410 Mr. F. Jenkin on the Retardation of 



less than that produced at the sending-end ; so that as the rapi- 

 dity with which the signals succeed one another is increased, the 

 variations in the received currents diminish, and above a certain 

 speed no sensible change whatever in the electrical currents in- 

 dicating a signal can be observed, however strong the currents 

 produced near the sending-end of the cable may be. 



3. The changes in the received currents corresponding to sig- 

 nals sent are not constant even on a given cable with a given 

 battery, but vary according to the electrical condition of the cir- 

 cuit, produced by the signals immediately preceding ; so that a 

 given signal at the sending-end does not produce an invariable 

 and constant effect at the receiving-end, but a variable and in- 

 constant effect, depending on preceding signals. Thus when the 

 ordinary electrical signals are sent through a long submarine 

 cable in quick succession, the received signals are not only re- 

 tarded, but, if sent too quickly, they are weaker than those that 

 would have been received through an aerial line of equal length, 

 resistance, and insulation, they are also confused and unintelli- 

 gible, and at a still higher speed are wholly obliterated. 



The phenomena, as exhibited in long submarine cables, are so 

 marked, and their commercial importance so great, that they 

 have been the subject of considerable study, although even now 

 the constants required for the rigorous application of Professor 

 W. Thomson's theory are imperfectly known*. 



Fig. 2, Plate III. is a reduced copy of the curve given by Pro- 

 fessor W. Thomson in the Philosophical Magazine for December 

 1855, and represents the gradual arrival of an electric current 

 at the further end of a long wire in connexion with the earth, 

 when the near or sending end of the line is connected with the 

 battery for a time sufficiently long to allow the current to reach 

 its maximum. The horizontal ordinates represent times in func- 

 tion of a certain quantity a, and the vertical ordinates the 

 strengths of the current at those times. Thus after a time 6a, 

 the current reaches about 65 per cent, of its maximum strength. 



The quantity a— — ^logj-j, where k = the resistance of 



the conductor per unit of length in absolute electrostatic mea- 

 sure, c its absolute electrostatic capacity per unit of length, and 

 / its length. I is of course known ; k can be pretty accurately 

 calculated from existing data ; c is imperfectly known for sub- 



* There are not many long and perfectly insulated cables in the world 

 on which the necessary experiments can be tried. The Malta Alexandria 

 cable, belonging to H.M. Government, was singularly adapted for these 

 experiments when first laid ; but no use whatever has been made of it for 

 this or any other scientific purpose. 



