Memoirs presented to the Cambridge philosophical society on the occasion of the jubilee of Sir George Gabriel Stokes, bart., Hon. LL. D., Hon. SC. D., Lucasian professor.

182 MESSRS GLAZEBROOK AND LODGE, EXPERIMENTS ON THE OSCILLATORY
The resulting value for the frequency is 2470 which is still below that found by
experiment, viz. 2560, but it has already appeared that the capacity assumed for the
cylinder condenser, viz. 5'5 metres, is too high. The assumption that the value was
51 which (p. 180) is required to reconcile with theory the experiments recorded in
Table VI. would also bring the results of this case into greater harmony.
On the same date (July 30) and immediately after the above experiment, the
condenser was reinoved and oscillations taken with the coil alone. In this case, assuming
the theory developed in Part III., we have
S1Lx2= - (1- ) = 238,
or if we suppose the capacity uniformly distributed along the coil,
SiLx2 =:3.
On substituting for S1 the value 16 metres, and for L 4-63 secohms, we find for the
frequency the values 3830 and 4300 respectively; the experimental result is 4630. In this
case theory and experiment would be reconciled by the assumption that the capacity of the
coil was 10 metres instead of 16, and this value fits, as we have seen, the experiments
just discussed in which the condenser was used.
If we adopt the first of the two formulae and take S, = 10 we find the theoretical
frequency is 4820,, while the second formula based on the assumption of a uniform
distribution of capacity leads to the value 5360. The observed value was 4630 which
agrees best with the first of these two theoretical values, being rather below it. It will be
observed however from the record on p. 177 that the experimental results are very variable.
Thus these three sets of experiments in which the condenser was connected to the
terminals B, C of the coil will be reconciled with theory by the assumption that when
the experiment is so conducted that there is a large potential difference between the
inner windings of the coil for each of which the electrostatic capacity is smaller than
for windings near the outer edge, the effective capacity of the coil S1 of the formula
is about 10 metres, possibly rather over 10 metres.
These results are given in Table VII. (a).
TABLE VII. (a).
Both coils A and B being used, but the coil capacity taken as 10 metres instead of 16.
Date'] Condenser CapaFrequency Frequency
Date Condenser Capacity calculated observed
Aug. 1 Cylinder and dise 10-5 1860 1830
July 30 Cylinder 5-5 2470 2560
1,, i Coil only 4820 4630