Maize seedlings (cultivars Bastion, Ace, Anko and Mirna) were raised for 15, 30 or 45 days in a glasshouse and planted in the field between early May and mid-June in 1989 and 1990, at 10 plants/m2. Samples of transplants and direct-seeded crops of the same cultivars were taken at regular intervals throughout the growing season for growth analyses. In both years, the 15-day-old transplants established more rapidly in the field than older transplants, and the patterns of leaf production of these young transplants were similar to those of direct-seeded maize. Transplants flowered earlier than direct-seeded maize from the same planting date. At most harvests, transplants showed significantly higher dry weights than their direct-seeded counterparts, and May-planted maize had significantly higher dry weights than June-planted crops, while cultivar effects were small. Final biomass yields were consistently higher in 1989 (17 t/ha) than in 1990 (11 t/ha). In both seasons, particularly in 1990, there was a trend for a lower harvest index in direct-seeded maize which reflected its immaturity at final harvest. The lowest harvest index (0·07) in 1990 was recorded for direct-seeded maize of the late cultivar, Mirna (June planting), the highest harvest index (0·65) for 30-day-old Mirna transplants (May-planted). A comparison of the experienced effective temperature sums during the vegetative, reproductive and grain-filling phases suggested that the transplant nursery period provides an advancement in thermal time which can affect grain-filling and crop maturity. Year to year variation in climate seems to be of primary importance for yield advantages of transplanted over direct-seeded maize.