Mr J. Aitken on Dust Particles. 
213 
1889 - 90 .] 
the transparency of the air, especially when one is constantly 
changing the position where the observations were made, and where 
the visibility of hills at different distances cannot he noted. Though 
these observations do lack accuracy, yet they point in a rough but 
perfectly distinct way to the conclusion that it is not necessary for 
the air to be saturated in order that condensation may take place ; 
they rather indicate that at almost all states of humidity the dust 
has some moisture condensed on it, and that as the humidity 
increases the amount of deposited moisture increases with it, and 
increases at a rate much quicker than the humidity, especially when 
the air is approaching the point of saturation. 
It may be remembered that in the Colmonell observations it was 
noticed that the transparency of the air depended on the number 
of dust particles in it; with 9000 particles per c.c. the air was 
extremely thick, with 5000 it was thick, and was clear only when 
the number fell below 1000. The Colmonell observations were 
made in winter, while the air was damp, and its humidity did not 
vary enough to have a great influence on the transparency. 
To illustrate the effect of humidity, let us turn to the observations 
recorded in this paper. Taking the Hyeres observations first, it 
will be at once seen that the humidity has a powerful modifying 
influence on the effect which the number of particles has on the 
transparency of the atmosphere. It is true that owing to local 
pollution the numbers observed at this station may be too high, and 
only correct for the air near the ground. We shall take only the 
Finouillet and La Plage observations as of any use, the others being 
too polluted from local causes. It will be seen from these that air 
with even more than 4000 particles per c.c. may be very clear and 
transparent if it is very dry — with that number of dust particles and 
a depression of the wet kulb of 10 degrees or more, the air was 
very transparent, as on the 21st March and 3rd April. It may be as 
well to note here that the observations taken on the 4th of April, 
when the number of particles was very great, are of no value for 
our present purpose, as on this day the wind brought the smoke of 
Toulon direct to the top of Finouillet, and while the air generally 
was clear, yet in the direction of Toulon it was thick. 
The effect of humidity is very clearly seen in the observations 
taken at Cannes. There was no very great difference in the number 
