268 Prof. Townsend and Mr. Morrell on Electric 



condenser C 2 was in one particular position, owing to the 

 damping introduced by the circuit A 2 C 2 B 2 when resonating 

 with A 1 C 2 B 1 . This difficulty was overcome by bridging the 

 rods with a second condenser C 3 , which stops the resonance. 



The parallel wires XiYj, X 2 Y 2 (fig. 1) were about 

 8 metres long, and connected at one end to a loop L for 

 coupling to the short-wave generator. A small condenser C 

 was connected between the ends X x X 2 , for the purpose of 

 adjusting the distribution of current in the wires so as to 

 have resonance when the bridge is near the ends Xi X 2 . 

 In order to avoid effects of surrounding objects the wires 

 wt»re stretched at a distance of about a metre above the 

 tables. With waves from 5 to 15 metres long resonating 

 lengths of the wires may be determined within a few milli- 

 metres by using a small incandescent lamp G to indicate the 

 current in the bridge between the wires. A higher degree 

 of accuracy may be obtained by using a thermo-junction, 

 instead of the lamp, and when the filament of the valve in 

 the generating circuit is maintained at a constant tempe- 

 rature the wave-lengths along the wires may be measured to 

 one part in 5000. 



In order to obtain accurate measurements the current in 

 the free ends GY 1? GY 2 should be very small when the 

 bridge is in a position of resonance. It is therefore necessary 

 that the system comprising the parallel wires should have 

 no free mode of oscillation which is nearly in resonance with 

 the short-wave generator when the bridge Gr is removed. The 

 simplest method of obtaining this result is to adjust the 

 condenser Cso that the detector indicates a maximum current 

 in the wires when the bridge is near the free endsY 1 Y 2 . 

 Another position of resonance of the bridge is required in order 

 to measure the wave-length, and this position may be within 

 one or two metres of the ends X-j X 2 . A wave nearly twice 

 the length of the parallel wires may be thus measured 

 accurately. 



When the effect of the free ends was thus eliminated, and a 

 current was induced in the oscillatory system comprising the 

 wires, by a wave of constant length, the distance measured 

 along the wires between the resonating positions of the bridge 

 remained accurately the same (within one part in 5000) when 

 the distance between the wires was changed from 2 to 10 

 centimetres, or the diameter of the wires changed from 1*2 to 

 2'05 millimetres, or the resistance of the thermo-junction 

 changed from *3 to 8 ohms. 



The measurement of the wave-length was not affected by 

 altering the shape of the loop L used for coupling to the 



