An enzymic method for the determination of L-malic acid in fruits and vegetables is described. Malate dehydrogenase (MDH) and diaphorase (DI) were co-immobilized on aminopropyl-, isothiocyanate-and N-hydroxysuccinimide ester (N-HSE)-modified controled-pore glass, which was then used for the construction of a packed-bed reactor. MDH catalyses the reaction between malate and NAD⁺ and produces NADH, which converts hexacyanoferrate(III) into hexacyanoferrate(II) in the presence of the DI. Amperometric measurements with pure graphite electrodes in a fully automated flow injection manifold were used to monitor the concentration of hexacyanoferrate(II) at +0.3 V versus an Ag/AgCl reference electrode. The electrochemical behaviour of hexacyanoferrate(III), 2,6-dichlorophenolindophenol and p-benzoquinone was tested with cyclic voltammetry. Parameters such as working and immobilization pH, flow rate, sample size, temperature and NAD⁺ and hexacyanoferrate(III) concentrations were investigated. The interference of various ions, amino acids and organic acids present in real samples was also studied. The calibration graph was linear for the range 2 × 10^–5–4 × 10^–4 mol l^–1 malate. The reactors remained active for a period of more than 3 months under specified storage conditions. The maximum sample throughput was 30 h^–1 and the relative standard deviation of the method was better than 1.5%(n= 5). The mean relative error was 4% for fruits and 2% for vegetables. The accuracy of the proposed method was tested by comparison of the results with those obtained by a reference method. Good correlation was attained. Recovery experiments showed results between 95 and 108%.