494 
MR. E. H. GRIFFITHS 0 ^ THE VALUE OF 
denotes by the letter m. This quantity appears to be of the same nature as that 
which we term d{= 0 — iii the platinum-thermometer formulae. This quantity m 
had to be determined from a difference of the order of 0°'05 C. between the air- and 
mercury-thermometers. The method adopted for finding the value of the analogous 
quantity d would appear much less liable to error. 
Howland himself was evidently anxious to throw more light on this portion of his 
investigation, for he performed, with great care, a series of experiments by the method 
of mixtures. These are given on pp. 125 to 130 ; he has not, however, tabulated his 
results. We give, with one exception (No. 10), the results of all these experimenis 
which did not involve temperatures above 30° 0. 
Table XLIV.—Specific Heat of Water by the method of Mixtui'e (Howland). 
Experiment. 
Rowland’s result. 
Mean coefficient 
of decrease. 
! 
1 
Mean sp. heat 0°—18°_ 
18° —27° ... . 
5, ,, -LU 1 
•000185 
2 
Above our range. 
— : 
3 
Mean sp. heat 0°—17° , 
„ „ 17°-29° .... 
■000165 ! 
i 
4 
Rejected bj Rowland. 
— 
5 
Mean sp. heat 0° —18° , nri«7 
„ „ 18°-30° • • 
•000447 
. 
G 
Rejected by Rowland. 
— . 
7 
Mean sp. heat 0°—16° i.nn'in 
„ „ 16°-24° 
•000083 
8 
Mean sp. heat 0°—17°_.nn .^7 
17°—25° ' ' ' 
,, ,, a. < -JL/ 
•000216 
9 
Meansp. heat 0° — 21° _ , r, 
21° —28° ^ a . . . . 
,, ,, .JJ- -,o 
•000321 
10 
Requires explanation, as it gives an in¬ 
creasing specific heat. Also the difference 
between the mean temperatures (2°'5 C.) 
is too small to enable us to expect an 
accurate result. The remaining experi- 
ments are above our range. 
— 
Mean .... 
•000236 
It is thus apparent that the results of Howland’s investigation by this method are 
in closer agreement with our conclusions than thev are with his own.* We do not, 
however, attach much importance to this point, since—as he justly remarks—the 
method by the mechanical equivalent is of far greater value. 
* Our valuo ’OOO^GG. 
