False Discharge of a Coiled Electric Cable. 209 



city of the near portions of the conductor has an effect analogous 

 to that of Fizeau's condenser in the Ruhmkorff coils ; and there 

 was little or no spark (none was observed, although it was 

 looked for, in the key) on breaking the battery circuit, and con- 

 sequently, as nearly as may be, the whole mechanical value of 

 the current left by the battery must have been expended in the 

 development of heat in the conductor itself, and by induced 

 currents in the iron of the sheath; and therefore we need not 

 wonder at the great length of time during which electric motion 

 remains in the cable. 



The first column of results for experiments Nos. 1, 2, and 3, 

 and the two columns for Nos. 4, 5, and 6, show that the con- 

 tinued flow of the main current through the cable, after the near 

 end is removed from the battery and kept insulated, is to reduce 

 its potential gradually from that of the battery (which for the 

 moment we may call positive), through zero, to negative,' in some 

 time less than five seconds, and to keep it negative ever after, if 

 it is kept insulated, as long as any trace of electro-dynamic 

 action remains*. It is probable that, at the same time, there 

 may be oscillations of current backwards and forwards again t, 

 and of potential to negative, and positive again, in some parts, 

 especially towards the middle, of the cable. The mathematical 

 theory of the whole action is very easily reduced to equations ; 

 but anything like a complete practical analysis of these equations 

 presents what may be safely called insuperable difficulties, 

 because of the mutual electro-magnetic influence of the differ- 

 ent parts of the cable with differently varying current through 

 them. These peculiar difficulties do not, theoretically viewed, 

 present any specially interesting features; and the problem 

 is of little practical importance when once practical electri- 

 cians are warned to avoid being misled by electro-magnetic in- 

 duction, in testing by discharge during either the manufacture, 

 the submergence, or lifting of a cable, and not to under-estimate 

 the rate of signalling through a long submarine cable to be 

 attained when it is laid, from trials through the same cable in 

 coils, when electro-magnetic induction must embarrass the sig- 

 nalling more or less according to the dimensions and disposition 

 of the coils, and probably does so in some cases to such an ex- 



* After what has been said in the text above, it is scarcely necessary to 

 point out that this effect is both opposed to, and much greater than, any- 

 thing producible by polarization of the earth-plates. 



t As in the oscillatory discharge of a Leyden phial, investigated mathe- 

 matically by Prof. Thomson ("Transient Electric Currents," Phil. Mag. 

 June 1853), and actually observed by Feddersen, in his beautiful photo- 

 graphic investigation of the electric spark (PoggendorfFs Ann. vol. cviii. 

 p. 497, probably year 1860; also second paper, year 1861). 



