196 AGGREGATION, COMPETITION AND CYCLES 



he finds the average cycle to be 9.7 years, while the sunspot cycle is 

 11.1, indicating that the two are independent. This is confirmed by 

 the lag of rabbit peaks after sunspot minima and maxima, as well as 

 by the low correlation between rabbit figures and sunspot numbers. 

 The abundance of the lynx was found to be correlated with that of its 

 chief prey, the hare, as shown by the earlier students of the subject. 

 It is interesting to find that the last year of great rabbit numbers for 

 the three recent cycles fell one year after the corresponding minimum, 

 namely, in 1914, 1924, and 1934, that the fluctuation from district to 

 district revolved about the year of minimum, 1933, and that peaks 

 appeared latest in the higher life-zones. 



MacLulich's work reaches higher quantitative values than any 

 previous study, as a result of the cross checking rendered possible by 

 his use of four methods, viz: trapping, censuses, hare transects, and 

 quadrats of droppings. This permitted assigning average numbers for 

 the rise and decline of the last maximum, as, for example, at Frank's 

 Bay, where the respective figures for hares were: 1932, 600; 1933, 

 1200; 1934, 100, and 1935, 70 per square mile. The estimates at Buck- 

 shot Lake show how rapid the drop may be in a single season, but at 

 the same time prove that the fall to the minimum requires 3-4 years 

 and that a fair breeding population must be left in or near an area 

 to produce the next rise. The correlation of numbers wdth the pos- 

 sible factors concerned placed the emphasis upon disease and food, 

 though it appears probable that both of these bear a complex relation 

 to weather and to each other. 



Cycles in Bird Numbers. As would be expected, studies of popula- 

 tion in birds are practically confined to those regarded as game, and 

 even for these the data are rarely quantitative in any strict sense. 

 The most notable exception is afforded by the Siberian nutcracker, 

 for which Simroth thought he had found a correlation between its 

 European invasions and sunspot maxima, but these were 2 and 3 

 years out of exact accord with the latter for the two occasions noted 

 (1908). According to Witherby (1920; cf. Elton, 1924:149), the sun- 

 spot cycle or its double is reflected in the massed migration of the 

 sandgrouse from the saline deserts of central Asia. Recently, Formo- 

 sov (1933) has pointed out a relation between nutcracker invasions 

 and decreased numbers in the squirrel skins taken in certain regions 

 of northern Russia, and he ascribes this to poor harvests of the seeds 

 of Pinus cembra. 



Cycles in game birds have been studied chiefly by Criddle (1930), 

 Leopold (1931), Leopold and Ball (1931), Middleton (1934), and 

 Wing (1935). The conclusions of Criddle as to the sharptailed grouse 



