44G PEOFESSOR W. WATSON ON A DETERMINATION OF THE VALUE OF 
either the silver voltameters, V, or a balancing resistance, R', can be inserted in the 
circuit. The resistance of the circuit could be roughly adjusted by means of the 
resistance, R''', which consisted of a box of manganine coils, adjustable by tenths of 
an ohm. The value of the current was roughly indicated by a Weston ammeter, A. 
The final adjustment of the resistance of the circuit was obtained by means of the 
three adjustable carbon resistances, E. Two of these resistances were in parallel, 
and tlie other in series with these. Each of them consisted of about 50 carbon 
jjlates in a narrow wooden box, the size of the plates being 9 centims. by 6 centims. 
and 0’6 centim. thick. By means of a screw passing through one end of the box 
the carbon plates could be compressed, and thus the resistance altered. 
The standard resistance coils, the potential between the terminals of which was 
kept equal to the E.M.F. of the standard cell, were j^laced in an oil bath at F. The 
leads to the coil in the magnetic experiments could be inserted in the circuit at H. 
A Pohl commutator inserted in these leads allowed of the current being reversed in 
the coil without its reversal in the resistance coils F. 
The potential circuit included the two standard resistance coils F, the cadmium 
cell C, a resistance of 10,000 ohms R'", a key J on the depression of which the 
circuit was broken and hence the galvanometer zero could be determined, and a 
galvanometer G. The galvanometer was one designed by the author and constructed 
in the Laboratory. It has a resistance of about 570 ohms, and has four coils each of 
half-an-inch in diameter. The needle system consists of two sets of magnets, the 
individual magnets being about 3 millims. long, suspended by a long quartz fibre. 
The magnet system was rendered astatic, so that the distance between the two sets 
of magnets only amounting to ‘5 inch, the field produced by the Helmholtz galvano¬ 
meter when at a distance of 3 metres produced no deflection. A small magnetized 
sewing needle was used to bring the needle system into any desired azimuth. The 
position of the magnet system was observed by means of a telescope and scale, and 
the sensitiveness of the arrangement was such that a change in the resistance of the 
main circuit of 1 in 12,000 produced a deflection of 6 scale divisions (millimetres). 
It was a matter of comjjarative ease, by manipulating the screws of the carbon 
resistances, to keep the current so constant that the galvanometer deflection never 
exceeded a millimetre, that is, to keep the current constant to within 1 part in 
120,000. 
By means of two resistance boxes and a single accumulator a potentiometer 
arrangement was provided by means of which the E.M.F.’s of the standard cells 
could be compared amongst themselves to within 1 part in 30,000. Such a 
comparison was always made before and after each silver depositioji. As no 
difterence, amounting to 1 part in 10,000, was ever detected, the E.M.F. of the cell 
actually used, which was changed from one experiment to the next, was taken as 
being the same as the mean E.M.F. of all the cells. 
In performing the magnetic experiments the two standard resistance coils F Avere 
