and the Communication of Heat. 
371 
1830.] 
preceding experiments. We have effected our object by means of a very simple 
arrangement, She idea of which is due to Mr. Leslie. 
When the thermometer, with naked bulb, cools in air, the total rate of cooling is 
made up of the rates due separately to contact of air and radiation. If these rates 
be put = v and v' the total rate will be v -f- v\ If again the thermometer have its 
bulb covered with silver leaf, the rate v, due to the contact of air, remains the same 
v' 
for the same excess of temperature, and v' becomes 5 707 1 since the ratio of the 
radiating powers of glass and silver is 5,707. The total effect of the cooling 
v' 
process, in the case of the silvered thermometer, will then be v -f- Hence 
it is easy to see, that to determine the loss of heat at every temperature, produced 
bv contact of air, it is sufficient to determine the total rates of cooling for the ther- 
mometer ; first, with its natural surface, and then with a silvered one. These rates 
of cooling being represented by a and b, we shall have 
= » + •»«* = » +-5^7 
whence 
5.707 
4.707 
Let us apply this formula to the results contained in the accompanying table : 
Excess^ of temp, 
of the therm. 
Total rates of cooling. 
Values of 
V. 
glass bulb. 
silvered bulb. 
260° 
21°, 42 
10°, 96 
8,10 
240 
21 ,12 
9 ,82 
7,41 
220 
17 ,92 
8 ,59 
6,61 
200 
15 ,30 
7 ,57 
5,S2 
180 
13 ,04 
6 ,57 
5,19 
160 
10 ,70 
5 ,59 
4,50 
140 
8,75 
4 ,61 
3,73 
120 
6 ,82 
3 ,80 
3,11 
100 
5 ,57 
3,06 
2,53 
8lt 
4 ,15 
2 ,32 
1,93 
The second and third columns contain the total effects of cooling of thermome- 
ters, with naked and with silvered bulb, for excesses of temperature, as contained 
in the first. The last column contains the corresponding values of v deduced from 
the above formula ; that is to say, the loss of heat which contact of air alone would 
occasion in each of the thermometers. Now, according to the preceding part of 
this paper, the law which loss of heat, proceeding from this cause follows, is ex- 
pressed by the equation 
V = wf 1 , 3 
in which « must always be determined for each particular case. In that we are 
considering « = 0,0057. If for t be successively substituted the numbers from 80 
to 260° we shall have corresponding values of v, which will not be found to differ 
much from those determined by experiment. This may be seen by comparing the 
second and third columns of the following table : 
Excess of 
temperature. 
Value of v 
from obsn. 
Value of v as 
calculated. 
260° 
8°, 10 
8°, 14 
240 
7 ,41 
7 ,38 
220 
6 ,61 
6 ,63 
200 
5 ,92 
5 ,37 
180 
5 ,19 
5,17 
160 
4 ,50 
4 ,47 
140 
120 
3 ,73 
3 ,11 
3 ,79 
3 ,14 
100 
2 ,53 
2 ,50 
80 
1 ,93 
1 ,90 
