The amount of reactive N in soils on the Australian continent appears to be increasing, mainly because of biological N-fixation by permanent pastures in the dryland farming zone. This gain is partly offset by N-mining by crops, which we estimate have removed between one-fifth and one-quarter of the original soil N. The vast areas of non-agricultural land and arid rangelands appear to be in neutral N balance and the relatively small area of intensive agriculture is in negative balance. There are regional N losses from the sugar and dairy industries to groundwater, estuaries and lagoons, including the Great Barrier Reef. Fertiliser N application is increasing and is likely to increase further, to compensate for the soil-N mining and to meet increasing crop yield potential, but fertiliser-N represents a relatively small fraction of the Australian N balance. The dryland farming zone utilises the largest amounts of native and fertiliser N. The average fertiliser application to dryland cereals and oilseeds, 45kg N ha-1, is low by international standards because of the small N-demand by dryland crops and because there are no subsidies on crops or fertiliser that promote overuse. The efficiency of N-use is relatively low, for example about 40% of fertiliser N is recovered in the aboveground parts of dryland wheat and the rest is retained in the soil, denitrified or otherwise lost. We suggest further research on fertiliser-application methods to increase crop recovery of fertiliser, as well as research to reduce the surplus N from permanent pasture.