Root decay and turnover of rhizodeposits in field-grown winter wheat and spring barley were estimated from the difference in C-14 distribution between a sampling carried out 3 wk after pulse-labelling and a sampling 15-20 days before crop harvest. Plants were labelled at tillering and at ear emergence on plots under conventional (CONV) and integrated (INT) management. The latter was characterised by reductions in nutrient input, soil tillage and use of biocides. Of the roots formed at the tillering stage, 43% had decayed by the end of crop growth in winter wheat, independent of the management practice. Root decay in spring barley was higher than in wheat, and was higher in CONV (55%) than in INT (49%). Since most of the roots were produced around tillering, these values were used as estimates of accumulated annual root decay. Of the total rhizodeposition of assimilates fixed at tillering 87% was respired by microbes by the end of the growing season with winter wheat, 69% with barley in CONV and 54% with barley in INT. Decay of roots formed at ear emergence and the associated microbial respiration of rhizodeposits was lower than that of roots produced at tillering, but the same crop and management effects were observed. The assumptions underlying the calculations are discussed and the results are compared to data from minirhizotron observations.