Analysis of heritable variation is important not only for grain yield and other agronomic characteristics, but also for nutritional and industrial purposes. Evaluation of maize (Zea mays L.) grain quality traits in improved genotypes across environments is essential to determine the potential of maize for value-added products. The objectives of this research were to characterize the level of oil, protein, starch, and extractable starch in 10 early maturing maize improved populations and to determine whether genetic relationships among genotypes can be inferred based on grain quality traits. Kernel samples of 500 g were taken from genotypes in experiments arranged in a 8x8 partially balanced lattice design with two replications across nine environments during 2003, 2004, and 2005. All samples across locations were quantified for oil, protein and starch using near-infrared spectroscopy. Highest values for grain quality traits were not different from commercial checks. Data from the 10 populations per se and the checks was also used to perform a cluster analysis based on the quality trait values and grain yield obtained from the field experiments. Three main groups were formed when the dendrogram was cut at a distance of 0.75. The first group was conformed by the genotypes from North Dakota and CGSS(S1-S2)C5, except NDSG(M)C15. The second group was formed by germplasm originated from Iowa (BS-5, BS-21 and BS-22), Leaming(S)C4, and CGL(S1-S2)C5. The third group included the commercial hybrids. Results showed the importance of North Dakota germplasm as a unique source of genetic diversity that could be useful to incorporate in breeding programs focusing on quality traits. This approach could also be helpful in the classification of maize genotypes into genetic clusters or groups of genetic diversity for quality traits.