428 Prof. J. A. Fleming on the Propagation of 
Concerning the figures in Table VI., the following remarks 
may be made :— 
In the first place, the wave-lengths for the five harmonic 
oscillations, which are equal to double the measured distance 
between the first and second nodes, are very nearly in the 
ratio of the numbers 3, 5, 7, 9, and 11, and the mean of the 
products 3 x 280, 5 x 160, 7x 116, 9x 88, and 11 x 72 is 807, 
or very nearly four times the length of the helix. 
On the other hand, the calculated frequencies (n) for the 
higher harmonics do not follow those of the fundamental and 
first harmonic in the ratio of the numbers 1, 3, 5, 7, 9, 11; 
and there is therefore a continual falling off in the product 
(zr) of the frequency in the condenser circuit and the wave- 
length on the spiral. In the case of the fundamental and 
first harmonic when the earth-wire was removed, this product 
agrees with the velocity of the wave along the spiral as cal- 
culated from the capacity and inductance per unit of length 
of the helix. The first suggestion which occurred on con- 
sidering this discrepancy was that the dielectric constant, 
and therefore the capacity of the glass condenser, might 
perhaps decrease as the frequency increased. It has been 
shown by Prof. J. J. Thomson and by M. R. Blondlot, that 
at a frequency of 25 x 10° the dielectric constant of glass falls 
to a value of 2°7 or 2°8*, and it seemed possible that this 
decadence might set in at a frequency as low as 10°. It has, 
however, been found that sulphur and ebonite do not exhibit 
this decrease of dielectric constant under increased frequency 
of electric force. Accordingly two other condensers were 
constructed of sheets of good ebonite, 3 mm. in thickness. 
These sheets were partly covered on both sides with rect- 
angles of tinfoil, each having an area of 15 x 17°5 centimetres, 
and two condensers were made up, each consisting of five of 
such coated plates. These were immersed in vaseline oil 
contained in an ebonite box. 
The capacity of these ebonite condensers was measured and 
found to be as follows :—The two ebonite condensers in 
parallel had a capacity of 0°0025 microfarad, and in series 
had a capacity of 0:00062 microfarad. 
Another source of error was at the same time discovered 
and correction made for it. It was found that the inductance 
of the spiral wire wound on a boxwood cylinder which was 
used as a variable inductance in the condenser circuit had a 
* See Prof. J. J. Thomson, Proc. Roy. Soc. vol. xlvi. p. 293 (1889), 
‘On Specific Inductive Capacities of Dielectrics under rapidly alternat- 
ing Electrie Forces” ; also M. R. Blondlot, Comptes Rendus, vol. exii. 
p. 1058 (1891). 
