Three experiments were carried out at Bahteem Agricultural Research Station during three successive seasons from 2005/2006 to 2007/2008. In 2005/2006 season using three wheat (Triticum aestivum L.) crosses namely; Gemmeiza 7 x Sids 1, Sakha 94 x Irena and Sids 1 x Sakha 94.  Six populations (P1, P2, F1, F2, BC1 and BC2) for each cross were used in this investigation. The data were recorded on an individual guarded plants for number of spikes / plant, number of kernels / spike, 100-grain weight and grain yield / plant. Significant heterotic effects were obtained for number of spikes / plant, grain yield / plant and number of kernels / spike in the first cross while, heterotic increase in grain weight, grain yield / plant and number of kernels / spike seemed to be accounted for the heterotic response observed in the second cross. Also, significant heterotic effect was found for number of spikes / plant, grain weight and grain yield / plant in the third cross. Inbreeding depression estimates were significant for all studied attributes except for number of spikes / plant in the first cross. Over dominance towards the higher parent for number of spikes / plant, grain weight, grain yield / plant and number of kernels / spike were observed in the first cross. Meanwhile, over or partial dominance towards the lower parent was obtained for all characters in the third cross. On the other hand, partial dominance was observed for all characters (number of spikes / plant, grain weight, grain yield /plant and number of kernels / spike) in the second cross. F2 deviations (E1) were significant for all studied characters in the three crosses, except for number of spikes / plant in the first and second crosses, and grain weight in the first cross only. Moreover, backcross deviations (E2) were significant for all characters studied in all crosses, except for number of spikes / plant and for grain weight in the first cross, grain yield / plant in the first and the second ones. The additive gene effects were significant for all studied characters in the second and the third crosses, except for number of spikes / plant, grain weight , grain yield / plant and number of kernels / spike in the cross (Gemmeiza 7 x Sids 1). These results suggest the potential for obtaining further improvement in most studied character. In addition, dominance epistasis was significant for some of the studied attributes (number of kernels / spike, grain yield/ plant in the first and second crosses; and grain weight in the first and third ones; and number of spikes / plant in the third one). High to medium values of heritability were associated with high and moderate genetic advance as percentage of F2 mean in most characters. These results indicated that selection for the studied characters could be useful in the early generations.