548 



NATURE 



[Ajfri/ 4, il 



Hence we get for the phase 83 this expression 

 cos mr _ sin mr 



tan 03=- 



sin mr 

 mr 



cos mr 



or, after transformation — 



l^ = mr - tan"^ mr. 



The line 83 of our Fig. S represents this function. One 

 sees that for this action the phase steadily increases direct from 

 the origin. Those phenomena, therefore, which indicate a 

 finite pace of propagation must make themselves apparent by 

 interferences even close to the vibrator. So it shows itself in 

 these experiments, and just herein consists the advantage which 

 we derived from this kind of interference experiment. But the 

 apparent velocity comes out greater in the neighbourhood of the 

 vibrator than at a distance, and it is not to be denied that the 

 phase of the interference must theoretically change less, but 

 notably more quickly than was experimentally the case. 



It appears to me probable that a more complete theory, one 

 which does not consider both conductors as vanishingly small 

 — perhaps, also, another estimate of the value of X — would here 

 afford a better correspondence. 



It is of importance that even on Maxwell's theory the experi- 

 ments cannot be explained without assuming a marked difference 

 between the velocity of waves along wire and their velocity in 

 free space. 



( To be continued. ) 



NOTE ON THE USE OF GEISSLER'S TUBES 

 FOR DETECTING ELECTRICAL OSCILLA- 

 TIONS. 

 A T the suggestion of Prof. Lodge, I undertook to repeat in 

 -^^ the Physical Laboratory of the University College, Liver- 

 pool, Hertz's celebrated experiments on electrical oscillations. 



In performing these experiments, I was searching for means 

 to make the effect of the electrical oscillations more easily ob- 

 servable, and I was induced to use for this purpose (i) Geissler's 

 tubes, in order to strengthen the visible effect ; and (2) the 

 chemical action of the oscillating currents (paper soaked in 

 solution of iodide of potash), in order to obtain a permanent 

 trace of them. 



For the present I will describe briefly the results of the use of 

 Geissler's tubes. 



In order to produce the electrical oscillations, I used a con- 

 ductor consisting of two zinc plates, about 41 "5 centimetres 

 square, suspended in the same plane 55 centimetres apart ; to 

 each plate was fastened a No. 6 copper wire, which was finished 

 off with a small brass knob. The two brass knobs were about 

 5 millimetres apart, and formed the sparking gap, as we shall 

 call it. As receiver of the oscillations, I used, like Hertz, circles 

 of No. 14 wire, 35 centimetres in radius. 



After the examole of Mr. F. T. Trouton (Nature, February 

 21, p. 391), I will call the first conductor a vibrator, and the 

 wire circles, or other receivers, resonators. 



The vibrator was connected with a small coil, 20 centimetres 

 long, supplied with an ordinary spring interrupter, and excited 

 by four secondary cells. 



If we connect one electrode of a convenient Geissler's tube 

 with either side of the sparking gap of the resonator, currents 

 pass through or into the tube, which lights up and so makes 

 the effect of the electrical oscillations on the resonator visible 

 even at a great distance. 



Of the few tubes which were at my disposal, I found that the 

 most convenient for this purpose was a small one with electrodes 

 8*5 centimetres apart, and filled with highly rarefied air. But 

 spectral tubes 20 centimetres long and filled with hydrogen, 

 oxygen, or nitrogen also gave good results. 



With the first mentioned tube I perceived a visible effect, when 

 the resonator was held horizontally in the plane containing the 

 wires of the vibrator, and with the sparking gap turned towards 

 it, at a distance of 4 metres from the vibrator. By this arrange- 

 ment all the phenomena described by Hertz ( Wiedemann's Ann. , 

 xxxiv. p. 160, 1888) about the direction of the electrical lines of 

 force can easily be shown. 



A very instructive experiment is to show the directions of these 

 lines by several resonators disposed round the vibrator. For 

 this purpose I suggest the following apparatus : — 



On a wooden frame mounted so as to be able to revolve on a 

 vertical axis standing under the sparking gap of the vibrator are 

 fastened several resonators, with their planes vertical and parallel 

 respectively to the directions of the lines of force and the spark- 

 ing gaps at the highest point. These resonators are supplied 

 with Geissler's tubes. In this position of the resonators all the 

 tubes will lighten up when the vibrator is working. But if the 

 frame with the resonators moves round the vertical axis, the 

 light of the tubes will become weaker, and, when the frame is 

 turned 90°, the tubes will become quite dark ; the planes of all 

 the vibrators in this position being perpendicular to the directions 

 of the lines of force. This change will occur inversely by 

 turning the frame from 90° to 180°. 



If, instead of one resonator, two are fastened to each pomt of 

 the frame, one perpendicularly to the other, both being vertical, 

 the changes in either of these will be contrary— that is to Say, 

 when the light in one set of the tubes becomes brighter it becomes 

 weaker in the set perpendicular to it and vice versA. Thus the 

 strength of the light is, so to say, proportional to the magnitude of 

 the components of the lines of force in the direction of the tubes. 

 If a disconnected Geissler's tube is held near the vibrator, it 

 begins in a short time to light up, owing to oscillatory currents 

 passing through it. The same effect is obtained if instead of 

 holding the tube by the hand it reposes on an insulating body. 

 This lighting occurs at all points near the vibrator, except about 

 the sparking gap. The tube becomes quite dark if the hand or 

 a conductor is interposed between it and the vibrator ; on the 

 contrary, the interposing of an insulating body causes no change 

 in the tube. The tube becomes more sensitive if a portion of 

 it is surrounded with tinfoil. 



In this way the existence of electrical oscillations in space can 

 be ascertained, and also the transparency of insulating bodies 

 and the opacity of conductors for electrical oscillations can be 

 demonstrated. 



When the two electrodes of a Geissler's tube are connected 

 with two different points of a resonator, the effect in the tube is 

 produced by the difference of potential of the two points. If 

 now we connect one point of the vibrator or the resonator with 

 one electrode of the tube, the other electrode hanging free in the 

 air or being earthed, we have an alternative current through the tube 

 whenever the potential of the point connected with the electrode 

 becomes different from zero, and thus the tube lights up. The 

 effect is strengthened if one portion of the tube is surrounded by 

 tinfoil. This is a very convenient arrangement for observing the 

 form of the electrical oscillation in conductors. 



If we investigate in this manner our circular resonator held 

 vertically before the vibrator, with its plane parallel to it and the 

 sparking gap upwards, we find that a tube hanging at the lower 

 end of the vertical diameter of the circle, opposite to the spark- 

 ing gap, remains quite dark, and lights up when moved to the 

 right or to the left of this point. The light becomes brighter till 

 the horizontal diameter is reached ; further on the light begins to 

 grow weaker till the sparking gap is attained, where the tube, 

 however, continues to lighten. The light becomes weaker when 

 the sparking gap is narrowed, and ceases when it is quite closed. 

 Thus we see that the circular resonator possesses one node at its 

 lowest point, two ventral segments at equal distances from the 

 node and the sparking gap, and two minima of oscillation one on 

 each side of the sparking gap. 



That a node is situated at the point opposite to the sparking 

 gap is also ascertained by observing that by touching this point 

 with the finger or by hanging from it a piece of wire or by con- 

 necting it to earth, no change is to be remarked in the spark of 

 the resonator. These manipulations, if applied to another point 

 of the resonator, diminish the spark. 



If the resonator is formed by a closed circle of wire, we find a 

 node at each end of the vertical, and a segment at each end of 

 the horizontal, diameter of the circle. The distance between 

 the two nodes being here no centimetres, the wave-length is 

 220 centimetres, while the length of the primary wave is about 

 880 centimetres. Thus the wave-length in the resonator 

 corresponds to the second higher octave of the fundamental 

 oscillation. 



If, instead of circular, we use linear resonators placed parallel 

 to the vibrator, we must be very careful to distinguish between 

 the effect produced directly from the vibrator in the Geissler tube 

 and the effect caused by the oscillations of the resonator. In 

 the case of the circular resonators, placed in the position above 

 described, one need not trouble much about the direct effect of 

 the vibrator, this being very small in the neighbourhood of the 

 vibrator' s sparking gap. 



