450 



DR BUCHAN AND MR OMOND ON THE 



means for the complete months at these nine stations, the difference between the highest 

 and lowest hourly value on the mean of the year being as follows : — 





Lat. N. 



Highest. 



Lowest. 



Range 



San Jose 



. 9°-56' 



+ •031 



-•046 



•077 



Hong Kong 



. 22 18 



+ •045 



-•044 



•089 



Trieste 



. 45 39 



+ •011 



-•014 



•025 



Magdeburg 



. 52 8 



+ ■012 



-013 



■025 



Ben Nevis . 



. 56 48 



+ ■009 



- -015 



•024 



Fort William . 



. 56 49 



+ •009 



-•010 



•019 



Sodankylii 



. 67 24 



+ •004 



-•006 



•010 



Bossekop 



. 69 57 



+ -005 



-•005 



•010 



Jan Mayen 



. 70 59 



+ •005 



-•006 



011 



This steady diminution of barometric range with increase of latitude, and the fact 

 that the double diurnal oscillation of the barometer appears at all these stations on the 

 mean of the year, indicates that they may be taken as a typical set of stations. At Ben 

 Nevis, being a mountain station, the afternoon minimum of pressure is small, but it does 

 not wholly disappear. This remark applies to all places more or less under the influence 

 of land. But in the Arctic Ocean in the summer months, at a distance from land, the 

 diurnal barometric curve shows only one maximum and minimum in the day. 



Turning now to the tables for each place giving the mean range on fine and cloudy 

 days, it appears that at the tropical stations there is little difference between the hourly 

 departures on either fine or cloudy days from the normal hourly curves for the complete 

 months, but at temperate and Arctic stations, the fine and cloudy days differ consider- 

 ably ; indeed, in the Arctic regions the fine and cloudy day curves are so distinct that 

 the difference between them is greater than the amount of the normal diurnal range of 

 the complete months. 



The following table gives the difference in thousandths of an inch between fine and 

 cloudy days at the nine stations on the mean of the year ; the heavy figures indicating 

 that the barometer is higher on fine than on cloudy days, and the italic figures the 

 reverse. 





1 



2 



3 



4 



5 



G 



7 



8 



9 



10 



11 



12 



13 



14 



15 



16 



17 



18 



19 



■20 



21 



22 



23 



Mid- 

 night. 



San Jose 



2 



3 



5 



6 



6 



7 



6 



7 



3 



1 



1 



3 



6 



8 



9 



8 



7 



5 



2 















2 



1 



Hong Kong . 



2 



4 



7 



9 



10 



9 



10 



10 



8 



5 



3 



1 



1 



5 



7 



10 



10 



m 



10 



8 



7 



6 



3 



2 



Trieste . 



9 



4 



1 



6 



9 



12 



14 



15 



12 



8 



6 



7 



6 



4 



1 



2 



5 



7 



11 



10 



11 



11 



13 



17 



Magdeburg . 



U 



11 



5 



1 



4 



10 



13 



17 



18 



17 



15 



11 



8 



4 



1 



3 



5 



7 



S 



9 



11 



13 



15 



19 



Ben Nevis 



19 



12 



'.) 



4 



2 



3 



6 



8 



12 



13 



14 



14 



14 



13 



10 



7 



5 



2 



3 



7 



10 



15 



19 



24 



Fort William 



18 



10 



,.' 



4 



10 



15 



19 



22 



23 



20 



17 



13 



8 



3 







4 



8 



10 



11 



IS 



15 



18 



20 



25 



Sodankyla . 



5 



1 



3 



7 



9 



10 



12 



9 



7 



7 



4 



1 



1 



1 



4 



5 



6 



5 



7 



5 



7 



6 



S 



10 



Bossekop 



15 



8 



1 



5 



10 



15 



18 



18 



17 



12 



11 



7 



4 



1 



1 



2 



6 



9 



12 



11 



11 



11 



13 



17 



Jan Mayen . 



37 



31 



24 



21 



15 



5 



3 



8 



15 



20 



24 



26 



27 



26 



25 



23 



19 



13 



9 



3 



10 



20 



33 



46 



From this it appears that the least difference between the diurnal barometric range 

 on fine and on cloudy days is at the tropical station of San Jose in the dry climate of 



