254 
same. The electrodes must be bright, as other- 
wise the phenomenon becomes very complicated. 
Furthermore the original polarization must 
often surpass a certain value if the residual 
polarization is to be contrary in sign; and 
there are other differences in detail for which 
there is no room here. Thus the rotational 
effect may proceed gradually to a maximum; 
an electromotive force zero may imply a very 
large residual polarization appearing on mo- 
tion. The charging of moving electrodes is 
an interesting case; etc. 
To elucidate this phenomenon, it suftices 
here to assume the occurrence of paired double 
layers a a’ and b’ b, one double layer at each 
electrode. One element, a’, b’, of each double 
layer is localized in the liquid and the other 
element a, b in the solid electrode, both of the 
double layers having the same direction; 7. e., 
being two condensers in series. Hence there 
are two interpenetrating electrostatic fields, 
one 
+\- +|- 
Axle 
——> + |- i) +|-——> 
Charging Current 
SS en ee 
ala’ Electrolyte b/ 
| 
0’ a’ localized in the liquid and the other a b in 
the electrodes. These fields are in a contrary 
direction and the liquid field must be very 
much stronger to correspond with the initial 
counter polarization. On rotating the elec- 
trodes, the field localized in the liquid }’ a’ is 
set free and its ions dissipated. The field 
localized in the solid, a b, however, remains 
and this constitutes the residual polarization 
in the direction of the charging current. Both 
fields decay in the lapse of time in the usual 
way. 
When rotation ceases a liquid field is re- 
established, but usually, though not always, to 
a smaller degree. Eventually a probably dis- 
charges a’ and 8, 0’, one of the fields passing 
through zero first, so that the effect of rotation 
finally vanishes. I have met both with marked 
polarization which on rotation vanished, as 
well as with an apparent absence of polariza- 
tion which on rotation became very marked. 
To obtain moving electrodes as free from 
polarization disturbances at the contact with 
SCIENCE 
[N. S. Von. XLVIII. No. 1236 
a liquid, it is therefore prudent to capture 
both fields; 2. e., to leave the electrodes entirely 
without interferences. This may be done by 
surrounding each with a porous cup, closed and 
completely filled with an electrolyte, the termi- 
nals passing out through an insulating tube. 
The electrodes should moreover be fixed rigidly 
to the cup. Again since zine electrodes soon 
tarnish in brine but remain bright in con- 
centrated zine sulphate solution, the latter is a 
preferable electrolyte and the cups may be 
submerged in brine or any other solution. 
I therefore constructed two cup electrodes of 
the kind in question and placed them in the 
rotational apparatus as before. The original 
potential difference of the zines was about 
A millivolt. After keeping the circuit closed 
over night this fell off to below .05 millivolt, 
and could be eliminated by exchanging the cup 
electrodes. Rotation of the appartus, 2 e., 
an external current in the brine surrounding 
the cups, produced no appreciable effect. The 
electrodes were then charged with a current 
of .2 am. for 80 sec. The polarization remain- 
ing was now much less than above, through- 
out, beginning with about 5 millivolts which 
fell to .5 m.yv. in 5 minutes and to .05 in a few 
hours. Rotation was ineffective through all 
stages of the decay. No doubt the simple 
electrode in which both the original and the 
residual polarizations haye vanished would 
often suffice, but with greater uncertainty, be- 
cause such electrodes can not be exchanged 
without danger as to modifying their value. 
Cart Barus 
Brown UNIVERSITY, 
PROVIDENCE, R. I, 
SCIENCE 
A Weekly Journal devoted to the Advancement of 
Science, publishing the official notices and pro- 
ceedings of the American Association for 
the Advancement of Science 
Published every Friday by 
THE SCIENCE PRESS 
LANCASTER, PA. GARRISON, N. Y. 
NEW YORK, N. Y. 
Entered in the post-office at Lancaster, Pa., as second class matter 
