416 
ON THE ELECTRICAL PRINCIPLES OF THE 
ATLANTIC TELEGRAPH. 
By Gr. C. FOSTER, Professor of Experimental Physics in University 
College, London. 
I T may be safely assumed that all readers of tliis Review 
Lave some knowledge of the peculiar properties acquired 
by a piece of copper wire, when it is used to form a connection 
between the two poles of a galvanic battery : bow it makes 
a magnetic needle suspended in its neighbourhood tend to 
place itself at right angles to it ; how, if coiled spirally round 
a piece of soft iron, it converts the iron into a magnet ; how, 
if it is cut across and the two cut ends are dipped into water, 
chemical decomposition of the water ensues ; how, if its 
dimensions are small relatively to the power of the battery, it 
becomes warm or hot, and may even be ignited and fused. Our 
readers are, doubtless, also perfectly well aware that no such 
properties are acquired by a glass rod, a piece of wood, or a 
strip of gutta-percha or india-rubber, when either of these 
substances is used, instead of a metallic wire, to join the poles 
of a battery. These familiar facts, which, with others of the 
same order, are summed up in the statement that copper is a 
conductor of electricity, and that glass, wood, india-rubber, &c., 
are non-conductors , form the basis of all the methods of electric 
telegraphy ; and if we can succeed in gaining a clear concep- 
tion of what this difference between conductors and non-con- 
ductors amounts to, we shall be in a position, not only to 
understand how submarine telegraphic communication between 
Ireland and Newfoundland is electrically possible, but also to 
appreciate a great part of the difficulties which had to be over- 
come before such a mode of communication could be established* 
When the electrical properties of different substances are 
accurately examined, it is found that various so-called “ con- 
ductors 33 differ greatly in their conducting power, and that 
even those which possess this property in the highest degree, 
such as silver and copper, are far from being absolutely perfect 
conductors : in other words, they oppose an appreciable resist- 
ance to that propagation of the electrical condition which, in 
