Microflora Activities and Decomposition 317 



production on measured respiration, but acknowledge that the difference 

 in soil respiration among microtopographic units is greater than that ex- 

 pected from differences in primary productivity alone. 



In taiga soils coefficient a^ and total respiration decrease with depth 

 in the profile, while the measured Q,o increases (Bunnell et al. 1977a). 

 Such findings suggest that the relative contribution from primary pro- 

 duction in the taiga decHnes with depth and that the effect of tempera- 

 ture on microbial populations increases with depth. In both tundra and 

 taiga, microorganisms in the litter layer continue respiration at lower 

 temperatures than do organisms at depth in the soil. Surface layers are 

 subject to wider ranges in both temperature and moisture than are deeper 

 layers. Thus it is not surprising to find deeper communities apparently 

 adapted to narrower ranges of temperature and moisture than surface 

 communities. The tendency within tundra soils for more linear relation- 

 ships with temperature may reflect an adaptation to low temperatures. 



The sensitivity analyses of the model based on the function gresp 

 that were conducted by Bunnell et al. (1977a) have been extended using 

 other substrates from tundra. In all cases the model is most sensitive to 

 coefficient a^, which determines the predicted Qio. Relative sensitivities 

 to other coefficients vary with the substrate. For those substrates where 

 the overall fit of the model is close (r^ 5= 0.80) we observe that the broad 



100 



96- 



c 



a 

 E 92 



0) 



q: 



a) 



88 



84 



Aug 

 '70 



Aug 

 '71 



Aug 

 '72 



Aug 

 '73 



FIGURE 9-10. Progression of weights of cellulose 

 placed in surface soil, litter and standing dead mate- 

 rial. Vertical bars give standard error. (After Bun- 

 nell et al. 1977b.) 



