518 DR BUCHAN ON THE 



Table II., Group III. — This group introduces a series of figures widely different 

 from the figures of Groups I. and II. The figures have been calculated wholly from the 

 observations of those days when the mean daily temperature at the two Observatories 

 did not differ more than 12°'0 from each other. This small difference of 12 o- or less is 

 one of the most striking accompaniments of the anticyclone. It may be here noted 

 that this small difference of temperatures is generally brought about by a relatively 

 greater elevation of temperature at the summit than at Fort- William — such as results 

 variously from, say, a raising of temperature at the summit but not at Fort- William, or 

 a decrease of temperature at Fort-William with no decrease at the summit at the same 

 time, etc. Conjoined with this reduced difference between the temperatures is a 

 great dryness of the atmosphere at the top, bringing with it necessarily the higher 

 specific gravity of such dry air. 



The number of days from which the figures of Group III. have been calculated 

 include 730 days, distributed somewhat irregularly through the year, the largest contribu- 

 tories being June 79 days, July 78 days, November 80 days, and December 79 days. 

 In March there were only 36 days, and in April 34 days. 



The characteristic feature of this group is that nearly the whole of the hourly values 

 are plus ; or, in other words, nearly the whole of the hourly reduced sea-level pressures 

 for the top exceed the sea-level pressures at Fort-William. The only exceptions occur 

 in. several of the months from about 9 a.m. to 4 p.m. In none of the months does the 

 minus sign appear from 6 p.m. to 7 a.m. The differences between the highest and 

 lowest mean values for the months are the following : — 



Inch Inch Inch Inch 



January, 0-013 

 February, 0-024 

 March, ' 0-028 



October, 0-027 

 November, 0-019 

 December, 0*011 



April, 0-027 July, 0-029 



May, 0-034 August, 0-029 



June, 0-031 September, 0-034 



Year, 0-024. 



These comparatively large figures are occasioned by the dry condition of the 

 atmosphere during anticyclones. 



Table II., Group IV. — This group includes the results of the cases when the 

 difference of the mean daily temperatures at the two Observatories is 18°'0 or greater. 

 Nearly all the days when the difference of the mean daily temperatures exceeded 21° *0 

 were not available for the inquiry, because in such cases the strong winds of the 

 cyclone accompanying these large differences caused ' pumping ' of the barometer, thus 

 rendering the barometric observations useless in this inquiry. It is further to be noted 

 that a very large number of the great differences of temperature were occasioned by a 

 depression of temperature occurring at the top, but no corresponding depression of 

 temperature at Fort- William at the time ; whilst in other cases an increase of temperature 

 occurred at Fort- William, accompanied by little or no corresponding increase at the top. 



The remarkable and important feature of Group IV. is that, excepting the four hours 

 in August from 9 p.m. to midnight, the whole of the hourly values for the months have 

 the minus sign ; in other words, virtually the whole of the hourly values of the sea-level 



