Expansion of Metals by Heat. 
379 
1831.] 
five feet long, prepared for the purpose by Captain Baker, superintendent of iron 
bridges ; unfortunately it only occurred to me towards the conclusion of the 
operations, or the series for iron might have equalled that for the other metals. 
I now proceed to the experiments themselves, which are here pro anted in a 
tabular form : 
Number of 
Range of 
Experiment. 
Degrees. 
1 . 
128°.2 ascending, 
2. 
125 .8 descending, 
3. 
128 .2 asc. 
4. rejected, 
126 .8 desc. 
5. 
125 .7 after 2 hours, 
6. 
125 .7 asc. 
7. 
125 .6 desc. 
8. 
128 .6 after 2 hours, 
9. 
128 .6 asc. 
10. 
128 .6 desc. 
Means 
Dilatations between 32° and ‘212°. 
Brass Rod. Lead Pipe. Iron Rod 
1.001921 
1.001904 
1.001880 
1.002953 
1.001855 
1.002914 
1.001896 
1.002929 
1.001919 
1.002913 
1.001910 
1.003028 
1.001911 
1.003040 
1.002921 
1.001278 
1.002933 
1.001234 
1.001906 
1.002954 
1.001256 
1.001278 
The expansions of these three metals, as given in Ure’s Chemical Du ♦ \on.u . > ' 
considerably according to the quality and form of the material, and nearly 
under the different trials of various experimenters : those which ac 
cumstances with the foregoing are as follows: , 
English plate-brass in rod, ..Roy, | * „, nft [ * •0019H 4 
Brass wire Smeaton, 1.001. U 
rass wire, ........ avoisier 1.001235 - 
Round iron, wire-drawn, Lavoisier, 
; ron Smeaton, 1.00 1-oa 
• * • Trouvhton 1.001440 
iron wire, iiougnton,. .. 9 
iron Dulong and Petit, 1.001 IS- 
’ '* Smeaton, 1.0028070} L0028587 
Lavoisier, 1.0028484 J 
The greatest difference observable occurs in the expansion o bcfo) .’ 0 est i m!lt ed: the 
in the present experiments to be one-thirtieth g reil ^ er tQ esta | )lish confidence 
very near agreement of the first two results en s presumptuous to say 
in the present mode of operating, although l e auctions of Roy, Smeaton, 
that it affords any confirmation of the more elaborate e t he correction 
and Lavoisier. At any rate it must be allone , ^ roc i 9t no influential error 
for temperature to the measurement of a base wit i me a^ ^ ^ tables of these three 
can be introduced by making use of the expansions gi ^ anonia i OQ8t and Dnlong’s 
philosophers. The iron wire ot Troughton appears • in mC rcury seeuia 
method of measuring the expansion of solid metals by munersm 
liable to objections. u he ascertained nithin « degree of 
If the temperature of the measuring rods con j tbe power of a careful 
the truth, which does not strike the mind as at a hafd , y be appreciable : 
operator, the error of the whole correction for iea an( i about two-thirds of an 
being less than half an inch per mile on an iron r0 ’ ‘ from ot her sources of 
inch for a brass rod, which is surely less than 
irregularity in any but the most refined measuremen ^ anX jous to avail myself 
While the present apparatus remained at commit ot ] xer metals ; I accord - 
^ the opportunity of verifying the tabular expans brasS} steel, iron, zi BC » 
in giy procured a series of wires of gold, silvei, co 
