284 



F. L. Bunnell et al. 



biomass or activity may play in modifying levels of soil nutrients, a prob- 

 lem that is particularly disconcerting when considering nitrogen. No sig- 

 nificant relationships were found between concentrations of ammonium 

 and abundance of either blue-green algae or fungi. 



Temperature, Moisture and Oxygen 



Relationships between microbial activity and temperature, moisture 

 and oxygen have been demonstrated in a variety of natural substrates 

 from tundra, including soil (Bunnell et al. 1977a, Bunnell and ScouUar 

 1981). Relationships between microbial biomass and these abiotic 

 variables are more equivocal. No relationship was found between fungal 

 density or biomass in the soil and soil temperature. However, rates of 

 fungal growth do show relationships with temperature. Relative growth 

 rate (rgr) is defined as: 



rgr = 



ln(D, /A-,) 

 57 



where In is the natural logarithm, D, and D,-i represent fungal densities 

 [m mycelia (gdw soil)"'] at times t and t-l, and At represents the time in 

 days between / and f-l. When values for relative growth rate are 

 stratified by temperature regardless of microtopographic unit, statistical- 

 ly different distributions of growth rates are observed within different 

 temperature classes based on the mean temperature during that 10-day 

 interval (Figure 8-12). Using the Smirnov one-sided test the probability 

 of temperatures 4 to 5°C and 5 to 6°C having the same distribution of 

 growth rates is < 0.1. A similar value is obtained when temperature 

 classes 5 to 6°C and 6+ °C are compared. The probability that classes 4 

 to 5 °C and 6+ °C have the same distribution of growth rates is < 0.(XX)1. 

 At higher temperatures a greater proportion of the observed changes in 



0.6 



01 



> 



0.4 



:; 0.2 



•I 



0*^ 



AtoSOQ 



-0.4 



-0T2 0.2 



Relative Growth Rote 



0.4 



FIGURE 8-12. The fre- 

 quency of the relative 

 growth rates of soil fungi 

 at different tempera- 

 tures. The temperatures 

 used were the means of 

 the lO-day interval. 

 (Laursen, Miller and 

 MacLean, unpubl.) 



