The primer-plant concept was tested for wheat (Triticum aestivum) grown on an alkaline sodic soil taken from the southern Mallee of Victoria. This concept relates to use of species of plants with high natural adaptation to hostile subsoils, being able to modify the soil environment and leave biopores for the benefit of subsequent annual crops. For the experiment reported here, wheat was sown into large (0.3 m diam. by 1.0 m length) intact soil cores (collected from a cropping paddock near Birchip in the southern Mallee region of Victoria, Australia) following either birdsfoot trefoil (Lotus corniculatus), canola (Brassica napus), chicory (Cichorium intybus), lucerne (Medicago sativa), safflower (Carthamus tinctorius), sulla (Hedysarum coronarium), or tall wheatgrass (Thinopyrum ponticum). At the conclusion of the priming phase [270 days after sowing (DAS)], all the different crops extracted c. 145 mm of stored water, the exception being canola (120 mm). Lucerne and birdsfoot trefoil produced the least above-ground biomass (26 g/pot), and safflower the most (115 g/pot). Greater early vigour and water extraction (49 mm) occurred for subsequent wheat crops after birdsfoot trefoil than with wheat after all other species (39 mm). This translated to a 15% yield advantage for wheat after birdsfoot trefoil compared with lucerne. Wheat after sulla yielded 12% more due to increased grain number and kernel size compared with wheat after lucerne. It was proposed that the difference in yield related to the root systems of species tested. Birdsfoot trefoil and sulla were characterised by intensive branching, which potentially produced a fine mosaic of residual biopores. Lucerne, in contrast, which was assumed to have similar break-crop effects, had a large taproot with fewer branches leaving fewer, larger residual root channels than either of the other legumes. It is believed that the fine biopores allowed more rapid and thorough exploration of the bulk soil by the crop roots.