Above- and belowground biomass of Idaho fescue Festuca idahoensis and bluebunch wheatgrass Agropyron spicatum and nematode densities under these plant species were sampled during the growing season inside and outside a 2-year old exclosure on Crystal Bench in Yellowstone National Park. Early in the growing season, grazed plants of both species had lower shoot and root biomass than ungrazed plants. Standing biomass of grazed plants was equal to ungrazed plants at the end of the growing season. Densities/g root biomass of phytophagous and bacterial feeding nematodes were higher under grazed than ungrazed plants of both plant species only early in the growing season. Foliar concentrations of nitrogen in grazed plants were higher than ungrazed plants but there was no difference in root nitrogen between grazed and ungrazed plants. The effects of ungulate grazing on the Northern winter range of Yellowstone National Park has recently received considerable attention (Frank 1990, Coughenour 1991, Singer 1992, Wallace submitted). Early interest in this topic centered around the question to cull or not to cull elk in the Park. However, as the concepts of "maintaining ecological processes" (Houston 1982) and "ecosystem management" (Keiter 1991) have gained acceptance in Park management, understanding the dynamics and interactions of a broader array of herbivores inhabiting the Park have become increasingly important. In this paper, we describe the results of a study which focused on the effects of aboveground herbivory on nematode density and trophic structure. Root-feeding nematodes are major herbivores in other grassland systems and may consume twice as much biomass as aboveground consumers (Ingham and Detling 1984, Stanton 1988). Houston (1982) reported that nothing is known about the effects of nematodes on the native grasses of the northern range especially in combination with aboveground grazers. We hypothesized that if spring grazing is intense, grazed plants would initially show a decline in root growth and phytophagous nematodes. Cessation of root growth is a common response of plants to grazing and may occur within the first 2-24 hours (Hodgkinson and Baas Becking 1977). Evidence to date supports the idea that phytophagous nematode densities are highest under moderate levels of grazing and low under heavily grazed and ungrazed plants (Stanton 1983, Stanton et al. 1984, Seastedt 1985, Seastedt et al. 1988). Because senescing roots, subsequent to grazing, provide increased substrates for decomposers, we also hypothesized that microbial activity and nitrogen mineralization should increase (Stanton et al. 1984). As a result, we expected to detect an increase in microbial feeding nematodes. As root regrowth occurred, we expected phytophagous nematodes to increase. However, we predicted that populations would not reach levels found under ungrazed plants because plants in grazed areas experience higher levels of nitrogen mineralization (Holland and Detling 1990) than ungrazed plants and may produce proportionally fewer numbers of root hairs (nutrient absorption organs) which serve as feeding sites for nematodes. Because of reduced densities of phytophagous nematodes and increased mineralization rates under grazed plants, we expected grazed plants to recoup their losses rapidly. The net result we predicted would be no detectable differences in aboveground or belowground biomass during years of normal rainfall. Thus, our study addressed 3 null hypotheses. First, root and shoot biomass of grazed and ungrazed plants will be similar at the end of the growing season. Second, density of phytophagous and microbial feeding nematodes will not differ between grazed and ungrazed plants. Finally, nitrogen concentration of roots and aboveground foliage will not be higher in grazed than in ungrazed plants. We focused our attention on bluebunch wheatgrass Agropyron spicatum and Idaho fescue Festuca idahoensis because of their importance as winter range forages and because Mueggler (1975) reported that bluebunch wheatgrass was more sensitive and recovered more slowly to heavy clipping than Idaho fescue.