130 
ME. C. F. YAELEY ON POLAEIZATION OF 
that in the other increases. During this change of surface-dimensions currents of elec- 
tricity will be found to pass from one to the other, the diminishing surface acting as the 
zinc plate, and the increasing surface as the copper plate of a voltaic couple. 
All attempts to polarize the mercury with oxygen have entirely failed. This experi- 
ment enables one to obtain a mercurial surface neutral to the fluid in which it is placed, 
i. e. free from polarization, or very nearly so indeed. 
All the specimens of mercury which were tested showed traces of this hydrogen 
polarization, but they can be almost absolutely depolarized (if the mercury be pure) by 
connecting it with the positive pole of a very feeble battery through a large resistance, 
and the aqueous solution with the negative pole, until the rocking to and fro of the vessel 
ceases to generate a current, or shows hardly any trace of one. The mercury-surfaces 
are then neutral, or very nearly so, to the fluid; if, however, they have been previously 
charged highly with hydrogen, this depolarizing process must be repeated over and over 
again until, after resting, the mercury is found not to yield a current by rocking. This 
“ neutral ” surface is likely to be of use in investigating the source of force in batteries. 
The following arrangements will be found to give a continuous electric current. The 
three vessels A, B, and C are arranged as shown in figure 2 (Plate II). The vessel A 
receives mercury which issues from the jet J into the acid and water in the middle 
vessel B. The mercury at the bottom of B runs through a siphon into the vessel C. 
The mercury in A and B is connected through a galvanometer (G) by means of plati- 
num wires. 
If the mercury be pure and not polarized, the running of the mercury from A into B 
gives rise to no current. If, now, the larger surface of mercury in B be polarized as 
before, it will share its polarization with each drop of mercury that falls from J, and 
thus produce a current from A through the galvanometer to B ; each drop as fast as it 
is polarized falls by gravitation into B, and so gives back the polarization it had received. 
By carrying the mercury from C back to A, this current is continued for a very long time. 
The following method (fig. 4) is a convenient one for showing the experiment on a 
larger scale ; and on this principle, several years ago, the author constructed an apparatus 
of twelve cells (which has never been published), which was worked by clock-work and 
gave a current of a potential rather greater than one cell of a Danjell’s battery. 
When clean platinum plates are polarized with hydrogen, and dipped into polarized 
mercury, they instantly amalgamate all over. 
If a gutta-percha trough of the form shown in fig. 3 in section be constructed, so as 
to contain two separate cells of mercury covered and united together by the supernatant 
aqueous solution, and a bunch of platinum plates be inserted in each parcel of mercury 
and amalgamated, the mercurial surface exposed to the fluid will be augmented when 
the amalgamated plates are partially withdrawn from the mercury, and will diminish 
when they are inserted again ; in this way, in a small space, a large extent of mercurial 
surface is easily obtained. If the platinum plates be well polarized, it will be found 
that, on raising those in the one cell and simultaneously immersing those in the other, 
