44 PRESIDENTS ADDRESS—SECTION A. 
velocity ; and this means they must occupy space for a finite 
time. But waves involve the motion of energy, so that we must 
admit that. energy can be located in space, and as a consequence 
of this we are, I think, just as certainly led to imagine a 
“plenum” of some kind as if we could touch it with our ‘hands, 
or smell it with our noses, or taste it with our tongues. The 
first deduction to be made from our interference exper iment then 
is that space is filled with a medium of some kind—unless we 
are prepared to admit that energy may exist fer se—which 
amounts to filling space with an idea merely. Experiments of 
the kind suggested have been actually performed by Hertz. In 
order to get waves of a manageable length, Sir William Thomson’s 
calculation shows that we must have the capacity and_ self- 
induction of the circuit small. Hertz’s first discovery was the 
means of getting this. Conductors were constructed either of 
plates or cylinders and were made symmetrical about a certain 
point. This point is the gap where the discharge occurs. To 
take a real case. In a repetition of Hertz’s earlier experiments 
by Trouton, a “vibrator” was used. consisting of two brass 
plates about 40 cm. square. These plates were suspended by 
silk threads, so that their plates were vertical and identical, and 
their edges 60 cm. apart. From each plate there ran a stout 
brass wire toward the other. Each wire carried a brass knob— 
presumably four or five centimetres in diameter 
between the knobs was about three millimetres, the terminals of 
an induction coil were brought respectively to each plate and the 
coil was set in action. At each break of the primary circuit of 
the coil, one of the plates becomes positively the other negatively 
Ghareed: As soon as the charging has progressed to a certain 
point the dielectric breaks down between, the knobs and. a spark 
occurs. If the resistance of the spark-path is not too great, we 
have the condition necessary for the setting up of oscillations, 
The frequency then depends on the self-induction and capacity of 
the plates, as has been said several times, and is in general so 
high that the connection of the plates through the coil becomes 
of no moment, the immense “ impedance” of the coil making it 
practically non-conducting for currents of this frequency. Some 
attention has to be paid to the condition of the surfaces between 
which the spark occurs. If the balls are not finely polished, or if 
the negative one is illuminated by ultra-violet rays, the spark 
will not be sudden enough, compared with the period of an 
oscillation, to enable the oscillatory motion to become established. 
Lenard and Wolf have lately shown that ultra-violet light causes 
the knobs (especially when negatively charged) to give off dust 
which is torn from thei surf: ing glow discharge 
When everything is well arranged there is a series of straight, 
bright, white sparks between the knobs at every discharge—very 
recognisable after they have once been attended to. The next 
