330 Dulong and Petit onthe Measure of Temperatures, [May, 
Excesses of| Velocities of 
temp.oftherm, cooling due to|/Ditto of hy-|Ditto of car-|Ditto of ole- 
‘above the sur-ithe contact of drogen. bonic acid. fiant gas, 
rounding fluid.jair. 
200° 5:48° —_ 0:25° 741° 
180 4-75 16°59 4:57 6°45 
160 4:17 14-26 4:04 5:41 
140 3°51 12-11 3°39 4°70 
120 2°90 10-10 2°82 3°84 
100 2°27 7°98 2°22 3°12 
80 hs 6:06 1-69 2°34 
On dividing the numbers in the third column by those in the 
second, we find for the ratios between the losses from hydrogen 
and those from air 
BAO hii 0 tial B42 | 62 cel BAB isso 8°48) ole'ors; SOA meng 43 
Now as it would be sufficient to render these ratios equal to 
alter the velocities which have served to determine them by 
quantities within the limits of the uncertainty to which all such 
experiments are exposed, we may conclude that the law is the 
same for hydrogen and for air. 
We shall come to a similar conclusion for the two other gases, 
if we take the ratios of the velocities of cooling which they pro- 
duce to the corresponding velocities produced by air. The 
numbers for carbonic acid are, 
O95 2. U'Y0e 12,0 S00 PD .. UTZ. ©. Oar? ea O00 
Those for olefiant gas are, 
PSG0699 BBG) Ve HBO a6: 1:93 ea eB? os Ae Stiunided 
The law of cooling produced by the sole contact of a gas is 
‘then independent of the nature and density of this gas ; and the 
comparison of the series given above, with an analogous series 
of cooling in vacuo, shows clearly that the law of which we are 
in search differs from that of radiation. After a great many 
trials, of which it would be superfluous to give an account, we 
have found that the velocities of cooling due to the sole contact 
of a gas vary with the excesses of temperature of the body, 
according to a law analogous to that which connects the cooling 
power of a fluid with its elasticity; that is to say, that the 
quantities of heat which a gas carries off from a body increase 
in a geometrical progression, while the excesses of temperature 
likewise increase in a geometrical progression. The ratio of this 
last progression being 2, that of the first is 2°35. We deduce 
likewise, by a calculation similar to those formerly employed, 
that the losses of heat due to the contact of a gas are propor- 
tional to the excesses of temperature of the body elevated to the 
power 1-233. ale 
