ATMOSPHERE OF GREAT BRITAIN AND ON THE CONTINENT. 651 



not till next day that the numbers became high on Ben Nevis. The results obtained 

 from this day's observations are so exceptional that they have been omitted from the 

 calculations to be referred to later. 



Coming now to the observations in Table XL , it will be noted that the decrease in the 

 limit of visibility is not quite regular. For instance, when the observations were made on 

 the 11th, the limit was much too great for the amount of dust. By reference to Table III. 

 it will be seen that there is only one observation during the period of the observations on 

 the 11th when the limit was so wide as 100 miles. The other observations gave only 50 

 miles. Further, the air was excessively dry, the wet-bulb depression being 10°, which may 

 in part have accounted for the great clearness. The limit on the 30th June was also too 

 wide. This seems to have been due to a change of wind taking place that day. It was 

 northerly the night before, and in the morning the air was pure and the limit of visi- 

 bility considerable in the forenoon, and it would appear that the impure easterly wind 

 had only imperfectly displaced the pure air, and the undue clearness on this day was due 

 to the upper air being still pure. This explanation is supported by the fact that on Ben 

 Nevis the numbers were low in the morning, rising in the afternoon, and became ver}' 

 high near midnight. Again, the limit on the 8th of July was too high. This may have 

 been clue to the limit entered in the tables being too high, as it was only so clear as to 

 give a limit of 50 miles for a short time, when the wet-bulb depression was as much as 

 10°. Half an hour before and half an hour after the 50-mile limit was taken, it was 

 only 26 miles. As the maximum limit observed has been generally used in these tables 

 it was thought better to keep to the rule, though in some cases it is not satisfactory. 

 On the other hand, on the 7th and 1st of July the air was too thick for the dust and 

 humidity. The cause of the thickness on the 7th was the heavy and lowering sky, with 

 little light ; further, the amount of dust at the high-level station was excessive. On the 

 1st July there were few clouds, but the amount of dust at high level was very great, 

 which partly explains the low limit of visibility on this day. 



The observations taken in 1892 have also been arranged in tables in a similar manner 

 to show the relation between the dust and the transparency. During the time observa- 

 tions were made in 1892 the weather was not very suitable for our present purpose, as 

 it was frequently raining, and thus fewer suitable observations were obtained than in 

 1893. The suitable observations for 1892 will be found arranged in Tables XII. and 

 XIII. There were only two observations taken when the wet-bulb depression was under 

 4°, and as the conditions did not remain settled for long, the observations have been 

 omitted. There is thus no table with observations with humidity under 4° in the 1892 

 observations. 



The observations in Table XII. for wet-bulb depression of from 4° to 7°, like the 

 observations in 1893, show the limit of visibility decreasing with the increase of dust. 

 Table XIII. also shows the same result. But the value of the limits of visibility are not 

 quite so satisfactory as they at first appear. For instance, taking Table XII. on the 

 18th July, with so small a number of particles as 126 per c.c, we are entitled to expect 



