THE USE OF INTERNAL MARKERS TO PREDICT TOTAL DIGESTIBILITY AND DUODENAL FLOW OF NUTRIENTS IN CATTLE GIVEN 6 DIFFERENT DIETS

Six ruminally and duodenally cannulated cattle were used in a 6 x 6 La tin square experiment with a 2 x 3 factorial arrangement of treatments . The cattle were fed diets based on silage or dried grass cut simulta neously from the same sward, each fed with 250, 500 or 750 g concentra te dry matter (DM) kg-1 total DM. The feed components investigated as potential markers to determine total DM digestibility (DMD) were acid insoluble ash (AIA) and acid detergent lignin (ADL), indigestible DM ( IDM41), neutral detergent fibre (INDF41) and acid detergent fibre (IAD F41) determined by a 288 h ruminal incubation in nylon bags with a por e size of 41 mum, and corresponding indigestible fractions determined by using bags with a pore size of 6 mum were IDM6, INDF6 and IADF6. IN DF and IADF were also determined by a 96 h in vitro incubation in rume n fluid (IVINDF, IVIADF). If the criterion of an ideal marker is to pr edict the digestibility and the differences between the diets, none of the markers met this criterion. Of the markers evaluated, the most su itable was AIA, followed by INDF6, INDF41 and IADF41. ADL was the leas t acceptable marker. Three double-marker systems were used to calculat e duodenal flows of organic matter (OM), non-ammonia N (NAN), microbia l N and NDF. Cr-mordanted straw, INDF41 and IADF41 were used as partic ulate markers in conjunction with Co-EDTA as a liquid phase marker. Du odenal OM flows were also estimated using a single-marker system. Doub le-marker systems reduced variability in duodenal OM flow estimates co mpared with single-marker systems. The choice of the particulate marke r affected (P<0.001) duodenal flow estimates but there were no indicat ions of marker x diet interactions. Neither was the statistical signif icance of dietary effects changed. When compared with the values obtai ned with Cr as a particulate marker, INDF estimated a greater relative increase in NDF flow (0.143) than in OM (0.062) or NAN flow (0.018). This suggests that there may be differences in the distribution of sol id phase markers in various particle size fractions. In such cases, if the digesta sample contains different proportions of large and small particles compared with the true digesta, then a double-marker system may not be sufficient to correct unrepresentativeness of digesta sampl ed from a simple T-shaped cannula.