ROLE OF PRE DAT ION 325 



Buffer Fluctuations 



Among game managers the term "buffer" has been applied to those species which serve 

 as the staple foods of predators and thereby lessen the pressure on grouse. In order to 

 serve in such a capacity a species must be present in far greater numbers than the game and 

 be relatively easy to catch. Also, to be most effective it should not compete seriously with 

 the other. The principal buffers with res])ect to grouse in the Northeast are the cottontail 

 rabbit, varying hare, squirrels, mice and shrews. Nevertheless, in connection with the 

 Cooper's and sharp-shinned hawks, the many small woodland birds must also be considered 

 in this category. 



Little is known of variations in abundance of these birds, but among the small mammals, 

 particularly the rodents, fluctuations are well recognized. The pronounced cyclic increases 

 and decreases in populations of the varying hare and certain mice, notably lemmings, have 

 been discussed by a number of authors'"'''". As with predators they have been most violent in 

 Canada and Alaska.* Yet their occurrence has not been uniform across the continent as some 

 have believed. In his account Seton states. "A Rabliit year in one part of the country is not 

 necessarily a Rabbit year in another, the condition being local." MacLulich''" also found 

 that years of peak abundance in hares varied in different districts. Undoubtedly the same 

 is true for mice although the pattern has not been worked out and the length nf their cycle 

 is much shorter. 



In New York the varying hare appears to exhil)it an ebb and fluw of abundance similar to 

 that displayed farther north but on a much smaller scale. Records indicate an interval of 

 roughly ten years which corresponds with that found in Canada'"". According to the report- 

 ed take by licensed hunters beginning in 1918 the years 192.3 and 19,32 marked the end of 

 periods of relative abundance and were followed by abrupt declines, while the years 1926 

 and 1936 re|)resented the troughs respectively. The magnitude of the fluctuation, however, is 

 much less. During "rabbit years" in Canada both Seton and MacLulich observed over 

 3,000 hares per square mile. No such density has ever been recorded here. The distribu- 

 tion of the species in this state also is at present limited, being found onlv in the Adiron- 

 dacks and a few localities in the Calskills. 



Data regarding mice are available for a much shorter period. Yet it is clear that they, and 

 probably shrews loo. are also subject to rhythmic increase and decrease in New York and 

 other parts of the Northeast as well as northward. To secure data on this problem the 

 Investigation established a series of half-mile trails on the Connecticut Hill study area in the 

 fall of 1933"^. During the following season and each winter thereafter tracks of the various 

 species were tallied along these trails after each fresh snowfall. Since the home ranges of 

 these animals are small this method appears to afford a reliable index of trends in abun- 

 dance. 



Analyzed statistically the records for mice and shrews indicate a definite cyclic tendency 

 on this area (figure 29). Unfortunately the period covered has not been long enough to more 

 than suggest a three to four year periodicitv of the cycle. This, however, corresponds quite 

 closely to the conclusions of Hamilton'" and Elton'"'. 



That there is probably little harmony in these fluctuations over any considerable area is 

 indicated bv comparing the available data for central New York. Thus the Investigation re- 

 corded peak populations on Connecticut Hill following the breeding seasons of 1934 and 1937, 

 while Hamilton"" working near Ithaca, less than 25 miles away, found 1935 to be high. Sim- 



* This refers only tn North American species. In Europe the Norwegian lemming (Lemmus lemmus} is a classical example of 



this behavior. 

 A See Appendix, p. 709. 



