In a feeding trial with 24 sheep, we used the alkanes, long-chain alcohols (LCOH) or both of these plant wax markers, to estimate the diet composition of animals offered diets comprising alkane-labelled cottonseed meal (CSM) together with up to four forages. The diets used were: Diet 1 subterranean clover (Trifolium subterraneum); Diet 2 subterranean clover + phalaris (Phalaris aquatica); Diet 3 subterranean clover, phalaris + annual ryegrass (Lolium rigidum); and Diet 4 subterranean clover, phalaris, annual ryegrass + wheat straw (Triticum aestivum). Estimates of diet composition were made following correction of faecal alkane or LCOH concentrations for incomplete faecal recovery, using recovery estimates derived from individual animals, mean recoveries for a given dietary treatment or grand mean recoveries. Estimated dietary proportions of CSM and known intakes of CSM were used to estimate forage intake. The LCOH concentrations of the diet components were much higher than their alkane concentrations, especially for phalaris. Multivariate analyses showed that the discriminatory information provided by the LCOH was additional to that provided by the alkanes, and that a combination of (LCOH + alkanes) discriminated better between diet components than either class of marker alone. Faecal recoveries of LCOH increased with increasing carbon-chain length; there were no differences in recovery attributable to diet. The most accurate estimates of diet composition were obtained with the combination of (LCOH + alkanes). Estimates of diet composition based on LCOH alone were not as good as alkanes alone, due to the high correlation between the LCOH profiles of phalaris and ryegrass. Total grass content of the diet was very accurately estimated using LCOH. Diet composition estimates provided estimates of whole-diet digestibility, which did not differ from the measured values. Trends in the accuracy of forage intake estimates reflected those found with diet composition and almost two-thirds of estimates based on (LCOH + alkanes) had lower error than those found with alkanes alone. The results confirm that supplements labelled with plant wax components can be used to estimate forage intake, and also show that the LCOH are useful markers for estimating diet composition. Intakes were also computed using a combination of natural LCOH concentrations in the diet and the daily dose rate of even-chain alkanes administered by intra-ruminal device. Differences between intakes so estimated and the measured intakes were closely related to the difference in faecal recovery between the LCOH/alkane pair used to estimate intake, by an amount close to that expected on theoretical grounds. It is concluded that the use of plant wax LCOH, especially in combination with alkanes, will result in improved estimates of diet composition and intake in grazing animals.