CLARK CELL AS A STANDARD OF ELECTROMOTIVE FORCE. 
573 
§ 7. The Clock 
The Laboratory chronometer, beating half-seconds, was used to give the time during 
which the current flowed, and the rate of the chronometer was determined by a series 
of comparisons with a clock belonging to the Cambridge Philosophical Society, which 
is rated from the Observatory. 
It was found that the chronometer was losing at the rate of about '16 seconds per 
hour. A small correction for this has been introduced in the final result. 
§8. The Resistance Boxes and Apparatus for the Comparison of Electromotive Forces. 
For the comparison of the electromotive forces two resistance boxes by Elliott Bros. 
were used. The coils in these boxes agreed together sufficiently nearly to introduce 
no serious error. The current from two Leclanche cells ran continuously through 
10,000 ohms, taken out of the two boxes in series. The standard Clark or the bottle 
cell was connected with one of the boxes through a sensitive galvanometer and a resist- 
ance coil of 10,000 ohms, and plugs transferred from one box to the other until there 
was no deflection on closing the galvanometer circuit. The Leclanches remained very 
steady during the whole course of the experiments. The resistance required in the main 
circuit to balance the Clark changed in about a fortnight from 5722 to 5784, or about 
1 per cent. This comparison was made from time to time during each individual 
experiment, to check the constancy of the Leclanches, with satisfactory results. On 
August 7 an extreme variation of 6 ohms occurred. On other occasions the change 
during the time an experiment lasted was only 2 or 3 ohms. In making the reduc¬ 
tions the mean value was taken. 
The galvanometer, battery, and resistance boxes were all insulated on separate 
paraffin blocks. 
The temperature of the bath containing the Clark was observed from time to time 
during the experiments. It usually rose slightly by about 0°'l C. 
The thermometers employed had been tested at Kew, and the necessary corrections 
are introduced. 
Fig. 1 gives a diagrammatic representation of this part of the apparatus; I. and II. 
are the two resistance boxes through which the current from the two Leclanches passes. 
A, B, C, D, L, M are six mercury cups in a block of paraffin. Of these L is 
connected through the galvanometer G with one end of the box I., while M is in com 
nection with the junction of the two boxes. The poles of the Clark are connected to 
A and C, and well insulated wires join B and D to the ends of the strip resistance B, 
Connections can be made as needed among the cups by means of pieces of wire bent 
into the required form and attached to pieces of sealing wax as insulating handles. 
When a current, i, is maintained in the resistance It, the difference of potential 
between B and D is Bi. By connecting L to B and M to D, and adjusting the plugs 
