ABSTRACTSunflower is an important oilseed crop due to its health related values. Induced mutations are often used to create new genetic variability where sufficient variation does not exist naturally as in the case of sunflower. This study was carried out at the laboratory and research farm of the Institute for Agricultural Research (IAR), Samaru, Zaria to assess the response of the sunflower seeds to different EMS doses and determine optimal doses for successful mutation; estimate genetic variability for agronomic traits among the M2 mutant lines and identify some desirable mutants for oil content and other agronomic traits; assess the degree of association among oil content, grain yield and other agronomic traits and assess the degree of similarity among the selected mutants. Two sunflower varieties (SAMSUN2 and SAMSUN4) obtained from Institute of Agricultural Research were treated with EMS concentrations of 0.1, 0.2, 0.3, 1.0 and 1.5% for 2, 4, 6 and 8 hrs with controls to ascertain optimum EMS concentration and exposure time to bring about desirable mutation. The laboratory study showed progressive decrease in seed germination as EMS concentration and exposure durations were increased. Optimum doses were observed at 0.1% EMS for 6 hrs, 0.2% EMS for 4 hrs, 0.2% EMS for 6 hrs, 0.3% EMS for 4 hrs and 0.3% EMS for 6 hrs for the two varieties out of which 0.2% EMS for 4 hrs, 0.2% EMS for 6 hrs, 0.3% EMS for 4 hrs and 0.3% EMS for 6 hrs were chosen to generate M1 population in the research farm. A total of 78 M1 plants with substantial seed set were used to raise M2 population in progeny rows. Surviving 418 mutant progenies as well as two checks were evaluated for desired agronomic traits in M2 generation using 16 x 5 lattices design with two replications. The analysis of variance for agronomic traits revealed significant (P<0.05) differences amongst the mutants progenies. Sixty (60) mutants professing higher seed yield per plant (11 - 23 g) were evaluated for oil content. The correlation study showed that grain yield per plant was significantly correlated with plant height (r = 0.68), stem diameter (r = 0.61), capitulum circumference (r = 0.72) and 100 seed weight (r = 0.81). Cluster analysis grouped the selected 60 mutants into four clusters with clusters I, II, III and IV consisting of 19, 4, 26 and 11 mutants, respectively, implying that crossing of mutants from these different clusters could enhance more variability for desirable traits for improvement. The present study identifies mutants T8M11-1-2, T4M18-3-1, T2M6-1-1, T8M11-4-1 and T8M23-1-2 as being superior for oil content (45%), stem diameter (4 cm), capitulum circumference (38 cm), seed weight (13 g) and seed yield per plant (26 g) respectively. Their inclusion in future breeding programs could result in the development of superior sunflower cultivars and as such, should be further evaluated in subsequent generations to stabilize the traits.
