402 
PROFESSOR O. REYNOLDS AND MR. W. H. MOORBT 
We thus get the following equation for R :— 
874,319,846 __ 1,752,718,746 
1,106,461 + 1872-5 R “ 2,235,731 + 1862-6 E ’ 
which, on solution, gives 
and, substituting for R, 
R = 9-33, 
K = 777-91. 
47. The loss of heat by radiation from the brake, as given in the Tables B, C, &c., 
was determined by multipljdng the difference of temperature between the brake and 
the air by the radiation constants, calculated as just described. 
The Tables B, C, and D, giving the results of trials 1 to 42 inclusive, should now- 
lie self-explanatory. 
The mean value of K given by the eight unjacketed trials I have mentioned was 
777-81. 
48. The best way of stating the values of K obtained throughout seemed to be as 
follows ;— 
The sums of the differences of the works and of the corrected heats were taken for 
each series of trials, and then a mean value of K for the series was found by dividing 
the first of these quantities by the second. 
The values of K given as the mean for each series in Table D have been calculated 
in this wav. 
49. A mean value of K can be obtained from the jacketed trials contained in 
Series II., III., VI., and VII. (Series V. being kept out of the determination on 
account of the possible error already noticed), by finding the sums of the respective 
differences of work and heat given with each of these series in Table D, and then 
dividing the vrork by the heat so obtained. 
The sum of the differences of work in Series II., III., VI., and YII. 
= 676,259,560, 
and the sum of the corresponding differences of heat- 
= 869,396 ; 
therefore the mean value of K given by the accepted jacketed trials so far considered is 
676,259,560 
869,396 
777'85. 
From this mean none of the values obtained from anv one of the above series differs 
c/ 
by as much as 0-03 per cent. 
Closer agreement than this could not possibW be expected, and it was consequently 
