300 Mr. G. A, Hemsalech on the Luminous Vapours 



current could have any influence in shaping the outline of 

 the fringe. I am at present unable to offer a satisfactory 

 explanation for this peculiarity of the red fringe emission,, 

 but wish to call attention to the fact that most luminous 

 phenomena which owe their origin to the flow of an electric 

 current at atmospheric pressure present a sharp outline, 

 which they retain even when acted upon by magnetic forces 

 or, mechanically, by air-currents. Thus the initial discharge 

 which marks the path of the condenser current in a capacity 

 spark passing between metal poles is a sharply-defined 

 luminous band stretching across the space between the- 

 electrodes *. (The more diffused aureole which envelops the 

 initial discharge is formed by luminous vapour of the metal,. 

 or more probably its compounds with oxygen and nitrogen. 

 This part of the phenomenon develops at a later stage than 

 the initial discharge and does not convey the electric current.)' 

 Again, the pilot spark and the luminous arcs formed when 

 the oscillations in a self-induction spark are separated by 

 means of an air-blast are all sharply defined f. But even 

 when a strong magnetic field is acting upon a high-tension 

 discharge, the drawn-out luminous figure thus produced 

 presents a very sharp contour line along its outer edge J ; 

 and, as I shall show in a subsequent communication, the path 

 figures described by electrified potassium particles under the 

 influence of magnetic forces are all characterized by their 

 well-defined outlines. Thus it would seem that the sharp 

 borders of these phenomena cannot be caused by the 

 magnetic fields set up by the electric current ; there must 

 be some other force which tends to keep the individual 

 constituents of a stream of electrons or ions confined within 

 certain definite bounds. But whatever the true explanation 

 may be, the sharply-defined boundary surface of the red 

 fringe is quite in keeping with the characteristic features of 

 a luminous phenomenon, the origin of which can be traced 

 back to the flow of an electric current. 



Plate VIII. represents the red end of the fringe spectrum 

 as obtained with an unprotected plate of graphite containing 

 an unusual amount of calcium as impurity. The actual 

 vertical distance to which the fringe and the luminous 

 vapour rise may be gauged by comparison with the edge of 

 the plate, which is 0*95 mm. across. The bright individual 

 band \ 626 which rises high above the plate coincides with 



* Schuster and Hemsalech, Philos. Trans. Koyal Society, vol. 193, 

 pp. 189-213 (1899). 



t Heinsalech, Comities JRendus de VAeademie des Sciences, t. Ill 

 p. 1103 (1905). 



t Hemsalech, ibid. t. 152. p. 1086 (1911). 



