BEN NEVIS AND EORT-WILLIAM OBSERVATORIES. 695 



short-time variations of temperature in the original tables, whether these be accidental 

 or are real fluctuations in the temperature changes of the year, but, even so, does not 

 give a smooth curve. A table of five-day temperature means for Edinburgh has been 

 computed by Mr Mossman ;* and if that table be " Bloxamed " it will be seen how much 

 more regular the increase and decrease of temperature is in it than in Table IV., due 

 to the longer period available, namely, one hundred years at Edinburgh, as against 

 twenty at Ben Nevis and Fort- William. In Table IV. the highest and lowest tempera- 

 tures at Fort- William are 57°'3 for the five days ending 4th July, and 38° "3 for the 

 five days ending 14th February, a range of 19 o, ; while at Ben Nevis they are 41°'8 

 also for the five days ending 4th July, and 23° *1 for the five days ending 6th March, 

 a range of 1 8° 7. These figures are probably nearer to the true average of the warmest 

 and coldest days at each place than the highest and lowest daily temperatures in 

 Tables I. and II. quoted above. 



Mean five-day temperatures have also been computed for both places for the ten- 

 year periods 1884-1893 and 1894-1903. It is unnecessary to give the results in 

 detail, but it may be said generally that minor irregularities of temperature were 

 greater in each of the ten-year series than in the twenty-year, that these irregular 

 changes were often in opposite directions at corresponding dates in the two ten-year 

 sets, showing that they were merely casual fluctuations, but that the main features of 

 the annual changes of temperature were common to both series. These features are 

 (1) the delayed low temperature in winter, no rise taking place till after the beginning 

 of March ; indeed, at Ben Nevis the five days ending 6th March are actually the coldest 

 five-day period of the year; at Fort- William the temperatures lie between 38° "3 and 

 40° "0 from 16th December to 11th March, and at Ben Nevis between 23°*1 and 25° *1 

 from 26th December to 26th March ; (2) a corresponding though shorter delay at the 

 time of maximum temperature; at Fort- William the temperatures lie between 56°'0 

 and 57°"3 from 24th June to 23rd August, and at Ben Nevis between 40°*3 and 41 c, 8 

 for the same period ; (3) the difference of temperature between Fort-William and Ben 

 Nevis is greatest in spring — 16° "4 to 17°'5 in April and May, and is least in winter — 

 14°*2 to 15° "2 in November, December, and January. There are indications of double 

 maxima and minima at both places, the former at the beginning of July and in the 

 second week of August, the latter at the beginning of January and at the middle of 

 February, with, for Ben Nevis, a third minimum at the beginning of March. The 

 greatness of the excess, about 1°, at the beginning of July is probably accidental, and 

 would diminish on the average of a longer period than twenty years ; but there can be 

 little doubt that this is a time of maximum temperature at both stations, and that a 

 second maximum occurs some time in the first half of August. The first maximum is 

 probably due mainly to solar radiation, and the second to the convective warming of 

 the atmosphere from the land and sea surfaces. 



During the twenty years dealt with, the warmest and coldest individual days were, 



* Trans. Roy. Soc. Edin., vol. xxxviii. p. 746. 



