BETWEEN THE FREEZING AND BOILING-POINTS. 
169 
The values obtained by Dr. Kahle at the Reichsanstalt are somewhat lower. 
They were obtained for the H-form of cell with negative electrode, zinc amalgam. 
The portable crystal cell in Table V. refers to the case where the metallic mercury 
for positive electrode is replaced by a flattened platinum wire amalgamated. The 
portable cells in Table VII., original paper, p. 152, made by the students in 1897, and 
from which the tests (in 1897) in Table Y. are taken, were compared with the old 
crystal cells rather too soon after setting up, to use in determining the true value of 
the E.M.F. of the old cells, as we have done for the latter cells and the sealed cells 
in 1896. The mean value of all these cells is very close to the crystal cells, but later 
tests showed that some of them gave too low a value at first. 
Ratio of Clark to Cadmium Cells .—In 1897, to check the value of the Clark cells 
made by us in the laboratory, several Weston cadmium cells were constructed. 
These were made in the inverted form, and one was made in the H-form after 
type III. described by Jaeger and Wachsmuth (‘ Wied. Ann.,’ vol. 59, p. 580, 1896) 
in their paper on the cadmium cell. All the cells had a cadmium amalgam of 1 to 6 
proportion, as recommended by Jaeger and Wachsmuth, except two, which were 
made after the B.O.T. “crystal” cell type with cadmium stick. These two cells, 
however, as was expected, gave much too high an E.M.F. and were only made as a 
matter of interest. I have described these cells in another place ( c Journ. Phys. 
Chem.,’ vol. 4, 1900), with comparisons which were obtained in 1897. 
The temperature coefficient obtained for these cells was a little in excess of that 
found by Jaeger and Wachsmuth for their cells, but is more in agreement with the 
value found by Dearlove (‘Electrician,’ vol. 31, p. 645, 1893) and the original 
value given by Weston. The expression is a linear one, and reads 
E, = E 15 - -086 (t - 15°), 
and holds with great accuracy over the range 15° to 40° C. At 15° a change of 
state occurs in the cadmium sulphate, so that no formula can be made to hold below 
that point. 
I made a determination of the ratio of these cadmium cells to the old crystal cells, 
by means of the cylinder potentiometer and 6000-ohm galvanometer which were used 
in the earlier comparisons of the Clark cell, given by Professor Callendar and 
myself in our original paper (p. 121). The potentiometer was repeatedly calibrated 
by comparison with the Thomson-Varley slide potentiometer, described in another 
place. The corrections for uniformity were somewhat large, but were exceed¬ 
ingly consistent, and were determined by myself, as well as by a large number 
of the students in the ordinary course of their work. The cells, both cadmium and 
Clark, were immersed at a constant temperature near 15° throughout the test. 
Table YI. contains the result of this test. 
* Callendar, ‘ Phil. Trans.,’ A, 1902, p. 63. 
VOL. CXCIX.—A, Z 
