16 



Vniversitij of California Publications in Zoology [Vol.15 



corrected for the error thus introduced. It is for this 

 purpose that the auxilliary thermometer, a, is pro- 

 vided. Its use depends upon the fact that, being an 

 ordinary thermometer, any one of its degree-scale 

 divisions always corresponds to the change in the 

 volume of mercury produced by a change of 1° C 

 in temperature, while each degree-scale division of the 

 reversing thermometer corresponds to the change in 

 the volume of the total amount of mercury produced 

 by a change of 1° C but not to that of the small frac- 

 tion of the column broken off when inverted. Thus, 

 the two thermometers, r and a, will give identical 

 readings only at the temperature at which the instru- 

 ment is reversed and the difference between their 

 respective readings at any other temperature will be 

 proportional to the difference between this temperature 

 and that at the time of reversal. Hence, by reading 

 both thermometers simultaneously the temperature at 

 tlie depth of reversal may be computed. "Without 

 going further into details, it may be shown that the 

 neces.sary correction, k, is given by the formula. 



k: 



(T—t) 



a+-) 



6300 



where T is the reading of the reversing thermometer, 

 / that of the auxiliary thermometer, v the volume of 

 the small bulb of the reversing thermometer, s the 

 volume of one degree-scale division of the reversing 

 thermometer, and 6300 a constant depending upon 

 the coefficient of expansion of mercury and the nature 

 of the glass. 



Ekmau (1905a) experimented on the time required 

 for the Richter thermometer to assume the tempera- 

 ture of the surrounding water. He gave the ther- 

 mometer a high initial temperature and then reversed 

 it in water 10° C colder. Conduction proved to be 

 ^•^^Z^ ver\- rapid on account of the fact that the large bulb 



Fig. 2. Eichter'sjs immersed in a jacket of mercury. The thermometer 

 reversing ther- 

 mometer, assumed the temperature of the water within an error 



