Semidwarf maize (Zea mays L.) could be grown in new areas of production or in alternative crop rotations. Our objectives were to determine (i) if genomewide selection is useful for the rapid improvement of an exotic × adapted cross, (ii) if genomewide selection is more effective than phenotypic backcrossing for a trait with major genes, and (iii) the extent to which the high grain yield of nondwarf maize can be combined with the reduced stature and adaptability to high plant population densities of semidwarf maize. We conducted four cycles of genomewide selection in two semidwarf × adapted crosses. Phenotypic backcrossing was also done until the BC4 with selection for short plants. The accuracy of genomewide predictions in Cycle 0 was 0.67 to 0.70 for plant height and 0.57 to 0.70 for grain yield. Genomewide selection from Cycle 1 until Cycle 5 either maintained or improved on the gains from phenotypic selection achieved in Cycle 1. Observed gains generally agreed with predicted gains. Cycle 5 did not always have the best mean performance. Unfavorable genetic correlations made it difficult to select for short plants with high grain yield. Linkage disequilibrium between markers declined as selection progressed. Compared with phenotypic backcrossing, genomewide selection led to better mean performance and a higher proportion of exotic germplasm introgressed. To our knowledge, this is the first empirical study on genomewide selection to improve an exotic × adapted cross.