294 THE DEPTHS OF THE SEA. [chap. vii. 



end is attached to the base of a short cylinder. 

 Any variation of temperature causes the coil to 

 wind or unwind, and its motion acts to rotate the 

 axial stem. This motion is magnified by multiply- 

 ing wheels, and is registered upon the dial of the 

 instrument by an index which pushes before it a 

 registering hand, moving with sutficient friction 

 merely to retain its place when thrust forward by 

 the index hand of the thermometer. The instru- 

 ment is graduated by trial. The brass and silver 

 portions are thickly gilt by the electrotype process 

 to prevent the action of sea-water upon them. The 

 box which covers the coil and indicatory part of the 

 thermometer is merely to protect it from accidental 

 injury, and is open so as to permit the free passage 

 of the sea-water. This instrument appears to answer 

 tolerably well for moderate depths, its error up to 

 600 fathoms not greatly exceeding 0°'5 C. ; at 1,500 

 fathoms, however, the error rises to 5°C., quite as 

 great as that of the unprotected Six's thermometers, 

 and the error is not so constant. It is evident 

 that under great pressure little confidence can be 

 placed upon instruments which give their indica- 

 tions through metal machinery. 



Before H.M.S. ' Porcupine ' started on her summer 

 crviise in 1869, a valuable series of experiments were 

 made upon the effect of pressure on various register- 

 ing thermometers at Woolwich, under the superni- 

 tendence of the Hydrographer and of the Deep-Sea 

 Committee of the E,oyal Society. The object was to 

 subject all the forms of deep-sea thermometers in use 

 to pressures in a hydraulic press, equivalent to the 

 pressures which they would encounter at different 



