H. V. Gill — Electric Discharge in Geisster Tubes. t05 



The electric discharge which gives rise to the stratified col- 

 umn being discontinuous produces periodic disturbances in the 

 gas; these disturbances are propagated as are sound w 

 studied in acoustics — exactly as in the case of Lodge's "mu- 

 sical sparks," if we may use the term. The stratifications are 

 the nodes caused by these waves. In other words, the phenome- 

 non of stratification is a form of Kundt's experiment in which 

 the heaps of powder which accumulate at the nodes are replaced 

 by the strata of molecules between which the discbarge is 

 taking place in a luminous form. 



This idea was suggested by the similarity of the aspect of a 

 stratified discharge and the arrangement of the powder in 

 Kundt's experiment, together with the fact of the oscillatory 

 nature of the electric spark. In subsequently examining the 

 literature of this question we found that an idea somewhat 

 similar to this had been indicated by more than one physicist,* 

 particularly by De la Rive. 



The following are some of the most striking points of resem- 

 blance between the stratified discharge and Kundt's experi- 

 ments : 



The appearance of the " concamerations " in Kundt's tube 

 resembles in a striking way a stratified discharge. " In a sec- 

 ond series of experiments M. Kundt has perfected his method 

 by replacing the lycopodium powder by silica in very fine 

 powder, obtained by the calcination of gelatinous silica ; it 

 forms at the nodes, no larger heaps, but " membranes" stretch- 

 ing across the whole section of the tube."f It is interesting to 

 note that the nodes formed in the case of singing flames when 

 the tube containing the flame is sounding by resonance in uni- 

 son with an exterior note, have been shown by Schaffgotsh to 

 be visible when a current of smoke ascends the tube. We may 

 mention that de la Kue had noticed this resemblance and re- 

 produced the figure from Tyndall's Sound, but he makes no 

 mention of a similarity of cause. 



Sometimes the powder in Kundt's tube takes a spiral form : 

 the same has been at times noticed in a stratified discharge. 



The diameter of the tube influences the distances betwi 

 the nodes in Kundt's tube because it influences the velocity of 

 transmission of the wave motion; the distance between the 

 strata in Geissler's tube is similarly affected. 



The general flow studied by Spottiswoode, above referred to, 

 seems to be excellently explained by this theory— an explana- 

 tion which, however, we propose with every reserve, A> has 



* Cf De la Rive, Ann'. Chim. Phys. [4], viii, p. 437, 1866. Gassoit PhiL I 

 1858-59, etc. Stafen, Journal de Physique, ii, p. 190, 1873. Neyr< oe if, Comptea 

 Rendus, lxxviii. p. 950, 1874: Ixxx, p. 1881, 1875; Ixxxn, L876. 



f Cf. Tyndall's " Sound." 



