472 REPORT—1904. 
water only caused a thirtyfold increase, while that due to ordinary Chambéry tap 
water was of only fifteen times the normal value. The maximum of effect was 
reached after half an hour's bubbling. 
The most active sediments which I could get (being collected only at a certain 
distance from the actual source) showed activity comparable to that of the fanghi 
of Battaglia and the sediments of Bad Nauheim, tested by Elster and Geitel. 
A viscous matter, partly organic, called Barégine, which floats on the water, 
showed the strongest activity; it is to be noted that this substance is formed at 
the spot where the waters emerge from the rock, and is then carried away by them 
as it floats. 
A great amount of emanation was obtained by repeatedly drawing air through 
some heated sediment; its activity decayed according to an exponential law, 
falling to half its value in 3-2 days. The same result was given by emanation 
obtained by bubbling air through the waters. 
A disc of tinfoil charged at — 600 volts and kept in a metal vessel containing 
some 100 gr. sediment showed excited activity, the rate of decay of which has not 
yet been precisely determined. 
5. Plan of a Combination of Atoms having the Properties of Polonium 
or Radium.' By Lord Ketvin, F.R.S. 
6. Electrical Insulation in Vacwwm.? By Lord Kevin, F.R.S. 
4. Electrical Conductivity of Flames. By Dr. H. A. Witson. 
8. The Electrical Properties of Hot Bodies. 
By Dr. O. W. Ricuarpson, 1.4. 
In every case of steady ionisation by hot bodies which has been examined up 
to the present, the quantity of electricity C discharged in unit time is connected 
with the temperature @ by the relation C= Aéve'/@, In this formula A and } 
are constants which depend on the nature of the body and on the state of its 
surface, and are different for positive and negative electricity. The numeric p 
also is a constant which does not vary sufficiently from zero to cause the variation 
of 4” with 6 to be comparable with that of the exponential term. me 
In cases like that of the negative leak from hot metals, where the ionisation 
in a good vacuum is a steady function of the temperature alone, a formula of this 
type can be deduced thermo-dynamically. The only assumption made in the proof 
is that the phenomenon is reversible, z.e., that the ions given off behave like a 
vapour, and can be in equilibrium with the metal at a definite pressure. Deduced 
in this way the formula for the quantity of negative electricity given off by hot 
bodies is found to be Ate-/6. This formula has been verified by the author ® 
for the cases of platinum, carbon, and sodium. Wehnelt,* who found that the 
negative leak from hot platinum was greatly increased by covering the surface 
with the oxides of barium, strontium, and calcium, showed that the same law 
held for the discharge from these substances, whilst more recently Owen® has 
found a similar relation for the negative leak from a Nernst filament. _ 
The case of the positive leak from hot substances is essentially different from 
that of the negative. Here the current is not a steady function of the tempera- 
ture, but decays asymptotically with time at constant temperature. The rate of 
) Printed in full in Ppit. Mag., vol. viii. p. 528 (1904). 
Printed in full in Phit. Mag., vol. viii. p. 534 (1904). 
3 Phil. Trans., A., vol. cci. p. 497. 4 Drude’s Ann., vol. xiv. p. 425. 
° Phil, Mag., V1., vol. viii. p. 230, 
