Alkaline calcareous or sodic soils represent an important proportion of the world's arable soils and are important for cereal production. For calcareous soils in general, despite high applications of P fertiliser for many years, P deficiency in cereals is common. Field experiments were conducted to test the relative ability of granular (e.g. DAP, MAP and TSP) and fluid fertilisers to supply P to wheat on grey calcareous and red brown calcareous sandy loam soils (Calcixerollic xerochrepts). A pot experiment was also conducted with these soils and with two non-calcareous alkaline soils to investigate the effects of placement on the efficiency of fertiliser performance. In 1998, fluid and granular sources of P, N and Zn were compared in the field by banding below the seed at sowing. In 1999, MAP applied as granular, and technical grade MAP applied as fluid, were compared as sources of P in rate response experiments. First year results showed that fluid sources of P, N and Zn produced significantly more grain than the granular product. In the following year, fluid fertilisers were found to produce significantly higher response curves for shoot dry weight, grain yield and P uptake in grain. At a commercial rate of 8 kg P ha(-1), fluid fertiliser produced between 22% and 27% more grain than the granular product. Soil moisture and fertiliser placement effects are implicated in the higher efficiency of fluid fertilisers.