In summary, breeding season runs of the bird communities sampled by the 

 roadside routes showed higher species number, more total individuals, higher 

 species diversity and richness, higher evenness values having less variability, 

 and lower one- or two-species dominance than the winter runs, with spring and 

 fall values generally showing a fairly smooth transition between summer and 

 winter. The five routes may be generally ranked from highest to lowest species 

 number, number of individuals, species diversity, species richness, and evenness, 

 and from lowest to highest single-species and two-species dominance as follows: 



1) Missouri River route; 



2) Prairie Elk route; 



3) Dreyer Ranch route; 



4) Circle route; and 



5) Flowing Well route. 



This is also approximately the order of decreasing structural diversity of habitat, 

 as indicated by the availability of tree and tall shrub vertical layering; decreas- 

 ing presence of western snowberry-prairie rose, wild hayland and marsh habitats; 

 and increasing presence of badland and big sagebrush-scabland habitats (table 5). 

 The separation among the five routes is most clearly evident in breeding-season 

 plots of species number and species richness (figure 40 and 43). It is interesting 

 that the Dreyer Ranch route is almost exactly intermediate between the Missouri 

 River and Flowing Well routes in these figures. This property will be especially 

 valuable in long-term monitoring, as mine- or plant-related changes in community 

 parameters of the Dreyer Ranch route will be reflected in a shift of the relative 

 positions of plots for this route with respect to the four standard routes, which 

 serve as controls. 



Beta Diversity . As mentioned above, the ranking of the five routes with re- 

 spect to several community parameters seems related to habitat characteristics 

 of areas sampled by routes. The question arises as to whether the differences 

 in bird community parameters (including such gamma diversity measures as S and H'; 

 are related more closely to differences in beta diversity (that is, differences 

 in the variety of habitats sampled by each route) than to differences in alpha 

 diversity (that is, diversity of each roadside stop sampled). Stated in another 

 way, is bird species number greater on the Missouri River route than the Flowing 

 Well route because the habitats sampled at each stop on the former are more diverse 

 than those sampled at each stop on the latter (differences in alpha diversity), 

 or because the former route simply cuts across a greater number of different types 

 of habitats than the later (differences in beta diversity)? 



The observed pattern of beta diversity among the five routes is quite var- 

 iable over the breeding season (table 35). When breeding-season averages are 

 considered, it is evident that the ranking of the five routes differs from that 

 discussed above. The relative positions of the Missouri River and the Prairie 

 Elk routes are reversed, as are those of the Circle and Flowing Well routes. 

 The Dreyer Ranch route remains intermediate. While the methods of the roadside 

 count do not sample a true coenocline (one of the conditions mentioned by 

 Whittaker 1972), three conclusions may be drawn from these data. First, the 



166 



