Metallic Vapours in an Exhausted Space. 1s 
weight of a carbon cathode in a mercury are ; the mechanical 
disintegration of the cathode, however, interferes with 
securing easily interpretable results. 
These experiments and the qualitative results in respect to 
Faraday’s law, are founded on the assumption that the ions 
move in only one direction, and would be entirely invalidated 
if the ions obtained positive charges at the anode and carried 
them back to the cathode. On the other hand, by following 
out closely the experiments and comparing the data obtained, 
I may be able to either confirm or else disprove the theory 
of unidirectional flow. 
In connexion with this, and with the theory of the are in 
general, the investigation of the nature of the luminous 
vapour which spreads out from the cathode into the parts of 
the tube lying outside the arc-path is of exceeding importance. 
A number of experiments have been carried out with the 
idea of ascertaining whether this luminous vapour carries 
electrical charges, or whether it represents a conductive state 
of mercury vapour, capable of taking up electric charges. 
I expect to publish in a continuation of this article the 
results of these experiments. 
The action of the magnetic field on the arc has also been 
investigated. It is very peculiar and not easily accounted 
for. The observed phenomena can be briefly described as 
follows : the field and are being both horizontal and perpen- 
dicular to each other, the are is deflected up or downward, 
according to the common rule of the action of a magnetic 
field on a current. The deflexion downward is accompanied 
by a motion of the bright cathode-spot along the surface of 
the cathode in the direction of the current ; the arc is thus 
lengthened, and when it reaches the wall of the tube it digs 
into the mercury where the latter is in contact with the 
walls of the tube. If the direction of the current in the arc 
is reversed, the arc is pushed upward and tries to become as 
short as it possibly can, so that the spot on the cathode moves 
in the same direction as before (as is easily seen by making 
a sketch). Changing the direction of the field changes of 
course the direction in which the arc is deflected, as well as 
that in which the cathode-spot moves. It is very strange, 
however, that when the arc contracts, the digging of the 
spot into the mercury also takes place. The behaviour of 
the arc in a magnetic field can be partly explained by assum- 
ing it to be a flexible conductor capable of changing its 
length, both ends being capable of sliding along two surfaces, 
and that on the anode being much less mobile than that on 
the cathode. It is difficult to imagine a mechanical model 
Phil. Mag. 8. 6. Vol. 7. No. 38. Feb. 1904. I 
