Processes causing crop establishment damage from ammonium (NH4+) fertilizer placed in close proximity to seed at sowing are generally poorly understood within farming communities of eastern Australia. Currently, the information used to assess establishment hazard includes nitrogen (N) tolerance for a limited range of crop species. Other factors include the N concentration of fertilizer products, with adjustment of the rate expected to be tolerated on the basis of soil moisture and application equipment. Current recommendations were adapted from North American research spanning a period from the start of the century until the 1990s and some recent Australian research (1960s to 1980s) for a limited range of crops and fertilizer types. The incidence of seedling damage from N fertilizer and fertilizer containing other nutrients appears to have grown in recent years. This may be attributed to use of a wider range of NH4+ and other fertilizers, a trend for at-sowing application in zero-tillage and expansion of areas of declining soil fertility, particularly N fertility. Other factors include the sowing of new crops with greater fertilizer sensitivity, low tolerance to establishment loss for high value genetically modified seed and modern designs of sowing and application equipment. The major objective of this research was to investigate ammonia (NH3) tolerance of 10 crop species of importance for eastern Australian cropping systems (maize, Zea mays L.; cotton, Gossypium hirsutum L.; wheat, Triticum aestivum L.; barley, Hordeum vulgare L.; chickpea, Cicer arietinum L.; sorghum, Sorghum bicolor (L.) Moench; canary, Phalaris canariensis L.; canola, Brassica napus L.; panicum, Setaria italica L. and sunflower, Helianthus annuus L.). Experiments were designed to highlight differences among crops in NH3 toxicity and osmotic damage potential for commonly used NH4+ fertilizers. Various strategies were then tested to maintain plant populations within commercially acceptable ranges when affected by NH3 toxicity and/or high osmotic pressure. Tolerance of seeds to NH3 toxicity was evaluated in the field and for atmospheric exposure. Response of various crop species to atmospheric-NH3 exposure showed that certain species responded differently in their germination, coleoptile growth and radicle growth in a closed system. Using these 3 parameters as indices of crop response to NH3 toxicity revealed different ranking for some species; the same species showed a different critical NH4+ concentration for each parameter. Exposing seeds above 200 x 10-4 M NH4OH for 72 h was sufficient to significantly reduce or inhibit germination of all 10 species tested. Seed of most species were unaffected by exposure above 20 x 10-4 M NH4OH. Species rank, combining tolerance for germination, coleoptile growth and radicle growth was established to relate to likely performance in the field. Decreasing order of tolerance for monocot species was: maize > sorghum > wheat = barley > panicum > canary and for dicot species chickpeas > cotton > sunflowers > canola. A range of physical and chemical seed characteristics was correlated with NH3 tolerance to investigate tolerance mechanisms. For monocot species, tolerance was related to the seed surface area/volume ratio suggesting that diffusion resistance was an important parameter whereas for dicot species N concentration of seed was negatively correlated with tolerance. In field experiments where NH4+-fertilizers were placed with seeds, difference between species in their tolerance of atmospheric-NH3 was insufficient to describe effects of NH4+-fertilizers on crop emergence. Crop species fell into 3 response categories; high (maize, sorghum, barley and wheat), medium (cotton, canary and sunflower) or low (canola, chickpea and panicum) tolerance to soil NH4+-N. Soil NH4+-N concentrations tolerated by the medium and low tolerance group was 50 % and 15 to 25 % respectively, that of the high tolerance group. Generally, NH3 tolerance response for species such as wheat, barley and sorghum was found similar to current recommendations for urea (~0.5 g/m N as urea) but there were significantly different responses to NH4+-N from different NH4+-fertilizer products, that are not recognised in current recommendations. Crop species were ranked for sensitivity to mono-ammonium phosphate (MAP), di-ammonium phosphate (DAP), triple superphosphate (TSP), urea and ammonium nitrate, and categorised according to the fertilizer rate at which significant establishment damage occurred. Ranking of crop species for NH3 toxicity was generally similar across experiments but the NH4+-N rate tolerated varied with experimental conditions. Urea and DAP caused larger reductions in establishment than equivalent NH4+-N rates from MAP or ammonium nitrate. The "safe" rate for ammonium nitrate (1 g/m NH4+-N) was approximately twice that of urea at equivalent NH4+-N rates. Usually between 20 and 30 % more NH4+-N was tolerated for MAP than for DAP. In the absence of NH3 toxicity, osmotic effects of fertilizer products delayed and occasionally inhibited germination. There was significant difference among species in osmotic tolerance; cotton, maize and sorghum (< -0.3 MPa) were more tolerant than sunflower or soybean (> -0.2 MPa). Strategies to improve crop establishment in the presence of NH4+ fertilizer such as increasing seeding rate, adding water to the seed furrow, changing fertilizer N source and chemically modifying hydrolysis of urea were identified and tested. For low to moderate rates of seed placed NH4+-N, increasing barley seeding rate from 25 to 40 kg/ha was found to be successful strategy to maintain establishment when urea rate was increased from 1.1 to 2.3 g/m of seed row. Changing the fertilizer N source and modifying hydrolysis of urea were successful in lowering soil NH4+ around the seed and thus reducing establishment losses, but osmotic effects also limit the maximum fertilizer application rates. The added cost of these strategies may prevent their widespread adoption. Complex interactions between crop species, fertilizer product, soil texture and moisture, and application equipment highlighted by the results of these experiments, suggest that simple decision tools are insufficient to provide fertilizer recommendations that meet the demands of modern agriculture. A computer based decision support programme, Fertsafe, was developed during this study from experiments conducted and papers reviewed, to provide "safe rate" recommendations to apply fertilizer at sowing for a range of crop species, fertilizers, soils and sowing conditions of eastern Australia. Changes to fertilizer application equipment, other crops and fertilizer products will require ongoing research continuously improve and update this decision support tool.