Fifteen hybrids were produced by crossing three male sterile-lines i.e., Tift 23A (L₁), Tift 23 E (L₂) and ICMA 89111 (L₃) and five pollinators of pearl millet i.e. Sudan-2 (T₁), ICMV 87101 (T₂), ICMV 155 (T₃), ICMV 87107 (T₄) and ICMV 87111 (T₅) according to line x tester mating system in 1998. Crosses  with their pollinators and two check varieties were evaluated in a randomized complete block design with four replications at four locations, Sakha, Gemmeiza, Ismailia and Sids Agricultural Research Stations during 1999 season. The aim of this study is to evaluate forage yield of pearl millet under different environments and estimates of  heterosis, combining ability and some other genetic parameters of forage yield and some related components.                 Results revealed that, Sids location gave the highest total fresh and dry forage yield (2670.5 and 346.1 g/plant, respectively) overall entries with significant differences than other locations. Cross (L₂ x T₁) produced the highest total fresh forage yield (1678 g/plant), while cross (L₂ x T₄) gave the highest total dry forage yield (278.3 g/plant) overall locations. The best crosses for fresh forage yield were cross (L₃ x T₁) at Sakha, cross (L₁ x T₂) at Gemmeiza and Ismailia and cross (L₁ x T₄) at Sids. This is the same trend for dry forage yield, approximately. High heritability values in broad sense (more than 80%) were obtained for number of tillers, leaf/stem ratio and forage yield. Number of tillers was the main trait affected forage yield. Significant positive correlation values between the two traits were obtained (0.756 and 0.768 for total fresh and dry forage yield, respectively were computed).                 Heterosis values expressed as a percentage of constant parent (male parent) ranged from negative to positive significant values. Cross (L₁ x T₂) had the highest heterosis values for most studied characters (55.35, 16.85, 30.81, 56.70 and 43.31% for number of tillers, fresh leaf/stem ratio, dry leaf/stem ratio, total fresh yield and total dry yield, respectively.                 The magnitude of general combining ability variance was generally more than specific combining ability variance for plant height, number of tillers, fresh and dry leaf/stem ratio, while, specific combining ability variance was more than general combining ability variance for stem diameter, total fresh yield and total dry yield.                 General combining ability effects indicated that male sterile-line 23A (L₁) proved to be the best combiner for stem diameter, number of tillers, and fresh and dry leaf/stem ratio, while male sterile-line 23E (L₂) could be consider a good combiner for fresh and dry forage yield, then male sterile-line ICMA 89111 (L₃) was the best combiner for plant height. Sudan-2 variety (T₂) may be consider the best combiner fore all studied characters except plant height which ICMV 87107 (T₄) was the best one for this trait. These parents (lines and testers) can therefore be used successfully in cross-combinations for the improvement of their respective traits. Cross (L₁ x T₃) had positive significant specific combining ability values for most studied characters, while, cross (L₃ x T₃) had negative significant values for the same traits.