ON THE SPECIFIC HEAT OF AQUEOUS SOLUTIONS. 
129 
11. The Capacity of the Calorimeter .—This was determined from the values of S„ 4 
for water, which it was the object of the former paper to determine. Since the date 
of that paper the results have been criticised by Callendar (Bakerian Lecture, 
‘Phil. Trans.,’ A, vol. 212, p. 1, 1'912), and the question arose as to which values 
should be adopted. For the reasons given more fully in a “ Note on the Specific 
Heat of Water,” which is communicated simultaneously with the present paper, 
(‘Boy. Soc. Proc.,’ A, vol. 93, 587, 1917), it has been thought that the values given 
in the former paper between 0° C. and 40° C. are preferable to those which result 
from the researches of Callendar and Barnes. Moreover, if, as Callendar 
suggests, the former values are affected by idiosyncrasies of the apparatus or method, 
there can be no doubt that the values now obtained for solutions will be similarly 
affected, and, therefore, for the purpose of comparing the specific heats of solutions 
with those for water, it will be best to work with the results for water given by 
the same method. For the intervals chosen for the present investigation the values 
of S a b for water derived from the former paper are (loc. cit., p. 234) 
0° C. to 13° C. 13° C. to 26° C. 26° C. to 39° C. 
4U93, 4-175, 4-175, 
the equality for the second and third periods resulting from the position of the 
minimum at about 25° C., whilst those which result from the work of Callendar 
and Barnes are 
4-199, 4-189, 4'184. 
The results of two sets each of six experiments are set out below in Table III. 
The figures tabulated are the values of 60U+ ^ (see Section 9), which result 
from the experiments of which the details of No. 516 have been given as an 
example. 
Thus, applying the series of values of S/ derived from the former paper, we get 
the regular series of capacities 
839, 857, 876, 
for the intervals. By applying the figures of Callendar and Barnes we get the 
irregular series 
824, 815, 849. 
Since the specific heat of glass steadily increases with temperature, it is cleai that 
the first series of capacities is to be preferred, and they are adopted for the purposes 
of this paper. 
s 2 
