ON THE SENSITIVE STATE OF VACUUM DISCHARGES. 
587 
and at the same time show how this clogging effect increases with the density of the 
gas in the tube, just as would be the case if streams of any kind of small particles 
were trying to force their way through it. 
In order to observe the effect of the resistance of the gaseous medium upon the 
molecular streams to which phosphorescence is due, it was necessary to have an 
arrangement by which these streams could be examined at various distances from 
their source without any alteration being made in the other circumstances of the 
discharge. To effect this the following experiment was devised. A tube was con¬ 
structed (Plate 27, fig. 10), having loose inside it a second piece of tube, whose external 
diameter was about 1^ inches, while the internal diameter of the main tube was about 
2 inches. When the tube was placed horizontally there was a distance between the 
two tubes of about half-an-incli on the upper side, decreasing down to zero on the 
under side, where the two tubes lay in contact. The arrangement adopted for pro¬ 
ducing the phosphorescence was that due to our assistant, Mr. Ward, referred to 
above, and with this arrangement we could, by moving the small patch of tinfoil 
upon the tube to a suitable spot, give to the streams of molecules thrown off from 
the interior surface of the outer tube any range we pleased from half-an-inch to zero. 
We then attached the tube to an Alvergniat air-pump and ascertained the maximum 
range at which relief-phosphorescence could be obtained at different pressures of the 
gas by moving the tinfoil about until we got to a position where phosphorescence just 
became visible on the outer surface of the inner tube. This gave us the distance 
through which the molecular streams occasioned by the impulsive inductive action on 
the exterior of the tube were able to force their way through the gas without having 
their velocity reduced below the limits necessary to produce phosphorescence. 
A series of precise numerical results would have involved an accurate determination 
not only of the pressure of the gas and the range of the molecular streams, but also 
of the quantity of elecfricity given off by the Holtz machine, as well as the length 
of air-spark used. But as our principal object w T as to show the existence of a 
maximum range dependent on the pressure, it will be sufficient to subjoin a few 
approximate results, which are, however, derived from a considerable number of actual 
observations. 
The pressures of gas (atmospheric air) and the corresponding maximum ranges at 
which phosphorescence could be obtained with the 12-plate Holtz running at 300 
revolutions per minute, and an air-spark of about half-an-inch in length, were as 
follows :— 
Pressure. Range. 
5 mi Him.12 millim. 
24 „ . 2 „ 
26 ,, .almost contact. 
