DUST, FOGS, AND CLOUDS. 355 
smoke and fog-producing sulphur, it would be well for us thoroughly to in- 
vestigate their saving properties and weigh their advantages, lest we substitute 
a great and hidden danger for an evident but less evil. 
While we look upon fires and all forms of combustion as fog-producers, yet 
we must remember there is ever present plenty of dust in the air to form clouds 
and even fogs; fires simply increase the amount of the dust. Now it is 
evident that as the rain is constantly washing the dust out of the air, fresh 
supplies must therefore be constantly added. 
We have every reason for supposing that there are immense quantities of 
very fine salt-dust ever floating in the air. This is evidenced by the ever- 
present sodium lime that at one time so troubled spectroscopists. One source 
of the supply of this salt-dust is evidently the ocean, and it affords us another 
example of how very closely the phenomena of nature are interlinked. The 
ocean, which under a tropical sun quietly yields up its waters to be carried 
away by the passing air, almost looks as if he repented the gift, when tossed 
and angry under tempestuous winds, as he sends forth his spray, which dried 
and disguised as fine dust becomes his messenger to cause the waters to cease 
from their vaporous wanderings, descend in fertilising showers, and again 
return to their liquid home. 
Parr LI. 
Since making my first communication to this Society on Dust, Fogs, and 
Clouds, many of the experiments have been repeated under different conditions 
and with improved arrangements of apparatus. I shall first give a short de- 
scription of the changes made in this direction, which seem to fill up some points 
wanting in the first paper, and shall then describe some experiments made in a 
department of the subject which I have only touched upon. 
We have seen that when steam is blown into dustless air there is no cloudy 
condensation, and that the vapour remained supersaturated till it came in 
contact with the sides of the receiver, on which it deposited itself. My next 
experiments were to determine to what extent dustless air can be super- 
saturated without the vapour condensing into drops—to determine whether 
vapour molecules can combine with one another to form a liquid, or whether 
they must have a nucleus to condense upon even when the vapour is very 
highly supersaturated. It is evidently very difficult to get a definite answer to 
this question, and I shall only describe the direction in which I sought to get 
an answer, the experiments not being sufficiently conclusive to settle the point. 
The first thing to be done was evidently to get quit of all “free surfaces ” 
of all nuclei of condensation, and the experiments have resolved themselves 
very much into questions of filtration, as I have not yet arranged any experi- 
