2l8 



DEPTHS OF THE OCEAN 



temperature of 20° C, 

 the mercury - thread 

 would have expanded a 

 Httle, giving a reading 

 perhaps of 2.25° C. in- 

 stead of 2.00° C. This 

 secondary change is 

 easily calculated when 

 the temperature of the 

 mercury at the reading- 

 off is known, and so 

 inside the protective 

 tube Richter has placed 

 a small auxiliary ther- 

 mometer (d), which 

 gives the reading tem- 

 perature, and thereby a 

 correction for the read- 

 ing. 



In many cases it is 

 necessary to have the 

 temperature determined 

 with the highest possible 

 degree of accuracy, and 

 Richter's reversing ther- 

 mometer is very satis- 

 factory in this respect. 

 During the " Michael 

 Sars" Atlantic Expedi- 

 tion the temperature 

 series were taken almost 

 exclusively by the aid of 

 these thermometers, and 

 in most instances two 

 thermometers were used 

 simultaneously, so as to 

 make quite sure of the 

 determinations. When 

 the readings were cor- 

 rected it was found that 

 the mean difference be- 

 tween the values given 

 by the two thermome- 



FiG. 156.— Richter's Reversing Thermometer. 

 The mercury breaks at e ; the figure on the left and the 

 upper one on the right show the position of the mercury 

 before reversing. The lower figure on the right repre- 

 sents part of the thermometer immediately after reversing. 



