76 PROFESSOR A. SCHUSTER ON THE PERIODICITIES OF SUNSPOTS. 
beginning of the second and more important part of the series, in spite of the greater 
labour of calculation which this would have involved. But it was necessary to be 
able to compare the phases of the oscillations which the two halves yield separately. 
On the whole the balance of advantage seemed to lie with the method adopted. 
Columns 3 and 4 of fable I\. give the numbers obtained for the periodogram when 
the whole range is divided into two portions, and fig. 2 represents the numbers in 
a graphical form, the curve A being the periodograph belonging to the last 75 years. 
The result is striking. The two curves have no resemblance to each other. Looking 
at them, we might think that they illustrate two entirely different phenomena. The 
truth is that between 1750 and 1820 the 11-year period, though existing, had 
a slight intensity, being quite overpowered by two others with periods of about 
13f and 9|- years. It may be argued that the observations during the eighteenth 
century were too scanty and uncertain to allow of any such definite conclusion, but 
I do not believe that the decisive verdict of the periodogram can be disposed of in 
this way. I he main features of the curves are determined by the times at which 
the maxima occurred, and to a minor extent by the estimate of the activity at these 
maxima. The uncertainty which probably does exist near the times of the minima 
does not appreciably affect the result, as the periods of maximum activity as well 
