654 MR R. C. MOSSMAN ON 



the mean of the day, and the hourly difference of pressure, temperature, and humidity 

 between Achariach and Fort- William. During the five days under discussion it will 

 be seen that at Achariach the temperature was 2° "4 higher than at Fort- William, being 

 greatest (3°'5) at 5 a.m., and least (l°*3) at 9 a.m. The reduced barometer was also 

 higher than the corresponding sea-level pressure at Fort- William by 0*012 inch, the 

 greatest difference being 0*023 inch at 5 a.m., and the least 0*001 inch at 10 a.m. The 

 mean relative humidity was 11 per cent, less, the maximum difference being 15 per 

 cent, at 2 p.m., and the least 7 per cent, at 9 a.m. and 10 p.m. The hours of greatest 

 temperature excess were also those showing a large increase of pressure and decrease 

 of humidity, as compared with Fort- William, while the hours of least temperature 

 difference showed a relatively small pressure and humidity departure from those at 

 the sea-level station. 



With regard to the diurnal variation in the elements, the barometric results cannot 

 be compared, because, as two of the five days under review contained a minimum of 

 pressure, the adjustment for "midnight difference" fails to eliminate this irregularity. 



The diurnal variation in the case of temperature and humidity is very small at all 

 three places, being much less than in the true Fohn, because the sky was completely 

 overcast and covered with dense rain-bearing clouds. The temperature shows a 

 variation from the mean of the day ranging from +0*8 to - l°*l on Ben Nevis; at 

 Fort- William the values range from +0°*9 to — 1°*1, and at Achariach from + 1°*1 

 to — l°'l. The curves of the temperature are thus closely accordant, the most interest- 

 ing feature being a slight rise of temperature at Achariach from 3 to 7 a.m., during 

 the hours when the barometer there showed the maximum excess over Fort-William. 

 The humidity curves for Fort-William and Achariach are in general the reverse of each 

 other, a fall of humidity at the one place synchronising with a rise at the other, and 

 vice versa. On the summit of Ben Nevis the air was saturated throughout. 



The greatest temperature excess during the five days under review occurred at 7 a.m. 

 on 28th October, when the difference between Fort- William and Achariach was 5°*6, and 

 on this day the difference exceeded 5° at four hours, the mean for the twelve hours 

 ending noon being 4°*6. If the ordinary laws governing the distribution of temperature 

 held good, the Glen should have been 0°*4 colder than Fort- William, which indicates 

 an anomaly of 5° for the first half of 28th October. These results are of considerable 

 interest, showing as they do how, even in very wet weather, dynamic heating goes 

 on in closed valleys, protected on the windward side from the rain-bearing winds by a 

 range of hills which trap the most of the rainfall. There are, doubtless, many regions 

 where this process goes on on a large scale, notably on the borderland dividing the 

 Argentine province of Santa Cruz from the neighbouring Chilian Territorio de Magal- 

 laneo. In this region the contrast between the precipitation on the windward and 

 leeward sides of the Andean Cordillera is very great. Thus, at Point Gallera, in lat. 

 40° S., long. 73° 44' W., the annual rainfall is 105 inches ; and at Junin, on the same 

 parallel of latitude, some 120 miles to the east, on the lee side of the Andes, only 31 



