Comparisons of Mercury and Platinum Thermometers. 33 



parisons at one temperature were completed, the Tonnelot 

 was removed from the tank and its zero quickly determined 

 (in a mixture of pure ice and distilled water) before any 

 recovery of the zero had taken place. In the zero deter- 

 minations the thermometer was adjusted vertical and the 

 axis of collimation of the micrometer-telescope horizontal. 

 Occasional barometer readings were taken to reduce the 

 indications of the Tonnelot thermometer to standard pressure. 

 While one observer was taking observations on the Tonnelot 

 thermometer, another observer was taking simultaneous obser- 

 vations of the resistance of the platinum thermometer. 



Results of Comparison. 

 h\ order to show in detail the methods used, &c, we give 

 below the reduction of a single series of observations at one 

 temperature. 



Observations : — 



Platinum f Coils FI ' °' F " 

 m, , < Bridge-wire reading = — 1*015. 



Inermometer I m to <> -P .mo 01 



^ lemperature ot coils 20 '61. 



Micrometer Readings. 



(1) (2) (3) 



f 20°"4 15-587 15-579 15-586 



Meniscus ... 15-109 15-102 15-100 



20°- 5 14-995 14-995 14-989 



Tonnelot . Barometer (brass scale)... 762-0 mm. 

 Ihermometers m < in0/ > 



Temperature 19 "6 



Zero determination : — 



0°-0 12-610 12-612 12-605 



■^Meniscus ... 12-428 12-430 12-426 



Reductions : — Zero. 



Stem- reading of Tonnelot 11801 ... 20-482 +0-031 



Calibration correction + 0-004 0*000 



Correction for external pressure due"} 



to 57 mm. of water + excess of > — 0*001 —0-001 



atmospheric pressure. J 



Internal pressure correction : ther- ^ 



mometer horizontal in comparison > 0*000 + 0008 



tank, vertical in zero tank. ) 



Zero correction —0*038 



No. 11801. 



20*447 

 Correction to fundamental interval. . . 0*000 



Phil, Mag. S. 5. Vol. 48. No. 290. July 1899. D 



