A field study was conducted on an experimental field of the Maize Research Institute Zemun Polje-Belgrade (latitude: 44° 49′ N), Yugoslavia over a 3 year period (1994–1996). The objective of this study was to find the optimal spatial arrangement of a maize–beans intercrop in irrigated and rainfed farming systems.

Plant arrangement patterns in an intercropping system did not significantly affect LAI values in maize compared with a sole crop, while irrigation had a greater positive influence on it. Leaf area values of beans were more sensitive to the same treatments. Microenvironmental conditions in maize–bean mixtures were more favourable for bean crop than for sole beans.

An intercropping system had a greater influence on yield components of maize. Component combination 1/2[ratio ]1/2 (maize[ratio ]bean) was most effective in all yield components of maize. Intraspecific competition appears to be more intense than interspecific competition in both crops. Yield component of bean was more sensitive to water regime of the site than to planting pattern in an intercrop. Irrigation increased all yield components of bean (especially pod number). The intercropping system decreased harvest index in both crops compared with monocrops. Maximum total grain yield was in 1994 in irrigated maize–bean intercrop 1/2[ratio ]1/2, with highest efficiency being in an intercrop in irrigation in 1995. The Land Equivalent Ratio (LER), based on grain yield, was consistently greater than 1·0 in an irrigation water regime in 1995. Proportion of maize[ratio ]bean = 1/2[ratio ]1/2 gave the highest increase of yield (LER = 1·54). Under conditions of chernozem soil type in Zemun Polje, with often-expressed dry periods, irrigation is a very important measure for increasing maize–beans intercrop productivity.