The present study of this investigation was aimed to use RAPD markers and phenotypic distances based on yield traits to assess the genetic diversity among six inbred lines (three Egyptian and three American inbred lines) widely used in maize breeding programs and to evaluate the association of the genetic diversity with their F₁ performance. Five primers succeeded to evaluate six inbred lines of maize using RAPD technique. This technique was efficient in detecting polymorphism with an average of 91.3% and determining genetic diversity among the six studied inbred lines. But they were not effective enough to distinguish some inbred lines by unique markers. The molecular distances and phenotypic distances were found to range from 0.355 to 0.600 and from 27.37 to 164.46, respectively. According to cluster analysis and principal coordinate analysis, the parental inbred lines divided into three and two groups based on molecular and phenotypic distances, respectively. Also, using principal coordinate analysis based on two types of distances observed that the separation between the American inbred lines was higher than separation between the Egyptian inbred lines. On the other hand, poor correlation (r = 0.297) among molecular and phenotypic distances were found. This poor correlation also found for two types of distance with specific combining ability and mid parents heterosis and better parent heterosis. Coefficient of determination for F₁ performance against molecular distance and phenotypic distances were extremely low and ranged from 0.0002 to 0.189 and from 0.0001 to 0.351, respectively. This result demonstrated the reliability impairment of regression models in predicting for F₁ performance.