302 Mr. Gr. A. Hemsalech on the Luminous Vapours 



raised to the necessary degree, there appeared in the place- 

 generally occupied by the continuous band of bluish vapour 

 a long series of approximately equidistant bluish patches or 

 striae. No vapour was seen above the plate except that 

 which rose vertically upwards from the small heap of car- 

 borundum. An upwards acting external magnetic field had 

 apparently no effect upon the striation. When the small 

 heap of carborundum was placed near the positive end of 

 the plate the bluish striae were observed only just beneath 

 that end, and did not extend towards the opposite one. 

 A contrary effect was observed with calcium carbide. 

 A small quantity of this material placed upon the negative 

 end produced only a short horizontal streak of bluish vapour 

 beneath that end. When, however, a little calcium carbide 

 was put near the positive end, flocculi of light grey vapour 

 formed beneath, and appeared to move towards the negative 

 end like a succession of closely-placed smoke puffs. When 

 larger quantities of carborundum are placed upon the plate 

 the individual striae are merged into a continuous bluish 

 band. Thus, when only a small heap of the substance is 

 used the vapour proceeding from it follows a narrow track, 

 the width of which is about equal to that of a stria. There 

 can be little doubt that the bluish striae are formed in a 

 similar manner as the bluish vapour — namely, by the 

 cooling of luminous vapour at the boundary surface of the 

 protected space. 



Further, the orderly arrangement of the striae seems to> 

 indicate that the encounters with cool air as the luminous 

 vapour moves towards the positive end of the graphite plate 

 take place at regular intervals. This fact suggests that the 

 particles concerned execute some uniform periodic motion,, 

 such as might result if the path of. the luminous vapour 

 proceeding from the small heap of carborundum formed a 

 helix round the plate. A path of this kind could be con- 

 ceived to be caused by the combined actions of the electric 

 and the magnetic fields set up by the heating current. Let 

 us consider the case of a negatively charged particle, such as 

 might originate from the electrolytic decomposition of a 

 compound molecule by the thermelectronic current in the 

 upper fringe, moving along the top surface towards the 

 positive end of the plate. In doing so the particle crosses 

 lines of magnetic force, and is deviated in the direction of 

 the magnetic south. Since the lines of magnetic force form 

 closed oblongs round the plate, the particle in moving 

 towards the south will tend to describe a similar figure ; but 

 as it also moves under the action of the electric field in a 



