The objective of present study was to compare of several methods for estimating the degree of heritability and calculating the number of genes using generation mean analysis of maize (Zea mays L.). The experiment was conducted at the field of Field Crop Dept. College of Agric / Univ. of Baghdad, for many seasons, spring and fall seasons 2009, 2010, spring 2011 and fall 2013.Six diverse inbred lines were crossed to produce F1,F2,BC1 and BC2 for four superior crosses.Broad-sense and narrow sense heritability estimates based on variance of different generations. The results showed that the four formulas used to estimate the heritability were different in estimating the values of heritability of different traits, as well as the different crosses used in the study. The broad sense heritability was differed according to the components of the formulas used to estimate. The number of genes that controlled the trait differed according to the different method of estimation and according to the studied trait, and to the all crosses. The highest number of genes (for all formulas) for plant yield, ranging from 15.85 (cross 4) to 155.7 (cross 2), because the yield is complex quantitative trait of several components, so the number of genes they control are the sum of genes that control the components. The inbreeding depression differed according to the parents involved in the crosses, and according to the studied trait, the inbreeding depression ranged from -7 for number of branches/tassel of cross 1 and leaf area of cross 3 to 68 for grain yield of cross 2. We can conclude that the best formula to estimate heritability depends on experimental design and methods of breeding.