In some instances, a particular TWINSPAN class included a plot 

 or plots that, based on field experience and ordination patterns, 

 appeared to be better placed in a different existing TWINSPAN 

 class. These plots were repositioned in the classification as 

 appropriate. 



In addition to helping refine the classification, ordination 

 assisted in describing and interpreting general patterns of vegeta- 

 tion communities and environment. For example, DCA extracts the 

 dominant compositional gradients from the species data matrix. The 

 environmental controls of these compositional gradients are then 

 interpreted based on comparisons with the site data. 



Taxonomic Considerations 



Nomenclature follows Hitchcock and Cronquist (1973) except for 

 Arabis f ecunda and Lesguerella paysonii which follows Rollins 

 (1984) and Rollins and Shaw (1973) , respectively, and Mertensia 

 lanceolata which follows Dorn (1984). Scientific names of all 

 species observed in this study and their code names are listed in 

 Volume II, Appendix B. 



RESULTS AND DISCUSSION 



Community Type Classification 



Classification of the 52 study plots resulted in the defini- 

 tion of 21 community types. These and 119 additional community 

 types documented in southwestern Montana (but not sampled in this 

 study) are listed in Volume II, Appendix C. 



A table of average species cover and constancy within each of 

 the 21 sampled community types is provided in Volume II, Appendix 

 D. Vegetation and site characteristics of each sample plot are 

 given in Volume II, Appendix E and F, respectively. 



Additional general information describing each sample plot is 

 provided as element occurrence records in Volume II, Appendix G. 

 Finally, Volume II, Appendix H provides a comprehensive listing of 

 condensed plant and community element occurrence records (from the 

 MTNHP database) on BLM lands within Beaverhead, Silver Bow, and 

 Madison counties. 



Vegetation-Environmental Relationships 



The DCA ordinations based on the 55 most characteristic 

 species in the data set are presented in Figures 4 through 7. 



Ordination of all 52 plots indicates that 45 plots lie near 

 the origin of the first and second axes and that the six riparian 

 community types occur as outliers (Figure 4). When the first and 



12 



