and the Absolute Scale of Electric Potential. 25 



tration of the point that, for equal differences between the 

 initial temperatures, the ratio between the rate of cooling down 

 from a temperature above that of the enclosure and the rate 

 of heating up from a temperature below that of the enclosure, 

 increases with the temperature-range. 



Difference of temperature. 



Eate of cooling. 



Eate of heating. 



o 



5 



10 



500 



1-039 



1-0797 



1-2153 



1-000 

 1-000 

 1-000 



Assuming as the lower temperature 0° C. (which is, how- 

 ever, not exact), it will be seen that the ratio of the rate of 

 cooling to the rate of heating is not very different from the 

 ratio of the initial and final absolute temperatures; for 



1 + -003665 x 5 = 1-018, 



1 + -003665 x 10 = 1-036, 



1 + -003665x500 = 2-832. 

 The great discrepancy of the last value is accounted for by the 

 impossibility of comparing a cooling from + 500° with a heat- 

 ing up from —500°. 



It is therefore clear that, though the rate of loss of heat is 

 not rigorously proportional to the absolute temperature, those 

 bodies whose absolute temperatures are highest part more 

 quickly with their heat for a given difference of temperature. 

 Suppose, therefore, two bodies of equal mass and surface, and 

 of equal specific heat, one heated, the other cooled, until the 

 excess of temperature of the one is exactly equal to the defect 

 of the other. If we had no other means of ascertaining which 

 was at the higher absolute temperature, we could be quite 

 certain, provided we only observed the rates of diminution in 

 their respective differences of temperature from that of the 

 mean-temperature enclosure in which they were placed; for that 

 one whose rate of change of temperature was the greater would 

 be the one which had initially a true excess of heat. 



It seems only legitimate, when we consider the innume- 

 rable analogies between the diffusion of heat and that of 

 electricity (especially in diffusion by conduction), to conclude 

 that, as the rate of loss of charge in " resinously " electri- 

 fied bodies is greater than that of " vitreously " electrified 

 bodies, the state of " resinous " electrification corresponds 

 in reality to that of excess or to a true plus electrification; 

 and this is the electrification possessed by our earth relatively 

 to the surrounding space. 

 University College, Bristol, 

 May 24, 1881, 



