Calorimetric Thermometers. 337 



tapping which was necessary to overcome the temporary 

 inertia of the bulb. The following, even more simple, expe- 

 riments are free from such an error, and bring; to light other 

 facts concerning this phenomenon. 



The method of procedure here consisted in suspending the 

 thermometer in a calorimeter containing a much larger volume 

 of water than the one generally used, taking the instrument 

 out, heating and cooling it alternately, when, on its being 

 replaced, readings of the same temperature were obtained, the 

 column having fallen or risen to that temperature respectively. 



A.s it was found to be impossible to keep the temperature 

 of the bath constant during the requisite time, any observed 

 reading, say when the column was rising to the temperature, 

 was compared with a theoretical reading consisting of the 

 mean of the previous and subsequent readings, both of which 

 were obtained when the column was falling to the temperature. 

 The differences here quoted will represent the total difference 

 between the readings with rising and falling columns. After 

 the thermometer had been replaced in the bath, it was tapped 

 continuously and read at intervals of one minute till it became 

 quite stationary ; between five and seven such readings were 

 made in the series here given, although it will be necessary to 

 quote the final reading only in each case. An interval of 

 about 8 minutes elapsed between these final readings, all the 

 intervals in each separate series being exactly equal. 



I. 



Kegistered. Theory. Difference. 



o o 



Immersed after being heated 10° 14-08 



cooled 10 14-08 14-103 -D23 



heated 10 14-125 14-118 + -097 



cooled 10 14-155 14185 --030 



heated 30 14-245 14233 4- "012 



cooled 10 14-310 14313 --Q03 



heated 10 14'380 



Mean -015 



II. 



Kegistered. Theory. Difference. 



o o 



Immersed after being cooled 10° 26 - 83 



heated 10 26-86 26-815 + "015 



cooled 10 2686 26908 --048 



heated 10 26-955 26918 4- -037 



cooled 10 26-975 27-008 --033 



heated 10 27-06 27-038 4- -022 



cooled 20 27-10 27'123 --023 



heated 20 27185 27-175 4- -010 



cooled 10 27-25 



Mean «J27 



Phil. Mag. S. 5. Vol. 21. No. 131. April 1886. 2 B 



