148 
On the Measure of Temperature. 
CMay 
III. — On the Measure of Temperature, and the Laws which regulate the 
communication of Heat. By M. M. Dulong and Petit. 
[From the Journal de I’Ecole Royaje Polytechniquo, T. xi.] 
§ 3. On the Expansion of Solids. 
If Tve compare the results of the preceding 1 table 1 with those which vre have re- 
corded in Table I. 2 , it will be perceived that the suspicions entertained, as to the 
correctness of the mercurial thermometer, were not without foundation. It will be 
found that the rate of expansion of the glass tube, and of the fluid contained in it, 
are very remarkably different in the case of a considerable change of temperature. 
When the air thermometer indicates 300, mercury, taken singly, would give 314,15 : 
while mercury in glass (the ordinary thermometer) would only mark 307,64. 
The results established in the preceding section are the more valuable, ’inasmuch 
as they enable us to obtain a very exact measure of the expansion of solid sub- 
stances. We have only, in fact, to measure the difference of expansion between 
mercury and each of these substances. 
Ihe application of this method to glass is sufficiently simple ; the difference in 
question, being the apparent expansion of mercury in a vessel formed of that 
material. Although this question has been made the subject of so man v experi- 
ments, we have thought it necessary to determine it also ourselves, and with every 
precaution that this sort of inquiry will admit of. A glass tube, about 6 deci- 
metres m length, (23,64 inch,) anil capable of holding about 700 grammes (1081 1 
giams) of mercury, was made use of. It was closed at one extremitv, and ;;t the 
other terminated in a capillary tube, the capacity of which, compared with the large 
one, was so small as to allow the correction forit to he safely disregarded. 
The tube being filled with mercury, and carefully purged of air and moisture, by 
repeated boilings, we determined the weight of mercury, driven out by a change of 
temperature from the freezing to the boiiingpoint. An idea may he obtained of the 
accuracy of which this method is susceptible, by considering, 'that the portion of 
mercury which is not heated is exceedingly small, while the horizontal position 
of the tube enables us to apply tbe correction of the boiling point, dependent on 
the change of barometric pressure. 
Five repetitions of this experiment, using different materials, have, given, very 
nearly, the same results as are expressed in the following table. No sensible differ- 
°o-T ed he !T en t . ubcs formed of different kinds of glass, whatever 
me size or thickness of their sides. 
fr J?V a '“ es of apparent expansion at 200° and 300», (cent ) hare been deduced 
from the comparisons previously made between the mercurial and air thermometers. 
Temperature 
as shewn by 
an air 
thermometer. 
Mean apparent 
expansion of 
mercury in 
glass. 
il« 
Absolute 
expansion 
of glass in 
volume. 
100 
1 
t» 4 80 
3SToo 
200 
itilTS 
3 6 3 oill 
300 
33 IS 
32 35? 
Temperature, as shewn | 
by a thermometer ; sup- 
posing the expansion of 
glass to be uniform. 
100 ° 
213,2 
352,9 
Z7cltn° c ?n thiS d,fr0re, ' C = • to 
t h aif tbey 1 b^Heved this numhTr TZl'Z ^ 
caution of boiling the mercurv Th» i Z 1 - y hr ; d neglected the pre- 
been, for the same reason iu' excess - ex]1ans ‘ on mercury must have 
in Table Thf> ti. 5 r.ri ’ • * , * and this is conformable to what we see 
ed. This «on is “otTmifoTm ‘t"*"*- determined, as above explain- 
0° and 100° • it ae-rpps w iti, ti, , i ntl ’ • Ufc . ,ncreases the temperature between 
tained in a more direct manner FinalK? 11 ^ T ^ Lavoisio V and L{t P !ace , ° b * 
Which would he « ffnraJa i . ! , „.." ally ’ ^e last column contains the indications 
see 
is 
which would he afforddT \hZT y ’ ^ last column contains the indication 
b.v the difference which is 1 fr ** 1 raadc elllire, y ° f glass. We may se 
frrnn beingu^formio^te expansion' tn “ » ,em P erature of *">•, how far ghfss i 
' ve c ? u,d determine the expansion of iron; 
= ■ a vessel composed of this metal; but our at. 
Vol. i. p. 303 1 a Vol. i. p. 35. 
