ATMOSPHERE IN GREAT BRITAIN AND ON THE CONTINENT. 25 



an end to all advance. We must, however, always bear in mind that our tests only 

 show the condition of a thin stratum of air resting on the ground and tell us little of 

 the condition higher up. Even on mountains we do not test the upper air, as the air 

 resting on the mountain face is often only the lower air more or less diluted with the 

 upper air. It must also be remembered, that though the lower stratum is very different 

 from the air immediately over it, yet it is from this lower layer that the upper 

 atmosphere receives most of its dust and humidity, and a study of these at their source 

 may tell something of their future. 



The Eigi Kulm observations for 1890 show very clearly the ascent of the lower air 

 to the mountain top during the day. When the mountain slopes are exposed to radia- 

 tion at night, the air resting on them gets cooled, and a downward current is produced. 

 This downward current draws its supplies more or less from the pure air above, and the 

 air on the mountain top in the morning is pure. But after the sun is up, the mountain 

 slopes get heated, the direction of the current is reversed, and the air from the valleys 

 drawn to the top of the mountain. The Ben Nevis dust observations show the daily 

 variation very well. The large dust-counter fitted in the tower of the Observatory 

 enables observations to be made at all hours of the night as well as day ; and by getting 

 observations before and after the sun has risen, important information has been obtained 

 on this point. An examination of the Rigi observations will show that on all days, 

 except the 19th May, the number of particles was lowest in the morning, and that they 

 increased as the day advanced. It will be observed that the valley air had generally 

 arrived at the top of the mountain by mid-day. Of course, wind and cloud will have 

 great influence on this up and down movement, both on its amount and on the hour of 

 its arrival at the top. The observations made in 1889 on the Rigi do not show the 

 day maximum well, except on the 22d May. The reason for this may have been that 

 during the period of the 1889 observations, the lower air was comparatively pure, so 

 that though it may have arrived at the top of the mountain it was not recognised, as 

 it bore no indication of having passed through the valleys. 



Pilatus Kulm. 

 The 21st of May, the day after the Vitznau observations were made, was wet nearly 

 the whole day, and the morning of the 2 2d opened dull and cloudy. As a complete 

 change had taken place in the weather since the Rigi observations were made, it was 

 thought some information might be gained by testing the air in its altered condition at 

 a high level. Instead, however, of returning to the Rigi Kulm, Pilatus was selected for 

 the purpose. On the way up the mountain, we passed through irregular masses of 

 cloud. The first of these were met at an elevation of about 1500 feet, and on the top 

 we were surrounded by dense clouds, which continued all day. The wind during the 

 visit was extremely light and variable. It looked as if the movements were due to the 

 clouds surging up the face of the mountain. They seemed to rise sometimes on one 

 side and sometimes on the other. Many tests were made during the day, but it has not 



VOL. XXXVII. PART I. (NO. 3). F 



