084 



MR ROBERT COCKBURN MOSSMAN ON 



The tables which accompany the text have been derived by immense condensation 

 from over one million observations, the re-tabulation of which was required as a 

 necessary preliminary. The values were entered on special forms having 366 columns 

 one for each day of the year and an extra one for February 29. After the daily and 

 annual means were computed, the excess or defect of each daily value from the mean 

 of the whole year was ascertained. The daily departures from the annual mean were 

 then smoothed by Bloxam's method, which consists in assuming the mean of the first, 

 second, and third days to be the true mean of the second day; and so on, the average of 

 continuous threes being calculated. Any sudden or accidental day to day difference 

 is obviously eliminated by this method. The results presented are based upon an 

 analysis of all the climatic elements ; a most essential condition, since from such 

 a comprehensive examination alone is it possible to show the mutual inter-relations 

 which exist among the weather elements. The results are graphically shown in the 

 plates which accompany the paper, while mean values for each of the seventy-three 

 five-day periods will be found in Table XXXII. 



Mean Barometric Pressure. (See Tables I. and II.) 

 The mean barometric pressure has been computed from observations made at the 

 rooms of the Royal Society, Mound, from 1840-51 and 1857-61. From 1862 to 1884 

 the readings were made at Leith, 2 miles N.E. of the former station, while from 1885 

 to 1894 the observations were again made in Edinburgh. Each observation was first 

 reduced to 32° and then to mean sea-level, while the instrumental correction, including 

 index-error, was applied to each of the 18,500 values. To reduce the observations 

 to standard gravity a further correction of +0*028 inch is necessary; this has not, 

 however, been applied. The reduction of the Leith observations was kindly made 

 under the direction of Mr R. H. Scott, F.R.S. No observations were made at this 

 station on Sundays until 1868, but the missing readings have been interpolated b) r 

 differentiating with Edinburgh. The hour of observation was 10 a.m. till 1861, 8 

 a.m. till 1884, and 9 a.m. since. The following are the extreme mean daily (not 

 absolute) pressures on the average of fifty years : — 



Month. 



Highest Inches. 



Date. 



Lowest Inches. 



Date. 



Range Inches. 



January, . 



February, 



March, 



April, 



May, . 



June, 



July, 



August, . 



September, 



October, . 



November, 



December, 





29-878 

 •921 

 •935 

 •983 

 30-013 

 29-992 

 •932 

 •890 

 •972 

 •913 

 •905 

 •948 



12 



23 



3 



30 



13 



16 



15 



18 



13 



4 



9 



27 



29-670 

 •704 

 •746 

 •764 



•815 

 •880 

 •820 

 •789 

 •772 

 •707 

 •628 

 •729 



30 



1 

 31 



2 

 18 



9 

 25 



8 

 29 

 24 

 26 

 19 



0-208 

 •217 

 •189 

 •219 

 •198 

 •112 

 •112 

 •101 

 •200 • 

 •206 

 •277 

 •219 





30-013 



May 13 



29-628 



Nov. 26 



0-385 



