In hybrid maize breeding, information on genetic diversity and heterotic groups is very useful in inbred line development and evaluation for planning crosses for hybrid cultivar development. The classification of inbred lines into different heterotic groups is an important task in any breeding program. The present investigation was undertaken with a view to: i). estimate the phenotypic and genetic diversity in 38 maize inbred lines developed from commercial hybrids, ii). their classification into existing heterotic groups and iii). highlighting the high yielding crosses between lines. The phenotypic diversity was determined by calculating the Euclidean distance between the lines based on 22 quantitative and qualitative traits, and the classification of inbreds into heterotic groups was performed using the Ward method of cluster analysis. The genotypic diversity was assessed by the effects of general and specific combining abillities. The Euclidean distance between lines ranged from 5.21 to 9.28. Observed inbred lines were grouped into 6 clusters, which predominantly correspond to the Lancaster (A619, Mo17), BSSS (MK390 and B73) and Iodent (MK01) heterotic groups. The 11 inbred lines were included into the Mo 17 germplasm group, the 7 inbreds - into A619 heterotic group, the 3 samples were asigned into MK390 heterotic group and a number of 5-6 inbred lines were classified into B73 and MK01 heterotic groups. For 6 lines, the germplasm group was not identified. They have formed a particular cluster located between the Lancaster-Mo17 and Iodent groups, and it is possible that they are recombinant genotypes between these groups. Based on the effects of the specific combining ability into Mo17 heterotic group the 6 lines were grouped, in the MK390 heterotic group - 2 inbred lines, in the A 619 germplasm group -3 inbreds and the 4 inbred lines were classified in the MK01 heterotic group. For the other lines (52.0%) the heterotic group, determined on the basis of SCA effects, did not correspond to that based on morphological markers. The correlation coefficient between the Euclidean distance and the SCA effects was insignificant (r = 0.3). It seems that both methods represent the distinct tools for diversity analysis. In each heterotic group by diallel analysis the genotypes with high GCA effects were noted and based on the high SCA effects the 8 high yielding crosses were performed, that corresponding to the Iodent - Lancaster, BSSS-Lancaster and Iodent - BSSS heterotic patterns. The results will allow for more accurate use of the genotypes in the breeding programs for new maize inbred lines and high performance maize hybrids development.