and poorest aspect for the dependent variable being predicted. The optimum 

 aspect is the aspect with the highest point along the curve. In figure 9 it is 

 17 degrees — indicating a general direction of north to northeast for the high- 

 est probability of stocking for the Abies grandis/Clintonia uniflora habitat 

 type. The poorest aspect is 180 degrees from the optimum aspect. 



Amplitude is the depth of the curve from the optimum aspect to the poor- 

 est aspect. The larger the amplitude, the larger the difference between the 

 optimum and poorest aspects. Amplitude can be calculated using any slope — 

 we use 30 percent throughout this paper. The amplitude shown in figure 9 

 is 0.34. This means the probability of stocking differs by 0.34 between the 

 optimum and poorest aspects for a 30 percent slope. 



There is another important point about the calculation of amplitude when 

 the form of the equation is 



P = d+e^^Pi^i')-' 



Amplitude is calculated with the curve centered on a probability of 0.5. 

 Since probabilities are asymptotic to 0.0 and 1.0, amplitudes are at their 

 maximum at a probability of 0.5. 



Analysis of data now excludes nonstocked plots. Given that a plot is stocked, 

 we are interested in predicting how many trees are established, species compo- 

 sition, and seedling heights. Distribution of the number of trees on stocked 

 plots is shown in figure 10. The highest percentage is for plots that have 1 tree, 

 followed in decreasing order by 2, 3, 4, and so on, up to 213 trees per plot. 

 The result is a reversed J-shape distribution. 



40. 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 



Number of Trees Per Stocked Plot 



Figure 10 — Distribution of the number of regeneration-size trees 

 per stocked plot. The tail of the distribution continues past 20 trees 

 up to 213 trees on a plot. 



Number of Trees 

 per Stocked Plot 



18 



