Meteorology. 43 



form of snow or sleet. The weather of the year was on the whole 

 of a changeable character, and tiiere was no protracted period of 

 drought such as was experienced in the previous year between 

 tlie 16th April and the 4th June, when less than an inch of rain 

 fell. The driest periods were eleven days between the 8th and 

 ■24th of May, and again between the 9th July and the 3rd of 

 August, during which only three-quarters of an inch fell. From 

 the 18tii October to the 10th November was also a dry period, 

 with only one day on which rain fell to the amount of 0-13 in. 



Hygrometer. — The mean of the dry bulb thermometer for the 

 year was 47-5 deg. ; wet bulb, 45 1 deg.; temperature of the dew 

 point, 42 6 deg. ; relative humidity (saturation = 100), 83 ; which 

 is about the average for the ten years of observation. The 

 monthly means of humidity ranged from 72 in May to 90 in 

 February and November. 



Thunderstorms were rather more frequent than usual, but, 

 with the exception of one on the 4th and 5th August and one in 

 December, were not severe or protracted. Besides these there 

 were three in June, one in July, and three in August. Hail fell 

 eight times as far as I observed, and lunar halos were seen seven 

 times. 



Wind. — With regard to the wind directions, the south-westerly 

 was as usual most prevalent. It blew 81^ days out of the 365. 

 The next most frequent was the easterly, which is down for 61^ 

 days ; southerly, 54 ; westerly, 47^ ; south-eastei'ly, 33|^ ; north- 

 westerly, 32 ; north-easterly, 24 ; northerly, 17 ; and calm or 

 variable, 14. 



Mr Andson said he observed from the Cargen record that the 

 number of days in which rain fell there was given as considerably 

 less than the number in his observations ; the respective figures 

 being 168 and 215. This was unusual, as the rainfall as a whole 

 was heavier at Cargen than in Dumfries ; but he was disposed to 

 think that the rain gauge in use there must be one which did not 

 measure hundredths or thousandths of an inch. The number of 

 days on which the amount of rainfall did not exceed one- 

 hundredth of an inch was 34, and on 13 days it did not exceed 

 two-hundi-edths of an incli. This made 47 days, which if added 

 to 168 would bring up the number to 215. Of course this did 

 not cause any difference in the aggregate amount of rainfall, as 

 the very small quantities not registered separately would be 



