140 THE DEPTHS OF -TP1E SEA. [CHAP. in. 



involve considerable difficulties. Still, the liability of the index 

 to slip, and the probability that the indication of the thermo 

 meters would be affected by the great pressure to which they 

 were exposed, rendered it very desirable to control their indica 

 tions by an independent method. 



" Two plans were proposed for this purpose, one by Sir Charles 

 Wheatstone, and one by Mr. Siemens. Both plans involved the 

 employment of a voltaic current, excited by a battery on deck ; 

 and required a cable for the conveyance of insulated wires. The 

 former plan depended upon the action of an immersed Breguet's 

 thermometer, which, by an electro-mechanical arrangement, was 

 read by an indicating instrument placed on deck. The latter 

 plan made the indication of temperature depend on the existence 

 of a thermal variation in the electric resistance of a conducting 

 wire. It rested on the equalization of the derived currents in 

 two perfectly similar partial circuits, containing each a copper 

 wire running the whole length of the cable, the sea, and a 

 resistance- coil of fine platinum wire ; the coil in the one circuit 

 being immersed in the sea at the end of the cable, and that in 

 the other being immersed in a vessel on deck, containing water 

 the temperature of which could be regulated by the addition of 

 hot or cold water, and determined by an ordinary thermometer. 



" The instruments required in Sir Charles Wheatstone's plan 

 were more expensive, and would take longer to construct ; and, 

 besides, the Committee were unwilling to risk the loss of a some 

 what costly instrument in case the cable were to break. On 

 these accounts they thought it best to adopt the simpler plan 

 proposed by Mi 1 . Siemens ; and the apparatus required for carry 

 ing the plan into execution is now completed, and in use in the 

 expedition. 



" Meanwhile a plan had been devised by Dr. Miller for 

 obviating the efi'ect of pressure on a minimum thermometer, 

 without preventing access to the stem for the purpose of setting 

 the index. It consists in enclosing the bulb in an outer bulb 

 rivetted on a little way up the stem, the interval between the 

 bulbs being partly filled with liquid, for the sake of quicker 

 conduction. The Committee have had a few minimum therino- v 



