Chemical Affinity in terms of Electromotive Force. 263 



both are perceptibly higher than the value obtained by 

 Joule by means of water-friction &c. in 1850 (Phil. Trans. 

 1850, p. 61); whilst this older value is completely corrobo- 

 rated by his recently made experiments on water-friction &c. 

 (Phil. Trans. 1878, p. 365). In 1850 Joule found that the 

 heat required to raise 1 lb. of water (weighed in vacuo) from 

 t to t + 1 (where t lies between 55° and 60° F.) corresponded 

 to the following number of foot-pounds at Manchester : — 



By friction of water . . . 772*692 

 „ „ mercury . . 774-083 

 „ „ cast iron . . 774*987 



Of these values the first was considered probably the most 

 accurate; and as a little vibration and sound was unavoidable 

 in the experiments, the round number 772 was adopted as the 

 most probable value. The 1878 value finally arrived at is 

 that the heat required to raise 1 lb. of water (weighed in vacuo) 

 from 60° to 61° F. corresponds to 772*55 foot-pounds, and is 

 hence sensibly identical with the former result. The 1867 

 values, however, were as follows, expressed in foot-grain- 

 second units: — 



Per grain degree F. 

 (water weighed in air). 



1st series. Mean of 10 experiments . . 25335 

 2nd „ „ 15 „ . . 25366 



3rd „ „ 30 „ . . 25217 



Of these the last value was taken as most exact ; when reduced 

 to water weighed in vacuo it becomes 25187. As in this series 

 the average temperature of the calorimeter was 18°* 63 C. = 

 65°*5 F., the final result is that, to raise the temperature of 1 

 gramme of water from 65° to 66° F., requires 25,187 foot- 

 grain-second units of work. 



34. When reduced to C.G.S. measure and degrees Centi- 

 grade the three values, 772*692 foot-pounds, 772*55 foot- 

 pounds, and 25,187 foot-grain-second work units, become 

 respectively 41*60, 41*59, and 42*14 megalergs per gramme- 

 degree, the value of g at Manchester being taken at 981*3. 

 (Compare illustrations of the C.G.S. system of units, by Prof. 

 Everett, p. 51.) Since, however, the specific heat of the water 

 at the three assigned temperature-limits, 55°-60°, 60°-61°, 

 and 65°-66° F., is not quite the same, on reduction to the 

 gramme-degree taken above as unit (the heat requisite to raise 

 1 gramme of water from 0° to 1° C), these values become 



41*60 ,_, K „ 41*59 ., K . , 42*14 , oin 

 = 41 ' 57 > i-T7wvq= 41 '54 ? and ^^=42*10 



1*0008 ' 1*0009 '1*0010 



