the Systems of Scientific Units of Measurement. 183 



referred to water, will have the specific heat •2381, we shall 

 then have 



1° C. =10^ C.Gr.S. (substance in question). 



Now Regnault's value for specific heat of air at constant 

 pressure is "2375, Wiedemann's value is •2389. The mean of 

 these is '2382. 



If therefore we adopt the specific heat of air at constant 

 pressure as the standard of specific heat, we have 



1° C. =10^ of 1 C.G.S. (air) unit. 

 The practical unit would be 10^^ of a C.G.S. unit, and there- 

 fore would be 10^^ of a degree Centigrade. 



In relation to the thermometer, air has other well-known 

 advantages over liquids, and is already the standard for specific 

 inductive capacity and magnetic permeability. 



This decimal relation of the Centigrade thermometer would 

 not exist if the coalition of the Electromagnetic and Electro- 

 static systems were carried out on the lines I have indicated, 

 viz. by making g'^ of a second the unit of time. The new 

 unit of temperature would then be 900 times that of the 

 electromagnetic quad rant- volt-second air unit. 



Hence 

 1 unit of temperature = 900 x 10^^ of 1 degree Centigrade. 

 „ „ =1620x10^^ „ Fahrenheit. 



„ „ = 720x10'^ „ Eeaumur. 



The endeavour to bring the horse-power and the unit of 

 power into decimal relation may be guided by throwing the 

 formulae into terms of length, time, and power ; and I have 

 accordingly given a table of the ideas involved on this basis, 

 in column VII. Any magnitude which appears in these 

 formulae unencumbered by the symbol of power will of course 

 be unaltered by a mere change in its unit. Hence, if we 

 adopted as the unit of power, '746 of the watt or a decimal 

 multiple of that quantity of power, we could bring the unit 

 of power into decimal relation with the horse-power without 

 altering the units in the practical system of time, length, 

 velocity, acceleration, resistance and conductivity, and ca- 

 pacity. Thus the expensive standards of our laboratories, of 

 resistance and capacity, our resistance-boxes and our micro- 

 farads, would remain to us. But our boxes of grammes would 

 fall in value. The unit of temperature, however, would not 

 be altered, so that Centigrade thermometers would remain. 

 Column VIII. expresses the results of such a change. 



No doubt if we did not suffer from an hereditary duodeci- 

 mal taint, we might discard the g^^g ^^ ^ ^^7? called a 

 second, and adopt the yoo'ooTJ ^^ ^ ^^7 ^^ °^^ ^^^^ ^^ time. 



