68 



cable in mid-ocean. For instance, a very slight abrasion 

 occurs in the insulation of the conductor at some point in 

 the ocean. It is just sufficient to allow enough of the 

 current to escape to prevent intelligent communication. 

 The test is made by disconnecting both ends of the cable 

 from all apparatus. A test battery and sensitive galva- 

 nometer are then applied to one end, and the resistance in 

 OHMS of that portion is obtained, which is the resistance 

 of the conductor to the fault, plus the resistance of the 

 fault itself to the earth. Then the same process is re- 

 peated from the other end. 



The resistance of the fault itself must be eliminated 

 from both tests, and as it will be the same in both cases, 

 the process will be as follows : 



To the known resistance of the whole cable add the 

 obtained resistance of one test, deduct from that the ob- 

 tained resistance of the other test, divide that result by 

 two and you have the resistance of the conductor in the 

 first test from the office to the fault, in ohms, which is 

 easily reduced to miles and fractions of a mile. 



The chart then gives its locality in the ocean. A 

 steamer then goes as near to the spot as possible and 

 drags for the cable until it is fished up. By cutting the 

 cable it is easily ascertained which side of the steamer 

 the fault lies. 



The cable is then picked up by machinery and passed 

 along across the deck as the steamer proceeds. By pass- 

 ing it through a tank of water connected by very delicate 

 test apparatus to the sea, the fault announces itself the 

 moment it reaches the tank of water on deck. It is then 

 cut out. Communication is established with both sides 

 of the ocean, the cable is joined, the insulation is com- 

 pleted, and it is dropped down into its bed of infusoria, 

 to throb again with those mysterious international im- 

 pulses. 



