The Temperate Pasture Sustainability Key Program (TPSKP) was established across southeastern Australia to test the hypotheses that an improved perennial grass content in pastures would result in fewer weeds, better water use (and hence lesser impacts on soil salinity), and lower soil acidification rates. Grazing tactics were seen as a means to enhance or maintain the perennial grass content. Soil and water sustainability experiments in summer and winter dominant rainfall environments showed fewer weeds, improvements in water use and less acidity under perennial versus annual grass pastures. Further work is needed to determine if these gains are sufficient to make perennial grass pastures sustainable in the long-term as some nitrate leakage still occurred at the winter rainfall site. Indicators were developed to rate the sustainability of treatments within experiments. A subset of these indicators was common across experiments and could readily be used by farmers to provide an initial assessment of the soil and water sustainability of their pasture systems. These are: the mineral nitrogen at the bottom of the root zone (40-60 cm); soil pH at the surface and bottom of the root zone and perennial grass content by species. Managing pastures through droughts is a critical aspect of grazing management in Australia. Experiments within the TPSKP demonstrated that perennial grasses survived during drought when maintained above critical lower biomass values. These values ranged from 0.5 to 1.5 t DM/ha depending upon species. Over all experiments, there was general support for the view that maintaining a higher level of biomass in pastures resulted in more sustainable systems. Twenty-three grazing experiments using an open communal grazing design showed that most perennial grasses were sensitive to grazing at some stage in their seasonal growth cycles. The exceptions were inconclusive for several reasons e.g. the grazing pressure may not have been high enough at those sites to elucidate any effects; they occurred where the perennial grass content was less than 10% or exceeded 70%, of the sward; or were confounded by interactions between species where the species under study was not dominant. After taking these exceptions into account, it was then possible to determine where grazing tactics could be expected to work. Species differed in their response to grazing. Some perennial grasses were more sensitive to grazing during periods of stress (e.g. dry summers) than when actively growing (e.g. cocksfoot), while the reverse applied with others (e.g. phalaris). Of the grasses sensitive to grazing when actively growing, sensitivity of some was largely confined to the reproductive period (e.g. perennial ryegrass). Across most experiments, continuous grazing resulted in either a decline in or no net benefit to, the perennial grass content. Microlaena stipoides was the only species to respond to increased grazing pressure - this only applied in spring. The experiments clearly showed that tactical rests were an important tool for grassland management. The effects recorded were predominantly expressed through impacts on vegetative growth and survival of existing plants. Short-term experiments and dry seasons did not enable recruitment processes to be studied. Within pastures, grazing tactics can influence many species. The challenge is to use the TPSKP outcomes to develop strategies that optimise the composition of these swards. Due to the short-term nature of these experiments the results were evaluated within a conservative framework and often simply on the absolute level of parameters. Techniques need to be developed to more effectively monitor the process (i.e. rates of change), rather than the consequences (i.e. ends). The information gained in this program needs to be incorporated into practical strategies for better management of pastures and tested at a commercial scale. The TPSKP was one of the largest, coordinated pasture programs ever attempted. Some major outcomes were the experience gained by a large number of grassland scientists in running such programs, the development and acceptance of standardised measurement protocols and a much stronger network among grazing systems scientists committed to achieving improved management systems.