Mr Scott o?i the Temperature of ' Shetland. 119 



phical Journal, and therefore the following only are submitted 

 to the public attention : — 



Table of the Mean Morning- and Evening- Observations Jbr 

 twelve Months y from June 1824. 



It appears almost unnecessary to explain how this result has 

 been obtained. It may be done, however, in very few words: 

 The temperature was registered at half-past seven o'clock every 

 morning, and half-past eight every evening. At the end of a 

 month, the morning observations were collected into one sum, 

 the evening into another, and each divided by the correspond- 

 ing number of observations. Thus the monthly mean tempera- 

 ture, at these hours, was obtained ; while that of the year 

 was, by a similar process, deduced from the mean observations 

 of twelve months. 



The times of registration were those proposed in 1823, by a 

 committee of the Royal Society of Edinburgh, who considered 

 the temperature of 7-J a.m., and 8^ p.m., as affording, several- 

 ly, a near approximation to the mean temperature of the day. 

 From the present observations, it appears, that, in the months 

 of June, July, and August, the morning mean exceeds the 

 evening by about 3° ; in September by V ; that in Oc- 

 tober and November they are equal; while in December 

 the evening exceeds the morning by 6' ; in January by 9! ; in 

 February by 1' ; after which the morning means again exceed 

 the evening, the difference increasing as the season advances. 

 The point at which the morning and evening mean tempera- 

 tures become equal, seems to be about 39° or 40° ; and if th*j 

 general winter range fell as much below this point as the sum- 

 mer rises above it, we should probably find the morning mean as 

 much less than the evening in winter, as it exceeds it in summer, 



