CHAP, vri.] 



DEEP-SEJ TEMPERJTUnES. 



291 



water on the outer shell, and thus protecting the 

 hulh within. 



Mr. Casella was directed to construct some ther- 

 mometers on this plan, only instead of 

 heing tilled with air, the outer shell 

 was nearly filled with alcohol warmed 

 to expel a portion of the remaining 

 air, and the chamber was then her- 

 metically sealed, leaving a bell of air 

 and vapour of alcohol to yield to the 

 pressure and relieve the bulb within. 

 The ' Miller - Casella ' thermometer 

 proved so nearly perfection that it was 

 decided to adopt it in future, and to 

 use it as a standard in a series of 

 experiments Avhich were undertaken 

 to test the ordinary Six's thermo- 

 meters of the Hydrographic Office 

 pattern. We depended upon this 

 thermometer alone in our subsequent 

 cruises in the 'Porcupine,' and we 

 found it most satisfactory. During the 

 summer of 1869 temperature observa- 

 tions were taken at upwards of ninety 

 stations, at depths varying from 10 to 

 2,435 fathoms. Two thermometers, 

 numbered 100 and 103 ^respectively 

 were sent down at every station, and 

 in no instance did they give the leas.t 

 reason to doubt their accuracy. Every 

 observation was taken by Captain Calver himself, 

 the lead with the thermometers attached being in 

 every single instance let down by his own hand, 



TJ 2 



I 



\\ 



fcfflO 



rSOs 



|.„ 



Fig. 53.— The Miller- 

 Casella modiiication 

 of Six's self-regis- 

 tering thermometer. 

 The large Iralb is 

 double, with a layer 

 of liquid and a bell 

 of vapour between 

 the shells, to relieve 

 pressure. 



