368 
MR, E. H. GRIFFITHS OH THE VALUE OF 
If E = 0, this becomes 
o- — p (6>i — 6>o) = 
\ l<r, p 
(6). 
Provided that the mechanical supply is regular, and 6q is kept constant, we can, by 
a sufficient number of observations over small ranges, determine the values of {ddijdt)^^^ 
for different values of and by substituting in equation 5, we obtain the value of 
E^/J.Pd.M' = A, where A is known. 
Assuming that over small ranges the values of R' and M' are linear functions of 
we have 
Pd = Pv{l + /v(6>, - ^)}+, and M' = M (1 + I {9, - 9)], 
hence 
_E_ _ A (7)' 
0)} . ^ 
Now the values of E and P at the standard temperature can be ascertained by 
comparisons with the standards, and the value of I' can. be ascertained by direct 
measurements of P' at different temperatures, hence 
J.M. U + l{9, - 9)} 
A.R.{1 + A;(^i - ffi} 
B, where B is known . (8), 
we have thus one equation connecting the three quantities, J, M, and 1. 
Let IV and be the capacities for heat of the water and the calorimeter respec¬ 
tively at the standard temperature 9, and let the temperature coefficients of their 
specific heats be f and g respectively, then 
M {1 + / — d)] {= ?(; [1 -\-f{9i — 9)} -f iv,f {1 -{-g [9^ — 9)], 
hence equation (8) becomes 
= B .... (9). 
If observations are taken with different weights ('?fq and u'^) of water, we obtain 
Bp B^ the corresponding values of B, thus 
J bh {1 +/(^i - 9)] + {1 + (7 (di - 9)}] = Bi 
and 
J [^^3 {1 +.m - 9)] + [1 -f g{9, - 0)}] = B, ; 
* In this, and siinilar cases, we use the suffix to denote the sources of heat, 
t The value of dO, during our experiments, was about 1° C. 
t The true vahie of R' = R {1 + + d — ^)} where /3 is the excess of the temperatui’e of the 
wire above dj the temperature of the calorimeter. See Section XIV., p. 478. 
