and total density of birds as follows: 1, 2, 3, 4, 5. This is also the approx- 

 imate order of decreasing habitat heterogeneity and vegetation foliage height 

 diversity (subjectively estimated) which other studies have indicated 

 are important factors determining species numbers and abundances of breeding 

 birds (Karr and Roth 1971, Roth 1976). The mosaic of grassland, low shrub, and 

 tall shrub habitats found on the silver sagebrush plot apparently provides 

 optimal conditions for bird species number and abundance of the plots studied. 

 With respect to species diversity and one-and two-species dominance, however, 

 the silver buffaloberry-red osier dogwood coulee plot assumes first rank, re- 

 flecting a more equitable distribution of individuals among species in this 

 highly diverse habitat, with fewer species represented by only one pair than the 

 silver sagebrush plot. It is interesting to note that a vegetation sampling 

 plot in the tall coulee shrubbery of this plot showed the highest plant species 

 diversity of all quadrats sampled (Prodgers 1978), and that a small mammal trap- 

 line in this same coulee had the highest biomass of all sampled. Total biomass 

 of breeding birds was greatest in the silver buffaloberry-silver sagebrush coulee 

 plot. Over 30% of the biomass of this plot was accounted for by a single ring- 

 necked pheasant pair; however, even if this pair is excluded from biomass cal- 

 culation, this plot ranks highest. Much of the remaining biomass in this plot 

 is attributable to the large number of nesting mourning doves found in the clumps 

 of silver buffaloberry within the coulee; the extent of re-nesting by this 

 species is unknown. The western wheatgrass-blue grama pasture ranked lowest 

 of the five plots with respect to all the above parameters; grassland develop- 

 ment was so poor on this plot in 1977 that little grass cover existed, and 

 shrubbery was limited to a few scattered big sagebrush plants very near the 

 edge of the plot. 



Adjusted number of species, total density, and total biomass of the grass- 

 land plots studied here are well within the range reported for censuses con- 

 ducted in shortgrass prairie and shrub-steppe habitats as part of the IBP Grassland 

 Biomass program (Weins 1973, 1974, Weins and Dyer 1975), although estimates of 

 total density and total biomass are somewhat less than the averages for shortgrass 

 prairie reported by Wiens and Dyer (1975). Adjusted species number, total den- 

 sity, and total biomass of the silver sagebrush plot, however, were much higher 

 than averages reported by Wiens and Dyer (1975) for various shrub-steppe hab- 

 itats (including sagebrush) or by Best (1972) and Feist (1968) for sagebrush 

 habitats in central Montana. This is probably due to the presence of scattered 

 clumps of buffaloberry and the high degree of interspersion of grassland and 

 various shrubland types in the present study. Data for comparison with the 

 coulee plots are not available. 



Roadside Wildlife Survey 



Data on bird species observed on the eleven monthly replicates of the four 

 standard roadside wildlife survey routes (excluding the Dreyer Ranch route) 

 have been summarized in table 8, and raw data for all 47 runs (four routes, 

 eleven months, plus one route, three months) are on file at DNRC, Energy Division, 

 Helena, Montana. 



Species Composition . Of the 139 bird species recorded during this study, 

 103 (74%) were recorded at some time on one or more of the four regular road- 

 side wildlife survey routes. An additional five species were recorded only on 



152 



