at Leith Fort every Hour of the Day in 1824 and 1825. 375 



only considers this as nearly the hour of mean temperature for 

 Edinburgh, but he regards the maximum as occurring " from 1 

 to half past 2, or even 3 o'clock * ;" and upon these principles he 

 has selected his three hours, viz. 8 A. M ., the time of maximum, 

 and 10 o'clock P. M. 



It appears, however, from Tables II., IV. and VI. that the 

 mean temperature of the 24 hours occurs at the following times : 

 * , 



Hours of Morning Hours of Evening 



Mean Temperature. Mean Temperature. 



1824, D' 13 8 26 



1825, 9 13 8 28 



Mean of two years, 913 8 27 



This very extraordinary agreement between the results of 

 1824 and 1825, shews how nearly we have approximated to the 

 true form of the daily curve, and how much confidence may be 

 placed in the general result. The following may therefore be 

 regarded as the leading points of the annual daily curve. 



M. ! ,3< 



Time of Minimum Temperature, a little before 5 A. M. 



Time of the Morning Mean Temperature, 9 13 A.M. 

 Time of Maximum Temperature, 2 40 P.M. 



Time of Evening Mean Temperature, 8 27 P.M. 



Interval between Minimum and following Maximum, 9 40 



Interval between Maximum and following Minimum, 14 20 



Interval between Morning and Evening Mean, 11 14 



Interval between Minimum and Morning Mean, 4 13 



Interval between Evening Mean and following Minimum, 8 33 



* Edinburgh Transactions, vol. iv. p. 214, and vol. v. p. 293, 294. Mr PLAY- 

 FAIR also adds, that 10 o'clock p. M. is " as near as circumstances will allow to the 

 time of greatest cold. 1 " These opinions prove how little was then known of the form 

 of the daily curve. It will be seen from Plate XVIII. that, both in the winter and 

 the summer curve, the maximum temperature is never before 3 o'clock, and that 

 10 o'clock at night is nearer the mean than the minimum temperature of the day. 



VOL,. X. P. II, 3 B 



