Long-term field trials suggest that yields are declining where maize (Zea maysL.) is grown continuously in some parts of Natal. Available evidence indicates that the decline is not ascribable to soil fertility or climatic constraints and the possibility exists that a build-up of soil-borne pathogens is responsible. The work reported here was conducted to test this hypothesis. A field trial incorporating four cultivars and plots either untreated or treated with a soil fungicide (sodium N-methyldithiocarbamate) was conducted over three seasons at Cedara. Each season, yields, soil and plant analytical data and estimates of root and stalk disease incidence were recorded. For all seasons and cultivars the average yield response to fumigation (680 kg ha−1) was highly significant, and there was evidence to indicate differential cultivar response. Yield responses were associated with marked decreases In the incidence of stalk and root rots and were not ascribable to nutritional effects. The evidence presented indicates that root rots are more detrimental to yield than stalk rots and the suggestion is made that these diseases warrant far more attention than they currently receive.