30 Messrs. C. W. Waidner and F. Mallory on 



Comparison of Platinum Thermometer with 

 Tonnelot 11801. 



This comparison was made by Dr. W. S. Day and the 

 authors, in order to bring the results of the different standards 

 of thermometry into comparison, and thus, if possible, to 

 account for the differences in the values of the capacity for 

 heat of water obtained by different observers. 



The comparison was carried out in a tank especially designed 

 by Dr. Day for the comparison of mercurial thermometers in a 

 horizontal position. This tank consists of a long rectangular 

 copper box 98 cm. long, 21 cm. wide, and 23 cm. deep, with 

 a movable cover, holding a piece of plate-glass, through which 

 the stem-readings of the thermometers could be observed by 

 means of a micrometer-telescope which slides along parallel 

 ways supported from the cover of the tank. Within this 

 copper tank was another small brass box, 10"5 cm. long, 8 cm. 

 wide, and 5 cm. deep, in which were placed side by side the 

 bulbs of the thermometers to be compared. This box shielded 

 the bulbs from the effects of convection-currents ; and in this 

 way the thermometers were kept at a constant temperature 

 for a considerable time, for it required an interval of some 

 minutes for a very appreciable change in the temperature of 

 the large mass of surrounding water to produce a minute 

 change in the temperature of the enclosed water owing to the 

 absence of convection-currents. This brass box was provided 

 with two movable lids on the top and two on the bottom, 

 which could be opened and closed from the outside of the 

 tank. The large tank was provided with three paddle-wheels, 

 operated from the outside, which produced efficient stirring. 

 The entire tank was then placed inside a wooden box, the 

 interspace between the walls of the tank and box being loosely 

 filled with cotton-wool. The greater portion of the lid, with 

 the exception of the plate-glass, was covered with a layer of 

 thick felt. The tank and its contents could be raised to any 

 desired temperature by allowing steam to flow through a 

 copper tube coiled along the bottom of the tank. 



As the platinum thermometer is peculiarly adapted to show 

 minute changes in temperature, the efficiency of this tank 

 was exposed to a severe test in these comparisons. It was 

 thus ascertained that if the lids of the inner brass box were 

 opened, the entire contents of the tank thoroughly stirred, 

 and the lids then tightly closed, the temperature would remain 

 constant to within o, 001 for at least several minutes, for a 

 difference of 20° 0. between the temperature of the room 

 and the water in the tank. 



