86 MR ROBERT COCKBURN MOSSMAN ON 



The mean annual temperature for the period was 47°'0, being highest, 59°'9, in July 

 and lowest, 36°'8, in January, a difference of 23°"1 between the mean monthly averages. 

 The warmest month during the five years was July 1734, 61 0, 8, and the coldest, February 

 1736, 33°'6, showing a range in the mean monthly temperatures of 28°'2. 



Rainfall. (Table XL IX.) 



The rainfall was measured from June 1731 to May 1735. The gauge was 28 

 inches in diameter, and was placed on the top of a garden wall. Precautions were 

 taken to prevent loss through evaporation, and the measurements were made, as a 

 rule, every day. 



The wettest month was March 1735, with 5'38 inches, and the driest, May 1733, 

 with only 08 inch of rain. 



Variability of Temperature. ( Table L. ) 



The mean daily temperature variability has been determined from the observations 

 made at 9 a.m. The average for the period was 3°'4, being greatest, 5°'3, in October 

 1731, and least, 2°"3, in September 1733. The mean varied from 4°'0 in December to 

 3°1 in May. 



Humidity. 



The hygrometer, or rather hygroscope, consisted of a piece of whip-cord with a 

 plummet appended. The cord was alternately baked in an oven and saturated with 

 moisture, before the scale was graduated. The operation was repeated four times until 

 the difference in the length of the cord when fully dried to its length when saturated 

 with moisture was constant at 4"5 inches. The point of greatest dryness on the scale 

 was fixed at five-tenths of an inch, the scale extending to five inches, which was the 

 point indicated in a completely saturated atmosphere. The instrument was inclosed in 

 the perforated cnse containing the thermometer. Although this method of observation 

 is crude, it may be of interest to give the results, as affording a tolerable approximation 

 to the seasonal distribution of this element of climate. The mean annual humidity on 

 this scale was 2*11, being at a maximum in December, viz., 2 '47 inches, and at a 

 minimum in May, viz., 170 inches. The seasonal variation in humidity was, therefore, 

 virtually the same as during the last thirty-five years. 



Wind Direction. 



The number of days the wind blew from the eight principal points of the compass is 

 shown in Table XXXII. which summarise the results of all the wind observations taken 

 in Edinburgh. During the five years under review, the mean percentage frequency was 



