G10 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
ordinary way, a second shadow of it can be produced by casting relief-pbospborescence 
upon it, and these two shadows will in general co-exist without to any great extent 
interfering with one another. In a similar way two shadows of the same object may 
be produced by placing two pieces of tinfoil connected to earth on the opposite side 
of the tube, one a little nearer the positive terminal and the other a little nearer the 
negative terminal than the object, so that the latter is within the relief-phosphor¬ 
escence produced by each of the pieces of tinfoil. Two oblique shadows in opposite 
directions will then be seen, though their definition is not so good as in the former 
case, since the two systems of molecular streams have under such circumstances a 
strong tendency to interfere with each other. 
XXII .—On the relief-effects in tubes of high exhaustion with a positive air-spark. 
II. Virtual shadows. 
If, while a discharge is passing through a tube of high exhaust with positive air- 
spark, we place the finger on the tube, the green light is seen to fade away from that 
part of the nearer side of the tube (i.e., the side on which the finger rests) which lies in 
the direction of the j)Ositive terminal, giving the effect of a shadow falling upon that 
part of the surface of the tube. As the shadows are produced, not by any object 
actually intervening in the path of the gaseous particles, but by a body affecting them 
from outside, we have termed them virtual shcidoivs. If the air-spark be small the 
region over which the shadow extends is bounded by a plane almost parallel to the 
tangent plane at the point where the finger rests and at a little distance from it 
(Plate 28 , fig. 18 ), but if the air-spark be large the bounding plane is inclined at a 
considerable angle to the tangent, and cuts the tube obliquely. But it is only in very 
rare cases that the virtual shadow extinguishes the green phosphorescence from the 
opposite side of the tube, or from the end of the tube round the positive terminal, 
although it will often diminish the phosphorescence in that portion of the positive end 
which lies towards the side upon which the finger is placed. 
A virtual shadow has a well-defined outline, the edge of which is generally brighter 
than the rest of the tube. It starts from the side of the finger nearest the negative 
terminal, but broadens considerably in the direction of the positive. Sometimes, when 
the air-spark is very large, it even seems to start from a point a little to the negative 
side of the finger, leaving that side of the finger surrounded by the outline of the 
shadow. The area of the shadow appears to be nearly, but not completely, deprived 
of the phosphorescent light. It is, however, difficult to judge how far the appearance 
of residual light within the shadow is due to reflexion, and how far to direct illumina¬ 
tion. It is probable on theoretical grounds that some phosphorescence remains within 
the shadow; and this is confirmed by observation, for when the discharge has recently 
commenced it is often difficult to trace the outline distinctly. It would seem as though 
the glass needed to lose a little of its sensitiveness to show the full influence of positive 
relief in producing this phosphorescent shadow. 
