THE METEOROLOGY OP EDINBURGH. 687 



In a recent number of the Scottish Meteorological Society's Journal (vol. ix. p. 

 224) Dr Buchan gave a paper on the "Mean Temperature of the N.E. of Scotland 

 from 1764-1894," the means being corrected to an elevation of 100 feet above the 

 sea. From 1764 to 1781 the observations were made by Mr Hoy, whose 8 a.m. readings 

 have been utilised for the decennium 1770-79. A comparison of Dr Buchan's adopted 

 means with the 8 a.m. values gave the following corrections to be applied in order to 

 bring them to the average of maximum and minimum thermometers at a height of 

 250 feet above mean sea-level. 



Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. 



+ l°-0 +l°-2 +0°-8 +0°-3 -0°-3 -0°-3 -0°-2 + 0°-3 + 0°-3 +0°7 +0°7 +0 o, 9 



The decennial means with the above corrections applied were then entered in the 

 table. Similarly the means of the 8 a.m. and 8 p.m. observations made from 1811 to 

 1820 were first compared with Forbes' adopted temperatures for the above ten years 

 and then with my own table of Edinburgh mean temperatures. The corrections, in 

 both cases, are virtually the same, showing that Forbes' interpolated values for the years 

 in question deduced by differentiation from the Dunfermline register, give a very close 

 approximation. The following were the corrections applied : — 



Jan. 



Feb. 



Mar. 



Apr. 



May. 



June. 



July. 



Aug. 



Sept. 



Oct. 



Nov. 



Dec. 



+ l°-2 



+ 1°"9 



+ 2°-4 



+ 3°-0 



+ 2°-8 



+ 2°-7 



+ 2°-4 



+ 2°-2 



+ 1°*8 



+ l°-6 



+ l°-4 



+ l°-3 



In view of the somewhat anomalous corrections given above a careful examination 

 was made of the adopted temperature values from which they were deduced. In this 

 connection the observations made at Gordon Castle, Kinfauns Castle, and other places 

 in Scotland were gone over and compared with the Edinburgh values, the accuracy of 

 which they confirmed. It will be seen that the above table of corrections differs 

 materially from that given for the reduction of Adie's 8 a.m. and 8 p.m. readings to 

 mean temperatures. The difference between the two tables may be accounted for by 

 the assumption that Adie's thermometer was slightly exposed to radiation during several 

 months of the year, while the Edinburgh Advertiser observations were probably made in 

 a shady court. 



The decennial means for the period 1841-50 given by Forbes have not been utilised, 

 as an examination of Adie's original register showed several blanks in the observations. 

 These were made good by interpolating values for the missing days from other Edin- 

 burgh registers. The decennial values for each day of the year were then tabulated and 

 the means entered in the table. Hence Table V., showing the mean temperature of each 

 day of the year at Edinburgh on the mean of 100 years, is made up of seventy years' 

 maximum and minimum readings obtained from registering thermometers, while during 

 twenty years observations were made at 8 a.m. and 8 p.m. from ordinary thermometers, 

 similar readings being taken for ten years at 8 a.m. only. These values, as already shown, 

 have been brought to the mean of the maxima and minima, so as to make the results 

 strictly comparable. 



