144 LIGHT SCIENCE FOR LEISURE HOURS. 



The resulting curve is remarkable in many respects. 

 The lowest point occurs in the first half of January 

 (with a singular, though slight, rise of temperature on 

 about the 10th or llth), between two equal depressions 

 a few days before and after. From the middle of 

 January there is a rapid rise to the beginning of the 

 last week, when there is a sudden sharp fall of tempera- 

 ture to the beginning of February. Then comes a fall 

 which causes one of the most marked depressions in 

 the whole curve, though strangely enough, when a 

 curve is run through the alternations in January and 

 February so as to leave as much space above as below, 

 the mean temperature for February is found to be 

 above rather than below the average. Still it remains 

 the case that the most marked interruption of the 

 upward rise occurs in February ; the greatest depression 

 corresponding to the time from February 10th to 

 February 14th. From February to the beginning of 

 April the rise is tolerably uniform, but from April 10th 

 to the 14th the curve which had been rising rather 

 sharply descends as sharply, and then ascends again, so 

 that this part of the curve is like the letter S placed 

 thus co. Again, the rise is steady or nearly so till the 

 beginning of May, but from the 9th to the 14th of 

 May we have a depression almost exactly like that 

 observed in the second week of April. Thence the rise 

 to about June 29 is uniform, though wave-like, i.e., 

 the slope is not constant, but, as we might expect, 

 diminishes as the crest of the wave is approached. In 

 the first week of July the curve is marked by two or 



