202 Mr. T. T. P. B. Warren on a Method of Testing 



Not to rely solely upon the electrometer for the measure- 

 ments of tensions, a modification of the condenser method pre- 

 viously described, and which will serve equally well to convey 

 continuity- signals to the ship, will supply a check-measurement 

 of these tensions, and in a way so as to give at once continuity- 

 indication to the ship and the tension in volts to the shore 

 observer. 



The charge abstracted from the cable in one condenser is sent 

 through a delicate galvanometer in one direction, simultaneously 

 that a condenser of similar capacity, charged to the same tension 

 from a constant battery, flows in the other direction; a current 

 will be perceptible on the galvanometer flowing from one con- 

 denser to the other, should they not be charged to the same 

 tension. The changes due to the galvanometer itself will not 

 interfere ; and thus not only will a considerable amount of irk- 

 some testing and calculating be avoided, but the results them- 

 selves be more accurate. 



- These tensions will be telegraphed to the ship at stated times. 

 The tension of the ship's battery will be carefully measured on a 

 delicate portable electrometer on board ship for comparison with 

 the results sent from the shore station. The fall of tension on 

 the cable can then be simultaneously noted on the shore as well 

 as on the ship ; and during this time the continuity-signals 

 need not be interrupted, as the sensibility of the ship's elec- 

 trometer will allow the small impulses to be noted, but, instead 

 of in a current form, either by regular and equal abstrac- 

 tions of charge, or by the shore attendant at stated times re- 

 storing the full tension. This, in addition to forming a perfect 

 continuity-signal, will at once indicate to the ship the loss ob- 

 served on shore during any specified space of time. In the same 

 way the ship may inform the shore observer ; and both will be 

 able to verify or compare their results. 



Any difference of tension can thus be instantly communicated 

 from shore to ship, or vice versa, without the necessity of sig- 

 nalling the results as noted by the sliding resistances or by the 

 condenser method ; the signal itself will convey all the data re- 

 quired by the ship, should this difference of tension arise from 

 -the existence of a fault. 



The reduction and. comparison of these tensions should in 

 every case be made with the constant cells of Mr. Latimer Clark. 

 Now the ship and the shore doing this independently of each 

 other, should still arrive at the same result, so long as every thing 

 remains right ; but in the event of a slight fault occurring, the 

 two results will not coincide, and the difference observed will 

 correspond to the magnitude of the fault ; and as the precise 

 tension at the fault can be. measured from either end, the 



