188 
nounced when the second method is used because of the 
multiple charges on the nuclei. Thus in a larger nuclei 
count, a smaller fraction of the nuclei are charged, 
because only a limited number of small ions are avail- 
able. As yet, there has been no quantitative explanation 
of this. 
2. There is a decrease of P with increasing humidity 
(Fig. 2). The nuclei, whose growth with humidity has 
amarked influence on P, have, according to Table III, 
radii of less than about 6 X 107 cm. Thus the be- 
havior of P reflects the growth of particles of this size. 
The quantity P gives us a good example of how 
careful we must be when we want to interpret statisti- 
cally determined relationships between the properties of 
natural aerosols and meteorological quantities. If, for 
example, we arrange measurements of the nuclei con- 
centrations and P values according to visibility, as 
Israél did in Frankfort, we obtain an increase in P 
values and a decrease of nuclei count with an increase 
of visibility. According to the relationship (1) described 
above we would have expected a decrease in P. How- 
ever, it 1s probable that, on the average, with mereasing 
visibility, both the number and size of the nuclei 
decrease, and the size appears, according to Table III, 
to have a dominant influence on P. 
Israél further points out that individual measure- 
ments of P show considerable fluctuations which are 
especially large in the vicinity of cities, that is, of the 
sources of nuclei [16]. He attributes this phenomenon to 
the slow establishment of the ionization equilibrium. 
From the preceding survey it is apparent that the 
study of the ionization equilibrium permits us to reach 
important conclusions concerning the aerosol. However, 
this possibility will be most fully realized only when we 
can take into account in the investigations the in- 
fluence on P of the size distribution in the nuclei 
spectrum. 
The Optical Effects of Nuclei 
Primarily effective for the scattering of light, and 
consequently the turbidity of the atmosphere, are the 
particles of the order of magnitude of the wave lengths 
of light or larger, that is, those with radii of approxi- 
mately r 2 4 X 10> cm. If the radius of the particle 
decreases, the optical effects diminish very rapidly. 
Thus, according to our method of notation, only the 
large nuclei, and not the Aitken-nuclei, are significant 
for the turbidity, unless the Aitken-nuclei are particu- 
larly numerous, or the atmosphere is extremely clear. 
Wright [46] used these facts in order to reach con- 
clusions concerning the composition of the aerosol, 
through extensive investigation of the visibility con- 
ditions at some English stations. He found that there 
were approximately 1000 table-salt nuclei with a radius 
of 2 X 10 cm in a cubic centimeter. This nuclei 
concentration varies relatively little with time and 
location and constitutes the maritime portion of the 
aerosol. Over land we have additional solid particles 
(such as soot) which can be detected with a dust 
counter. These can become so numerous, especially 
during wintertime and in densely populated areas, that 
CLOUD PHYSICS 
their optical effects are dominant. The many small 
nuclei of combustion (Aitken—nuclei) have hardly any 
effect on visibility. 
Dessens [8] found good proof for these conclusions. 
He measured the attenuation coefficients for various 
wave lengths and also calculated them from the simul- 
taneous determinations of the size distribution of the 
haze droplets, according to the formulas of Stratton and 
Houghton. The agreement between observed and com- 
puted values is good and at the same time shows that 
his method of measurement yields a quantitatively 
correct sample of the haze droplets. 
Recently, and independently, Siedentopf [35] and 
others have deduced the optical effects of the aerosol 
from Mie’s theory on a much more inclusive and exact 
basis. Siedentopf bases his calculations on an atmos- 
phere in which the visibility is 20 km, and attempts to 
find the composition of the aerosol which would cor- 
respond most closely to the dependence of the absorp- 
tion on the wave length, as well as to the scattering 
function (dependence of the scattered light on the angle 
of the incident ray). With pure dust aerosols (com- 
pletely opaque, index of refraction n = ©) or with 
pure droplet aerosols (n = 44), no agreement could be 
reached. This was possible only with ‘mixed aerosol,” 
composed of particles of both types in the proportion 
of 20 dust particles per cubic centimeter to 500 haze © 
droplets per cubic centimeter, with an average radius 
from 2 to3 X 10-* em. The addition of several thousand 
Aitken—nuclei with r = 5 X 10-* em has no effect on 
the results. These findings are in good agreement with 
the results of both Dessens and Wright, a fact which 
seems to indicate that these haze droplets, found far 
inland, are essentially of maritime origin. 
It is therefore clear why previous imvestigations 
yielded no definite relationship between the Aitken— 
nuclei count and visibility (according to [23]). The 
situation is different, however, for dust particles, as was 
shown by Rétschke [33]. 
The effect on visibility of the growth of haze droplets 
with humidity has already been discussed. Wright [46, 
47] has established a similar relationship between hu- - 
midity and visibility when the optical effect of the dust 
particles predominated. This can be explained by pos- 
tulating either a considerable accumulation of such 
particles during very humid weather conditions or a 
coagulative effect of humidity. 
Processes of Nuclei Destruction 
Aside from condensation itself, the following processes 
essentially diminish the number of nuclei: 
1. Sedimentation. 
2. Diffusion (adhesion to surfaces of all kinds). 
3. Coagulation. 
In all these processes, the mobility of the particles plays 
a role. The mobility of particles that are smaller than, 
or comparable to, the length of the mean free path of 
the air molecules (approximately 10~ cm), is 
