Fractional passage rates form a fundamental element within modern feed evaluation systems for ruminants, but knowledge on feed-specific fractional passage is largely lacking. Commonly applied tracer techniques based on externally applied markers, such as chromium-mordanted neutral detergent fibre (Cr-NDF), have been criticised for behaving differently to feed particles. This study describes the use of the carbon stable isotope ratio (13C : 12C) as an internal digesta marker to quantify the fractional passage rate of concentrates through the digestive tract of dairy cows. In a crossover study, five dairy cows were fed low (24.6%) and high (52.6%) levels of concentrates (dry matter (DM) basis) and received a pulse-dosed Cr-NDF and 13C isotopes. The latter was administered orally by exchanging part of the dietary concentrates of low 13C natural abundance with a pulse dose of maize bran-based concentrates of high 13C natural abundance. Fractional passage rates from the rumen (K 1) and from the large intestine (K 2) were determined from faecal marker concentrations of Cr-NDF and of 13C in the DM (13C-DM), NDF (13C-NDF) and neutral detergent soluble (13C-NDS). No differences in K 1 estimates were found for the two concentrate levels fed but significant differences between markers (P<0.001) were observed. Faecal Cr-NDF excretions gave lower K 1 estimates (0.037-0.039/h) than 13C-DM (0.054-0.056/h) and 13C-NDF (0.061-0.063/h). The 13C-NDS was calculated by the difference of 13C in the DM and NDF, and K 1 values (0.039-0.043/h) were comparable to Cr-NDF. Total mean retention time was considerably higher for Cr-NDF (40.9-42.0 h) as compared to 13C-DM and 13C-NDF (32.0-33.5 h; P<0.001). The accuracy of the curve fits for Cr-NDF and 13C-DM and 13C-NDF was overall good (mean prediction error of 9.9-13.9%). Fractional passage rate of Cr-NDF was comparable to studies where this marker was assumed to represent the fractional passage of roughages. However, K 1 estimates based on the 13C : 12C ratio varied considerably from studies based on external markers. Our results suggest that the use of 13C isotopes as digesta passage markers can provide feed component-specific K 1 estimates for concentrates and provides new insight into passage kinetics of NDF from technologically treated compound feed.