Tropical Agricultural Research Vol. 101998,425432 Combining Ability and Heterosis of Some Productivity Traits in Pearl Millet [Pennisetum glaucum (L.) R. Br.] P. Azhaguvel and N. Jayaraman Department of Millets, Centre for Plant Breeding and Genetics Tamil Nadu Agricultural University Coimbatore, India ABSTRACT. Nine male sterile lines belonging to three diversified source such asAl,A2 and 732 A were crossed with ten restorer lines. The resulting Fl hybrids were evaluated for different productivity traits such as plant height, 50% flowering, number of productive tillers, panicle length, panicle gink, 100 grain weight and grain yield per plant. From these observations combining ability and heterosis were worked out to choose the best performing parents as well as hybrids. Predominance of non-additive gene action was reported for all the character in the present study. The parents such as 851A.ICMA 92777 A, 852 A, PT5573 and PT 5576 were considered as general combiners. The crosses 3024 *PT5573 and852 A *PT5567were recommended for high grain yield through heterosis breeding. INTRODUCTION Pearl millet [Pennisetum glaucum (L.) R. Br.] is the most widely grown cereal crop under short growing seasons in the arid and semi arid regions of the world. It is a protogynous and cross pollinated crop amenable for die development of heterozygous populations, which can be utilized for the production of high grain yielding hybrids. The discovery of cytoplasmic male sterility (CMS) has made a major break through in the development of hybrids. With the introduction of Georgia male sterility in pearl millet through Tift 23 A to India in 1962 and subsequently the release of the HB 1 reportedly doubled the millet production between 196S and 1970 (Burton, 1983). The knowledge of combining ability effects and the corresponding variances is important in the choice of selecting parents and it can be further used for exploiting heterosis to produce high performing new recombinations. The research on the diversification of CMS lines gained momentum because of genetic vulnerability of the female parents and the hybrids carrying the same cytoplasm. Hence the utilization of diverse ̂ cytoplasmic male sterile lines has been attempted for breeding of best hybrids. The present study Azhaguvcl & Jayaraman MATERIALS AND METHODS Nine male sterile lines, namely ICMA 91444 A, ICMA 92444 A, ICMA 92666 A, ICMA 92777 A, 851 A, 852 A, 302 A, 306 A and 732 A, which included three cytoplasmic groups Al, A2 and 732 A sources were crossed with ten restorers in a Line x Tester fashion in Rabi (winter season) 1996. The crossed seeds were sown as Fl seeds in Kharif (wet season) 1997 with two replications as a randomized block design at Millet Breeding Station, Centre for Plant Breeding and Genetics, Tamil Nadu Agricultural University, Coimbatore, India. Each entry was represented by two rows of three meters length with a spacing of 50 cm. The plant to plant spacing with in the row was 15 cm. All the agronomic practices were made as for commercial crop. The data with respect to seven characters viz., 50% flowering, plant height, number of productive tillers, panicle length, panicle girth, 100 grain weight and grain yield per plant were recorded on the hybrids and the parents. These data were subjected to Line x Tester analysis and the general combining abilities (gca) of the parents and specific combining abilities (sea) of the crosses were worked out, as suggested by Kempthrone (1957). The degree of heterosis in Fl over mid parent, better parent and standard heterosis were calculated and expressed in percentage (Turner, 1953). RESULTS AND DISCUSSION The mean performance of the hybrid lines and the testers with respect to the measured parameters is shown in Table 1. The hybrids and the Line x Tester interaction was significant for all the characters studied. The variance due to lines was significant for all the characters except for 50% flowering and grain yield per plant. The over all variances for the testers were also significant for all the traits except for number of productive tillers. SCA variance was more pronounced than that of GCA variance, so non-additive gene actions were noticed. This result is in confirmity with the findings of Pethani and Kapoor (1995). Parents with high negative gca effect coupled with negative mean per se for 50% flowering was considered the best for exploiting earliness in hybrids. The lines ICMA 92666 A, ICMA 92444 A and the testers PT 5573, 426 reports the heterosis of the hybrids and the combining ability analysis of pearl millet (Pennisetum glaucum (L.) R. Br.) male sterile lines taken from three systems of male sterility and the testers. Combining Ability and Heterosis in Pearl Millet PT 5574, PT 5575 expressed high significant negative gca effects (Table 2). The highest negative significant sea effect was registered in the crosses 851 A x PT 5570 and ICMA 92777 x PT 5572. Best crosses showing significant specific combining ability effects for all the characters are shown in Table 2. The crosses ICMA 9277 A x PT 5572 and ICMA 92666 A x PT 5574 recorded significant .negative relative heterosis (di), heterobeltiosis (dii) and standard heterosis (diii), at 50% flowering. Crosses showing high heterosis for all the characters are shown in Table 3. Table 1. Mean performance and combining ability effects of parents. Parents DF PH NPT PL PG GW GY U n a ICMA9U44A 103.9 50.5 4.0 23.9 8.8 1.2 19.1 ICMA92444A 111.2 50.0 5.1 16.0 8.7 0.9 13.5 ICMA92666A 129.3 50.5 5.0 17.9 8.7 0.9 19.2 ICMA92777A 112.6 52.0 2.4 15.6 9.2 1.0' 26.9 85IA I3S.9 52.5 5.6 19.1 8.2 0.8 20.3 852A 139.3 51.5 3.9 16.1 7.6 0.8 13.2 302A 148.2 52.0 3.9 13.5 4.4 1.0 12.0 306A 139.8 52.5 4.3 13.5 7.4 0.7 18.0 732A 77.2 53.0 2.9 18.0 7.0 0.9 16.5 Tester PT5567 149.S 51.0 5.0 20.4 8.7 0.9 38.5 PT5568 13S.3 SI.S 3.7 18.4 8.9 0.7 19.0 PT5S69 144.9 51.0 2.8 18.7 8.3 0.8 23.8 PTSS70 140.4 53.5 3.6 15.7 8.5 0.7 20.5 PT5571 183.6 53.5 5.2 24.5 9.6 1.0' 40.0 PT5572 147.5 51.5 3.4 24.8 16.8 0.8 16.2 PT5573 178.0 49.5 2.9 20.5 11.3 1.4 26.9 PTS574 96.1 48.5 3.9 17.6 7.9 0.8 19.3 PTSJ75 147.2 50.5 3.2 19.4 7.2 0.5 28.8 PT5576 162.2 51.0 3.3 22.9 10.7 1.0 21.8 SE 97.6 0.87 0.32 3.1 1.6 0.01 35.6 DF: days to 50% flowering; PH: plant height (cm); NPT: number of productive tillers; PL: panicle length (cm); PG: panicle girth (cm); GW: 100 grain weight (g); GY: grain yield (g per plant); SE: standard error of the means. The parents 851 A and PT 5571 had positive gca effects. Out of this, PT 5571 expressed high per se performance for plant height. The parent with 427 Azhaguvel & Jayaraman Table 2. Estimates of general combinding ability effects. Pucntj DF PH NPT PL PG GW GY FEMALE PARENTS ICMA9I444A 0.7 -3.5 -0.2 0.1 4.4' or •5.2" ICMA 92444 A 4.6' -4.2 -0.3" -1.5" 0.7" -o.r 2.2 1CMA92666A 4 .8" -1.2 0.3" •1.7" 4.1 0.03 -1.5 ICMA92777A 4.4 1.3 0.2 4.7 o r 0.1" 1.0 131 A 0.6" 13.0" 0.1 3.8" 4 . 3" 0.02 5.6" 852 A 0.9- -3.4- -0.1 4.3 4.1 0.1" 4.5" 302 A 4.1 4.6' 4.02 1.5" 0.0 4 . 1 ' -0.4 306 A 0.01 -0.1 4.1 -I.S 0.3" 0.1 2.4 732 A 0.4 0.34 0.12 0.3 4 .2" 4 .2" -4.1' SE 0.2 1.6 0.09 0.32 0.16 0.02 1.13 CD(p-0.05)' 0.5 4.6 0.25 0.89 0.44 0.05 3.16 CD(p-0.0l)" 0.7 6.04 0.33 1.19 0.59 0.07 4.18 MALE PARENTS PT5S67 4.5 -4.0 0.4" -2.4" 4 .6" 4 . 1 " 0.1 PTSS6S o r 01 4.2 4.2 0.5 4 . 1 " 0.1 PTSS69 1.4" O.S 0.2 -Iff 0.1 4.1 2.1 PT5S70 i.4~ 2.5 0.1 4.6 4.4 4.01 4.1' PT557I 1.7" 23 5 4 . 1 " 2.5" 0.3 4.1" -7.1" PT5572 -1.0" -4.6 0.1 0.4 0.1 0.2" 3.0 PTS573 1.9" -5.9 4.6 4.2 4.2 0.2" 4.4' PTJ574 -1.4" -14.5" 4.2 4.2 4.2 4.1* 0.2 PT5575 -1.1" -4.6 4.2 O.S 4.1 0.00 3.0 PTSS76 0.6" 6.6 O.S" 1.2* 0.8" 4.01 2.3 SE 0.2 1.73 0.09 0.34 0.17 0.02 1.2 CDtp-OOS)' 0.S6 4.85 0.25 0.95 0.47 0.05 3.36 dXp-OOl)" 0.74 6.41 0.33 1.26 0.63 0.07 4.44 DF: days to 50% flowering; PH: plant height (cm); NPT: number of productive tillers; PL: panicle length (cm); PG: panicle girth (cm); GW: 100 grain weight (g); GY: grain yield (g per plant); SE: standard error of the means. high mean per se plant height and high gca can be advantageously utilised in pedigree breeding programmes depending on the need to develop tall varieties. The cross 851 A * PT5568 and 306 A * PT5570 recorded highest sea effect for plant height. The cross 851 A * PT 5568 showed significant heterobeltiosis and standard heterosis. Among the parents ICMA 92666 A, PT 5567 and PT 5576 showed significant gca effect for number of productive tillers. The positive gca effect of the tester PT 5567 was also associated with the high mean performance and it will be useful to evolve high tillering hybrids. The crosses ICMA 92777 A x PT 5576 and 302 A * PT 5572 428 Combining Ability and Heterosis in Pearl Millet showed significant sea effect for the number of productive tillers (Table 3). The crosses 851 A x PT 5576 and 302 A x PT 5572 recorded significant relative heterosis, heterobeltiosis and standard heterosis. Table 3. Crosses showing significant specific combining ability effect for different characters. Characters Crosses Days to 50% flowering 851A APT 5570 ICMA 92777AJTPT 5572 306AJTPT5567 Plant height (cm) 851 AJTPT5568 306 A APT 5570 852 A APT 5572 Number of productive tillers ICMA 92777AAPT 5576 302AJfPT5572 851 AJfPT 5571 306AJTPT5567 851 AJTPT5576 Panicle length (cm) 306 A APT 5575 851 AJfPT5576 ICMA 92777A*PT 5568 Panicle girth (cm) ICMA 92777A A PT 5569 ICMA 95666A A'PT 5575 306AJfPT5572 732A*PT5570 100 Grain weight (g) 851 A A'PT 5570 851 AXPT 5568 92444 A ATT 5573 732 S APT 5572 Grain yield (g per plant) 302 A AT PT 5573 95444AAT5573 ICMA 91444A APT 5569 306 A A" PT 5571 852AJfPT5567 429 Azhaguvel & Jayaraman Table 4. Crosses showing significant heterosis. Ch. Relative heterosis Heterobeltiosis Standard heterosis DF 1CMA92777A APT 5572 ICMA 92666 A APT 5574 306 A APT 5572 ICMA 92666A APT 5574 ICMA 62777A APT 5572 306A APT 5578 ICMA92777A APT 5572 ICMA 92666 A APT 5574 851 AAPT 5570 PH 852 A APT 5568 851 A APT 5572 732 A APT 5573 851 A APT 5568 306 A APT 5572 852 A APT 5569 851 A APT 5568 851 AAPT 5570 732 A APT 5572 NPT 851 A APT 5576 851 A APT 5572 306 A APT 5572 851 A APT 5576 302 A APT 5572 732 A APT 5572 851 A APT5576 302 A APT 5567 ICMA92666A APT5576 PL 306 A APT 5575 851 A APT 5576 732 A APT 5572 306 A APT 5575 852 A APT 5572 302 A APT 5573 852 AAPT 5571 306 AAPT 5572 ICMA92777A APT 5569 PG ICMA A 92777 APT 5570 852 A APT 5573 306 A APT 5572 851 AAPT 5568 ICMA 92777A APT 5569 732 A APT 5572 ICMA 92777A APT 5569 852AAPT5571 851 AAPT 5570 GW 851A APT 5570 732 A APT 5572 ICMA 92777A APT 5568 851 AAPT 5570 732 A APT 5572 852 A APT 5572 732 A APT 5572 851 AAPT 5570 306 A APT 5573 GY 302 A APT 5573 851 A APT 5568 306 A APT 5572 302 A APT 5573 732 A APT 5572 ICMA 92777A APT 5576 302 A APT 5573 851 AAPT 5570 306 AAPT 5572 DF: days to 50% flowering; PH: plant height (cm); NPT: number of productive tillers; PL: panicle length (cm); PG: panicle girth (cm); GW: 100 grain weight (g); GY: grain yield (g per plant); SE: standard error of the mean.. 430 The parents such as 851 A, 302 A, PT 5575 and PT 5576 recorded highly significant gca effect for panicle length (Table 2). The highly significant positive effect observed in PT 5571, PT 5576 pollinators were also associated with high mean per se performance. Such parents could be utilized profitably to increase the earhead length which in turn will improve grain yield. The crosses 306 A * PT 5575 and 306 A x PT 5570 showed significant gca effects. The 306 A * PT 5575 cross recorded significance for all three types of heterosis (di, dii and diii). Combining Ability and Heterosis in Pearl Millet CONCLUSIONS The parents such as 851 A, ICMA 92777 A, 852 A, PT 5576 and PT 5570 were the recommended parents for future crossing programme for improving the yield and yield contributing characters. The crosses 302 A x PT 5573, 851 A x PT 5570, 306 A x PT 5572, ICMA 92777 A x PT 5569, 306 A x PT 5575, 732 A x PT 5572, ICMA 92777 A x PT 5576 and 851 A x PT 5568 were considered as good performing hybrids. REFERENCES Burton, G.W. (1983). Breeding pearl millet. PI. Breed. Rev. 1:162-182. Kempthome, 0.(1957). An Introduction to Genetic Statistics. John Wiley and Sons. Inc. New York. Turner, A. (1953). A study of heterosis in upland cotton. I. Yield of hybrids compared with varieties. Agron. J. 45:485-486. 431 Among the parents ICMA 92444 A, ICMA 92777 A, 306 A, PT 5576 and PT 5568 showed high gca effects for panicle girth. These parents could be useful for evolving varieties with thicker earheads. The crosses ICMA 92777 A x PT 5569 and 306 A x PT 5572 recorded high sea effects. The cross ICMA 92777 A x FT 5569 performed better with a significant expression of relative heterosis, heterobeltiosis and standard heterosis. For the character 100 grain weight, the parents ICMA 91444 A and PT 5573 recorded high positive significant gca effects coupled with high mean per se performance. The crosses 851 A x PT 5568, 732 A x PT 5572 and 92444 A x PT 5573 expressed high sea effect. The crosses 851 A x PT 5570 and 732 A x FT 5572 recorded high heterosis, for 100 grain weight character (Table 4). The ultimate objective in any breeding programme is the improvement of grain yield and hence breeding procedures are to be focused and carefuly formulated in increasing the potentiality of this complex trait. Among the parents 851 A 852 A and PT 5570 possessed high gca effects associated with moderate per se grain yield. The crosses 302 A x PT 5573 showed high gca effect with significant heterosis. Theses crosses may be further utilized in the breeding programme for improving other desirable agronomical characters. Azhaguvel & Jayaraman Pethani, K.V. and Kapoor, R.L. (1995). Combining ability analysis for yield components in pearl millet. Gujarat Agri. Uni. Res. J. 20:87-93. 432