14 SATURATED STEAM, AND OT11ER VAPORS. 



temperature of the boiler. Tin- n -suits were then plotted as before, and 

 compared with the heats of the liquid derived from Regnault's mean specific 

 beats uocorrectcd. The points by the corrected method were a little more 

 regularly arranged than the points obtained by assuming the specific heat to 

 be unity at low temperatures; but tin- improvement was inconsiderable. The 

 inequality of the specific heat at low temperatures is seldom so much as the 

 unavoidable errors of the meth< - 1 . 



It appeared, that if the specific heat was assumed to be constant. from 

 40 to 45, from 45 to 155, and from 155 to 200 C., the straight lines thus 

 drawn represented the experimental values as recalculated quite nearly : .ml. 

 further, they represented the unconnected experinu -ntal values more marly 

 than Reguault's equation. 



Specific Heat of Water. The combination of Rowland's and Regnault's 

 experiments on the heat of the liquid by the in< -thod des< -nl.ed gives the 

 specific heats set down in the following table, Centigrade scale : 



8PE< II l l II-: AT. 



From 0to 5 C. 32 to 41 1 . . I.OOT-J 



5 10 41 50 ... 1.0044 



10 15 50 59 ... 1.0016 



15 20 59 68 . . 1. 



20 25 68 77 ... 0.9984 



25 30 77 86 ... <>.:> f J48 



30 35 86 95 ... 0.9964 



35 40 95 104 . . . 0.9982 



40 45 104 113 . . . 1. 



45 155 113 311 . . . 1.008 



155 200 311 392 . . . 1.046 



Thermal Unit Heat is measured in calories, or British thermal units 

 (ETU). A calorie commonly is defined as the heat required to raise one 

 kilogramme of water from freezing point to 1 C. ; and a British thermal 

 unit, that required to raise one pound from 32 to 33 F. Nothing is known 

 about the specific heat of water from to 2 C. ; consequently the commonly 

 accepted value of the thermal unit is an ideal quantity inferred from the 

 behavior of water at higher temperatures. It is more scientific total 

 easily verified quantity for the standard ; and there is a practical convenience 

 in choosing 62 F. for the standard temperature, because it is near the mean 

 temperature of the air during experimental work. Therefore, it is near t In- 

 mean temperature in the calorimeter during ordinary work with that instru- 

 ment; and the specific heat of water for the range of temperature in the 

 calorimeter may usually be considered to be unity without error, unless great 

 refinement is desired. 



The BTU in Tables I and II is taken to be the heat required to raise 



