intended to reduce the effects of Inductive Action, 335 



proposed to include two wires in the same gutta-peicha sheath, 

 of course separated from each other by an intervening stratum 

 of the same material, and to use one as a return wire. As this 

 proposition has been the subject of some discussion amongst 

 telegraph engineers, and as there are certain conditions set up 

 by the arrangement which do not appear to me to have received 

 sufficient attention, I have thought it necessary to offer a few 

 remarks upon it. 



The circumstances attending the action of two separate sub- 

 marine cables and two insulated wires enclosed in a single sheath 

 are altogether different. In a single cable, the whole of the in- 

 ternal surface of the dielectric induces upon the whole of the 

 external surface, each being as an entirety in an opposite elec- 

 trical state, thus correctly representing a Leyden jar. 



In a cable, however, containing two conductors carrying 

 reverse currents, the conditions are different, and will be repre- 

 sented by a Ley den jar having a glass diaphragm across its 

 centre, dividing it, in fact, into two semi-cylindrical jars op])0- 

 sitely charged, in which the respective coatings of each will be 

 partly inducing upon the respective opposed halves of the ex- 

 ternal coating, and partly upon each other through the sectional 

 diaphragm ; thus dividing the system longitudinally through 

 the centre into two arrangements analogous in action, but oppo- 

 site in character, and exerting a mutual influence on each other. 



Let us now consider attentively the phaenomena here involved. 

 First, then, with regard to the portions of the two wires which 

 include the dielecti-ic between them. This resolves itselfinto a 

 Leyden arrangement, the surfaces of whose conductors will be 

 represented by the longitudinal sectional area of the respective 

 wires, and whose dielectric will approximate in thickness to the 

 mean distance between the two oj)posing halves of the respective 

 wires. This portion of the dielectric, involving the action of 

 one-half of each wire, will have a greater tendency to take up 

 charge than the portions of the dielectric surrounding the outer 

 halves of each wire. Now, although the wire itself may exhibit 

 and indicate the tension of the surface with which it is in con- 

 tact, yet this tension is no evidence whatever of the quantity 

 absolutely existing as charge upon that surface, since very diffei'- 

 ent quantities of electricity, exhibiting the same amount of ten- 

 sion, will be taken up by similar surfaces of a dielectric whose 

 thickness varies in different parts. I will illustrate this by an 

 example. Take a thick Leyden jar or plate, and charge it until 

 the spark is capable of overleaping a given interval in discharge. 

 I call the length of this discharge the degree of tension. Let 

 the quantity of electricity necessary to produce this effect, 

 • wether measuicd by turns of the machine, by sparks of a givcu 



