80 Proceedings of Royal Society of Edinburgh. [sess. 
low hill about four miles distant, situated a little north of west ; 
second, the Ochils, at a distance of from 12 to 15 miles to the north ; 
and third, Ben Ledi, at a distance of 25 miles to the N.W. It is 
fortunate that a hill so far away as 25 miles is visible, but the great 
defect of this situation for observing haze is that the view is open 
only in a northerly direction from N.W. to N.E. In all other 
directions the ground rises and shuts out the distant view. The 
great disadvantage of this is, that while all estimates of haze should 
be made in the line of the wind, either up or down it, so as always 
to look through air coming from the same area, at Falkirk the 
limit of visibility of all clear air has to be estimated on the haze 
observed on Ben Ledi — that is, looking in a N.W. direction. So 
that, supposing the wind at the time is from S.W., we shall be 
looking across the direction of the wind, and though the observer 
may be surrounded by impure air, that impure air may not extend 
to Ben Ledi, which receives its air from the pure area to the west 
of Greenock when the wind is from the S.W. This evidently will 
give rise to an over-estimate of the purity of the air coming from 
the westerly part of the S.W. quadrant. 
As the wind gradually veers from S.W. to W., the northern limit 
of the impure air gradually comes nearer Fall^irk ; and when the 
wind has gone due W., all impure air passes to the south of it. 
This gradual approach of the northern limit of the impure air has 
been frequently observed, and may be seen whenever the wind 
veers from S.W. to W., if the air be at all dry. On looking N.W., 
Ben Ledi often becomes visible before the Ochils are seen, though it 
is further away, but it is seen through a shorter distance of impure 
air. After the Ochils have become visible, and while the wind is 
still south of west, the air to the N.E. is still very much hazed. 
The reason for this difference is, that while looking at the Ochils 
we see them across a short length of impure air between us and the 
purer air beyond, while in looking N.E. we are seeing through a 
greater length of the impure air. It is therefore evident that the 
estimates made of the haze in S.W. and particularly in W.S.W. 
winds must be too small, and the limit of visibility given in the 
tables too great. It seems also probable, for the same reason, that 
the estimates of the haze in N.E. winds are also too small. 
The boundary between the pure air from the uninhabited and 
