132 William C. McComb and Robert L. Rumsey 



(P<0.05) with any habitat characteristic. Dueser and Shugart (1978) 

 suggested that since T. striatus is diurnal and P. leucopus is nocturnal, 

 and since they differ morphologically, they can coexist with little com- 

 petition. Microtus pinetorum was found most frequently on M91 on 

 northfacing slopes, and was intolerant of other sites. Understory and 

 overstory structure was important to capture of M. pinetorum, and 

 these characteristics differed from characteristics important to the other 

 rodents sampled. Variation in vegetative structure seemed to allow coex- 

 istence of M. pinetorum with other rodents on northfacing slopes. If 

 competition is occurring within the small mammal community on our 

 study areas, then it is probably most intense on northfacing slopes and 

 least intense on ridgetops. 



Differences in relative abundance and habitat selection among dif- 

 ferent physiographic sites may allow a more complete understanding of 

 small mammal species distribution within a watershed, and may increase 

 our understanding of niche overlap and competitive exclusion among 

 sympatric small mammals. 



Management Implications 



An increase in small mammal abundance may be desired as a non- 

 game management practice or to produce small mammal biomass for 

 predatory game or furbearers. If nongame management is an objective, 

 then plots should be small and widely spaced to maximize edge. Habitat 

 alterations should include providing habitat for species intolerant of 

 habitat variety, such as O. nuttalli (clearcut) and M. pinetorum (herbi- 

 cide, preferably similar to M-3864). Resulting increases in rock expo- 

 sure and microarthropod populations through increased leaf decompo- 

 sition on herbicide and clearcut plots (Bormann and Likens 1979; 

 Gottschalk and Shure 1979), and in log abundance, stump abundance, 

 and understory density, will benefit the six species studied. 



Small mammals were 1.5 times more abundant on clearcuts than 

 on herbicide plots, and twice as abundant on herbicide-treated plots 

 than on uncut plots. Pelleted herbicide application is less expensive than 

 cutting (Dewey 1980), but the resulting increase in small mammal bio- 

 mass following herbicide application up to 91 kg/ ha is also less than 

 results from cutting. Limited accessibility, ruggedness of the terrain, 

 and/ or increasing fuel and labor costs may influence managers to use 

 herbicides to create clearings for wildlife. 



ACKNOWLEDGMENTS.— We wish to thank R. L. Anderson, D. 

 M. Allen, and M. J. Immel for advice on statistical analyses; J. B. Davis 

 and G. M. Gigante for assistance in field work; Dow Chemical Co. for 

 providing herbicides; and B. A. Thielges, S. B. Carpenter, W. H. Davis, 



