1907.] on Rpcmt Contributions to Electrir Wave Telegraphy. 6ft3 



in the antenna and the wave-length of the radiated waves. I have 

 devised instrnments of tliis type covering the whole range of wave- 

 length measnrement from 50 to loo feet up to 20,ooo feet or more. 

 An instrument of the same kind, but with a more sensitive oscillation 

 detector than a neon tube, can be used to measure the wave-length 

 of waves being received on the antenna. The cymometer has other 

 uses besides wave-length measurement. One of these is to draw a 

 resonance curve and thence deduce the rate of decay of the oscilla- 

 tions in a train and their number. In a train of oscillations the 

 time period occupied by each oscillation, whether of current or 

 potential, is the same, but the amplitudes die away in geometric ratio. 

 Hence the ratio of two successive amplitudes or oscillations is con- 

 stant, and the natural logarithm of this ratio is called the decrement. 

 We can determine this decrement when we know the frequency of 

 the oscillations in the primary circuit and the current induced in any 

 secondary oscillation circuit, placed near to the first, when the latter 

 is in exact syntonism, and also shghtly out of syntonism, with the 

 primary. Employing a formula of Bjerknes, we can find the sum of 

 the decrements D and d of the primary and secondary circuits by the 

 formula 



D + <^ = .(1+|)^^/^ 



where a is the current in the secondary circuit when it is tuned to a 

 frequency n, and A is the maximum current when the secondary 



0*7 



0-S 



1-1 



1-3x04 



9nN 1-0 

 Frequency n. 



Fig. 4. 

 Resonance Curve of Loosely Coupled Oscillatory Circuit. 



circuit is tuned to agree with the frequency N of the primary circuit. 

 For this purpose I modified the cymometer by including in the bar 



