ELECTRICAL PRINCIPLES OF THE ATLANTIC TELEGRAPH. 419 
by water, which, if the current could reach it, would afford a 
passage back to the battery, offering practically no resistance ; 
while the area of its cross section is only O' 01 63 square inch. 
The current may therefore be considered as having the choice 
of two channels, one of them 425,000 square feet in width, the 
other less than of a square inch, and the question is to 
make it take the narrower channel in preference to the wider 
one. It is for this purpose that the copper conducting wire of 
the Atlantic cable, instead of being allowed to come into direct 
contact with the water, is separated from it by being coated 
throughout its whole length by four layers of gutta-percha and 
Chattertoffis compound,” of a total thickness of 0T6 of an 
inch. The electrical resistance of gutta-percha is so enormous, 
that the current finds a much easier passage along the copper 
wire, more than 2,000 English miles in length, and only about -gV 
of a square inch in section, than through this coating, notwith- 
standing that its thickness is what we have just stated (little 
more than i of an inch), and that the wire is in contact with 
it over a surface of about 425,000 square feet. 
Let us try to put this comparison between the resistance of 
the conductor and that of its insulating coating into a more 
exact form. We have already estimated the former at about 
7,500 B. A. -units : a statement by Mr. Latimer Clark, in the 
Mechanics 3 Magazine for the 10th August, enables us to esti- 
mate the latter. According to this statement, the resistance 
to the escape of electricity from each knot of the conductor 
through the gutta-percha covering into the ocean may be 
represented by 2,200,000,000 B.A. -units ; consequently, the 
resistance to the escape of electricity from the whole 1,858 
knots of the cable will be equal to tsYs of 2,200,000,000, or 
1,180,000 B. A. -units; or it is about 157 times more difficult 
for an electric current to find its way back to the battery by 
taking a short cut across the gutta-percha into the water, than 
by traversing the whole distance from one end of the cable to 
the other and back by the way prepared for it. The same 
comparison may be otherwise stated as follows : of a given 
current which enters the Irish end of the cable at Yalentia, 
nearly 99*4 per cent, would arrive at Newfoundland, supposing 
it were there received upon an instrument of no appreciable 
resistance. 
But difficulties of insulation and resistance are not the only 
difficulties of an electrical kind by which such an undertaking 
as the Atlantic Telegraph is attended. Difficulties scarcely 
less formidable are occasioned by the electrical phenomenon 
known as “ induction ” ; but we shall be better able to under- 
stand the way in which these make themselves felt, if we have 
previously briefly considered the manner in which electrical 
