CHEMISTRY: LEWIS AND MINE 
637 
creased resistance be ascribed to a greater resistance to the motion of 
the electron in those regions of the mercury which contain particles 
of the dissolved metal, these particles or regions, according to the law 
of action and reaction, will be impelled in the direction in which the 
electrons are moving. This explanation is borne out by the fact that 
potassium, which produces a greater increase in the resistance of mer- 
cury than an equivalent amount of sodium, is also transferred to a 
greater extent in the direction of the negative current. 
We may also suggest here another relationship of considerable in- 
terest. Attention has been called by Kraus^ to the very large influence 
upon the conductivity of liquid mercury of a change of volume through 
pressure. Now it has been shown by Maey^ that the addition of the 
alkali metals produces a very considerable change in the volume of 
mercury. This change in volume follows the same order as the electri- 
cal resistance of the amalgams given in the preceding tables. It is 
interesting therefore to see how the resistance of mercury would change 
upon the addition of the three alkali metals, if this addition were made, 
not at constant pressure, but with such change of pressure as would 
keep constant the average atomic volume, that is, the volume occupied 
by 1 gram of solvent and solute together. 
If R is the resistance of pure mercury, V and the atomic volumes 
respectively of pure mercury and of amalgams, P the pressure in kilo- 
grams per square centimeter, and N the atomic fraction of the dis- 
solved metal, then, according to the measurements of Bridgman,^ 
d\nR/dP = — 3.34 X 10~^, while from the measurements of Richards, of 
Buchanan, and of Bridgman,^ d\nV/dP = — 3.8 X 10~^. Hence 
dhiR/dhi F = 8.79. In Table IV the second column gives the value 
of dlnV'/dN calculated from the work of Maey, and the third column 
the values of dlnR/dN, which are obtained from the second column 
through multiplication by 8.79. These figures therefore show the 
change in the resistance of mercury which would be produced by the 
three metals, assuming that the effect of the dissolved metal is due 
solely to the accompanying volume change. The fourth column gives 
the values of dlnR/dN obtained directly from our measurements, by 
fmding the slope of the resistance curve at iV = 0. These figures will 
vary somewhat according to the relative weight given to the measure- 
ments at the lowest concentration. The fifth column gives the values 
of (d\nR/dN)v' namely, the fractional change in the resistance per 
gram-atom of dissolved metal when the average atomic volume is kept 
constant. These figures are obtained by subtracting the second column 
from the third. 
