at Low Pressures. 
395 
When making a measurement of the Hall Effect the stopper 
G was first rotated until the electrometer indicated that E and F 
were at the same potential. The magnetic field was then applied 
and reversed, and the change in the P. D. between E and F on 
reversing the field was measured by the electrometer. 
The results obtained are shown in the accompanying diagrams 
Nos. 1, 2, 3 and 4 (pages 396, 397). An examination of the 
curves in these diagrams shows that the variation of the Hall 
Effect along the discharge is very similar to that of the electric 
intensity 1 . 
In the previous paper it is shown that in a uniform part of 
the discharge 
Z = \FLX {k 2 — k x ). 
According to this equation Z is proportional to X so that the 
curves for Z and X would be exactly similar if this equation held 
true in all parts of the discharge. 
In the striated positive column the Hall Effect falls to an 
almost zero value between the striae, whereas the electric intensity 
only falls to a comparatively small extent. 
It appears therefore that the difference between k 2 and k x in 
the dark spaces between the striae is very small. It is probable 
that when a molecule is ionised a free electron or negative 
corpuscle is split off so that the negative ion is of very small 
mass compared with the positive ion ; k 2 — k x is therefore at first 
very large. After a short time however the negative corpuscle 
probably becomes attached to a neutral molecule, so that the two 
ions are then of almost equal mass and k 2 — k x is very small. It 
seems therefore probable that ions are formed largely in the 
striae, and that by the time they get into the dark spaces, the 
negative ions which were corpuscles at first have become attached 
to molecules. 
This view is in agreement with the idea that each stria and 
its adjacent dark space constitute a to some extent independent 
portion of the discharge analogous to that formed by the negative 
glow and Faraday dark space. 
In conclusion I wish to say that my best thanks are due to 
Prof. J. J. Thomson for his kindly interest and advice during the 
carrying out of these experiments in the Cavendish Laboratory. 
1 For curves showing the variation of the electric intensity see H. A. Wilson, 
Phil. Mag., June, 1900. 
