EXCESS OR DEFICIENCY OF TEMPERATURE. 423 



positive or negative column according as the positive or 

 negative values are in excess for the year. If the theory 

 alluded to were correct, the figures in these two columns 

 ought to be zero. The 6th and 7th columns give the 

 number of months above or below the average in each year 

 respectively. 



During the 92 years there were 23 in which the excess 

 or deficiency of temperature found by summing up the 

 monthly differences amounted to over 20° for the year. 

 There were 28 years in which this excess amounted to from 

 10° to 20°. There were 19 years in which it amounted to at 

 least 5°, and 22 in which the amount was less than 5°. 



There are individual years in which the excess of tem- 

 perature during one part of the year is nearly compensated 

 by a corresponding deficiency during the remainder of the 

 year; but these years are very few in number, there 

 being only 5 in which the amount does not exceed 5°. 

 Taking the entire period, the observations distinctly prove 

 that, during a series of years, excess or deficiency of tem- 

 perature is not generally compensated during the same 

 year. 



Table II. shows the number of years which have a given 

 number of months above or below the average. 



This Table is to be read as follows : — For example, taking 

 4 months, we have 4 years with 4 months out of the 12 

 above the average; 11 years with 4 months below the 

 average, and 15 years with 4 months above or below the 

 average, taken together. 



In order to show the variations better, I have traced the 

 figures given in columns 4 and 5 of Table I., for each 

 year, in the form of a curve (Plate XIII.), taking the years 

 as abscissae, and the number of degrees excess or deficiency 

 as ordinates. 



The central horizontal line in the diagram represents 

 what the curve would become if the theory propounded 



SER. III. VOL. II. 2 F 



