December 2, 1909] 



NA TURE 



14: 



by a well-known theorem, be resolved into the sum of 

 two oscillations of different frequencies. Hence each 

 pendulum may be said to possess two rates of vibration. 

 The same thing happens in the case of two closely coupled 

 syntonic electric currents. If one circuit has free oscilla- 

 tions set up in it, the action and reaction of the circuits 

 generates oscillations of two frequencies. Accordingly, 

 when an antenna circuit is coupled to a condenser circuit 

 we have oscillations of two frequencies set up in it, and 

 waves of two wave-lengths radiated from the antenna. 

 The presence of these two waves can be detected either 

 by measurements made with the cymometer or by an 

 oscillograph vacuum tube. In the first case, all that is 

 necessary is to place a cymometer in proximity to the 

 antenna and vary its oscillation constant. It will be found 

 that there are two settings of the handle for which the 

 neon tube glows brightly, and the scale of the instrument 

 will indicate the wave-lengths of the two waves 

 respectively. 



Some instructive measurements of this kind have been 

 made by Prof. W. G. Pierce in a recent research, and he 

 has shown that the wave-length given by the formula; 

 which can be deduced from the theory of the operations 

 are in agreement with actual measurements (Fig. 6). 

 Another striking confirmation can be obtained by the 

 oscillograph vacuurnFtube, invented by Dr. Gehrcke, of the 

 Reichsanstalt, Berlin. This consists of a glass tube having 



200 400 600 800 1000 1200 1400 i6jo 

 Ai Metres. 

 Fig. 6.— Pierce's Experiments oil Inductive Coupling. 



two Strip electrodes in it nearly touching, which are made 

 of nickel or aluminium. The tube is filled with pure 

 nitrogen and exhausted to a pressure of about 10 to 20 mm. 

 If such a tube has a high voltage applied to its terminals 

 a glow light extends along the electrodes, the length of 

 which varies with the electromotive force. Hence, if the 

 tube is connected to a circuit in which an oscillatory dis- 

 charge is taking place, the glow light along the tube will 

 rapidly extend and contract. If the electrodes are 

 examined in a revolving mirror, making from fifty to a 

 hundred turns a second, the images of the glowing elec- 

 trodes corresponding to each oscillation will be separated 

 out, and if the oscillations are persistent or undamped we 

 see a series of short bright lines alternately above and 

 below a central line. If, however, the oscillations are 

 damped, then we see in the mirror a train of images each 

 decreasing in length (Fig. 7). On applying such an 

 oscillograph vacuum tube to the circuit of an inductively 

 coupled antenna, and examining in a revolving mirror the 

 image of the electrodes, they will be seen to present an 

 appearance as in Fig. 8, taken from photographs kindly 

 given me by Herr Hans Boas, of Berlin. These oscillo- 

 grams indicate that there are two oscillations present of 

 different frequency, producing an effect similar to beats in 

 music. Owing to the difference in frequency, the oscilla- 

 tions alternately reinforce and extinguish each other 

 NO. 2092, VOL. 82] 



throughout the period, and as this type of oscillograrn is 

 only obtained with an inductively coupled antenna, it is 

 a proof that in such a case there are two oscillations 

 present of different frequencies. A similar result has been 

 obtained by Prof. E. Taylor-Jones with low-frequency 

 oscillations in coupled inductive circuits by means of an 

 electrostatic oscillogram of his own invention. Looking at 

 these photographs, it will be seen that each represents a 

 single train of damped oscillations gradually dying away, 



Pi,-, y. — oscillogram of Damped Oscillation (Antenna not connected ) taken 

 with the Gehrcke Oscillograph Vacuum Tube. 



but that in each train of oscillations there is an alternate 

 waxing and waning of the amplitude, which indicates that 

 it may be considered to be composed of two superimposed 

 oscillations of different frequency (Fig. 9). 



.Accordingly, in the case of wireless telegraph antennse 

 inductively coupled, we have in general two waves radiated 

 of different lengths, and either of these can be made to 

 affect suitably tuned receiving circuits. These vvaves have 

 different damping and different maximum amplitudes. 



One of the disadvantages of close inductive coupling is. 



Fig. 3.— Oscillogram of Secondary Oscillation (A. 

 with Gehrcke Vacuum Tub 



onnected; taken 



therefore, that we must divide the energy given to the 

 antenna between two waves of different length. As the 

 receiving antenna is generally only tuned to one of these 

 wave-lengths, we then capture and absorb only the energy 

 conveyed by the waves of that wave-length. To meet this 

 difticultv it'has been the custom to employ a feeble coupling 

 between the circuits of the oscillation transforiner, so as 

 to generate waves of only one wave-length. The objection 

 j then arises that the energy conveyed to the antenna is 



