34 
On the Measure of Temperature. 
law is true at every temperature. So thatthe changes of elasticity, which heat occa. 
sions in a gas the volume of which is constant, are subject to the same law as the 
changes of volume would he, were the pressure to continue the same. 
\t’e shall no. v give the mean results, deduced from a great number of experi- 
ments, made by both of the preceding methods. These results will be found in the 
following table, which includes the full scale of mercury, from its freezing toils 
boiling, that is to say an interval of about 400 degrees. 
Temperature* 
as shown by the 
mercurial thermometer. 
Corresponding- 
volumes of a mass 
of air. 
Temperatures which would be 
sluMvn by an air thermometer 
corrected for the expansion of 
glass. 
36 
0.8650 
36. 
0 
1.0000 
0. 
100 
1.3750 
100 
150 
1 5570 
148 70 
200 
1 7389 
197-05 
250 
1 9189 
245.05 
300 
2.0976 
292.70 
Boiling point of 
mercury, 3C0 
23125 
350.00 
The numbers in the third column have been corrected for the expansion of glass, 
which we shall presently establish*. 
Thdre is a considerable difference amongst the numbers given by different expe- 
rimenters, as the boiling point of mercury on its own scale. This in part arises from 
the different degrees of care, with which different observers construct their instru- 
ments, but principally on account of the want of precision, in the correction made 
for that part of the stem not inserted in the liquid. The method we have followed, 
and which furnished the result given in the preceding table, enables us to dispense 
altogether with this correction Instead of measuring directly the augmentation 
of volume of a constant mass of matter, as is done in the ordinary thermometers, 
we determined the loss of weight which a mass of mercury, capable of filling a 
certain vessel at 0°, sustains when this vessel is completely 'immersed in boiling 
mercury. Knowing at the same time the apparent expansion' of mereurvin glass for 
the first 100 we may by a very simple calculation find the corresm'mding tem- 
perature on the mercurial thermometer, the stem of which is of the same tempera- 
ture with the I »lb+. In order to prevent the contents of the vessel from entering 
also into ebullition, we had taken the precaution to make it terminate bv an up- 
right narrow tube about six centimetres (in. 2,36) in length. The column of mer- 
cury ,t contained .lid not constitute the m^th part of the total mass, vet bv the 
pressure it exerted within the vessel, it completely prevented the formation of 
vapour. It is scarcely necessary to say, that everv precaution had been taken to 
expel completely every trace of air or of humidity. 
The corresponding temperature of the air thermometer, has been calculated by 
means analogous to those we h ive constantly employed, in our experiment on the 
expansion ot ras. The number inserted in the table is the mean of four results 
the extremes of which do not differ one degree. OI IOUI ,esuJts ’ 
From the beautiful observation of M. Guv Lass 
... i / husSiic, that all elastic fluids pxnaiiil 
precisely m the same manner between 0" and 100, it was at least orohahhPtb.t 
this uniformity would also he found to hold in more elevated temperatures, and 
consequently that the preceding table derived from experiments 
mon air, would also hold true of every kind of gas. But to “ave uotbinl H T'Z' 
ful m a subject of so much importance, we made an experiment also witl , g doubt ' 
which, as is known, differs the most in some of its prS^a^os J£ri2S 
• As we are averse to extend this memoir to an undue length . ... _ 
entering into the details of each particular experiment. " \Ve ntedfrom 
tent ourselves, in what follows, to give the results simoresVin. - ti the refore c °“‘ 
terminataons and inu-rmedinte’calculatious that ImVii'lSff to them part,cular de * 
«t>' that is Stp^edVy &lh U ‘ e Vessel at °*» P 
/ 4-64 8 0. p 
it — p 
