CYCLES 579 



All this raises the question of just what the term "cycle" should be considered to mean. 

 In discussing the characteristics of cycles in connection with weather phenomena Marvin°'° 

 wrote: 



Mere recurrences of natural phenomena without reference to the time interval between 

 recurrences do not necessarily constitute cycles. To be cyclic there must be systematic 

 recurrence ... If, as some claim, the variations observed in length and amplitude of 

 the cycle — such, for example, as the differing lengths and intensities of the sunspot 

 cycle — are not of an accidental character but are controlled by law. then the proponent 

 of such a view nmst not only prove the reality of the climatic cvcle he claims, but the 

 nature and reality of the systematic fluctuations in length and intensity must be fully set 

 forth . . . Proof of cycles by verification of predictions is the ultimate criterion, but 

 its utilization is obviously beset with difficulties. If climatic features are the composite 

 of the operation of several cyclical effects then past features must he capable of syn- 

 thetic reproduction from cyclical elements, thus affording proof of the cycles." 

 The same analysis could be applied equally well to the use of the term with respect to the 

 fluctuations in abundance of grouse and other wildlife. If a definition such as the above 

 were adhered to, grouse and many other species could not be classed as cyclic on the basis of 

 existing data. 



On the other hand, the fluctuations of grouse, for example, have shown a distinct tendency 

 toward rhythmic behavior, especially in certain regions. Furthermore, if accurate quantita- 

 tive data were available, the degree of variability might be found to have been even less. 

 In view of the absence of such data the possibility remains that the trend has followed a sys- 

 tematic pattern and. to prove or disprove it. highly accurate records will have to be kept for 

 some time to come. The data marshalled by Elton'"' with respect to mice, lemmings and foxes 

 afford some idea of what is needed for such a purpose. 



Likewise, correlation or lack of correlation with the fluctuations of other species or events 

 will require long series of data in each case. This was shown by MacLulich"" who found 

 that the relatively accurate records for the varying hare in Canada* for the 87 years from 

 1847 to 1934 were insufficient to demonstrate their lack of correlation with the sunspot 

 trend although this could be proven through the use of similar records for the lynx extend- 

 ing over 174 years (1751-1925). But it must be enqihasized that mere correlation of the 

 periodicities of two trends, even though quite perfect, does not mean they are related with 

 respect to cause and effect. For instance, mice near Ithaca, New York were found to exhibit 

 a cycle in harmony with that of Presidential elections'"". Some link connecting the two must 

 be identified. 



For the present, however, the element of variability cannot be plotted in advance. The 

 next "high" or "low" may occur a year or two before or after that representing the average 

 interval. Amplitude also may vary, as happened in New York in 1936-37, so that a seg- 

 ment of the cycle is largely obliterated. Furthermore, local populations have often been out 

 of step with the general average and the relative status of such populations is subject to 

 change at any time in response to changes in the local environment. 



With these limitations in mind, continued application of the cyclic concept to grouse may 

 be warranted. But predictions regarding the prospects for fall abundance in any particular 

 year cannot be made with any degree of exactitude on the basis of data assembled to date. 

 Rather such prospects, for the time being at least, must be judged during the same season on 

 the basis of a knowledge of the spring population level and of the degree of breeding and 

 rearing success. 



* Principally tliose of the Hudson's Bay Company. 



