of descriptions reported for the subspecies vaseyana. However, there- seems to be little 

 morphological difference between la, lb, and Ic that can be used consistently to sepa- 

 rate them. Odor differences are not sufficiently characteristic to' distinguish groups 

 within subspecies. Considerable variation in leaf color has been observed, but similar 

 shades can be found in each subgroup. Leaf size appears to be slightly reduced from 

 Ic to la to lb as the elevation decreases (fig. 35). However, this is believed to be 

 mostly a reflection of the favorability of the site since the difference does not al- 

 ways persist when all types are grown under similar environmental conditions. Leaves 

 of subgroup Id {A. tvidentata subsp. Wyoming ensis) are cuneate in form and are the 

 smallest in group I (fig. 36). 



Similarly, Ila, lib, and lie are probably different ecotypes within the subspecies 

 tridentata. Leaves of Ila and lib are cuneate, whereas those of lie are longer and 

 oblanceolate (fig. 37). Subgroups Ila and lib appear to be morphologically similar. 

 However, differences have been observed between them and lie. Plants of subgroup lie 

 are commonly observed along fence rows, gullies, and similar places throughout the 

 northern Great Basin, but are more common to the south. These plants are typically 

 much taller than those of Ila and lib, frequently attaining heights of 12 to 15 feet. 

 Although individual plants of large populations of lie are not generally as tall as 

 those growing along fence rows or in other protected places, they are somewhat taller 

 than similar populations of Ila or lib. The color of lie is quite distinct and indi- 

 viduals of this group can usually be distinguished from Ila and lib by observers 

 acquainted with color variations in sagebrush populations. Plants of subgroups Ila 

 and lib usually exhibit varying shades of gray green, whereas those of lie are bluish 

 gray. 



The collections studied may not have included representatives from either A. 

 tridentata subsp. vaseyana f. spiaiforviis or .4. tridentata subsp. tridentata f. 

 parishii (Beetle 1960). However, a few long-leafed, high-elevation specimens of A. 

 tridentata subsp. vaseyana were chromatographed . These collections were chromatograph- 

 ically similar to shorter- leafed A. tridentata subsp. vaseyana types and are included 

 in vaseyana Ic. Furthermore, plants of large stature with drooping inflorescences, a 

 prominent characteristic of A. tridentata subsp. tridentata f. parishii, were collected 

 from the sandy areas of northwestern Nevada. Most of these were chromatographi cal ly 

 similar to tridentata Ila. We have assumed that these were probably not true A. 

 tridentata subsp. tridentata f. parishii collections. 



Distribution in relation to grazing preferenoe.--Vne two major chromatographic 

 groups, I and II, also show a pronounced difference in palatability. Almost without 

 exception, collections from individual shrubs or populations that normally show signs 

 of heavy grazing by deer and livestock, especially on winter ranges, are included in 

 group I. Grazing preference has been evident for many plants collected from widely 

 scattered areas under heavy utilization at the time of collection. On the other hand, 

 all collections from populations observed to be relatively unpalatable are included in 

 group II. The preference for group I plants, as contrasted with those of group II, was 

 particularly evident in the area from which wyomingensis Id and tridentata Ila were 

 collected in Nevada and where these two types were growing together as an intermixed 

 population. Form wyomingensis Id was highly palatable to cattle that grazed the 

 area, but form tridentata Ila was grazed very little. Sheep and deer exhibited the 

 same partiality for wyomingensis Id. Similar selectivity of group I over group II has 

 been observed where types vaseyana lb and tridentata Ila come together and intermix in 

 lower foothill areas in the Great Basin. Under these circumstances, plants of vaseyana 

 lb are grazed much more extensively than those of tridentata. lib. Intermediates result- 

 ing from apparent hybridization between these two strains exhibit considerable variation 

 in the degree to which they are grazed, but are usually preferred to the group II plants. 



Table 5 illustrates the differential browsing selectivity of deer during April of 

 1972 for the foliage of several sources from different intermountain areas. Plants from 



17 



