Intelligence and Miscellaneous Articles. 289 



translation in the direction of the discharge, we beg to make the 

 following historical observations : — 



1. When von Zahn investigated the spectral lines of a discharge- 

 tube, in one case in a direction at right angles to the axis, and then 

 in the direction of the axis itself, the lines were not displaced, 

 although he was able to distinguish even the -^ of the distance of 

 the D lines. This displacement would have corresponded to a 

 velocity of one geographical mile in the direction of the axis *. 



2. Preliminary experiments of Tait, in which the spectra of a 

 tube containing bromide of carbon were found to be the same both 

 when parallel and at right angles to the discharge-tube, showed 

 that the velocity must be far greater, about 90 geographical miles 

 in a second. 



3. But that it can neither attain the value obtained by von 

 Zahn, still less a higher one, follows from the experiments of E. 

 Wiedemann and H. Ebert f , which were obtained both by the 

 method of Fraunhofer's minima of the second class, as well as by 

 the method of high interferences. 



4. The method employed by Prof. Trowbridge, as he himself 

 states, shows changes in the wave-length of 1/4000 ; that used by 

 Wiedemann and Ebert, of which, as well as of the other methods 

 mentioned above, Professor Trowbridge makes no mention, shows 

 a change of 1/834,000 of the value. The latter is therefore about 

 200 times as sensitive as the former. 



Erlangen, January 1891. 



ON HERTZ S ELECTRICAL VIBRATIONS IN AIR. BY E. SARASIN 

 AND L. DE LA RIVE. 



A very thin lead plate 2*95 metres broad and 2-8 metres high, 

 was suspended vertically. The primary conductor was horizontal, 

 and placed either at 5*7 or 9 metres distance from the lead plate, and 

 parallel to it ; its spark was formed in the perpendicular to the 

 centre of the plate. The space in front of the lead plate was ex- 

 plored by means of a circular resonator, which was displaced along 

 a divided scale on the normal to the lead plate. The scintilla at the 

 break is, as observed by Hertz, stronger when the continuous part 

 of the circuit in which the induction preponderates is in a loop, 

 and the break is in a node, whereby for each node the break 

 can be turned to or away from the reflecting lead plate. Close to 

 the reflector is a node. With various primary conductors and 

 resonators of 1, 0*75, 0*5, # 36, 0*25, and 0*20 metre in diameter, 

 almost the same intermediate spaces were found as along wires. 

 The velocity of propagation through air is thus essentially the same 

 as with wires. In this case also the resonance is a multiple one ; 



* Wiedemann's ' Electricity,' vol. iv. p. 580. 

 t Wiedemann's Annalen, vol. xxxvi. p. 553, 



1889. 



