782 MR J. Y. BUCHANAN ON THE 



The principle on which the dates were selected was: — For foggy weather, to take 

 spells of three or more whole days of continuous foggy weather, and continuous foggy 

 weather is defined by twenty-four consecutive entries of fog in the log of each day. 

 The suppty of foggy days seemed to be so abundant that the minimum length of spell 

 was able to be fixed at three whole days. When it became a question of selecting the 

 spells of clear weather it was necessary to adopt the hour as unit, and twenty-four 

 consecutive hours during which fog was not once entered in the log was adopted as the 

 specification of the spell of clear weather of minimum duration. It was not possible to 

 enforce the limitation that the twenty-four hours should all belong to the same day. 

 It will be observed that a clear day only means twenty-four hours free of fog, and 

 implies nothing with regard to the presence or absence of cloud overhead. 



Summit and Base. — Although observations have been made at the base of the 

 mountain since the observatory on the summit was established, and for the last ten 

 years a first-class observatory has existed at Fort- William, in this paper no account is 

 taken of the observations made at the base. After the meteorology of the summit 

 has been thoroughly studied by itself, and that of the base by itself, there will be 

 greater light for the study of the combined observations and more assurance of 

 the validity of the conclusions arrived at. 



Details of Method of Selection. — A list of dates selected from the years 1885-1897 

 w r as made. The principle of the selection was, in the case of foggy weather, to pick 

 out every block of at least three days of continuous fog ; in the case of clear weather, 

 to pick out every spell of at least twenty-four hours continuous clear weather. By a 

 spell of clear weather is meant one in which no fog was logged. The blocks of foggy 

 weather vary in length from three days to eleven days ; the spells of clear weather vary 

 in length from twenty-four hours to two hundred and eighty-three hours. 



This list of dates having been made, the barometer observations for every hour of 

 the days involved were copied from the Ben Nevis sheets into a series of note-books : 

 one note-book containing all the data belonging to a particular month for all the years. 

 Thus, one note-book contains all the January blocks of foggy and spells of clear 

 weather ; another, the February, and so on — a series of twelve books being thus made 

 up. In the case of the spells of clear weather the entries for the entire day in which 

 the spell begins and for that on which it ends were copied out (besides those for the 

 intermediate days), and afterwards the exact hours at which the spell begins and ends 

 were marked. Thus, in the case of a spell of clear weather extending from nine o'clock 

 on 15th August to two o'clock on 19th August, all the entries for 15th, 16th, 17th, 

 18th, and 19th August were made. The advantage of this arrangement is that it shows 

 plainly the barometric conditions prevailing at epochs of transition from one kind of 

 weather to another, and it has greater completeness. The same thing was done in the 

 case of the temperature, rainfall, tension of aqueous vapour, wind, cloud, and sunshine ; 

 so that we have a series of eighty-four note-books, containing records of the 

 meteorological elements mentioned above, for the selected list of dates. The dates of 



