MR. W. CROOKES ON MOLECULAR PHYSICS IN HIGH VACUA. 
649 
a e e\ a //', &c., the rays a ff and a g g' forming the shadow f'g’. Now let us 
suppose the idle pole b to be joined to earth : its positive charge is now very small, 
and its attraction on the negatively electrified molecular streams correspondingly less; 
they are, therefore, scarcely at all converged from the direction originally impressed 
on them by the pole a, and they follow the directions a f f", a g g", &c., the shadow¬ 
forming rays a f and a g now proceeding to f" and g", forming a wider shadow. The 
effect of the umbra and penumbra is caused by the idle pole not being permanently at 
the very low positive potential, but being rapidly charged and discharged, the wide 
and narrow shadows following each other so cpfickly that they appear to be simul¬ 
taneous. 
Fig. 9. 
a 
605. Experiments were tried with an idle pole and shadow tube whilst the exhaus¬ 
tion was going on. At such a rarefaction that the shadow can just be made out, it is 
quite sharp; touching the idle pole causes a small penumbra to appear round its shadow. 
When the exhaustion is at the best point for obtaining the green phosphorescence 
on the glass, the shadow is very sharp and well defined; and connecting the idle pole 
with earth gives a much wider penumbra, the width of the penumbra increasing with 
the degree of rarefaction. When the vacuum is so high that the spark has difficulty 
in passing, the penumbra (which becomes visible on uninsulating the idle pole) is much 
wider than before, and apparently eight or ten times as wide as it was at the lowest 
exhaustion at which observations were taken. 
If the object whose shadow is cast on the screen is a non-conductor (such as a piece 
of glass rod), its shadow remains constant at all exhaustions, no penumbra being 
visible, as it cannot be uninsulated. 
606. [Professor Stokes, whose suggestions throughout the course of this research 
have been most valuable, considered that much information might be gained by 
experimenting with an apparatus constructed in the following manner: The two poles 
of the tube (fig. 10) are at a and b. At c is a fluorescent screen ; d is a fixed bar of 
aluminium, and e is another aluminium bar hanging from a platinum pole f by a 
metal chain. The bar and pendulum are on opposite sides of the horizontal axis of the 
tube, as shown in the plan, so that when properly exhausted and the pole a made 
negative, the shadows of bar and pendulum shall fall side by side on the screen, as 
shown in fig. 10 a. On swinging the pendulum, the shadow alternately overlaps and 
recedes from the shadow of the bar (figs. 10b and 10c). 
4 o 2 
