A three-year (1989-91) field study was conducted to assess the effect of cowpea-melon populations on crop productivity and land equivalent ratios (LERs) in a cassava-maize system in Uyo, south-eastern Nigeria. A split-plot design replicated thrice was used. Cropping systems constituted the main plot and crop populations the sub-plots. Cassava and maize were planted at populations of 10 x 103 plants and 20 x 103 plants ha-1, respectively, cowpea at 20, 33, and 50 x 103 plants ha-1, and melon at 5, 10, and 15 x 103 plants ha-1. A higher number of cassava storage roots per plant and fresh root yield, number of maize grain seed per cob, and grain yield in the intercrop were produced at the low cowpea population (20 x 103 plants ha-1) than at the higher populations. However, cowpea and melon populations showed no significant effect on cassava and maize harvest indices irrespective of the season of planting. Intercropping generally reduced the number of cowpea pods by 35% and also the number of seeds per pod. Melon pod number per plant and seed number per pod were significantly (P < 0.05) affected by intercropping. Cowpea grain and melon seed yields increased with increase in populations. As in cassava and maize, cowpea and melon indices were not affected by cowpea and melon populations. Grain yield and seed yield components of cowpea and melon were significantly (P < 0.05) higher in the sole crop due to the absence of competition from companion crops. The cassava + maize + cowpea + melon system resulted in the highest LERs of 2.29, 2.62, and 3.12 in 1989, 1990, and 1991, respectively, due to the low population combination of 20 000 cowpea plants with 5000-10000 melon plants ha-1. The lowest LERs (2.19, 2.12, and 2.31 for the three respective years) were obtained from the highest population combinations of 50 000 cowpea plants and 15 000 melon plants ha-1. This suggested an effectively-balanced competition among the crop species leading to 37-63%, 41-49%, and 26-70% greater LER for 1989, 1990, and 1991, respectively, than if the crops were grown sole. This LER is of utmost significance for the study area where the available arable land is seriously being threatened by rapid urbanization and increased production density.