INTER-RELATIONS OF BAROMETRIC PRESSURE. 519 



pressures for the top are lower than the sea-level values for Fort-William. The 

 differences between the lowest and the highest mean hourly values for the months are 

 these : — 



Inch Inch Inch Inch 



January, 0-007 

 February, 0-019 

 March, 0-023 



April, 0-025 

 May, 0-026 

 June, 0-022 



July, 0-018 



August, 0-018 

 September, 16 



October, 0-009 

 November, 0-008 

 December, - 015 



Year, 0-015 inch. 





These figures show a considerably smaller range of daily differences of the pressures 

 than the corresponding figures of Group III., the reason being that in Group IV. the 

 accompanying weather conditions are cyclonic and therefore the vapour in the atmo- 

 sphere is much larger and cloud more prevalent than during an ticy clonic weather, 

 thus reducing the daily range of temperature. 



The figures in Table II. are simply the results of the method of reduction employed 

 in reducing the barometric observations of the summit to sea-level. In every case of 

 reduction the temperature employed was the simple arithmetical mean of the two 

 temperatures observed at the hour, one at Fort- William and the other at the top of 

 the mountain. Now, as regards the separate hours of the day, the mean of the tempera- 

 tures about 9 a.m. and 9 p.m. may be accepted as approximately near the mean 

 temperature of the day at the two Observatories, particularly in this case, as the 

 temperatures used are the mean temperatures deduced from many days' observations. 



But for the other times of the day this is not so. During the day, owing to the 

 effects of solar radiation, the temperature at both Observatories is raised above the 

 mean temperature of the day, the increase rising to a maximum about 3 p.m. and 

 falling to the average about 9 p.m. Now this increase of temperature, due to the 

 heated ground, extends only to a comparatively restricted depth of the atmosphere at 

 each Observatory, leaving the great mass of the atmosphere occupying the interspace 

 from the level of Fort- William to the summit of Ben Nevis very little affected. It 

 follows that during this time of the day, the mean of the two temperatures at the base 

 and the summit of the mountain do not represent the mean temperature of the whole 

 atmosphere between the two Observatories, but a temperature higher than this. Hence, 

 for this higher temperature, a correction too small is applied by the Table of Corrections 

 to reduce the barometric observations of the summit to sea-level, resulting in too low 

 sea-level pressures. 



On the other hand, during the other hours of the day, when, under the influence of 

 terrestrial radiation, the temperature of the air falls from the early evening to the daily 

 minimum shortly after midnight and then rises to the mean of the day in the morning 

 about 9 o'clock. Since the temperature of the ground falls under the influence of 

 nocturnal radiation in a much greater degree than the air above it, and as the degree 

 of cooling of the air diminishes rapidly with vertical distance from the ground, it follows 

 that the great mass of the atmosphere occupying the space from Fort- William to the 



