194 MR. A. NORMAN SHAW: A DETERMINATION OF THE ELECTROMOTIVE 
It will be seen that this differs from the earlier value of the resistance hy only 
5 parts in 200,000. 
In this investigation the two ohms were used at 23°'12 C. At this temperature 
their resistance, obtained by using the temperature coefficient + 0‘000248, is 
2’00368 international ohms. 
During the deflection observations a current of about half an ampere was flowing 
through these coils. If the oil was kept well stirred it was found that the temperature 
of the coils was given to a sufficient degree of accuracy by the temperature of the oil. 
IX. The “ Ratio of the Currents ” in the Two Systems of Coils 
DURING THE DEFLECTION EXPERIMENTS. 
In § VII. (c) it will be seen that the ratio of the currents is determined by the 
ratio of the standard two ohms to a resistance composed of two standard manganin 
tenth ohms (Haider, Nos. 4389 and 4388). This ratio was measured by means of the 
Kelvin-Varley slide (calibrated for refined measurements and used as described in the 
appendix). In each case the measurements were taken with a current of about half 
an ampere flowing through the coils to be compared ; and a current of five amperes 
had been allowed to flow through the tenth ohms for some time preceding the 
measurement. In this way the actual conditions of the deflection experiment were 
closely approached, and any error due to the inaccuracy of the manganin temperature 
coefficient was eliminated. 
Taking the temperature coefficient of 0'000248 for the two ohms and 0'000018 for 
the manganin resistances, and reducing the value of the ratio to that for the condition 
when the two ohms were at a temperature of 23 T2 C. and the tenths at a temperature 
of 28°'0 C. we got the following mean results for the ratio in two separate series of 
measurements:— 
(1) 10’0357, (2) 10’0357. 
Mean 10'03 57. 
This result is equivalent to a value of 0'199625 international ohms for the two- 
tenths at 20° C. Thermal or contact effects in the measurement were eliminated in 
the usual way, and each determination was checked with the Kelvin-Yarley slide 
reversed. 
The four windings of the suspended coil were arranged in parallel, and as errors due 
to any unsimilarity of the windings were eliminated by means of the series of 
observations involved in the deflection mean, and also by the method of determining 
the constant “ g 1 ,” we could assume that the effect was the same as that of a quarter 
of the current flowing through the four windings in series, and hence the actual ratio 
of the currents under the conditions of the deflection experiment Avas 
40T428. 
