The serum response of beta-carotene as an indicator of bioavailability was compared after feeding beta-carotene (0.8 mg/kg body weight) either from grass meal or a synthetic beadlet preparation (Lucarotin). Both were each given without or with added dietary vegetable fat (2-2.5% vs. 6.6% fat in dry matter) in a Latin square design with four horses. The nutritionally complete diet was supplemented with alpha-tocopherol (4 mg/kg body weight). Each treatment period (4 weeks, two serum samples) was followed by a washout period of 4 weeks with low intakes of beta-carotene (traces) and alpha-tocopherol (0.5 mg/kg body weight). Within 4 weeks of supplementation, serum beta-carotene increased about 10-fold, from a mean initial concentration of 0.05-0.53 micromol/l. There was no effect of beta-carotene source and of fat addition, respectively. Faecal excretion of beta-carotene ranged from 55 to 81% of intake. No beta-carotene was detected in any urine sample. Serum alpha-tocopherol (across all time points and animals, n=64) was 14.5 micromol/l. During supplementation, the values were significantly higher than during washout-periods. Additional dietary fat did not affect the serum response. Faecal excretion of alpha-tocopherol ranged from 69 to 121% of intake. Fat addition resulted in a significant decrease of serum cholesterol. In conclusion, the natural and the synthetic source of beta-carotene showed significant and identical bioavailability independent of additional fat.