CAUSES OF ANIMAL CYCLES 191 



emigrants tliemselves, they belong almost wholly to the new generation 

 of the lennning (sec p. 192). 



The Sunspot Cycle. The evidence for a solar cycle in the weather 

 of the globe and in related biological phenomena is now so strong 

 that this must be regarded as by far the most probable primary cause 

 involved. It furnishes much the best explanation of the fluctuations in 

 numbers of plants and animals, and practically all recent studies of 

 these are concerned with it. As indicated earlier, the data as to 

 animal numbers in particular are mostly based upon the opinions of 

 untrained observers and are narrowly limited in both space and time, 

 so that they often show wide discrepancies as well as contradictions. 

 The solar cycle alone seems capable of bringing order out of the con- 

 fusion and of assigning proper values to the primary and secondary 

 causes concerned. The most definite expression of this cycle is found 

 in the fluctuation of sunspots, and hence the sunspot cycle is first 

 considered as the necessary foundation for organizing biological cycles 

 in a consistent fashion. 



It is generally understood that the sunspot cycle comprises two 

 extremes, a minimum and a maximum, separated by respective phases 

 of increase and decrease wdth variable and unequal intervals. The 

 average length is generally taken as 11.2 years, but within the period 

 of definite record beginning with 1749, this has ranged from 7 years 

 (1830-37) to 17 years (1787-1804). Since the last date given, the 

 variation has been much less, namely, from 10 to 13 years. In addi- 

 tion to the more or less regular progression between extremes that 

 characterizes it, the intensity varies greatly at both maxima and 

 minima. The highest maximum recorded was an average of 154.4 

 spot numbers for the year 1778, the lowest 45.8 for 1816; for the 

 minimum the respective values were 11.4 for 1766 and 0.0 for 1810. 

 The five maxima from 1750 to 1787 varied from 83.4 to 154.4, and 

 the three from 1804 to 1830 from 45.8 to 71; the highest successive 

 maxima were 138.3 for 1837, 124.3 for 1848, 95.7 for 1860, and 139.1 

 for 1870. From this date, relatively low and high maxima have 

 alternated, viz., 63.7 in 1883, 84.9 in 1893, 63.5 in 1905, 103.9 in 1917, 

 and 77.8 in 1928, indicating a double cycle. Furthermore, the sunspot 

 cycle differs in the length of the two phases, the interval from mini- 

 mum to maximum being usually several years shorter than for the 

 reverse phase (cf. Clements, 1916, 1929). 



For more than a century the relation between sunspots and ter- 

 restrial processes has been a subject of speculation and assumption, 

 but investigations of it were few and scattered until about 25 years 

 ago. New momentum came from Douglass's study of tree rings in 



