The goal of this work was to quantify the effects of eucalypt woodland blocks on the productivity of native pastures. This research was conducted on the Southern Tablelands of New South Wales. Tree planting or retention is seen by many as an important tool in addressing the problems of soil degradation resulting from clearing and pasture improvement that threaten the sustainability of pasture systems. In particular these are dry land salinity and erosion, both of which affect large areas of agricultural lands in the south east of Australia. Whilst native tree cover remains over substantial portions of Australian pasture lands, mainly on steeper slopes and poorer soils, little has been done to measure the effects of trees on pasture productivity and soil fertility on the Southern Tablelands. Previous studies in other areas have shown a range of effects�from facilitation to inhibition�of pasture growth in the presence of trees. Soil fertility beneath trees has been shown by a number of workers to be elevated in comparison with situations in the open. Given that the range of effects may be highly site dependent, application of results from one area to another may not be valid. Thus it is necessary to measure tree effects on a regional scale if results are to be reliable. Pasture productivity was assessed over a two year period on four sites in the vicinity of Bungendore, New South Wales. A pair of plots was selected on each site, one plot in a block of eucalypt woodland, and the other nearby in an exposed, open situation. Plots were chosen to be as similar to each other as possible with the exception of tree cover. Treed plots had a tree basal area of between 10 and 20 m2 ha-1 and plots had an area of 900 m2. Two of the sites were on granitic soils and had a tree cover consisting predominantly of Eucalyptus pauciflora. The remaining two sites were on soil derived from sedimentary rocks with tree cover consisting mainly of E. mannifera, E. dives and E. melliodora. Perennial native pasture species present were similar across all sites, although their relative contributions to standing biomass varied between sites. As the plots were grazed during the period of measurement, productivity and offtake were measured seasonally using exclosure cages on each plot. Pasture standing biomass was assessed using the comparative yield technique. Microclimate was monitored in each plot by automatic weather stations. Soil moisture to a depth of 45 cm was measured by time domain reflectometry using permanent probes in each plot. Ten additional survey plots on each site, covering the range of tree basal area from 0 - 30 m2 ha-1, were assessed each season in the second year for standing biomass, soil fertility and pasture quality; expressed by nitrogen content and dry matter digestibility. Pasture floristics were measured using the dry-weight-rank method. These additional plots were chosen to be as representative of the paddocks as possible. Over the two years that productivity was measured, it was found to be higher under trees than in the open. This was predominantly due to higher winter and spring growth within treed plots. Grazing offtake was also found to be higher under trees, partly accounting for lower standing biomass found in the treed plots. Wind run, evapotranspiration and photosynthetically active radiation were all reduced by the presence of trees. Beneficial effects of shelter from winds may largely explain the higher productivity observed in the treed plots, and could outweigh negative effects of below ground competition and radiation interception by tree canopies at low to moderate tree densities. Soil moisture was not affected by the presence of trees. Soil fertility also did not differ between treed and open plots nor was there any difference in pasture nitrogen content or dry matter digestibility. On the sites where soils were derived from sedimentary rocks, pasture floristics were found to be related to tree basal area. Themeda ausfralis biomass was negatively related to tree basal area, and was partially replaced by large tussock species such as Poa sieberiana and Chionochloa pallida. A reduction of pasture quality resulted, particularly as the latter species is not grazed to any significant extent. Given the desirability of having deep rooted perennial components in grazing lands, the results of this study indicate that it may be possible to utilise trees to assist in preventing or reducing a range of adverse environmental consequences arising from agricultural activities, without unduly compromising pasture productivity. Additionally, the wide range of environmental conditions provided by a mix of treed and open pasture promotes a higher degree of heterogeneity of the herbaceous layer. This may assist in maintaining productivity over a greater range of climatic conditions than would be the case with a more homogeneous pasture.