These are estimates from the mean of the two closest 

 official weather stations in the same vegetation type 

 (Reese River Valley and Austin). 



One tree-harvested and one nonharvested plot were es- 

 tabUshed adjacent to each other on north (N. 20° E.), 

 west (S. 84° W.), and south (S. 16° E.) aspects. Square 

 tree-harvested plots (0.25 acre [0.1 ha] in size) were 

 cleared of all trees 3.3 ft (1 m) in height. Cut trees, in- 

 cluding slash, were removed from the plot. Adjacent 

 nonharvested plots (0.1 ha in size) were left imdisturbed, 

 and both tree-harvested and nonharvested plots were 

 fenced to exclude hvestock. Sampled aspects were within 

 1.29 mi (2 km) of each other on 14 to 18 percent slopes 

 at a mean elevation nf 7,580 ±100 ft (2 310 ±30 m). 



The soil surface in the woodland was a mosaic of soil 

 microsites. Tree litter (duff) greater than 0.3 inch 

 (0.5 cm) in depth occurred under the tree crown. A tran- 

 sition zone of hght needle cover (less than 0.5 cm in 

 depth) formed a halo at the crown edge. And bare 

 mineral soil occurred in the interspaces between trees 

 (Everett and Sharrow 1983). Microsites with needle 

 cover (duff and transition) occupied 50, 72, and 70 per- 

 cent of the ground surface of north, west, and south 

 aspects, respectively. 



The three plant assemblages sampled were: Pinus 

 monophyllal Purshia tridentata (Pursh) D.C. (antelope 

 bitterbrush)/i<'estaca idahoensis Elmer (Idaho fescue)/ 

 Lupinus caudatus Kellogg (tailcup lupine) on the north 

 aspect; Pinus monophyllal Artemisia arbuscula Nutt. (low 

 sagebrush)/Poa sandbergii [Steud.] Vasey (Sandberg blue- 

 grass)ITrifolium gymnocarpon Nutt. (hoUyleaf clover) on 

 the west aspect; and Pinus monophylla/Artemisia triden- 

 tata ssp. wyomingensis Nutt. (Wyoming big sage- 

 brush)/Poa sandbergiilMicrosteris gracilis (Hook.) Greene 

 (microsteris) on the south aspect (Everett and others 

 1984). At the time of tree harvest the ratio of tree to 

 grass cover was 28/3 percent, 61/2 percent, and 54/1 per- 

 cent on north, west, and south aspects, respectively. 



Cover, Density, and Biomass 



In 1979 species cover and plant density of Sandberg 

 bluegrass, squii-reltail (Sitanion hystrix [Nutt.] J. G. 

 Smith), Idaho fescue, and junegrass (Koelaria cristata 

 [L.] Pers.) were estimated on harvested plots immedi- 

 ately before trees were removed. Samphng was repeated 

 on both tree-harvested and nonharvested plots in 1981 

 and 1983. Crown cover and number of plants for each 

 grass species were estimated with 20- by 20-inch (50- by 

 50-cm) frames laid at every meter mark on five perma- 

 nent parallel transects of 66 ft (20 m) in length and 

 16 ft (5 m) apart in each tree-harvested and nonhar- 

 vested plot. In 1981 grass yield was estimated on these 

 permanent transects in nonharvested and tree- harvested 

 plots. Leaf weight estimates were made separately for 

 each grass species in each frame using the weight esti- 

 mate double sample method (Pechanec and Pickford 

 1937; Wikn and others 1944). Of each grass species, 20 

 samples were clipped at maturity, seed heads discarded, 

 and estimates made of dry weight. Samples were oven- 



dried, weighed, and regression equations derived (r^ = 

 0.81 to 0.96). Yield (ovendry weight) was calculated from 

 the regression of weight estimates made in the field. 



Forage Quality 



In June 1980 we collected at random 20 plants of each 

 species in each tree-harvested and nonharvested plot 

 where they occurred in abundance. All species were sam- 

 pled at the anthesis stage of development. Sampling was 

 refined in 1981, and eight plants of each species (anthe- 

 sis stage) were harvested from each of the soil micro- 

 sites, duff, transition, and interspace, on each tree- 

 harvested and nonharvested plot. Grass samples were 

 clipped at 0.4 inch (1 cm) height and seed heads were 

 removed. Leaves were ovendried at 117 °F (47 °C) and 

 ground to pass through a 0.5-mm sieve. 



Plant materials collected in 1980 were run in dupUcate 

 through in vitro digestibility trials (TiUey and Terry 

 1963) using rumen inoculum from heifers maintained on 

 a grass hay diet. Plant materials for 1981 were analyzed 

 in dupUcate for in vitro digestibHity, total Kjeldahl 

 nitrogen-salicylic acid modification (Eastin 1976), and 

 phosphorus (sulfuric acid digest-colorimetric procedure 

 using ascorbic acid indicator: Watanabe and Olsen 1965). 

 Duplicate samples not within 10 percent of their mean 

 value were rerun. A standard forage sample was in- 

 cluded in each run and each run was adjusted to every 

 other run via the common standard. Gross energy of 

 each species was determined from four composite sub- 

 samples with a Parr adiabatic bomb calorimeter. Digesti- 

 ble energy (DE) was computed by microsite and whole 

 plots using the formula DE = Production (i) * Gross 

 Energy (i) * Dry Matter DigestibiUty (i) for each (i) 

 species, as suggested by Conroy and others (1982). 



Analysis 



The -experimental unit was the individual plant (20-24 

 repUcates) when we tested for differences in in vitro di- 

 gestibility, percentage of phosphorus, and percentage of 

 nitrogen between harvested and nonharvested plots. In 

 comparisons of the above parameters among soil micro- 

 sites there were eight rephcates per harvested and non- 

 harvested plot. Belt transects (five rephcates) were sub- 

 divided into individual soil microsite components, duff, 

 transition, and interspace. Microsite area per transect 

 served as the experimental unit. Differences in yield and 

 forage quaUty among microsites were evaluated on a per- 

 unit-area basis. The three rephcates of paired harvested 

 and nonharvested plots served as the experimental units 

 in the comparison of the composite microsite change in 

 forage yield and quahty following tree harvest. 



Analysis of variance and Hartley's sequential method 

 of testing (Snedecor 1956) were used collectively to test 

 for differences in total grass cover among years and in- 

 dividual species differences in yield, percentage nitrogen 

 and phosphorus, in vitro digestibihty, and plant density. 

 Orthogonal contrasts were used to test for differences in 

 forage quahty (digestible dry matter, digestible energy, 

 protein [6.25 x %N] and. phosphorus) among microsites 

 and tree-harvested treatments. 



2 



