged with the two solutions and con- 
nected w ith a differential galvanometer wdth currents in opposite directions, 
to ascertain which gave the strongest currents. 
On examining the results it is observed that the effects are not manifestly 
related to the chemical nature of the solutions. It is difficult to prove how 
far chemical action occurred as a cause of currents, though it cannot have 
been great. The strength of the solution clearly affected the quantity, and 
in some instances the direction of the current. Stirring, and previous 
heating of the solution, also influenced the deflection. It appeared that the 
currents were due to heat acting on and altering particular molecular 
structures, and that the difference of electromotive power was not due to 
chemical variety, but to differences of molecular arrangements. 
Measurements by means of Thomson's Galvanometers are not, according to 
M. Gaiffe, as repoited in the Comptes Rendus for January, found to be 
proportional to the value of the currents measured, but to be exaggerated as 
those values increase. This error arises from the fact that the angles of 
