6 Proceedings of the Royal Physical Society. 



indicated a temperature of 51°'3, a reading equalled last year 

 (1896) on May 6, while the 22-inch thermometer read 51°0, 

 which was passed in 1896 on May 9. With the close of 

 June terminated the long period of deficient sunshine, only 

 712 hours, or one-fifth of the possible, having been recorded 

 since the beginning of September 1896. During the last 

 four months of 1896 the sun shone for only 150 hours, and 

 of the 562 hours recorded during the first half of this year, 

 April and May had 330. There were hardly any fine days 

 in the period under review, 104, or one- third, having no 

 sunshine, while 94 of the days on which the sun shone had 

 less than 20 per cent, of the total possible. Only one day 

 in nine during the period under review had more than half 

 the possible, and there were only nine days with more than 

 70 per cent, of the total possible sunshine. 



July. — The weather of July was unusually fine and warm, 

 especially during the first eighteen days. The conditions 

 during tliis time were exceptionally brilliant, the sunshine 

 recorder showing a total of 153 hours; the greatest previously 

 registered for the period since the record commenced in 1861 

 being 139 hours, in 1876. The rainfall for the first eighteen 

 days was only 0'15 inch, the lowest since 1869. The mean 

 temperature for the whole month, 5 9° '2, was the highest 

 since 1887, while the amount of bright sunshine, 232 hours, 

 was the greatest since 1885. Unusual nocturnal warmth 

 prevailed on the night of the 23rd to 24th, the minimum 

 recorded being 61°*4. During the last fifty-seven years only 

 ten warmer nights have been experienced.^ No rain fell 

 from the 9th to the 18th inclusive. 



August. — Although a good deal of rain — principally 

 thunderstorm — fell, the weather of August was both warmer 

 and sunnier than usual. Very warm weather prevailed 

 during the first twelve days, the mean of the maxima being 

 71°'9, and of the minimum 55°"3, these values being respec- 

 tively 6°'3 and 4°'5 above the average for the period under 

 review. A good deal of rain fell during thunderstorms, 

 which were rather frequent during the first eleven days. 

 Mean pressure was the lowest, with one exception (1891), 



^ Trans. Roy. Soc. Edin., vol. xxxix. p. 134. 



