May 13, 1915] 
NATURE 
287 


: \ : 
the number of which varies greatly from time to + a temperature change due to the mere alteration of 
time, can be removed by filtration of the air through 
cotton-wool, or, in closed vessels, by settlement with 
the drops formed by expansions. In general, these 
nuclei are electrically uncharged, and whatever their 
nature, are conveniently known as dust particles. 
C. T. R. Wilson has shown that in air recently 
freed from dust, with increasing supersaturations, 
the first visible condensation takes place on the small 
ions. It is now known that the circumstances of the 
condensation remain unchanged during intervals of 
time extending to days after the removal of the dust. 
The intermediate and large ions are eminently suitable 
nuclei for the condensation of water vapour, as their 
mobilities are largely affected by changes in the 
hygrometric condition of the air, so the results just 
mentioned indicate not only that these ions are re- 
moved with dust particles, but also that they are not 
produced in air once made dust-free. There is no 
doubt that the large ions are present in ordinary 
saturated air; it appears, then, that filtration removes 
some rigid nucleus without which at least the large 
jon cannot be developed. 
From the facts which have been stated, the picture 
of the large ion most readily formed is that of a dust 
particle round which water molecules are adsorbed 
to an extent depending on the vapour pressure, the 
whole being electrified by the attachment of a small 
ion. 
Some idea of the nature of the relation between 
mobility and vapour pressure which is to be expected 
in connection with such an ion, may be obtained by 
comparing, on simple thermodynamic lines, the work- 
ing of two Carnot’s engines, one with unit mass of 
a mixture of ions and water vapour as the working 
substance, and the other with unit mass of water 
and its vapour. The vapours are to be taken as per- 
fect gases, and it is to be assumed that the density 
of a vapour is small compared with that of the sub- 
stance in the corresponding denser state. With these 
assumptions the result is readily obtained that 
(pA, / Bowe = (P,/P.)”, when only the change of 
state is being considered. p and P are the values 
of the vapour pressures in the two engines at the 
same temperature, and 1 is the ratio of the latent heat 
of vaporisation of water to that of the fluid surround- 
ing the nucleus of the ions. It is convenient here to 
take m as the mass of the denser part of the sub- 
‘stance. The expression, which holds for all cases of 
adsorption, states that at two temperatures the mass 
adsorbed will be the same if the ratio of the vapour 
pressures, in equilibrium with the adsorbed fluid, is 
the nth root of the ratio of the saturated vapour pres- 
sures at those temperatures. It is the formula of 
reduction for adsorption observations taken at different 
temperatures, and a clue to the condition of the 
adsorbed moisture is to be obtained from the value of 
gv found necessary to make the observations fall into 
line. As the mobility of the ions under consideration, 
at constant temperature and air pressure, is constant 
if the mass of the adsorbed fluid remains the same, 
the formula is directly applicable to mobility deter- 
minations if m is taken to refer to the mobility re- 
‘duced to constant air density. 
Trouton, and Masson and Richards, find that the 
mass of contained moisture in the case of flannel 
and cotton-wool is a function of the relative humidity. 
‘This means that n is unity in the preceding expres- 
sion. m is also unmistakably unity in connection with 
the large ion, the determinations of mobility only 
falling into line if plotted against the relative humidi- 
ties. The result of such a plot is shown in Fig. 1. 
No heat change due to a variation of surface energy 
is involved in the value of n, so in these cases where 
w=1, as the heat per unit mass necessary to annul * value of the vapour pressure. 
NO. 2376, VOL. 95| 
state is the same as that required to keep the tem- 
perature constant when water evaporates, it may be 
definitely concluded that the molecules in the con- 
tained or adsorbed fluid are in the same condition of 
aggregation as those of water. 
In the case of the intermediate ions the determina- 
tions of mobility are not accordant enough to allow 
the value of to be found in this way with any 
accuracy, but the fit of the points to a line is on the 
whole better if the mobilities are plotted against 
vapour pressures than when set out against the rela- 
tive humidities.. This, according to the preceding 
expression, corresponds to the physically extreme case 
when » is equal to some large number, though, so 
far as could be inferred from the plot, m might not 
be greater than some small integer. In any case, 
here the latent heat of vaporisation of water is some- 
times greater than that of the adsorbed fluid. 
The result of the preceding line of argument, 
though not conclusive in the present instance, at least 
suggests the idea that the intermediate ion consists 
of a rigid core enveloped by a collection of water 
HUMIDITY 
RELATIVE 


MOBILITY— RECIPROCAL 
Fic. 1.—The relation between the reciprocal of the mobility of the 
large ion and the relative humidity. 
molecules existing as a dense vapour rather than in 
the liquid condition. 
Trouton, in 1907, made the interesting discovery 
that there are two modes of condensation of water 
vapour on rigid surfaces. If special precautions are 
taken in drying the surfaces, on exposure to water 
vapour adsorption occurs as a dense atmosphere of 
water molecules, in a state, perhaps, intermediate 
between that of a gas and that of a liquid. At any 
rate, a change to the liquid condition somewhat 
abruptly takes place in these circumstances when, 
according to Trouton, the humidity is about 50 per 
cent. in the case of glass, and about 90 per cent. in 
‘that of shellac. 
The fluid surrounding the nucleus of the inter- 
mediate ion is, no doubt, in a state corresponding 
to that of the moisture condensed at low pressures 
on carefully dried surfaces in Trouton’s experiments. 
Further evidence supports the preceding view of the 
ion. Fig. 2 shows the relation between the reciprocal 
of the mobility of the intermediate ion and the vapour 
pressure as deduced from a plot of the determinations. 
At a pressure of about fifteen millimetres the 
mobility decreases very rapidly with increase in the 
) Simvitaneous observa- 
