474 MEMOIRS OF THE AMERICAN ACADEMY 



made upon the Red Sea cable while it was coiled in iron tanks, and he found that, 

 although the electromotive force had no effect upon the velocity, the rate of transmis- 

 sion was inversely as the square of the length. When Quetelet and Airy despatched 

 siirnals between Greenwich and Brussels, in order to ascertain the difference of 

 longitude, the observers exchanged stations with each other, during a part of the 

 operation, for the purpose of eliminating the Personal Equation. The length of the 

 line was 270 miles, of which about 180 miles consisted of a subterranean coated wire. 



7 



The transmission time (one tenth of a second *) was comparatively large for so short 

 a line. This was due, doubtless, to the great resistance and electrostatical capacity of 

 the core, but I do not possess sufficiently accurate knowledge of its character to cal- 

 culate their precise values. The same difficulty applies to Faraday's experiment 2 in 

 which he obtained two seconds as the transmission time of electricity in passing over 

 1,500 miles of coated wire. Moreover, the galvanometer which he used was not the 

 same as those employed on the transatlantic cables. Varley experimented upon 1,600 

 miles of the same wire, in 1854, and obtained a transmission time of three seconds, 

 the arrival of the electrical current being recognized by its chemical effect. 3 Mr. Lati- 

 mer Clark found the transmission time over 768 miles of coated wire to be two thirds 

 of one second. In this last case the dimensions of the copper wire and the core are 

 given. The formula, by which the times are compared, when the lengths, diameters, 

 and electrostatical capacities of two lines are known, would lead us to expect in Mr. 

 Clark's experiment a time only one half as great as the transmission time between 

 Brest and St. Pierre. In reality it was quite as large ; but no allowance could be 

 made for the difference of galvanometers, or the specific conductivities of the wire. 

 It should also be kept in mind, that in some of the various experiments made upon 

 the transmission time of the current, the conducting wire was simply covered with 

 gutta-percha, whereas, in other cases, it had, outside of the gutta-percha, a protecting 

 armor of iron. Now it is easily conceivable, as was suggested by Mr. Varley, that 

 magnetic induction in the iron would not be without its influence on the rate of 

 transmission. Moreover, the theory of Ohm was framed before the discovery of 

 the extra current induced in the conductor, and could not, therefore, have taken 



a # 



into its account any effect which that may exert upon the propagation of the primi- 

 tive current from the battery. Furthermore, it should be remembered that, thus far, 

 I have supposed that the wire, if a naked one, was uninfluenced by the surrounding 



DfBUm 



* Experimental Researches in Electricity, III. 512. 

 1 Proceedings of the Royal Society, London, XIL 2 





