100 
DE. C. CIIEEE; SOME PHENOMENA OF SUNSPOTS AND OF 
180 for the 3-year groups. To assist the eye, vertical lines divide the columns into 
blocks of five. The last column gives the mean value of the character figure from all 
days of the year dealt with. These annual means are nearly, but not quite, the same 
as the means that would be derived by combining the figures from the 41 previous 
columns. Entries in the table which exceed the annual means are in heavy type. 
Considering first the 11-year data, we have the primary pulse anticipated, with its 
crest at day n, the value for this day being more than double the mean from all days 
of the period. The 11-year mean is exceeded in the six columns n — 2 to ?^-t-3, and 
columns n—3, n + i, and n + b are also obviously afiected by the pulse. From columns 
+ 6 to w + 23 inclusive we have a practically dead-level value, with only such 
fluctuations as would naturally arise accidentally. In column n + 24 there begins a 
well-marked pulse, covering columns n-f-24 to n-f 33. The seven columns n-\-2b to 
n + ol give values above the ll-j^ear mean, and the crest comes between columns 
n + 27 and n + 2'^. Ihe values in these two columns are respectively 34 and 21 
per cent, above the 11-year mean. 
A coiisideration of the numerical results in the first line of Table XI, or their 
graphical representation in fig. 2, will probably remove the doubt which has I think 
