MR. J. E. H. GORDON ON MEASUREMENTS OF ELECTRICAL CONSTANTS. 423 
per second—say 100, as the card would check the speed slightly. This determination 
was made with a battery which had been charged 24 hours, and had done a good deal 
of work. 
Estimation of electromotive force of secondary' current. 
The spark given by the coil with the wheel-break and Leclanche battery is about 
•04 inch. Let N be the number of cells this corresponds to. Now Messrs. De La Rue 
and Muller have * enunciated the following formula for the spark produced by any 
number of their cells :— 
Length of spark = 
0033N 2 
600 2 
Hence we have 
Here length =’04. 
600 3 
N 3 = *0 4-~ ; = 4,300,000 
N — about 2050. 
The electromotive force is then equal to that of about 2050 cells of a chloride of 
silver battery. 
This determination was made with the coil disconnected from the induction balance. 
When it is connected to it the spark obtainable is much shorter, owing, possibly, to the 
leakage, or more probably to the capacity of the induction balance and electrometer, a 
certain proportion of the electricity being required to charge these at every make and 
break. I had no means of accurately measuring the lengths of very short sparks. 
Connections a,nd general method of working. 
The connections are shown in figs. 4 and 5. The wheel-break and Leclanche 
battery were connected to the coil primary in the usual way. The secondary current 
after passing the secondary reversing engine came to the induction balance. 
One of the wires carrying the secondary current was connected to the centre plate c 
(fig. 4), the other to the two outer ones, a and e ; the smaller plates h and d were con¬ 
nected to the quadrants of the electrometer respectively. Now, for a moment, let us 
suppose the electrification to be produced by a battery acting constantly in the same 
direction, and the needle of the electrometer to have a permanent charge; then when 
the plates are all placed symetrically and with only air between them, however strong 
the battery may be, the spot of light will remain at zero. This may be seen by noting 
in fig. 4 the actions of the large plates on the small ones respectively. If, however, 
a is moved away so as to make the distance a h greater than d e, the needle will be 
deflected in one direction. If instead of moving a away we had placed a dielectric 
plate in the space a h, there would have been a greater action from a to b, and the 
* ‘ Proc. Roy. Soc.,’ 1875-76, p. 170. 
