The Tilley and Terry (1963) technique as modified by 

 Moore (1970) was used to determine in vitro dry matter 

 digestibility (IVDMD). Rumen fluids from elk and domes- 

 tic sheep, both maintained on alfalfa hay and pellets, 

 were used for the IVDMD analysis. 



The concentrations of nitrogen, calcium, phosphorus, 

 dry matter, and ash were determined for each sample. 

 The macro-Kjeldahl technique (A.O.A.C. 1965) was used 

 to determine nitrogen. Plant samples for determination 

 of calcium and phosphorus were ashed at 500 °C, put into 

 solution in an acid medium, and filtered. Phosphorus was 

 determined with a spectrophotometer using the ammo- 

 nium molybdate-ANSA method (Harris 1970), and cal- 

 cium was determined using the EDTA titration technique 

 (Allen and others 1974). 



We subjected these data to an analysis of variance. 

 There were three replications (sites), two treatments 

 (burned and unbumed), nine plant species, and several 

 collection dates. 



The major comparison for the first postbum year (1982) 

 consisted of one date and five species; the major compari- 

 son for the second postbum year (1983) consisted of three 

 dates and nine species. The five species analyzed from 

 August 22, 1982, were: aspen, serviceberry {Amelanchier 

 alnifolia), snowbrush (Ceanothus velutinus), chokecherry 

 {Prunus virginiana), and pinegrass {Calamagrostis rubes- 

 cens). Another test for the first field season used two 

 dates (August 2 and 22) but only two species, aspen and 

 pinegrass. 



In 1983, the three dates were July 15, August 15, and 

 September 15. The nine species included the same five 

 species as in 1982 plus blue wildrye {Elymus glaucus), 

 sticky geranium {Geranium viscossissimum), blue lupine 

 {Lupinus argenteus), and heairt-leaf arnica (Arnica 

 cordifolia). 



For a year-to-year comparison, the five species fi-om 

 August 22, 1982, were matched with the same species 

 fi-om August 15, 1983. We also made a comparison using 

 only aspen and pinegrass for both August 2 and 22, 1982; 

 and July 15 and August 15, 1983. 



Additional analyses using the least squares estimator, 

 or in case of no overall statistical significance, the ShefFe 

 LSD test, were used to test for differences between species 

 means (Neter and Wasserman 1974). 



All differences discussed in this paper are significant at 

 least at the 0.05 percent probability level. 



RESULTS AND DISCUSSION 



Plant phenological stages were observed to be similar 

 on both burned and unbumed areas. Forbs bloomed 

 throughout the season. Thus, several phenological stages 

 of forbs overlapped within and between the bumed and 

 unbumed areas. Pinegrass produced many culms and 

 flowers in the burned areas but only a few in the un- 

 bumed areas. Shrubs on the burned areas neither flow- 

 ered nor set seed during either summer. So, dates instead 

 of phenological stages are used for comparisons. 



Irrespective of buming, changes in forage quality oc- 

 curred as the season progressed. In the second year, for 

 example, the earliest sampling yielded the highest quality 

 (table 1). Also, over the entire season, there were marked 

 differences among species (table 2). Forbs had higher 

 digestibilities than shrubs or grasses. 



Rre improved forage quality during the first postburn 

 growing season (1982). Aspen on the bumed areas on 

 August 2, 1982, had a better calcium/phosphorus ratio, 

 higher elk digestibility, higher crude protein and phospho- 

 rus levels, and a lower calcium content than on the un- 

 bumed sites. However, 3 weeks later only phosphorus 

 and crude protein levels were greater on burned sites 

 (table 3). At that time, all shrubs on the bumed areas 

 had higher crude protein levels. Two (serviceberry and 

 chokecherry) had lower phosphorus levels, and choke- 

 cherry had a lower calcium level (table 4). In the second 

 growing season (1983), there were no improvements in 

 quality of either shrubs or forbs that could be attributed 

 to burning. 



Some of the changes in quality of shmbs could be due to 

 changes in growth form on the burns. In early August of 

 the first year, most shrubs on the bumed areas were less 

 than 1 ft (30 cm) tall and had fewer stems and larger 

 leaves than on unburned areas. By late August, secon- 

 dary branching had appeared and the plants were up to 

 2 ft (60 cm) tall. In the second summer, these shrubs 

 were 3 to 5 ft (100 to 150 cm) tall, with more branches and 

 smaller leaves. Thus, shrubs in early August of the first 

 postburn year had lower stem-to-leaf ratios than they did 

 later in that summer or in the second year. 



Higher levels of phosphorus in plants from bumed ar- 

 eas were found only in aspen and only on August 2 of the 

 first year. All of the other samples from the burns had 

 either lower levels of phosphorus or did not differ from the 

 unbumed sites (tables 3, 4, and 5). Our findings are simi- 

 lar to those of others (Halls and others 1952; Meenely and 

 Schemnitz 1981; Swank 1958). 



Table 1 — IVDMD and nutrient concentrations of all forage species combined on three dates in 1983 



IVDMD 



Date Sheep Elk Protein Calcium Phosphorus 



- - Percent 



July 15 '65.9 60.2 M6.3 '1.28 '0.33 



August 15 586 61.6 125 1.57 .27 



September 15 57.5 60.1 10.9 1.70 .29 



'Significantly (ptO.001) different from values on other two dates. 



3 



