CYCLES 129 



plotting in the early times of sunspot maxima and minima, according 

 to Wolfer, we find that the telescope was invented and spots observed 

 just in time to show that it has always been the same maximum of the 

 double-crested tree-cycle that came at sunspot minimum. This in 

 itself is an interesting fact, for it intimates that the 11 -year cycle can 

 be called a well-defined period which the sunspots do not always follow 

 exactly. Apparently, the 11.30-year period and the sunspot cycle are 

 two different things. 



Seven years and multiples — There is further information in the 

 tree-records which perhaps adds light but does not fully solve the 

 solar puzzle. The Flagstaff area mean curve in figure 18 has some 

 large variations which are roughly solved without difficulty. A 21- 

 year cycle is very prominent and a 14-year and a 7-year cycle easily 

 evident. These values seem to be very close to 7.0 and its multiples. 

 The time of maximum of the shorter periods is about 1910 and for the 

 21-year period possibly 4 years later. This, however, is not a rigorous 

 solution. The amplitudes (from the mean value) increase from 5 or 

 10 per cent in the 7-year to double that in the 14-year and triple in the 

 21-year periods. This group of multiples of 7.0 becomes evident about 

 1663 with a large maximum of the 21-year type. It rather fails in the 

 1680 's, but after 1700 comes in regularly. Its beginning is thus con- 

 nected with the great dearth of sunspots described by Maunder (1922). 

 A single maximum of this apparent type occurred in 1479. 



Nine-year-plus cycle — A very crude graphic synthesis of these 

 periods has been made (and extended to 1980) whose resemblance to 

 the original curve is fair. This is shown in the central curve in figure 

 18. So a set of residuals between it and the original was plotted 

 and two interesting features appeared, as shown in the third curve of 

 Figure 18. A set of crests came in 1747, 1758, 1766, 1777, 1786, and 

 1794, all of which except the last came close to the sunspot minima 

 during that unique interval when the sunspot cycle averaged about 

 9.3 years in length. (The minima were 1755, 1766, 1776, 1784, 1797.) 

 The length derived from these crests is 9.4 years, which thus gives us 

 a terrestrial cycle related immediately to a definite solar cycle. It is 

 possible that the fairly common climatic cycle of 19 years is the double 

 of this solar cycle. From 1800 to 1880 the agreement between the 

 natural and synthetic curves is good, except for the extreme minimum 

 growth in 1847 and 1880, 33 years apart, and from 1880 to 1905 the 

 7-year cycle is practically absent, reappearing again subsequently. 



Historical changes — In a general way it is safe to say that the 

 sunspot cycle and its double and triple values are very common. The 

 double value has persisted in Arizona for 600 years with interruption 

 from 1630 to 1850 or thereabout, and in some North European locali- 

 ties it shows for the last century and a half covered by our tree groups. 



