282 



note to the table for each month are given : — 1. The mean tempe- 

 rature of the coldest day of that month, with the day of the month 

 and the year, from 1814 to 1851; 2. the mean temperature of the 

 hottest day of that month, with the day of the month and year, and 

 the extreme difference of mean temperature of two days in that mouth; 



3. The day of the month on which the mean temperature was sub- 

 jected to the greatest change, with the minimum and maximum 

 mean temjDeratures, the year of the minimum and of the maximum ; 



4. the day of the month on which the mean temperature was sub- 

 jected to the least change, with the minimum and maximum m-ean 

 temperatures, the year of the minimum and of the maximum. These 

 results are embodied in the opposite table : — 



The author then treats of the method adopted to deduce the most 

 probable true mean temperature due to every day in the year ; and 

 concludes his paper by observing that there are periods of some dura- 

 tion which are very remarkable on account of the difficulty of as- 

 signing a physical cause for the anomalies apparent in the mean 

 temperature. Starting from the lowest temperature, in January, it 

 increases till the beginning of March, when, between the 3rd and 

 10th, not only is the increase checked, but there is a remarkable 

 depression of temperature. After the 10th, the increase proceeds 

 and is very rapid towards the end of April and the beginning of May ; 

 this rapid increase is rather suddenly checked, and followed by a 

 period of cold towards the middle of May : this period is very marked. 

 As remarkable a depression of temperature at this time of the year 

 seems to have taken place in France, having been noted in Paris 

 and at various locahties, some situated near the coast ; but it does 

 not appear that the equally remarkable rise at the end of April has 

 been noted. After the middle of May the numbers steadily increase 

 till the 5th of July, when they attain their maximum value. The 

 decline in the temperature towards the end of July is followed by 

 an increase at the beginning of August, after which the decline of 

 temperature is very regular till towards the end of November, when 

 a sudden and considerable increase of temperature takes place ; after 

 this the curve declines to its lowest point on the Sth of January. 



