A simulation model was constructed to ascertain the extent of benefits from spatial diversification when an intensive form of use is assumed for the additional (alluvial) land. Although expected profit increased and coefficient of variation decreased, the return to additional capital was only 2.7 per cent. Spatial diversification successfully overcomes winter-spring feed uncertainties, but appears economically doubtful under present assumptions. Some alternatives and modifications are suggested which, although untested, might improve the economics of diversification.