Gitt— Theory of the Stratified Discharge im Geissler Tubes. 419 
These results seemed to prove that the cause of this curious 
formation, whatever it might be, was something more than the 
effect produced by air currents simply passing over the powder. In 
order to ascertain if this was really the case I repeated the experi- 
ment already described, having replaced the powder by smoke. 
The tube I made use of was 30 inches long and } inch in 
diameter. When the reed-pipe was made to sound at one end, 
the other end being closed, the smoke was seen to acquire the same 
formation as the powder. The ridges or strata of smoke were + 
or % inch apart, and sometimes nearer. ‘They were not all the 
same distance apart, but it was to be remarked that while one series 
consisting of, perhaps, a dozen, were separated by more or less 
equal distances, another series were, though at equal distances, not 
separated by intervals of the same length as in the first series. 
It required a strong steady note to produce these smoke strata. 
The instant the note ceased there was a rapid movement of less 
than an inch in the direction from which the sound was proceeding, 
just as if a pressure had been removed. This experiment, which 
I have not seen noticed before, seems to show that there are “rib- 
like ” formations produced in the body of the gas. ‘The resem- 
blance to the stratified discharge is striking (11.). 
Finally, I endeavoured to procure the powder strata in a 
Geissler tube, by means of a spark discharge, which was itself 
stratified. In order to obtain this result, it was necessary to 
employ a strong discharge. On former occasions I was unable 
to obtain this effect. However, on connecting the discharging 
pillars of a coil, giving in air, a six-inch spark with the insides 
of Leyden jars, the outsides of the jars being also in connexion, I 
obtained a discharge in a Geissler tube which was stratified. There 
was some lycopodium powder in this tube, which, by its move- 
ments, showed that the disturbance caused by the spark was 
considerable. The powder was arranged evenly along the tube, 
and, after a few sparks, it was found to be divided up into the 
usual “ rib-like” strata. These powder-heaps were not the same 
distance apart as were the luminous strata. The tube was so made 
that one of the electrodes was placed about 3 inches from the 
extremity of the tube, with the object of seeing the effect of the 
discharge on the powder not in the direct line of discharge. The 
powder in this space was influenced by the discharge in the same 
SCIENT: PROC. R.D.S., VOL. IX., PART IV. 2H 
