416 Scientific Proceedings, Royal Dublin Society. 
half-wave-length of the note. An example of the same pheno-— 
menon is found in the case of a glass rod caused to vibrate longitu- 
dinally by being rubbed with a damp cloth. The residue of 
water on the rod is seen to divide up into little rings very near to 
each other, and such rods have frequently been found to divide 
up into little glass dises, showing that this formation was due to 
the vibrations of the rod. In general, bodies in a state of longi- 
tudinal vibration seem to show a like state of division and sub- 
division. 
In the paper already alluded to I suggested the theory that 
the stria might be the true nodes formed by the explosive effect 
of the spark, and illuminated by the light of the same spark :— 
“Having proved that there are gas waves in the Geissler tube, 
we can deduce the existence of the stratified discharge. Since the 
gas is strongly electrified the discharge takes place between the 
molecules (or, as some hold, between the atoms) : these inter-mole- 
cular discharges produce the illumination in the body of the gas. 
Add now the effect of waves in the gas: the gas at a condensation 
will be in a different condition from that at a rarefaction. When 
the molecules are crowded together at a condensation the number 
of inter-molecular discharges will be very much greater than at 
a rarefaction. There will hence be a greater illumination at a 
node; this is a stratum, but as the state of condensation at a node 
is intermittent the illunination will be so too.’”! 
The first part of that paper dealt with the mechanical effects 
produced by the explosion of the spark. The chief among these 
is the strongly-marked sound wave which proceeds from the seat 
of the disturbance. I then described some experiments showing 
the similarity between the ‘“ Kundt-tube”’ effect produced by 
sparks and the stratified discharge. 
The object of this note is to state the results of some further 
experiments, and to show the connexion between the results and 
some of the results obtained in Kundt’s experiment. We have 
seen that in addition to the well-marked nodes which correspond 
to the wave length of the note, there are also a series of dust heaps 
which are independent of the note period. Nowa further con- 
sideration of the conditions of the stratified discharge make it 
1 Loe. cit. p. 415, 
