Salvage-logging and artificial seeding of grass following wildfire are common practices in coniferous forests of the western United States, yet few studies have quantified the ecological effects of these post-fire activities. The effects of post-wildfire salvage-logging and grass-seeding on vegetation composition, aboveground biomass, and growth and survival of Pinus ponderosa and Purshia tridentata were quantified on the area burned by the 1992 Lone Pine Fire, Winema National Forest in the Klamath Basin, Oregon. Prior to the fire, the area was dominated by uneven-aged stands of Pinus ponderosa with Purshia tridentata and Stipa occidentalis in the understory. The fire was a stand-replacement disturbance, where the majority of trees, the herbaceous component, and crowns of understory shrubs were killed. Salvage logging resulted in a significant decrease in understory biomass, species richness, species diversity, and growth and survival of P. ponderosa and P. tridentata. In addition plant community composition was shifted from native forb dominance to grass dominance. In 1993, the understory biomass of salvage-logged sites was 38% of the aboveground biomass produced on nonsalvaged sites (322 kg ha^-1 vs 843 kg ha^-1). In 1994, salvage-logged sites produced only 27% of the biomass produced on nonsalvaged sites (402 kg ha^-1 vs 1468 kg ha^-1). Salvage-logging reduced species richness, species diversity, and altered species composition. The first and second years following logging, species richness was reduced by 13% (20 versus 23), and 30% (15 versus 22), respectively. In 1993 and 1994, native forb frequency on nonsalvaged sites was 80% and 77% respectively, while salvage-logged sites recorded 68% and 31% respectively. Conversely, graminoid frequency was significantly higher on salvage-logged sites. In 1994, native graminoid frequency was 35% in nonsalvaged sites and> 61% on salvage-logged sites. During the first two years following salvage-logging, mean height growth of naturally-regenerated Pinus ponderosa was significantly lower on salvaged sites (9.4 versus 7.8 cm yr^-1) as was density of natural Purshia tridentata seedlings (313 versus 530 seedlings ha^-1). Survival in salvage-logged treatments was 22% lower for planted Purshia tridentata seedlings (57% versus 45%). Height growth in salvage-logged treatments was 16% lower for planted Pinus ponderosa (4.4 versus 3.7 cm). Total aboveground biomass on nonsalvaged burned sites (controls) averaged 843 kg ha^-1 in 1993, and 1473 kg ha^-1 in 1994. In 1993, the first year following the fire, sites seeded to Secale cereale produced = 1995 kg ha^-1 total aboveground biomass; most of the biomass was S. cereale. Those seeded sites produced 89% less native forb biomass than controls (82 versus 780 kg ha^-1), and = 80% less native forb biomass than areas seeded to the native grasses Sitanion hystrix and Festuca idahoensis. In 1994, the second post-fire year, biomass on sites seeded to S. cereale was>1653 kg ha^-1, dominated by S. cereale. In 1994, no differences in total aboveground biomass were detected between S. cereale, F idahoensis, S. hystrix, and the Control, indicating that the erosion reducing benefits of grass-seeding did not last for more than one year. Treatments seeded with S. cereale produced 58% less native forb biomass than controls (350 versus 825 kg ha^-1). Seeding S. cereale also reduced by 69% the biomass ofStipa occidentalis, the most common native grass in the area (311 kg ha^-1 on control sites versus 96 kg ha^-1 on S. cereale sites). In 1993 and 1994, fewer species were recorded on S. cereale sites than on all other sites. Grass-seeded sites exhibited no significant differences in frequency of noxious weeds in either of the two years following wildfire and seeding, rejecting the hypothesis that grass-seeding reduces noxious weeds. However, native graminoid frequency was reduced on all grass-seeded treatments. There were no significant differences between grass-seeding treatments in height and density of natural P. ponderosa and P. tridentata seedlings or growth and survival of planted P. ponderosa and P. tridentata seedlings. While salvage-logging provides an economic benefit to local communities through the extraction of commercially valuable timber, there is an ecological cost. Reduction in ecosystem structure, production, species richness, diversity and alterations in species composition are important ecological consequences of salvage logging. In addition, this study demonstrates that salvage-logging retards the re-establishment and early growth of Pinus ponderosa and Purshia tridentata, two important wildfire restoration priorities. Grass-seeding also results in significant ecological changes. Alterations in ecosystem production, species richness, diversity and species composition are important ecological consequences of grass-seeding. Managers should consider these long-term influences on ecosystem composition and structure when faced with decisions concerning post-fire rehabilitation and management.