170 
DR. H. T. BARNES ON THE CAPACITY FOR HEAT OF WATER 
Table VI. — Comparison of Clark and Cadmium. March 6. 1897. 
Cell. 
Potentiometer reading 
corrected to 15°. 
Corrected for 
uniformity. 
X] .... 
68225 
68290 
Cd 3 .... 
48453 
48575 
Cd 4 .... 
48460 
48582 
Cd, .... 
48458 
48580 
Cd„ . . . . 
48460 
48582 
Correcting reading of X } to mean of old crystal cells and reducing mean cadmium 
reading to 20° by the formula, 
E, = E 15 - -086 (t - 15°), 
the ratio of Clark to cadmium becomes 
Clark 15°_ 68294 
cadmium 20° 48558 
1-40644. 
The ratio obtained by Kahle for the cells in the possession of the Reichsanstalt 
was (‘ Wied. Ann.,’ vol. 67, p. 35, 1899), 
Clark 15° 
cadmium 20° 
1-40663. 
The value of our ratio is somewhat lower than the value given by Kahle, which 
may be explained by either assuming the cadmium cells too high or the Clark cells 
too low. We have seen, however, that the mean of the old crystal cells is lower 
than the most probably true Clark-cell value obtained by comparison with later tests 
by "14 mvt., or 1 part in 10,000. 
Correcting the ratio by this amount, it becomes 1*40658, a value nearly identical 
with the value obtained by Kahle. 
The Absolute Value of e .—The assignment of the true value of e to the cells used 
in the present work is, at present, somewhat difficult. Glazebrook and Skixner 
found on standardizing the B.O.T. form of test-tube cell by means of the silver 
voltameter, and assuming the value "001118 gram.-sec. for the electro-chemical 
equivalent of silver as determined by Lord Rayleigh and Mrs. Sedgwick, that the 
value was 1"4342 international volt at 15° C. More recently we have the 
measurements made by Dr. Kahle at the Reichsanstalt with the Helmholtz Electro- 
dynamometer (‘Wied. Ann.,’ vol. 59, p. 532, 1896, and ‘ Zeit. fiirlnstk.,’ June, 1898), 
which give a result independent of the value assigned to the silver voltameter. We 
have also the value obtained recently by Professors Carhart and Guthe, at Ann 
