330 Frederick Guthrie on certain 
hydrochloric acid, and subsequent evaporization and weighing, 
give a control upon the hydrogen translation of the alka- 
line metal. The mercury is thereupon left nearly ready for 
weighing. 
On the other hand, I have not yet been able to establish a 
column of mercury having an unlimited stock of pure cold 
alkaline metal above pure mercury at the same temperature 
below. I do not see the possibility of it. Granted that 
when such metals as tin, or lead, or gold, or silver dissolve 
in mercury heat may move, such movement of heat is, 
I should think, swamped in its power of causing convec- 
tion-currents by the conductivity of the mass. But in the 
case of the alkaline metals the first contact of the two 
metals is accompanied by so much heat that the condi- 
tions obtainable with other metals are here far more difficult. 
Perhaps mercury and sodium brought into contact at a tem- 
perature far below the freezing-point of mercury might give 
the required starting-point. If their contact were real and 
the elevation of temperature very gradual and well controlled, 
we might have a trustworthy condition ; but scarcely at a 
single temperature. 
Such a condition would represent a certain fixed sodium 
potential (not infinite, because the sodium has to be disin- 
tegrated), on the one hand, and a lower, but not zero, on the 
other; and between the two the integral of the resistances 
of the various amalgams after the first contact. 
§ 26. This being so, I elected to employ sodium amalgam 
and potassium amalgam rather than the free metals. 
On mixing sodium with mercury, the two combine with 
great energy and liberate so much heat as to point to a 
loss of volume. Is this loss of volume, if it take place under 
any circumstance, so great as to give rise to an amalgam 
having a greater density than mercury itself? 
If — be the density of mercury and — 2 that of sodium, 
and if v 3 be the volume of the amalgam, then the density of 
the amalgam would be equal to that of the "mercury, if 
^h + m 2 _ m \ 
or 
v m x J 
If v 3 should be less than this for any ratio between the con- 
stituents, the convection-currents of sodium would at all 
events begin to flow down if such an amalgam were at the top 
of the mercurial column. 
