186 
Proceedings of the Royal Society of Edinburgh. [Sess. 
two most powerful factors in these phenomena. If the wet-bulb depression 
be more than 2°, the thickening is seldom great, and decreases with increase 
of dryness as the day advances. The worst conditions of all are : cloudless 
sunrise, wet-bulb depression 1° or under, and absence of wind after 
blowing from an impure direction. Under these conditions the air in- 
variably thickens either to a very thick haze or a dense fog, according to the 
humidity. If the air be nearly saturated, the sun soon gets hazed out and 
a dense fog is formed. Of course, on some mornings there are already 
dense fogs formed before sunrise, due possibly to night radiation, or they 
may be a heritage from the previous day ; and on these occasions, though 
the sun may not be responsible for their appearance, yet it causes them to 
become denser. Again, there may be fogs at sunrise, and these may tend 
to clear though the sun be shining if the wind should happen to rise and 
clear away the impure air. 
The history of these sun fogs varies greatly. If the wind does not rise 
they generally remain all day if the sky keeps cloudless. Sometimes they 
only thicken for a time, and as the day advances and the air gets drier they 
begin to dissolve, and sometimes the wind comes and clears them away. 
Returning to Table I., it will be noticed that no wet-bulb readings are 
given when the temperature is under 32°, as readings under these conditions 
are of little value. It will be seen from the column for velocity of wind that 
the velocity was on most occasions small, generally only a fraction of 1 on 
the usual scale of wind velocities. It should also be noted that, as shown in 
the columns for temperatures, the temperature was always higher at the hour 
at which the observations were made than it had been during the night. 
The fogging therefore would not be due to the air becoming colder. An 
examination of the observations in the table shows that on the days on 
which there was no sunshine there was no thickening, even when the other 
conditions were favourable, and that winds from west to north do not thicken, 
as they bring to the place of observation pure air, blowing as they do from 
thinly populated districts. The air, however, from all other directions 
comes from densely populated areas, and contains a large amount of the 
products of combustion of coal, and other impurities. I have shown in a 
previous paper * that at the place of observation the air is nine times more 
hazed when the wind is southerly than when it blows from the north-west 
quadrant, this being due to the impurities thrown into the air in its 
passage over the densely inhabited parts. The effect of sunshine on these 
impurities is shown in all the observations given in Table I. when the wind 
was southerly, the limit of visibility being in every case greatly reduced when 
* Proc. Roy. Soc. Edin ., January 1893. 
