mMUMOSPHERIC: ELECTRICITY ewer 
can hardly disengage any great quantity of radium emanation), 
the variations in radium content were not large, but of the same 
order of magnitude as observed over the sea. 
(b) 3. Natural ionisation in-closed vessels. 
The paper before mentioned contains the results obtained 
on the voyage of the Terra Nova from England to New Zea- 
land, of which the section on Natural Ionisation may be sum- 
marised as follows. 
(i) Variations in natural ionisation are due primarily to 
varying amounts of radioactive products in the air (disengaged, 
chiefly at least, from land surfaces). 
(11) These radioactive products are too diffusely distributed 
in the atmosphere to have any direct effect on the natural ionisa- 
tion, and only become operative when deposited in the neighbour- 
hood of the experimental station by precipitation, or by the 
earth’s electric field (potential gradient). 
(iii) There exists a minimum value to this natural ionisation 
(about 4 ions per c.c. per sec.) which has not by any method been 
reduced in value. This minimum would seem to be independent 
of the size or material of the retaining vessel, and may therefore 
be best ascribed to a spontaneous breakdown of the enclosed gas, 
very similar to the spontaneous breakdown of radioactive 
substances. 
From the above one can see that in places where the radium 
content of the air is very small (as over the sea), the variations 
in natural ionisation will also be small. Further work on the 
minimum value in the Antarctic (carried on in an ice cave at con- 
stant temperature) gave a value very slightly lower than that 
found over the sea, and showed, with a self-recording instrument, 
no variations greater than the probable errors of observation. 
For further details of the above, and in respect to measure- 
ments of the ionisation of the air, the reader is referred to the 
paper before mentioned. 
Samples of sea water were also collected from various depths 
for radium analysis, and will be worked out by Prof. Joly, but the 
results are not yet available. 
(c) Pendulum observations. 
It will be known to the general reader that the weight of any 
substance as measured by the pull of the earth upon it is not an 
