588 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
This shows that although these molecular streams exist at a pressure of an inch of 
mercury, they are unable to force their way through the gas except to a very short 
distance, and that if the pressure be reduced the distance to which they can penetrate 
rapidly increases. So far, then, from being exempt from the action of the surrounding 
gas, they are very highly susceptible to its influence. 
Another experiment which shows how completely these molecular streams are 
subject to the ordinary laws of gaseous resistance was made by us while working with 
the tube described on page 24, into which air was allowed to enter in order to discover 
the superior limits of the pressure within the tube at which we could obtain phos¬ 
phorescence. When the phosphorescence had entirely disappeared it occurred to us 
to examine the effect of a magnet placed beneath the tube with its axis pointing in 
the direction of the tinfoil, so that the molecular streams, if any existed, would be 
moving towards it in directions nearly parallel to its axis. We knew that the effect 
of a mag-net in such a case is to constrict the molecular streams and cause them to 
move in a more compact body, so that if the disappearance of the phosphorescence was 
merely the effect of the loss of velocity of the particles through their having to pass 
through gas of such considerable density, the magnet might have the effect of enabling 
them to penetrate to the other side of the tube so as to produce phosphorescence. 
Accordingly when phosphorescence had completely ceased to be visible a strong electro¬ 
magnet was placed with its pole near the tube, diametrically opposite to the place 
where the wire from the positive terminal of the machine rested upon it. The 
experiment proved the justness of the conjecture, for while the magnet was in action 
a small bright and well-defined green patch was observed in the place where the 
phosphorescence would naturally appear, and this disappeared as soon as the current 
within the magnet was stopped. We then connected the tube with an A lvergniat’s 
air-pump fitted with a siphon gauge to measure the pressure within the tube, and 
repeated the experiments while the tube was in connexion with the pump. We 
found that without the assistance of the magnet we could produce phosphorescence at 
a pressure of a quarter of an inch of mercury, and with the assistance of the magnet 
at a pressure of at least three-eighths of an inch. Considering the very large 
diameter of the tube (something more than two inches) and the moderate magnetic 
power which we were using, these measurements, as well as those previously given, 
strongly confirm the estimate of the pressure in the case of the tube with potash 
mentioned above. 
There is another interesting experiment of a different kind which shows clearly 
how readily the moving particles lose their velocity on passing through the gas in the 
tube. A tube containing a number of loose films of glass of extreme tenuity was 
exhausted till it gave very fine stake, soft in outline, and also gave, with an air- 
spark, good phosphorescence. A discharge with a long positive air-spark was made 
to pass through it. On touching the tube with the finger (which, as we have 
already mentioned, has the effect of causing these molecular streams to pour off from 
