THE MECHANICAL EQUIVALENT OE HEAT. 3G5 
Again, the thermometry was not of a sufficiently satisfactory nature. The ther¬ 
mometers were standardized by direct comparison with B.A. thermometers calibrated 
at Kew * As these B.A. thermometers were only divided into xq of a degree, the 
YXo 0^" ^ degree had to be estimated,! and the order of accuracy was not sufficient. 
1890. —Throughout this year I devoted my attention entirely to the measurement 
of temperatures by platinum thermometers, and I, at that time, proposed to use such 
thermometers in place of mercurial ones during my J experiments. Excellent, 
however, as platinum thermometers are, for the accurate determination of tem¬ 
perature, they are not suitable when the observations have to be taken at exact 
intervals of time. A thermometer, whose readings are at all times visible, must be 
used in such a case. 
1891. —By the kind permission of the Master and Fellows of Sidney College, the 
Chemical Laboratory was placed at my disposal during this summer. This building 
was unsuitable for physical work, for the traffic in the adjoining street was so great 
that galvanometers became at times unmanageable, and many of the observations had 
to be taken during the night, under conditions somewhat conducive to personal 
errors. 
The steel regulating chamber, described in Section III., was first used in this year 
(1891), and the results are therefore of more value than the preceding ones; but 
the same persistent error in the measurement of B still remained, and tlie form of 
stirrer adopted was unsatisfactory. The summer’s work was, however, of some value, 
as it gave sufficient data for calibrating the mercury thermometer tvhen rising. 
A platinum wire coil, having a single coating of amber varnish, was used, and it is 
reasonable to suppose that its increase of temperature above the surrounding water 
was of a similar order to that ascertained to be the case in the coil of 1892. Assu¬ 
ming the increase of its resistance to bear the same ratio to its total resistance as in 
the case of the 1892 coil, and, applying a similar correction to our results, we obtain 
J = 4T88 X 10^ 
expressed in terms of water at 18°'8 8 C, and assuming the coefficient of the change in 
specific heat of water obtained in 1892, we get 
4T92 X lO^j; 
as the value expressed in terms of the thermal unit as previously defined. 
* The accuracy of the Kew corrections for the inequality of the bore has been pi-eviously discussed, 
see ‘ B.A. Report,’ 1890. 
t I find that many observers, by means of such thermometers, determine temperatures (without using 
a micrometer scale) to -roVo ^ degree. I envy, but cannot lay claim to, such powers of observation. 
+ The wires connecting the lid of the calorimeter with the steel chamber (see fig. 2, infra) had (in 
1891) a slightly higher resistance than those used in 1892. I am unable to make the necessary correc¬ 
tion, as 1 have not sufficient data. The effect of the correction (which would certainly be less than 
in 6000) would be to slightly increase the value of J here given. 
