190 P.J. Webber et al. 



o 



^ 



Low 

 Basins 



Ponds' 



Moist Ponds' 

 Meadows vvet 



Msadowi 



Dry Troughs 



Meadows 



Tops' 



Rims 



Tops' 



LO 



moisture 



W LO soluble phosphote H/ 



FIGURE 6-3. The distribution of microtopographic units within the 

 three principal axes of the ordination. The axes are soil moisture, soil 

 odor of hydrogen sulfide, and soluble soil phosphate. Low basins are the 

 centers of low-centered polygons. Tops' are the centers of high-centered 

 polygons with little or no peat, Tops^ are the centers of high-centered 

 polygons with thick peat, Ponds' are those with no significant flow of 

 water, and Ponds^ are those with flowing water. (After Webber 1978.) 



and rims and basins of low-centered polygons. Two categories of ponds — 

 those with and without flowing water — and three types of meadows — 

 dry, moist and wet — were distinguished. Two categories of polygon 

 tops — those with little or no peat at the surface and those with thick sur- 

 face peat — were distinguished. 



The different microtopographic units can be plotted within the axes 

 of the indirect ordination (Figure 6-3). The sequence of units along the 

 soil moisture gradient from low moi':ture to high is: tops of high-centered 

 polygons with shallow soil, polygon rims, dry meadows, moist meadows, 

 basins of low-centered polygons, tops of high-centered polygons with 

 thick peat, polygon troughs, and wet meadows. The sequence along a 

 gradient of soil hydrogen sulfide from low to high is: tops with thick 

 peat, rims, tops with shallow soil, dry meadows, troughs, wet meadows, 

 moist meadows and basins. The sequence along a gradient of soil soluble 

 phosphate from low to high is: ponds with no flow, basins, dry mea- 

 dows, moist meadows, rims, tops, troughs, and wet meadows. Basins of 

 low-centered polygons and polygon troughs have the greatest snow ac- 

 cumulation and are the last microtopographic units to be free of snow, 

 while polygon rims and tops of high-centered polygons have only a thin 

 snow cover and are first to be snow-free. The duration of snow cover af- 

 fects the length of the growing season, which varied from 30 to 42 days in 

 1973 (Table 6-1). 



These observations coincide with the conclusions of Wiggins (1951) 

 and Britton (1957) who emphasized the control by the microrelief of the 

 substrate conditions, which in turn control the distribution of plants. In 

 contrast to the results of most tundra ordinations (e.g. Webber 1971, 



