F. vesca; Polemonium folio si ssimum and Polemonium 

 occidentale as P. foliosissimum; and Delphinium 

 occidentale and Delphinium barbeyi as D. occidentale. 

 Nomenclature ambiguities forced a somewhat arbitrary 

 selection of names for some species. Thus the Agropyron 

 trachycaulum-Agropyron subsecundum-Agropyron 

 caninum complex is treated as A. trachycaulum; Bromus 

 marginatus and Bromus polyanthus are included with 

 Bromus carinatus; and Stipa columbiana and Stipa 

 nelsonii are lumped with Stipa occidentalis. The tradi- 

 tional name of Koeleria cristata is used to include 

 Koeleria nitida. Considerable confusion revolves around 

 the separation of Geranium viscosissimum and 

 Geranium fremontii. This confusion is reflected by differ- 

 ent floras in adjacent States that seldom treat both spe- 

 cies and that may indicate that G. fremontii is synony- 

 mous with G. viscosissimum. We have arbitrarily chosen 

 to call this uncertain, pink-flowered complex 



G. viscosissimum. 



THE CLASSIFICATION: VEGETATION 

 KEY 



This classification partitions those forests in Utah 

 where aspen comprises at least 50 percent of the tree 

 canopy into six cover types based upon the dominant 

 and codominant trees. These cover types are then sepa- 

 rated into 36 community types based upon prominent 

 indicator species in the undergrowth. Table 3 is the 

 vegetation key for the classification. 



Although aspen generally has relatively broad environ- 

 mental tolerances, it typically is less shade tolerant and 

 shorter lived than most conifers. Thus, aspen stands 

 that contain a substantial element of conifers are consid- 

 ered to be at a serai stage leading toward a conifer cH- 

 max. They are categorized as aspen-conifer cover types. 

 The aspen-conifer cover types are separated by minimal 

 amounts of conifer occurrence, giving sequential con- 

 sideration first to those conifers that require the less 

 stressful moist sites, and proceeding eventually to those 

 that are able to occupy the driest sites. Pinus contorta, 



an exception, is taken out last because it is usually con- 

 sidered serai to other conifers. At least 5 percent canopy 

 cover of Abies lasiocarpa, alone or combined with Picea 

 engelmannii, qualifies a stand for the Populus 

 tremuloides-Abies lasiocarpa cover type. At least 10 per- 

 cent canopy cover is required for Abies concolor, 

 Pseudotsuga menziesii, and Pinus contorta; only 5 per- 

 cent is required for Pinus ponderosa. These minimal 

 cover requirements are subjectively judged to be more 

 than accidental and to approximate the level required for 

 the conifer species to validly indicate site differences, as 

 well as being indicative of successional trends. 



The Populus tremuloides cover type is further sub- 

 divided into three undergrowth types based upon vegeta- 

 tional structure. Those communities possessing a dis- 

 tinct tall shrub component are placed in the tall shrub 

 undergrowth type. Those without a tall shrub layer but 

 that possess a distinct layer of low shrubs are placed in 

 the low shrub undergrowth type. Community types that 

 lack a well-defined layer of shrubs are placed within the 

 herb undergrowth type. 



The largest number of community types (21) occurred 

 within the Populus tremuloides cover type. The P. 

 tremuloides- A. lasiocarpa cover type contained seven 

 community types; the remaining four cover types con- 

 tained only two community types each. Community 

 types were based on as few as two stands in one case 

 where composition and environment were unique (and 

 the type had been reported elsewhere), to as many as 

 152 stands in the most common type. Over 40 percent of 

 the stands classified fell into only four community types: 

 P. tremuloides/Symphoricarpos oreophilus/Senecio serra 

 c.t.; P. tremuloides/S. serra c.t.; P. tremuloides/Prunus 

 virginiana/S. serra c.t.; and P. tremuloides- 

 A. lasiocarpa/S. serra c.t. Over half of the community 

 types were infrequent but eire considered valid because 

 of their repeated occurrence in the large number of 

 stands that served as our data base. A Usting of all com- 

 munity types is given in table 1. Less than 5 percent of 

 all stands sampled could not be matched with any of the 

 36 community types listed. 



8 



