The potential use of faecal n-alkanes for estimation of intake and diet composition over periods of 1-2 days was assessed in two experiments. The aim was to determine the accuracy with which intake and diet composition could be estimated by characterizing faecal excretion of n-alkanes following a discrete dose as opposed to steady state kinetics used in previous work. In the first experiment, 16 sheep were fed mixtures of spinach (rich in C31-alkane) and cabbage (rich in C29-alkane) in known proportions and amounts for two days. Artificial n-alkanes (C28- and C32-alkane) were dosed on four occasions during this time. Total intakes were controlled at 0.2, 0.3, 0.4 or 0.5 kg dry matter (DM) per day and nominal amounts of spinach offered (as a proportion of the total diet) were 0.00, 0.15, 0.30 and 0.45. Each sheep received a unique combination of intake and dietary proportions (four intake rates x four proportions). Sheep were fed fresh grass (timothy, Phleum pratense) before and after feeding spinach and cabbage. Sequential rectal grab samples of faeces were collected at regular intervals and total faecal collections were carried out over 144 h from the start of the spinach/cabbage feeding period to obtain samples for n-alkane analysis. In the second experiment, fresh grass was sprayed with two combinations of artificial n-alkanes (C24- and C32- or C28- and C36-alkane) and fed to 16 sheep over a 24 h period. Each sheep received one of four intake rates (0.8, 1.0, 1.2 or 1.4 kg DM/day) and, within intake rates, each sheep received one of four different proportions of the herbage sprayed with the combinations of n-alkanes (0.2, 0.4, 0.6 or 0.8), in a similar fashion to the first experiment. In order to estimate intake, C26- and C34-alkanes were dosed at the start of the feeding period. Faecal sampling procedures were the same as those in the first experiment. Different parameters of faecal excretion curves of dosed and natural n-alkanes were used to estimate dietary proportions and intake. Parameters tested included area under the excretion curve and curve maximum. Dietary proportions were calculated using an iterative minimization procedure employing faecal and herbage n-alkane concentrations. Intakes were estimated using ratios of dosed: natural faecal n-alkanes. The best estimates of dietary proportions were obtained using faecal concentrations at a single point in time in both experiments (> 80% variation explained for regressions of estimated v. actual proportions). Intake estimates required the calculation of the area under the excretion curve to obtain acceptable estimates (70-90% variance explained for regressions of estimated v. actual intakes in Expt 2). The experiments demonstrate that precise estimates of diet composition can be obtained using single faecal samples following consumption of simple herbage mixtures over 24-48 h. Intake may also be estimated using this technique provided that a series of faecal samples are collected over 4-5 days following the period of ingestion.