590 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
known as the negative glow, with sufficient force to affect the radiometer.* Thus it 
would appear that all are agreed upon the point. It seems, however, to be thought that 
the phenomena in tubes of extremely high vacua show a freedom from this retardation. 
But it must be remembered that these vacua are estimated to be equivalent, or at all 
events comparable, to one millionth of an atmosphere. Now we have shown that at a 
pressure of a quarter of an inch or thereabouts we can get bright phosphorescence at a 
distance of two inches from the origin of the molecular streams. Is it a matter of 
wonder, then, that at a pressure of ten thousand times less than this we should find 
that these streams move through a distance of a few inches without appreciable 
retardation, especially when we consider that we have no certain means of detecting 
whether they are retarded or not ? The conclusion to be drawn from the above is, we 
think, that whether or not the ordinary gaseous laws suffer any modification in high 
vacua there is nothing in the phenomenon of phosphorescence in such vacua which 
entitles us to suppose that they do so.t 
* It must be borne in mind tbat neither tbe negative glow nor positive luminosity necessarily bar the 
passage of these molecular streams. They often (as has been mentioned in connexion with some of the 
previous experiments) pass through bright positive luminosity for a considerable distance, and very 
frequently penetrate through a clearly marked negative glow and render phosphorescent the glass behind 
it. On the other hand, the experiment with the tube containing the glass films shows that it is not 
necessary that there should be luminous matter in the tube in order to stop the molecules. When a 
sufficient air-spark was used there was no positive luminosity at many of the places in the tube where the 
experiments were made, and there was only a very faint haze in the remainder, and yet the retardation of 
which we have spoken was clearly manifested. 
t In a letter published since the reading of this paper Mr. Crookes has made a further statement of 
his views on the existence of a fourth or ultra-gaseous state of matter. 
We have never expressed any opinion as to the possibility of such a state, and have only dealt with the 
question whether the phenomenon of “ molecular streams ” furnishes evidence of its actual existence. 
It may readily be conceded that if we could “ by some extraneous force infuse order into the apparently 
disorderly jostling of the molecules in every direction by coercing them into a methodical rectilinear 
movement,” we should fundamentally alter the physical properties of a gas. But our experiments fur¬ 
nish no evidence that any such action as this takes place in the formation of molecular streams. Before 
the discharge the particles of the gas are moving about in a perfectly irregular manner, and the effect of 
the discharge is to impress on them a very rapid proper motion in a definite direction. But we see no 
ground for supposing that the lateral motions, and the collisions consequent thereon, are in any way 
affected. Every wind furnishes us with an instance of gas the particles of which have an average proper 
motion, but no one would contend that such proper motion lessened the number of collisions in the gas or 
interfered with its gaseity. And we can see no reason for regarding a molecular stream as anything else 
than an exaggerated form of the well-known electric wind, or a mass of gas with an extremely rapid 
proper motion the magnitude of which is evidenced by the heat imparted to the body on which the gas 
impinges. 
It is shown in the text that molecular streams can be produced with an intermittent discharge in tubes 
at comparatively high pressures where the gas is certainly in its ordinary state, and it may be added that 
in an intermittent discharge the periods of action are in all probability very small in comparison with the 
periods that separate them. Thus in all probability the greater part of the molecular stream would be 
composed of gas which had not been subjected to the direct action of the electrode, and which, therefore, 
