520 DR BUCHAN ON THE 



summit has its temperature only in a comparatively small degree affected by terrestrial 

 radiation. Hence, during this time of the day, the mean of the two temperatures at the 

 base and the summit do not represent the mean temperature of the atmosphere between 

 the two Observatories, but a temperature lower than this. It thus happens that for 

 this lower temperature during the night a greater correction is supplied by the Table 

 of Corrections for reducing to sea-level the barometric observations of the summit, thus 

 giving too great a sea-level temperature for the top. 



Table III. — In the data given in the four groups of Table II., the figures simply show 

 the mean arithmetical differences between the sea-level observations at Fort- William 

 and the observations at the summit reduced to sea-level as explained above. But in 

 Table III., similarly arranged in four groups, as has been done in Tables I. and II., the 

 hourly means of each month are compared directly with the means of that month which 

 are given at the foot of the Table, together with the number of days' observations 

 from which they are computed. Thus the figures show the hourly departures from 

 the monthly means. The results of this comparison, set forth in the four groups of 

 Table III., exhibit the influence of the diurnal and nocturnal radiation on the 

 phenomena under consideration. Here again Groups I. and II. closely correspond, since 

 the differences of the observed temperature approximate to the observed mean data for 

 the thirteen years. 



But it is quite different with Groups III. and IV., which show the results obtained 

 during anticyclonic and cyclonic weather respectively. A comparison of the two 

 Groups as they appear in Table III. is very instructive. In Group III. the figures were 

 obtained solely from the days when the differences of the mean daily temperatures 

 at the two Observatories were only 12 o, or less, inclusive of the days when the 

 temperature at the summit was higher than at Fort- William. On such occasions it has 

 been shown that the humidity was small, sunshine large, and rainfall and cloud little or 

 none. On such days it is seen that the departures of the hourly values from their monthly 

 means very greatly exceed the departures shown in Groups I., II., and IV. On the 

 other hand, in Group IV. the figures were obtained from those days when the differences of 

 the mean daily temperatures at the two Observatories were 18° - or more. On such 

 occasions the humidity was large, saturation being of frequent occurrence, sunshine 

 small or nil, rainfall generally large, cloud large, and sky often continuously clouded. 

 The departures of the hourly values from their monthly means were small, the contrast 

 of Group IV. with Group III. being in this respect very striking. 



Results and Remarks. — The broad results are these : — (1) when the difference of the 

 mean temperatures of the day is only 12 o, or less, the calculated sea-level pressure for 

 the top of the mountain is markedly greater than at Fort- William, and the accompanying 

 meteorological conditions are anticyclonic, the weather being clear, dry, and practically 

 rainless ; (2) when the difference of the mean temperature of the day is 18°"0 or more, 



