THE METEOROLOGY OF EDINBURGH. 67 



the Pleasance, and, for a short time, at Mertown. They have been reduced and other- 

 wise corrected to the mean of the maximum and minimum by Dr Buchan, so that it was 

 only necessary to correct them to a height of 250 feet by applying a reduction equal to 

 1° for each 270 feet. After having the small correction of 0*6° applied they were 

 entered in the table. 



Considerable labour was involved in the reduction of the observations taken from 

 June 1781 to December 1821. It will, therefore, be necessary to go into the processes 

 involved in the reduction of the earlier registers with some degree of elaboration. The 

 best observations throughout this period are undoubtedly those made by Adie in 

 Merchant Court from 1795 to June 1805, the hours being 8 a.m. and 8 p.m. 



The uncorrected values for the months and the years are given by Forbes. They 

 have been brought to the mean of the maximum and minimum by applying the correc- 

 tions given in the first part of this inquiry.* The corrections there given were tested 

 by a number of methods, but the values were so accordant that it was decided not to 

 make any alteration. A comparison of the Edinburgh Advertiser 8 a.m. and 8 p.m. 

 readings from 1795 to 1804 with Adie's corrected mean values gave the following plus 

 corrections, which were applied to the Edinburgh Advertiser record from 1787 to 1806. 



Jan. 



Feb. 



Mar. 



Apr. 



May. 



June. 



July. 



Aug. 



Sept. 



Oct. 



Nov. 



Dec. 



0°-9 



l°-4 



r-8 



2°-l 



2°-2 



2'-2 



2°-0 



r-5 



r-o 



0°-5 



0°-4 



0°'6 



Some change was made in the exposure of the instruments in 1806, the corrections 

 applied from that time till 1821 being those already given. t The means for the period 

 1787 to 1831 have been computed, and are given in Table VIII. Another change took 

 place in the instruments or their exposure in 1824, but a fresh table of corrections was 

 not made, as the observations were not utilised after 1821. 



Another register is available for the period 1785 to 1816, the temperatures taken 

 "before sunrise" and "at noon" being given in extenso, in the Edinburgh Magazine 

 and afterwards in the Scots Magazine. The station was at Duddingston, near the foot 

 of Arthur's Seat, from 1785 to January 1793, "within one mile of the Castle of Edin- 

 burgh" from 1793 to May 1798, and then at Barnton, three aud a half miles west of 

 Edinburgh, till 1816. The means have been computed and are given in Tables IX. 

 and X. The averages utilised for the calculation of mean temperatures are those taken 

 before sunrise, some little doubt attaching to the noon observations, especially in hot, 

 sunny weather. The corrections were obtained by a comparison with Adie's and the 

 Edinburgh Advertiser records, the latter being the values for the five years 1788 to 

 1792. The corrections thus obtained were applied to the observations at Duddingston 

 from 1785 to January 1793. The observations taken within one mile of the Castle 

 from February 1793 to May 1798 were corrected by means of a comparison with Adie 

 for the three years 1795 to 1797, and those taken at Barnton till 1816 from a com- 

 parison with Adie for the five years 1800 to 1804. 



* Trans. Roy. Soc. Edin., vol. xxxviii. p. 686. t Trans., vol. xxxviii. p. 687. 



