CYCLES 131 



1650 for more than 100 years, accompanied by a 28-year cycle, of 

 which traces are found to continue even in the late half (1775 to 

 1900), in which the sunspot cycle and its double prevail. The latter 

 condition extends from about 1750 to the present time. In the early 

 half also a 20-year cycle is faintly shown by a distinct alignment, as 

 marked in the explanation diagram. So in this record also we find the 

 11 -year cycle replaced by 20- and 28-year cycles during the dearth 

 of sunspots near 1700 and for a brief time after, that is, to about 1750. 



Cyclogram No. 5 gives an analysis of the sequoia record in four 

 trees, D-3, 12, 20, 23, which were selected for their excellence in show- 

 ing the solar cycle. The interval covered is the 400 years from 1450 

 to 1850 at a set period of 23 years, represented by the thread. The 

 change from the double sunspot cycle to the 10, 20-year cycle took 

 place near 1630. At about 1700 all three cycles (10, 20, 23, and 28) 

 begin to show. In the last half century or so, the 20-year cycle domin- 

 ates, which agrees with the "arcigram" of the sequoias mentioned a 

 few pages above. The dearth cycles (20 and 28) were forming by 1550 

 more or less, and they are the ones which prevail during the absence of 

 sunspots near 1700. 



The Flagstaff evidence of dearth cycles is shown in cyclogram 6. 

 Here it is easy to trace the double sunspot cycle from 1400 to its 

 end near the center at 1650. The 14, 28-year cycle enters at about 

 1550, but after 1700 it is practically lost, due to smoothing and the 

 great dominance of the 21-year variation, which continues to the end. 

 The 35-year variant begins not far from 1700. This cyclogram was 

 taken in 1921 from the original Flagstaff group, smoothed by 5-year 

 overlapping means; all the others shown are from original unsmoothed 

 plots or from Hanned curves. 



Flagstaff long record — Cyclograms 6, 7, and 8 show the analysis 

 of the long Flagstaff record (500 years used here) at three different 

 settings for cycle-length, 22.1, 14.0, and 7.0 years. The first, as just 

 described, shows the main features of the sunspot cycle to 1650 and 

 the 21-year cycle since 1700. The second gives more detail. The 14- 

 year cycle enters near 1500 and continues to the end. The 11-year 

 period, often double, may be traced from 1400 to well after 1600. 

 A 9- to 10-year cycle is evident from about 1650 to 1775 or so. Thus 

 the "extra" cycles (10 and 14) are clearly found connected with the 

 dearth of sunspots about 1700. 



The Flagstaff analysis at 7.0 years is given in cyclogram 8, but the 

 numerous short cycles shown are not so important and sure as the 

 longer ones already described. 



Arizona drought cycles — There is no doubt that a demonstration 

 of the periodic action of droughts would be of great value to the South- 

 west. Accordingly, in 1925 a "skeleton" plot of Arizona droughts, 



