614 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
tive and a piece of tinfoil immediately over the tetrahedron was connected with the 
negative terminal, thus forming the negative special which, as we shall see (and as is 
otherwise obvious) produces precisely the same phenomenon as the relief-positive. 
Just as before, the shadow of the wires of the tetrahedron was cast, sharply and 
clearly, upon the bright patch of phosphorescence which was formed thereby. A con¬ 
ductor was then brought, as before, in contact with the tube at one of the corners of 
the tetrahedron. No bulging out followed. The relief afforded by the conductor only 
permitted positive electricity to stream from the wires of the tetrahedron, and as this 
was unaccompanied by any molecular streams it was powerless to divert those that 
were passing across the tube.* 
Secondly. It is experimentally evident that the shadow is produced by a depression 
or deviation of the streams of molecules in question which come from the negative 
terminal. An examination of the appearance is well nigh enough to establish this, 
for it is easy to see how the rays are bent down, and how the parts upon which they 
are constrained to fall grow brighter thereby.! But there is an experiment which 
puts this beyond a doubt. In a tube kindly lent by Mr. Crookes there is an inter¬ 
mediate terminal, nearer to one end of the tube than to the other, partly composed 
of a flat piece of aluminium, the flat sides of which are turned towards the terminals 
of the tube, and in this flat piece of aluminium there chance to be some little holes 
(Plate 28, fig. 23). When the more distant terminal is made negative, a bright image 
of these small holes appears on the side of the tube in the midst of the general shadow 
of the intermediate terminal. When the tube is touched on the side on which this 
image appears, but at a point on the negative side of the image, it is found that the 
image is splayed out, part being thrown down or farther along the side of the tube. 
If the finger be moved a little to one side the splaying out moves towards the other 
side. j. This seems to show distinctly that the effect of the finger is to push away from 
it the rays going to form the image. 
But it is possible to prove in an even more direct way that the virtual shadow is 
* An interesting variation of these experiments is obtained by placing the finger beneath the tetra¬ 
hedron and casting the shadow of the apex on the opposite side of the tube. If the tetrahedron be a 
non-conductor a somewhat magnified shadow will be seen; but if it be a conductor this will be splayed 
out to an enormous extent, for the finger which supplies the negative electricity which permits the pro¬ 
duction of the relief phosphorescence also enables the edges of the tetrahedron to become sources of nega¬ 
tive discharge, and thus to splay out the shadows which they cast upon the relief-phosphorescence. 
t This is clearly shown when we form a vii’tual shadow at a spot that lies within an already existing 
virtual shadow. If the former would, if it alone existed, extend to any part of the surface beyond the 
limits of the latter it will still do so, and the bright line that marks the edge of the virtual show will 
have an angle at the spot where their boundaries would cross and will thereafter follow the outline of 
the outer shadow. This shows that each relieving system produces its effect independently of the other. 
J Similar effects are produced upon the edges of the shadow of the intermediate terminal itself, but the 
phenomenon is best observed in the way described in the text as the displacement or distortion of the 
small isolated bright spots which form the image of the little holes in the aluminium is more capable of 
being accurately observed. 
