(PASSMI) in vegetation composition and soil parameters with pH ranging from 8.0 to 8.4 and 

 texture being primarily loamy. Sarcobatus-dominatQd sites in our study may be contrasted with 

 those sampled by Brown (1971) in Rosebud and Powder River Counties on badlands derived from 

 the Fort Union Formation (sedimentary rocks ranging from sandstones to lignite coal beds but 

 dominated by clay and silt shales). Brown's Sarcobatus Community had only 16% total vegetative 

 canopy cover and on a percentage basis only 3 % of which was grass; thus the community 

 represented in Brown's sampling was considerable more depauperate, especially in the graminoid 

 component. The depauperate vegetation of Brown's sites probably reflect the fact that they were the 

 most stressful of the badland sites he sampled in terms of soil chemistry (average conductivity 7.2, 

 pH < 8.0 and sodium concentration of 9.1 me/ 100 g indicate saline-alkali soils), osmotic stress and 

 lack of aeration. The S. vermiculatus / Pseudoroegneria spicata type of Hansen and Hoffman 

 (1988), though possessing higher vegetative cover, is a close match for the sites described by Brown 

 (1970) within these same two counties. 



Comments: SARVER/ELYLAN was described from the Yellow Triangle area of eastern Montana 

 by Jorgensen (1979), but has not been reported elsewhere. The ecologically similar (equivalent?) 

 SARVER/PASSMI, with Pascopyrum smithii replacing Elymus lanceolatus as the dominant grass, 

 has been described by numerous studies. Bottomland SARVER/PASSMI communities are 

 documented from north-central (Mackie 1970) and northeastern (DeVelice et al. 1995), southeastern 

 (Hansen and Hoffman 1988, Ecological Consulting Service 1975, Olsen-Elliott and Associates 1980) 

 and southwestern Montana (Mueggler and Stewart 1980, Cooper et al. 1995) as well as western 

 North Dakota (Jensen et al. 1992), Wyoming, South Dakota, Nebraska, and Saskatchewan (The 

 Nature Conservancy 1997). One of these studies (Ecological Consuhing Service 1975) identified 

 SARVER/PASSMI in Carter County in the Thompson Creek drainage not far from one of our plots 

 of SARVER/ELYLAN. Due to the difficulty of distinguishing between Elymus lanceolatus and 

 Pascopyrum smithii and their apparent overlapping ecological distributions it may be best to 

 consider communities with either grass dominant in the same plant association, but we have 

 designated the type SARVER/ELYLAN to emphasize local trends in floristic composition. Upland 

 greasewood commimities are less commonly reported in the literature. A badland Sarcobatus 

 vermiculatus community was described from Rosebud and Powder River Counties, Montana with 

 slopes ranging from to 80%, however, neither Elymus lanceolatus nor Pascopyrum smithii or any 

 other herb approach dominance (Brovm 1971); ecologically this community is closer to the S. 

 vermiculatus / Pseudoroegneria spicata topoedaphic climax habitat type described by Hanson and 

 Hoffman (1988) as occurring on shale-derived small-scale terraces. 



49 



