TRANSACTIONS OF SECTION A. 407 
result indicates that the doubly charged ions carry double mass, while it 
would agree as well with the new theory of Sir J. J. Thomson and Wellisch, 
from which follows that a double charge reduces the coefficient of 
diffusion, even when the mass remains unaltered. In this case the co- 
efficient of diffusion may be still smaller than the measured value. 
Witha, 8, y rays and point discharge no doubly charged ions were found, 
In the case of point discharge, the formation of big charged complexes, due 
t» chemical processes, was proved to occur. 
10, The Recombination of Ions in Air at Different Temperatures. 
By Dr. P. Pures. 
Langevin has shown that McClung’s original experiments on this subject 
were quite spoiled by diffusion owing to the electrodes being too near 
together, and to the intense ionisation which existed at the surface of the 
electrodes, 
L. L. Hendren has shown experimentally that diffusion is almost 
negligible at the temperature of the laboratory and at atmospheric pressure 
if the distance between parallel electrodes is at least 2 ems. and the 
ionisation is uniform between them. 
In the present experiment Langevin’s method for finding the recombina- 
tion is adopted. The rays produced by a single discharge in a Rontgen 
bulb ionise a layer of air between two parallel electrodes, one of which 
is connected with a Dolezalek electrometer, and the other raised to any 
desired potential. The parallel electrodes are about 3 cms. apart, and 
the layer of ionised air about 1°5 cm. thick, thus leaving spaces of about 
‘75 cm. between the surfaces of the ionised layer and the electrodes. 
Under these circumstances, it is probable that diffusion may be safely 
neglected, but in any case more rapid diffusion would here make the 
recombination appear slower. ° 
The electrodes are surrounded by a jacket through which vapours from 
liquids boiling at known temperatures may be circulated. 
In essence the experiment consists in measuring the charges received 
by the electrometer when different fields are established between the 
electrodes. This is done first at the temperature of the laboratory and 
then at the temperature of the boiling liquid. From a comparison of 
the two series of readings, the ratio of the coefficients of recombination at 
the two temperatures may be calculated, the coefficient at the temperature 
of the laboratory being taken as unity. 
The following results have been obtained up to the present, and they 
show that a decreases somewhat rapidly with a rise of temperature : 
Temperature a Mr, Erikson’s values 
15° C, 1-00 ; 1:00 
100 “50 BL 
155 40 405 
178 "36 *38 (extrapolating a little) 
Since completing the above experiment, Mr. Erikson has published a 
paper on the same subject. Using ionisation from radium, and with an 
entirely different method, he gets almost identical results. These are shown 
in the third column of the table for comparison. 
11. The Terminal Velocity of Fall of Small Spheres in Air. 
By Professor Joun ZauEeny and Li. A. McKnenan. 
To test Stokes’s formula for air, the size, density, and the terminal 
velocity of fall of some spherical spores were determined, 
