SUBMARINE TELEGRAPHY 249 



ADDENDA. 

 Explanation of Diagram of Shore Tests Atlantic Expedition. 



S is a swatch by which the cable can be connected at pleasure 

 with the studs 1, 2, and 3. 



The cable is always connected with Smith's resistance, through 

 which a feeble current passes, deflecting galvanometer G, and show- 

 ing by its magnitude the state of insulation of cable, or, more strictly 

 speaking, the potential at S. 



When S is connected with stud 1, as drawn, condenser 1 is 

 charged by the cable. It can be discharged through galvanometer G,. 

 This discharge is a second test of the potential at S. It will cause an 

 onward movement in the current applied by the ship's battery, and a 

 consequent momentary change of deflection in the ship's galvanometer. 

 This will be a continuity test for the ship, proving the copper unbroken. 

 This test might have been used as an arrangement for speaking from 

 shore to ship. Two condensers of different sizes would have given two 

 signals, from which an alphabet could be formed. 



When S is connected to 2 and the second switch P to 1, as drawn, 

 the cable will charge one plate of condenser 2 ; and conversely, if 

 the potential of the other plate be changed by being connected 

 through the reversing key, K,, alternately with each pole of the bat- 

 tery, C Z, this will induce alternate positive and negative charges in 

 the cable, which, being superadded to or withdrawn from the original 

 charge produced by the ship's battery, will cause alternately back- 

 ward and forward currents which will show as signals on the ship's 

 galvanometer. Similarly, if switch P connect the second plate of 

 condenser 2 to the galvanometer G 2 , any changes of potential in the 

 first plate of condenser 2 caused by the ship's battery will induce 

 corresponding changes in the charge of the second plate, and there- 

 fore currents to and fro through galvanometer G 2 , which currents 

 may be used as signals. 



The cable at S may be connected through stud 3 with the electro- 

 meter, and its potential directly measured by comparison with the 

 potential of some point of a set of slide resistance coils, through 

 which a current is flowing from a local battery. This current is main- 

 tained constant by reducing the resistance in the resistance coils, so 

 as to maintain a constant deflection in the galvanometer G 3 . The 

 potential at the successive studs of the slide 1 may be called in an 

 arbitrary unit, 10, 20, 30, .... 100 ; and if the electrometer were 

 directly connected with these studs, it would show with an accuracy 

 of ten parts in a hundred what was the potential of the cable in the 



