The experiment comprised of five lines, three testers and their 15 F1s. They were grown in randomized complete block design under late sown conditions. The line x tester analysis of Kempthorne (1957) was carried out to estimate general combining ability (gca) and specific combining ability (sca) effects. The line ‘OL 22-1’ and tester ‘OLC 10’ recorded significantly positive gca effects for both number of capsules per plant and 1000-seed weight. They excelled others and may be involved in a crossing programme resulting in identification of superior genotypes with higher capsule number with moderate 1000-seed weight leading to higher seed yield. The crosses ‘OL 2-3 x Padmini’ and ‘OL 4-1 x OLC 10’ recorded significantly positive sca effects for capsule number, 1000-seed weight and seed yield. They may be exploited in heterosis breeding.