Maize breeders are concerned with the narrow base of Corn Belt germplasm and the wide use of one heterotic pattern, ‘Reid’ × ‘Lancaster’, in maize (Zea mays L.). Knowledge of other heterotic patterns can contribute to a broadening of the commercial germplasm base. The study was carried out to evaluate heterotic combinations among seven yellow‐endosperm populations—‘BSSS(R)C10’,‘BSK(HI)C8 Syn 3’, ‘BS16(S)C3 Syn 2’, ‘Mexican Dent’, ‘Leaming’, ‘Midland’, among three white‐endosperm populations—‘Pride of Sameline’, ‘Johnson County White’, ‘Potchefstroom Pearl’, and among all 10 populations. In 1985, the 10 populations, their diallel crosses, and 26 checks were evaluated in a 9 × 9 partially balanced lattice with two replications in seven locations. Sources of variation for entries, populations, and general combining ability (GCA) were significant for root lodging, stalk lodging, ear height, days‐to‐pollen shed, days‐to‐silking, moisture content, and yield. Specific combining ability (SCA) mean squares were significant only for ear height. BSSS(R)C10 had the highest GCA effect for yield of all the populations tested. The BSSS(R)C10 × Mexican Dent had the highest yield of 8.77 Mg ha−1, although the four top‐yielding crosses did not differ significantly. This yield was not significantly higher than the check B73 × Mo17, but was significantly higher than BSSS(R)C10 × Lancaster, which represents the heterotic pattern Reid × Lancaster. There were no significant yield differences among the white‐endosperm population crosses. BSSS(R)C10 × Mexican Dent and Midland × BSK(HI)C8 Syn 3 should be evaluated for possible hybrid combinations.