ABSOLUTE THERMAL CONDUCTIVITY OF NICKEL. 



367 



sivity was increased before the tarnish on the surface became even perceptible. The 

 following table will show this, and it provides the means of allowing for it. - 



Date of Experiment. 



1894 



July 12 



. u 13 



„ 17 



„ 18 



August 2 



„ 3 



.. 7 



Time of 





Time of 





Time of 





Time of 





Cooling 



Tempera- 



Cooling 



Tempera- 



Cooling 



Tempera- 



Cooling 



Tempera- 



200° to 150° 



ture of Air. 



150° to 100° 



ture of Air. 



100° to 70° 



ture ot Air. 



70° to 40 Q 



ture of Air. 



in Secorids. 





in Seconds. 





in Seconds. 





in Seconds. 





1375 



20-4° 



2210 



20-3° 



2295 



20-3° 



5125 



20-2° 



1355 



19-5° 



2175 



19-3° 



2280 



19-3" 











2087 



i9-r 



2193 



19-1° 



4795 



19-r 







2105 



19-1° 











1348 



18-8° 



2150 



18-7° 



2201 



18-6" 



4660 



18-55° 











2191. 



J8-7" 



4503 



18-7° 







2178 



18-3° 



2214 



18-5° 







1341 



192° 



2149 



19-2° 



2233 



19-1° 



4740 



19-1° 



The table also shows irregularities in the cooling, due possibly to changes in the 

 state of the atmosphere, or to variations in the unavoidable draughts of the second order 

 of magnitude. The readings obtained on 8th August gave the best curves, and they 

 have been used to determine the rates of cooling given in a later table. A reduction of 

 2 per cent, was made in the rates of cooling found, in order to allow for the increase in 

 the emissivity which had taken place by 8th August. A correction is necessary in. the 

 cooling experiment for the exposed stem of the thermometer. In the statical experi- 

 ment the stem correction was made by adding to the observed reading V. the product 

 000113 V 2 . Using the same form of correction in the cooling experiment, let V. be the 

 )bserved, and 6 the true temperature, then since = V. + '000113 V 2 ., the true rate of 

 :ooling dO/dt is equal to (1 + '000226 V.) times the apparent rate of cooling dv/dt got 

 rom the curves in which stem exposure is not allowed for. Since the thermometer 

 )arts with its heat to the cooling bar, its temperature must always during the cooling 

 >e higher than that of the bar ; in other words, the thermometer lags behind the bar by 

 n amount depending on the rate of cooling. No attempt has been made to allow for 

 hat in these experiments. Some mercury was put into the hole for the thermometer 

 o give good thermal contact between the bar and the thermometer. The following 

 able gives the rates of cooling of the short bar found after applying the corrections 

 eferred to above, except that due to lag, and that due to loss of heat at the ends of the 

 ar. 



Temperature. 



Rate of Cooling. 



Temperature. 



Rate of Cooling. 



Temperature. 



Rate of Cooling. 



40 



•00362 



90 



•0149 



150 



•0308 



50 



•00556 



100 



0174 



160 



•0336 



60 



•00770 



110 



•0200 



170 



•0363 



70 



•01005 



120 



•0225 ■ 



180 



0392 



80 



•01250 



130 



•0251 



190 



•0423 







140 



•0279 



200 



•0455 















