Among more than 50 provitamin carotenoids, beta-carotene is the most metabolically active source of retinol. Despite diets rich in fruits and vegetables containing beta-carotene, vitamin A deficiency is the leading cause of blindness and childhood mortality in developing countries. In addition, the uncertainty of beta-carotene bioconversion into vitamin A suggests that new data are needed to update the nutritional guidelines in developed countries. Previously, we reported the development of a carotene/retinol plateau isotopic enrichment method (CarRet PIE) for the determination of beta-carotene bioavailability and bioconversion into retinol, which utilizes positive ion atmospheric pressure chemical ionization (APCI) liquid chromatography/mass spectrometry (LC/MS). While seeking to validate the CarRet PIE using a mass balance approach requiring fecal measurements of beta-carotene and retinol, interference was encountered that required substantial modifications of the LC/MS assay. Here we report a new LC/MS assay that is based on the detection of molecular anions of beta-carotene using negative ion APCI with a reversed-phase C30 column for HPLC separation. Sample preparation required saponification to eliminate interfering triglycerides. The limit of detection (LOD) of beta-carotene was 0.25 pmol calculated on the basis of an injection of 20 microL of 0.0125 microM beta-carotene, and the limit of quantitation (LOQ) was 1.0 pmol based on the injection of 20 microL of 0.050 microM beta-carotene. The linear range was 1.1 to 2179 pmol on-column. The wide linear range and low LOD and LOQ of this assay facilitated the sensitive and selective quantitative analysis of beta-carotene in both serum and fecal samples in support of an on-going clinical investigation of beta-carotene bioavailability and bioconversion into vitamin A.