TRANSACTIONS OF SECTION A. 473 
decay has been found to increase rapidly with the temperature, so that at low 
enough temperatures the rate of decay is inappreciable, and the leak appears to be 
a definite function of the temperature of the wire. In these cases, where the 
amount of the positive leak does not decrease perceptibly during the course of the 
experiments, it also is found to obey the above formula, ‘This is found to be the 
ease on reducing the observations given by Strutt’ for the positive leak from 
silver and copper in air and silver and copper oxide in hydrogen between the 
temperature limits of 175° C, and 331°C., and has been fully confirmed by a 
large series of numbers obtained by the author in investigating the emissibility 
imparted to a platinum wire by the luminous discharge. It also holds for the 
numbers given by Wehnelt? for the positive leak from the alkaline earths, and 
those given by Owen® for the positive leak from the Nernst filament. Generally 
speaking, the numbers given for the positive leak are not so accurate as those for 
the negative, owing to the complication caused by the falling off with time; and 
as we have no theoretical guidance here, we cannot be certain that in this case 
the index p in the formula A6ve—"/6 is equal to one-half. 
None of the results for the ionisation from hot bodies appear to be in disagree- 
ment with this formula. It holds for all temperatures where the ionisation is a 
definite function of the temperature, and its applicability extends over a far wider 
range than those cases where it can be shown to hold theoretically. In cases 
where the thermodynamic reasoning applies, the quantity 4 has a definite meaning, 
and is equal to twice the energy in calories required to set free a gram molecular 
weight of the ions from the hot substance. As to its magnitude, 6/2 is always of 
the order 10°, and varies from 2°67 x 10* for the positive leak from silver in air 
(Strutt, Joc. cit.) to 1°32 x 10° calories for the negative leak from hot platinum 
(H. A, Wilson, ‘ Phil. Trans.” A., vol. excvii. p. 415). 
It is not without significance that the rate of production of ions by hot 
substances is connected with temperature by the same formula as that which 
expresses the temperature variation of the velocity constant of almost all known 
chemical reactions. 
9. The Production of Radio-active Surfaces. By C. E. 8S. PHILuips. 
The opacity of radium bromide to its own radiations (and especially to the 
X-rays) makes it necessary, in order to get the greatest action from a specimen, 
that the material should be spread evenly over a considerable surface. And, in 
addition, it has been thought advisable to find some means of distributing the 
radium upon a surface capable of ready sterilisation, so that, after its application 
for therapeutic purposes, it may be rendered completely innocuous. The following 
method has been found to be rapid and effective. 
A short length of platinum wire was coated with a layer of radio-active 
crystals by dipping it into a concentrated solution of radium bromide, and after 
drying over a flame this wire was sealed into a long straight vacuum tube. A 
cylinder of thin mica was then slipped into the tube, so that the wire stood axially 
within it. Having sealed a second wire (to serve as anode) into the tube, the 
apparatus was exhausted, while the radium-coated electrode remained connected 
with the negative terminal of an induction coil. The passage of the discharge 
deposited in a few moments a fairly uniform radio-active metallic layer upon the 
mica. A surface so prepared may be cleansed without diminishing its radio-active 
properties by raising it momentarily to a red heat, and appears to be a suitable 
means of applying comparatively small quantities of radium for therapeutic or 
other uses where a considerable area is required to be radiated. 
10. The Kinetic Theory: Determination of the Size of Molecules. 
By J. H. Jeans. 
Phil, Mag. , VI., vol. iv. p. 98 2 Loe, cit. 8 Loe, cit. 
