61 
ance to the passage of an electric current between two metal 
plates in a liquid. Thus I have been informed by Mr. France, 
of the Submarine and Mediterranean Companies, who has had 
long experience in testing and w'orking submarine cables, that 
he has frequently observed, when applying constant electro- 
motive force to one end of a submerged cable in which there is 
a bad defect of insulation, that the indicating needle of his 
galvanometer has continued oscillating through nearly the 
whole range of its scale without any apparent cause. Phe- 
nomena of the same kind, to a greater or less degree ai’e, 1 
believe, familiar to all careful observers who have been engaged 
in submarine telegraphing. Another very remarkable feature 
of the insulation of gutta percha covered wire, is the difference 
in the effects of positive and negative electrifications. 
“ It is well known that a fault of insulation in an actually 
submerged cable causes a much greater loss of current when 
the wire is negatively, than when it is positively electrified, 
and that if after the wire has been left to itself, or has been 
negatively electrified for some time, a positive electrification 
be applied and maintained, the insulating power (resistance to 
loss) gradually rises, and continues rising, minute after minute, 
and sometimes even sensibly for hours ; as is shown by the 
current from the battery at one end of the cable, gradually 
diminishing, while the current through the other end, if put to 
earth, gradually rises in strength. On the fourth day after 
the end of the cable was landed here, I found that a positive 
current entering from ten cells of a constant battery fell in 
the course of a few minutes to half strength. When the 
battery was next suddenly reversed, the negative current rose, 
and remained after that nearly constant, at about the same 
degree of strength as that at which the positive current had 
commenced. The same kind of action is, I have learned, 
certainly observed in cables actually submerged, and known 
to have faults in the gutta percha, by which the conductor 
becomes exposed to the water, and this has been attributed to 
