448 
PROFESSOR W. WATSON ON A DETERMINATION OF THE VALUE OF 
Fig. 5. 
a standard thermometer graduated in tenths of a degree which has been tested at the 
Reichsanstalt. 
Since the maximum current which could be used in the magnetic experiment was 
such that, if this current had been used w^hen making the silver depositions, then in 
order to obtain a sufficiently heavy deposit of silver the deposition would have taken 
eight hours, an arrangement was employed by which the two resistance coils F could 
be placed in series wffien making the magnetic measurements, and in parallel when 
making the silver depositions. Thus the resistance was four times as great during 
the magnetic experiments as during the silver depositions. The resistances of the 
two coils were so very nearly equal that no error was caused by taking their resistance 
in parallel as equal to a quarter of the sum of their resistances. In order to make 
the change from the series arrangement to 
the parallel arrangement Avithout disturbing 
the coils the connector shown in fig. 5 was 
used. Stout copper bars AF, BC, DE 
(section 3 sq. centims.) were screwed to a 
marble base. The terminals of one of the 
resistance coils fitted into two cups A, B 
and those of the other into E, F, The 
terminals were soldered into these cups by amalgamating tliem and also the cups and 
then fusing some Wood’s metal in the cup. Two mercury cups C and D could be 
connected by a short copper rod. Wires M, L and K were soldered into the cups E, 
B, A, and were used to connect the potential circuit. When the coils were used in 
series, the connector between C and D was removed, and the main circuit connected 
to the binding screws H and J, the wires L and M being used to connect the 
potential circuit. When the coils were to be used in parallel the connector was 
inserted and the leads of the main circuit moved to the binding screws G and J, the 
wires K and M being used for the potential circuit. Experiment showed that the 
increase in the resistance of one arm, when the coils are in parallel, due to the 
connector l^etween C and D, is inappreciable. 
The Cadmium Cells. —It had been originally intended to use Clark cells, and two 
dozen cells Avere prepared in 1897 according to the specification giAmn by Kahle.^ 
An equal number of cadmium cells was prepared at the same time, accordin 
to instructions given by Jager and Wachsmuth.I' The materials used 
preparing both kinds of cells Avere obtained from Kahlbaum of Berlin. When, in 
the autumn of 1900, the Clark cells AAmre conq^ared amongst themsehms, it AA'as 
found that although the cells Avere kept in an ice safe for tAvo weeks, they did not 
agree, the differences amounting to about 4 parts in 10,000. The cadmium cells, 
on the other hand, in no case exhibited a differences of 1 part in 15,000, and this 
‘ Zeits. fur Instrumentenkunde,’ 1893, p. 191. 
t ‘Wied. Ann.,’ vol. 59, 575, 1896, 
O' 
o 
in 
