650 
MR. W. CROOKES OK MOLECULAR PHYSICS IK HIGH VACUA. 
This apparatus was tried many times with an induction coil, and also with a Holtz 
machine; but the results were not sufficiently definite to render it safe to draw any 
inference from them. By the kindness of Mr. De La Bate, I have lately had the 
opportunity of experimenting with his large chloride of silver battery, and the results 
now come out with great sharpness and with none of the flickering and indecision met 
with when working with an induction coil. 
The tube was so adjusted that the pendulum hung free, and a narrow line of mole¬ 
cular discharge passed between the edges of the bar and the pendulum, forming a line 
of fight between the two shadows on the screen (fig. 10a). When the pendulum 
was set swinging, and the idle pole f connected with it was kept insulated, the regular 
appearance of the moving and fixed shadows was very slightly interfered with. That 
is to say, the shadows followed the successive positions between those shown in figs. 
10 b and 10c almost as if they had been cast by a luminous point in place of the nega¬ 
tive pole. As the shadow of the swinging pendulum came very near that of the bar, 
the latter shadow seemed to shrink away, showing that the pendulum itself exerted 
slight repulsion on the molecules which passed close to its edge. 
The pendulum was again set stationary, as shown on the plan (fig. 11), the fine of 
fight separating the two being at f so that the appearance on the screen was as shown 
at fig. 10a. The pendulum pole was then connected with earth, and instantly the 
fine of fight which separated the poles moved from f to g through an angle, measured 
from e, of about 30°, the shadow widening out and getting indistinct at the same time. 
When the pole a was negative and b positive, the bar cl and pendulum e were each 
