128 Mr. G. V. Mac Lean on the 



In the original experiment the area of the tinfoil was just 

 one-fourth of that on the condenser which formed the capacity 

 for the spark photographed, whereas the length of copper 

 wire was 9*62 cm., being 9'62/1051'l of what was used in 

 the photographic process. Accordingly the oscillation-time 

 of the spark would be 1'976 x 10 -8 sees. This is the value of 

 T. We have already found A, the wave-length, to be 591'5 cm. 



And since \ = VT, .•. V, the velocity of the electric wave, =y 



= 591-5/1 976 X 10~ 8 = 2-991 x 10 10 centimetres per second or 

 about 186365 miles per second. 



Professor Trowbridge and Dr. Duane in 1895 found the 

 velocity of electric waves along wires to be 2 - 996 x 10 10 cm., 

 though the mean of seven of their results gave the velocity to 

 be 3-0024 x 10 10 cm. per second. 



To make sure that the coherer and oscillator were in tune 

 with each other, the capacity of the oscillator was altered by 

 doubling the area of the tinfoil sheets on the glass plate. 

 Sparking at the oscillator was again begun, and readings 

 taken, whereby it was seen that the positions of the original 

 nodal and antinodal points were changed. The same thing 

 took place upon altering the self-induction. Also, separately, 

 the self-induction and capacity of the coherer were altered, 

 with a similar change in the position of the original nodal and 

 antinodal points. When the wave-length had been determined, 

 it seemed desirable to test the truth of the theory of Poincare 

 and Bjerknes. 



It is a known fact in the photography of electric oscillations 

 that it is possible to damp out all the oscillations except the 

 fundamental, by replacing part of the self-induction by a 

 self-induction having a suitable resistance. Accordingly a 

 cylinder of graphite 5 cm. long was ground down till it had the 

 same diameter as the copper wire it was to replace. The ends 

 of this rod of graphite were next electroplated with copper. 

 5 cm. of one of the copper wires forming the self-induction 

 of the coherer were cut out and replaced by soldering in the 

 prepared graphite. The coherer and oscillator circuits had 

 now the same capacity and self-induction as before, only the 

 resistance of the self-induction of the coherer-circuit had been 

 increased. The resistance of the graphite was 12 ohms. 

 Work was now undertaken under these new conditions, with 

 the result that the fifty observations thus made were identical 

 with those taken under the original conditions. The following 

 table gives the measurements taken when the graphite was 

 in the circuit : — 



