Maize (Zea mays)-mucuna (Mucuna pruriens) systems have been promoted to the smallholder farmers of the Los Tuxtlas region of southeastern Veracruz, Mexico. To determine on-farm performance, an agronomic assessment was conducted in 1995–97 replicating farmer conditions in four fields. Treatments were first- and second-season maize with first-season mucuna (system Zm-Mp/Zm), first-season maize with first- and second-season mucuna (system Zm-Mp/Mp), second-season maize following first-season mucuna (system Mp/Zm), and first- and second-season maize, no mucuna control. Data on mucuna biomass amount and quality as well as maize yield, yield components, and nutrient status were collected. Highest mucuna biomass was obtained in system Mp/Zm (leaf-stem-mulch biomass in 1996/97, 7.34 t ha−1, 147 kg ha−1 N), followed by systems Zm-Mp/Mp (5.06 t ha−1, 101 kg N ha−1) and Zm-Mp/Zm (2.75 t ha−1, 50 kg N ha−1). Second-season maize yield was increased over that of the control by 45–58% (0.15–0.23 t ha−1) in system Zm-Mp/Zm and by 118% (0.60 t ha−1) in system Mp/Zm. Mucuna did not increase first-season maize yield. Climatic constraints make second-season maize production risky and yield increases due to mucuna are low in absolute terms, perhaps not offsetting labour costs (systems Zm-Mp/Zm and Mp/Zm) or loss of first-season maize (Mp/Zm).