MOLECULAR PRESSURE, AND THE TRAJECTORY OF MOLECULES. 147 
exhaustion is such that the spectral and other luminous appearances in the tube 
are characteristic of the special gas under examination, the viscosity is little different 
from what it is at full atmospheric pressure. When, however, the spectral and other 
characteristics of the gas begin to disappear, the viscosity also commences to decline, 
and at an exhaustion at which the green phosphorescence is most brilliant, the 
viscosity has rapidly sunk to an insignificant amount. 
523. Another special characteristic of a gas at exhaustions between 250 M and 
35 to 40 IV!, is the molecular pressure generated, when light falls on a black surface 
immersed in it. In the diagram just referred to, the variation in the force of repul¬ 
sion is also given. This is seen to decline rapidly and almost to die out along with 
the viscosity: the phosphorescent phenomena become more brilliant as these special 
characteristics of the gas disappear. 
PROJECTION OF MOLECULAR SHADOWS. 
524. In ordinary vacuum tubes, illuminated by the induction current, the luminous 
phenomena follow the tube through any amount of curves and angles ; a hollow 
spiral becomes illuminated just as well as if the tube were in a straight line. Not 
so, however, the phenomena of green phosphorescence observed at these high ex¬ 
haustions. The molecular ray which gives birth to green light absolutely refuses to 
turn a corner, and radiates from the negative pole in straight lines, casting strong and 
sharply-defined shadows* of anything which happens to be in its path. In a U tube 
with poles at each end, one leg will be bright green and the other almost dark, the 
light being cut off sharply by the bend of the glass, a shadow being projected on the 
curvature. I can detect no trace of polarisation in the green phosphorescent light on 
the surface of the glass, except, of course, when it emerges at an angle through the 
side of the glass tube. 
525. The projection from the negative pole of a shadow rendered visible by 
a sharply-defined image on the side of the glass, seemed worthy of more close 
examination. A tube was accordingly made, as shown in fig. 13. In the centre, 
dividing the tube into nearly ecpial parts, is a screen of thin mica, a a, loosely fitting 
into a groove blown round the tube. A flat plate of uranium glass, b, about half a 
millimetre thick, is rivetted to the mica on one side, c is a star-shaped piece of 
aluminium foil attached to a platinum terminal, and d is a similar star made of mica. 
At each end of the tube are two terminals, e and h, being flat aluminium disks, and 
/and g, aluminium points. 
With this apparatus experiments were carried on during exhaustion. When 
the exhaustion is moderate (say 1 or 2 millims.), so as to show stratifications and the 
ordinary phenomena of vacuum tubes, the luminosity extends from one pole to the 
other. Thus, if e and g are the two poles, the light extends the whole length of the 
* ( See note, page 143. 
u 2 
