204 
Proceedings of the Royal Society of Edinburgh. [Sess. 
no nuclei after being exposed to either light or radio-activity. But if we 
now place some S0 2 in S while the peroxide is in E, then on the gases mixing 
a dense condensation takes place on expansion, and there is also a dense 
condensation without expansion, showing that there has been produced an 
immense number of nuclei with a strong affinity for water. This is probably 
due to the formation of S0 3 by the action of the peroxide. We shall see 
later that it is possibly peroxide of hydrogen that is responsible for the 
thickness of the sun-fogs. Of course, all the peroxide formed by the 
combustion of the coal will combine at once with the S0 2 and be responsible 
for part of the early morning haze, but the oxidising agents formed during 
the combustion do not seem to be enough to oxidise all the S0 2 . 
Ozone. 
Ozone is another constituent of the atmosphere requiring attention. It 
has been found that if we introduce ozonised air, from an ordinary ozone 
tube, into the apparatus, it reacts powerfully on the S0 2 and produces a 
very dense condensation after being sunned. 
Products of Combustion. 
Having shown how unstable S0 2 is when prepared by artificial 
processes, let us now examine its character when produced along with 
the other gases in our fires and furnaces during the combustion of 
coal. In testing these products the same apparatus was used as before, 
and shown in fig. 1. The products were collected from a coal fire and put 
into a large glass flask about 1 foot in diameter. For collecting the gases 
a long glass tube about 5 mm. diameter was used. The tube was held with 
one end over a bright, smokeless part of the fire, while the upper end 
entered the glass flask. The natural heat convection carried the gases into 
the flask and displaced the air. The flask was then removed to the 
apparatus, and a tube connected with the open end of the filter dipped into 
the flask among the products, the top of the flask being stopped with a 
plug of cotton-wool to prevent the escape of the products and at the 
same time allow air to enter. The pump was then worked to draw the 
products through the filter into the apparatus in the usual way. The 
following are the results of the tests. If very strong products were used — 
that is, only air that had passed through the fire unmixed with other air, — 
then there was constant condensation in the test-flask without light and 
without expansion. That is, if strong products, even when free from dust, 
are provided with plenty of moisture, they seem to be able to set up a 
