short-beaked sedge {Carex simulata), and Kelsey's phlox 

 (P. kelseyi). The stream is closely bordered by clumped 

 communities of Kentucky bluegrass, beaked sedge (C. ros- 

 trata), and Baltic rush {Juncus balticus). Associated forbs 

 and graminoids include mannagrass {Glyceria spp.), wa- 

 ter sedge (C. aquatilis), Nebraska sedge (C. nehraskensis), 

 American bistort {Polygonum histortoides), and large- 

 leaved avens (Geum macrophylum). 



The study area is located largely on public lands ad- 

 ministered by the Bureau of Land Management, U.S. 

 Department of the Interior. Recent (1976 to 1989) stock- 

 ing levels have varied from about 1,000 to 2,000 animal 

 unit months (AUM's), with a grazing season from about 

 mid-May to late October (Hale 1989). Stocking levels and 

 grazing periods are adjusted annually on the basis of 

 current resource conditions. 



METHODS 



Two 9-ha plots, one in the upper (westernmost) section 

 of the exclosure and the other in the adjacent (upstream) 

 grazed riparian area, were censused for breeding birds 

 using the spot-map method (International Bird Census 

 Committee 1970). Plot locations were selected on the 

 basis of similarities in topography and vegetation between 

 the grazed and ungrazed environments. The census plots, 

 600 by 150 m, were oriented lengthwise along Summit 

 Creek and gridded with transects crossing the stream 

 channel. Both plots straddled the riparian zone and in- 

 cluded part of the extensive uplands. Grid points were 

 surveyed and marked with numbered stakes at 25-m 

 intervals. 



Eleven census visits were made to each plot from 

 May 17 to June 29, 1989. The same observer (DEM) 

 conducted the censuses on both plots. Most of the spot- 

 mapping was done from sunrise to early afternoon when 

 birds were most active. Census routes were varied by 

 choosing different routes through the plot, with different 

 starting and ending points. To ensure complete coverage, 

 the plot was censused by walking within 25 m of all points 

 on the grid. Observations and registrations extended well 

 beyond plot boundaries. 



At the end of the sampling period, clusters of observa- 

 tions and coded activity patterns on species maps were 

 circled as indicating areas of activity or approximate terri- 

 tories (International Bird Census Committee 1970). PVac- 

 tional parts of boundary territories were included. Oelke 

 (1981) summarized methodological difficulties and other 

 special problems of the mapping method. We followed 

 Hill (1973) for estimates of species diversity. 



A 1.7-ha trapping grid was located in each of the grazed 

 and ungrazed study plots to estimate small mammal 

 populations. Trapping grids were placed near the center 

 of the 9-ha plots established to census bird populations. 

 Each grid measured 225 by 75 m and consisted of 40 trap- 

 ping stations systematically spaced at 25-m intervals in 

 10 rows and 4 columns. The rectangular grids were posi- 

 tioned lengthwise along Summit Creek and straddled the 

 stream channel. Two Museum Special mouse traps and 

 one Victor rat trap were placed near each trapping sta- 

 tion. Traps were baited with a mixture of peanut butter 



and rolled oats and examined daily for 5 consecutive days 

 from August 3 to August 7, 1988, and from August 17 to 

 August 21, 1989. 



Vegetation and other features of the grazed and un- 

 grazed study plots were measured from July 17 to August 

 30, 1989. Twenty sample locations were established 

 within each of the three plant community types for a total 

 sample size of 60 per study plot. A 50- by 50-cm (0.25-m^) 

 quadrat was located at each of the systematically posi- 

 tioned sample locations. Canopy cover (Daubenmire 

 1959) was ocularly estimated for the total of each plant 

 life form (graminoid, forb, shrub) and recorded as the 

 midpoint of one of eight percent-cover classes (0-1, 1-5, 

 5-10, 10-25, 25-50, 50-75, 75-95, 95-100). Percentages of 

 litter, rock, bare ground, and lichen-moss were similarly 

 estimated. The vegetative height (excluding flower and 

 seed-head heights) of each graminoid, forb, and shrub 

 nearest the center of each quadrat was measured. 



Biomass of graminoids, forbs, and shrubs was deter- 

 mined by clipping vegetation from ground level upward 

 within a vertical projection from the 0.25-m'^ quadrats. 

 Clipped materials were bagged, ovendried, and weighed. 



Plant names follow Hitchcock and Cronquist (1973). 

 Bird nomenclature is from the 1983 AOU Check -list 

 (American Ornithologists' Union 1983). Scientific and 

 common names of mammals follow Jones and others 

 (1986). 



RESULTS AND DISCUSSION 



We found structural (physiognomic) differences in the 

 vegetation between grazed and ungrazed habitats on 

 Summit Creek. Those differences were apparently re- 

 flected in the organization of associated breeding bird and 

 small mammal communities. 



Vegetation 



The most evident structural difference in the vegetation 

 was in the height values where graminoid, forb, and 

 shrub height means were significantly reduced on the 

 grazed site (table 1). Other differences were primarily in 

 the herbaceous layer where graminoid and forb biomass 

 and graminoid canopy cover values were lower on the 

 grazed site. Graminoid biomass on the grazed plot was 

 only about one-seventh that inside the exclosure. Esti- 

 mates of forb and shrub cover were similar on the grazed 

 and ungrazed areas. There was significantly more rock 

 coverage on the grazed plot. Shrub biomass and shrub, 

 bare ground, and litter coverage were similar. Lichen- 

 moss cover values were slightly higher on the grazed site. 

 There were no tall shrubs or trees on the study plots. 



Birds 



We recorded eight species of birds breeding on the Sum- 

 mit Creek study site; seven species bred on the grazed 

 plot and four species bred on the ungrazed plot (table 2). 

 Vesper sparrows, savannah sparrows, and western mead- 

 owlarks were found as breeding birds on both the grazed 

 and ungrazed plots. Killdeer, willets, long-billed curlews, 



2 



