506 M. B. W. Feddersen on-the Electric Spark. 



the same resistance, the observed bands becoming longer, or 

 dwindhng to extinction, or sometimes bounded by a new spark, 

 or short series of sparks (fig. 18). The cases where a long 

 quite regular series of sparks was observed, occurred less often, 

 but still they were sufficiently numerous for observation. 



These two kinds of discharge offer also a difference worthy 

 of notice for the direct consideration of the spark discharge ; 

 and pi'obably we may venture to hope for the explanation 

 of many remarkable changes of light and colour in the electric 

 spark by a closer study of the manner of discharge. The con- 

 tinuous discharge, which plays a very suborchuate part by great 

 resistance, and must therefore be observed by means of a small 

 column of water, appears in such cases as a homogeneous spark of 

 moderate diameter, and is recognized by the car as a single dull 

 blow*. The intermittent discharge, on the contrary, gives rise 

 invariably to a hissing sound, and generally, instead of a single 

 spark, a broad pencil f of sparks often with distinct intervals. 



The separate bands of light in this pencil are clearly the 

 partial sparks passing one after the other, but which pass so 

 rapidly that they appear simultaneous to the ej^e J. The proof 

 of this is sufficiently obvious in viewing the image of the series 

 of sparks in the rotating mirror. 



Invariability of the course of the sparks. — If the spark-micro- 

 meter is not protected from the motion of the air caused by the 

 rotating apparatus, the first sparks almost always appear quite 

 straight, but they soon begin to bend towards the side to which 

 the draught blows, and such bending increases as the discharge 

 continues. The change of form is a continuous one, derived by 

 the one spark from its predecessor. Even when the line of the 

 spark is abruptly broken at one place, the angular bend is 

 assumed by the whole series of sparks (fig. 20). Such a 

 bending maj^, during the passage of the sparks, suffer a small 

 vertical displacement, so that it appears as an oblique line through 

 the series of sparks (fig. 21): this is undoubtedly attributable to 

 a small vertical motion of the air. All this appears to me to be 



* The concordant observation of Riess in liis latest investigations has 

 reminded me that I had neglected, in my first description of the experi- 

 ment, to draw attention to the difference of colour in the two species of 

 discharge. The spark of the continuous discharge (employing the smallest 

 resistances) assumed a yellowish colour, — while that of the intermittent 

 was red, with perhaps a shade of violet. 



t The term pencil {Buschel) is strictly inappropriate, because it ajipears 

 in reality as though the successive s])arks all lay nearly in the same j)lane. 



+ The discontinuity of the lightning observed by Dove on a rotating 

 coloured top, depends probably on nothing else than the occurrence on a 

 large scale of the above-described intermittent discharge. 



