Wheat (cv. Gutha) and barley (cv. O'Connor) were grown as field crops on a shallow duplex soil (sand over clay) in Western Australia with their root systems contained within pvc columns. At four stages during growth, the shoots were pulse-labelled for 1.5h with14CO2; immediately prior to labelling, the soil was isolated from the shoot atmosphere by pvc sheets. After labelling, the soil atmosphere was pumped through NaOH to trap respired CO2 and after 2.5, 5, 7.5 and 24 h from the start of labelling, columns were destructively sampled to recover14C from the roots, soil and shoot. Both species showed similar patterns of14C distribution and changes in distribution through the growing season. During early tillering, 15-25% of the14C recovered after 24 h had been respired by the roots and rhizosphere, 17-27% was retained in the roots, 0.4-1.8% was recovered as water-soluble14C in the soil and the remainder (45-67%) was present in the shoot. These percentages changed during growth so that during grain filling only 2-3% of the14C recovered after 24 h was as respired CO2, 2-6% was in the roots, 0.2% was in the soil and over 90% was in the shoot. The distribution of14C in components of the soil-plant system changed during the 24 h after labelling with the most rapid changes occurring generally during the first 7.5 h after labelling. Using growth measurements from adjacent plots, the amounts of C added to the soil were estimated for the whole season. Carbon input to the soil was about 48 gC m-2 for wheat and 58 gC m-2 for barley; the crops produced total shoot dry matter of 494 (wheat) and 735 g m-2 (barley). Of the C input to the soil, 27.8% (wheat) and 40.3% (barley) was as respired C and only 3.3 (wheat) and 4.1% (barley) was collected as exudate (water-soluble material).