Introduction

            Maize (Zea mays L.) is referred to as the “Queen of cereals” as it has got the highest potential among cereal crops. Maize is an important crop after rice and wheat in terms of acreage but ranks second in terms of total production and yield. In India average maize area is 6.93mha, total output is about 13.30 million tones and productivity is 2018 kg/ha. Maize is a potential source of protein for humans and animals. According to the analysis report by FAO (2005)(11), the maize kernel contains approximately 70-75% carbohydrates, 9-31.5% crude protein, 2-3-5 % crude fiber, 3-5% lipids per 100g of kernels. However, like most cereal proteins, maize proteins are nutritionally poor because of their reduced content of essential amino acids. They contain <35 lysine, but the recommended concentration is 5.5% for human nutrition. The normal maize protein is of very poor quality owing to deficiency in two essential amino acids lysine and tryptophan, high leucine – isoleucine ratio, and low biological value. The discovery of the mutant opaque (2) enhances the nutritional quality of the maize. The lysine value of opaque (2) maize is 3.3 to 4.0 g/ 100g of protein, which is more than twice that of endosperm from the normal maze (1.3g/lysine/100g protein). The decreased level of zein (5-27%) in opaque (2) maize is in contrast to normal endosperm quality. Thus the nutritive value of normal endosperm maize is corrected by a genetically improved and biofortified form known as QPM (Quality protein Maize).

            An understanding of the genetic architecture of parents and their mode of inheritance will greatly assist the breeder to formulate appropriate breeding methodologies to incorporate the traits in question into an otherwise desirable variety. The present study was, therefore conducted to estimate sca and gca for hybrids and parents (QPM lines and normal inbred lines) respectively, for qualitative traits and the extent of heterosis for hybrids combined with yield as well as the nature and magnitude of heterosis.

Materials and Methods

            The present investigation was carried out at the Agricultural College & Research Institute, Madurai, Tamil Nadu Agricultural University during the year 2002-2005. Seven diverse white color seed and high lysine and tryptophan content QPM maize inbreds were crossed to five UMI (University Maize Lines) maize inbreds testers in a Line x tester mating design to generate experimental hybrid for this study (Table 1).

All the twelve parents (7 lines and  5 testers), 35 F₁’s with the check Co 1 were grown in randomized block design with three replications in a single row plot of 5 m length having 60 x 25 cm crop geometry. Border rows were raised all along the field to avoid environmental influence to the experimental material to a certain extent. The yield characters studied were days to 50% tasselling, days to 50% silking, plant height, ear height, cob length, cob breadth, number of grain rows per cob, number of grains per row, ear weight, 100-grain weight and grain yield per plant. The quality characteristics like total sugars content were recorded by the anthrone reagent method (25) in percent followed by total starch content recorded in percentage by the anthrone reagent method (5). The total nitrogen content was estimated by the microkjeldahl method (15) and b carotene content as by Rodriquez et al. (1976). The total oil content of dry seed was determined by the Nuclear Magnetic Resonance (NMR) technique (32). The important quality character lysine and tryptophan content were estimated by Rainin Amino HPLC column. 

Results and Discussion

            Analysis of variance showed highly significant differences among genotypes for all characters (yield and quality traits) (Table 2). The variance due to parents, hybrids, and parent-hybrid interactions was highly significant for all 19 characters. Mean squares due to females were larger in magnitude than male parents for all the characters.

            There was a significant difference among hybrids for all the traits. Mean squares due to parents versus hybrids were also highly significant for all the characters. The relative estimates of variances due to specific combining ability (S²­_gca) effects were predominant for all the characters studied indicating the predominance of additive components, except for ear weight. The ratio of variances (S²_gca/S²_gca) due to general and specific combining ability effects ranged from 0.01 to 3.5. These findings, in general, are in conformity with the results reported earlier (29). In the present study, the analysis of GCA and GCA variances showed that yield, yield contributing characters and quality characters were controlled preponderantly by non-additive gene action. These observations suggest that a breeding method that can incorporate both additive and non-additive genetic components would be a useful strategy. The recurrent selection method, which provides a better opportunity for selection, recombination, and accumulation of desired genes, should help to increase fixable as well as non-fixable types of gene action. It could be expected, when parents possessing high gca effects were combined by hybridization, a large proportion of progenies would have high value for the trait concerned, facilitating easy selection for that trait.

            The magnitude of genetic variance was given in Table 3 for all the yield characters and quality characters. The magnitude of s_D² variances was higher than s_A² variances for all the traits (Figure 1). The ratio between s_A² and s_D² ranged from 0.014:1 (100 seed weight) to 0.39:1 (days to 50% tasseling). The same fact was reflected in s²_gca and s²_gca ratio. The magnitude of s_D² variance was higher than s_A²for all the quality characters. The ratio between s_D² and sranged from 0.012: 1 (Total sugar) to 3.5 : 1 (carotene).

            The preponderance of specific combining ability variance (SCA) was greater than general combining ability variance (GCA) indicating the non-additive gene action was reported for yield (Bhatnagar et al., 2004). For plant height (8)  for ear height (1);  for days to 50% silking (35);  for cob length (14),  for a number of grains per row (20); for cob breadth(14); for 100 seed weight (12);  for starch content for carotene (9); for lysine and tryptophan (33); for Protein.        In the present study proportional contribution of lines, testers, and hybrids to total genetic variance;

i) Yield

            The contribution of female x male interaction was highest for all the yield contributing characters. The proportional contribution by females and female x male interaction was almost equal for days to 50% tasselling and days to 50% silking. The contribution of lines was higher than the testers for all characters except for ear height, and single plant yield (Table 3). Compare to female and female x male contribution, males contribution was meager.

ii) Quality

            The contribution of females was higher than males for all the quality characters except for protein content. The proportional contribution of females and females x males contribution was almost equal for total sugar and starch content. The contribution of the female x male component was highest for all the characters studied except for oil content (Table 3).

            The knowledge of combing ability coupled with per se performance of parents would be of great value in selecting suitable parents for the hybridization program. In the present study, the association between per se performance and gca effects was not evident for all the traits, except days to silking (late silking). A similar result was reported by (23).

            Recent development in biometrical genetics has led to the formation of a number of statistical procedures for the genetic analysis of quantitative characters. The Line x Tester method of analysis was developed in order to give information on combining the ability of parents and the genetic mechanism controlling various traits. Experimentally it is a systematic approach to the identification of potential parents and the crosses with the best selection potential in the earliest generation possible (Johnson, 1963). Successful breeding approaches are a direct consequence of the gene action prevalent in the breeding population under consideration. The relative importance of additive or non-additive effects for grain yield is an indication of the type of gene action (1).

            The general combining ability effects (gca) of the parents (Tables 4 and 5) indicated that the parents (CML 145 and CML 146) with low mean values and negative gca effects for days to 50% tasseling could be utilized in a breeding program for earliness. The parent’s (CML 141 and UMI 524) high mean values combine with high positively significant gca effects for days to tasselling and silking might be used for developing long-duration crops with a higher yield. The parents (CML 142, CML 144,UMI 524 and UMI 814) for plant height, CML 143 and UMI 814 for ear height, CML 144, CML 146 and UMI 524 for cob length, CML 144 and UMI 524 for cob breadth; CML 144, CML 147 and UMI 524 for a number of kernel rows; CML 145 and UMI 814 for number of kernels per row, CML 144, CML 145 and UMI 524 for ear weight; CML 144 and UMI 524 for 100 seed weight, CML 144 and UMI 524 for single plant yield; CML 144, CML 146 and UMI 524 for total sugar; CML 144 and UMI 524 for starch, CML 144 and UMI 427 for protein, CML 144 and UMI 524 for oil; CML 144 and UMI 427 for carotene, CML 142, CML 145 and UMI 524 for lysine, CML 144,CML 145 and UMI 426  for tryptophan possessed high per se with significant gca effect (Table 5).

            The results of the present study indicated that it would be desirable to develop multiple crosses to select desirable segregants for yields and favorably associated traits in the advanced generations, as no parents was a good general combiner for all the traits as reported by, (27) and (23). However, CML 144 and UMI 524 might be utilized as potential parents since they possessed high per se with significant gca effects for most of the traits under study. Since maize has enormous variability for all traits; a single genotype with all desirable combinations can recombination breeding through recurrent selection.

            The basic idea in any hybridization program is to accumulate favorable genes present in different parents into a single genotype. The hybrids thus obtained are utilized in two ways.

1. Forwarding to further generation and selecting superior individuals in the 

   Segregating generations with desirable gene combinations.

            2. Directly using the F=0to exploit hybrid vigor.

            The utilization of hybrids in any of the two ways will depend upon the genetic constitution of the parents as well as the hybrids. The hybrids obtained by Line x Tester fashion in the present investigation were evaluated for their performance based on mean, heterosis, and gca effects both individually and in combinations. To exploit the hybrid vigor, the parameters like per se performance of hybrids, gca effects, and standard heterosis of hybrids are to be taken into account (7,19,21,18,,31)

            The primary criterion to evaluate a hybrid is its mean performance. (17,21) reported that the per se performance of hybrids appeared to be a useful index for judging the hybrids.  (13) suggested that parents with good per se performance would result in good hybrids.

            For the formulation of a sound breeding program, it is essential to assess the nature of inheritance of yield and yield contributing characters and the potentialities of parents in hybrid combinations. When the character is unidirectionally controlled by a set of alleles with additives, the choice of parents on the basis of per se may be effective. However, in the case when non-allelic interactions determine the character, selection based on per se performance may vitiate the results and so the combining ability estimates must be considered for parental choice (28).

            In the present study, the first top-ranking hybrid that exhibited maximum sca effects is presented in Table 6. The ranking based on sca effects of the hybrids showed reasonable deviation from that based on the magnitude of heterosis for most of the traits. This could be expected since the sca effects are only estimates. The large gca effect need not necessarily result in the exceptional performance of a cross as represented by (3)

            The hybrid CML 142 x UMI 426 recorded the maximum sca effect for 17 traits out of the 20 studied. Three reason for the high sca effect exhibited by this particular hybrid was due to the high or moderate gca effects of one of the parents or both the parents and the per se performance of either one or both the parents were higher. Based on the sca effect and moderate per se performance CML 142 x UMI 426 might be used for a further breeding program.

            The sca effects along with the per se performance of hybrids gave an idea about the practical utility of hybrid combinations for heterosis breeding as was reported by (23). Crosses recorded considerably higher per se performance with high sca effects as could be seen in the combination CML 145 x UMI 189 which was late to tasseling, CML 147 x UMI 426 for ear height, CML 146 x UMI 814 for cob length, CML 147 x UMI 189 for single plant yield and CML 146 x UMI 426, CML 142 x UMI 189 for lysine and tryptophan respectively, for the other characters Number of kernel row, ear weight, total sugar, starch, protein oil, and carotene the combinations with high per se values recorded positively significant sca effect.

            In the present study CML 147 x UMI 426 for days to 50% tasselling, plant height, CML 147 x UMI 427 for starch and oil, CML 147 x UMI 814 and CML 147 x UMI 524 for tryptophan, one of the parents recorded highly significant gca, while the other recorded highly non- significant gca. Parents with high x high gca effects indicating the presence of additive x additive type of gene action between favorable alleles contributed to be of fixable nature (30). Thus these hybrids would be very much useful for further testing.

            Significantly positive gca effects in one of the parents and negative (or) non-significant gca effects on the other parent leading to significantly positive sca effect in the hybrids could be observed for most of the traits. Some of the combinations which fulfill the above condition like CML 145 x UMI 427 for days to 50% silking, CML 146 x UMI 426 for plant height, CML 143 x UMI 427 for starch, and CML 144 x UMI 524 for tryptophan. Parents with low x high or high x low gca effects produced crosses with significant sca effects indicating the presence of additive x dominance type of gene interaction. Therefore, the above crosses could throw desirable transgressive segregants. The additive genetic systems are present in one general combiner and complementary epistatic effects on others act in the same direction to maximize the desirable plant attributes (11). In general, this indicated that a combination of good general combiners was not necessarily the best cross combination nor was a low x low cross always a poor combination. (13) reported similar results.

            A breakthrough in maize yield and quality improvement may be possible through hybrid maize breeding programs. Maize being an allogamous crop endowed with the mechanism of protandry and monoecy has lent itself feasible for heterosis breeding and a number of commercial hybrids have been released for cultivation.

            The extent of heterosis has been measured as the superiority of hybrids over their mid-parent (relative heterosis) and standard check (Standard heterosis). Among the two heteroses, standard heterosis is given importance for the exploitation of heterotic vigor. The frequency of heterotic combinations is presented in Table 7.

Frequency of positive significant heterotic combinations

Over mid parent

            A high percentage of cross combinations was found to have significant positive relative heterosis for the eighteen characters studied (Table 7). The observed number of combinations over mid-parent was highest in carotene (27) followed by lysine (26), while it was least in plant height (5).

Over standard parent   

            In comparison with the standard parent CO 1, crosses are found to have significant positive standard heterosis for all the characters. The number of combinations over standard parent was highest in 100 seed weight (24) and carotene (24).  While it was least in Days to 50% tasselling (5).

            Among hybrids   CML 143 x UMI 189 (CML 143 x UMI 189) recorded significant standard heterosis for eleven traits it was followed by CML 141 x UMI 427 (CML 141 x UMI 427), CML 146 x UMI 426 (CML 146 x UMI 426) and CML 147 x UMI 189 (CML 147 x UMI 189) which showed significant heterosis for ten different traits. The hybrids CML 141 x UMI 524, CML 141 x UMI 814 recorded early tasselling and silking and the standard heterosis for these cross combinations could serve as a source population for developing early flowering segregants.

            Regarding ear height (25.36) and cob length (41.66) the hybrids CML 142 x UMI 427 (CML 142 x UMI 427) had high heterosis. For plant height CML 144 x UMI 189 (42.75) (CML 144 x UMI 189) and CMCML 14146 x UMI 189 (CML 146 x UMI 189) (29.49) had maximum standard heterosis. The number of kernel rows and number of kernels per row had maximum standard heterosis in CML 145 x UMI 524 (24.54) and CML 143 x UMI 524 (54.91) hybrids respectively. Among the hybrids, CML 141 x UMI 426 (30.08) and CML 144 x UMI 524(17.52) had significant standard heterosis for ear weight and 100 seed weight. Maximum yield heterosis in the present study, over standard check, was 59.99 percent it was observed in the combination CML 141 x UMI 427. This was followed by CML 141 x UMI 189 with a heterosis of 55.65 percent. These combinations (CML 141 x UMI 427 and (CML 141 x UMI 189) also recorded a high magnitude of heterosis for the single plant yield (55.65) and also yield components viz., number of kernels per row, number of kernel rows and 100 seed weight.

            Hybrid CML 146 x UMI 524 recorded highly significant standard heterosis for total sugar (32.11) and starch content (12.49). The hybrid combination CML 146 x UMI 189 recorded significant standard heterosis for protein (36.41) and oil content (40.67)

            Among the hybrids CML 143 x UMI189 recorded highly significant standard heterosis for carotene (41.66) lysine (13.29) and tryptophan (34.29). (33) reported heterosis for carotene, lysine, and tryptophan and also this hybrid was considered to be the best one for moderate yield (31.25 %) and moderate starch (4.92) and low protein(1.63%), which indicates that heterosis, in general, can be commercially exploited especially in maize to improve nutritional qualities along with yield to suit the farmers and industrial needs.

Acknowledgment

It’s my duty to thank the Tamil Nadu Agricultural University for their financial support of the TNAU merit scholarship/research assistantship award throughout my doctoral study.

References

1. Bhatnagar. S.,  F.J. Betran and W. Rooney (2004). Combining abilities of Quality Protein Maize inbreds. Crop Science.44: 1997-2005.
2. Bjarnason, M.K. Short, S.K. Vasal and E.Villegas. 1992. Genetic improvement of various Quality  Protein Maize (QPM) population. Agronomy  Abstr. Pp. 74.
3. Chandramohan,P.(1999). Inheritance of earliness and yield components in a sic parental diallel cross of rice (Oryza sativa L.) M.Sc (Ag.) thesis. Tamil Nadu Agric. University. Coimbatore.
4. Choukan, R. 2000. Estimation of combining ability, additive and dominance variances of characters using line x tester crosses of maize inbred lines. Seed and Plant. 15(1): 47-55
5. Clegg, K.M. 1956. Application of anthrone reagent for estimation of starch content in cereals. Theor. Appl. Genet., 49: 117-122.
6. Dass, S., M. Singh, H. T. Sehya, D .R. Aneja and A.Kumar.1992. Genetics of harvest index in maize. Agrl. Sci. Digest (Karnal)., 12: 117-120.
7. Debnath, S.C. and  K.R. Sarkar.1990. Combining ability analysis of grain yield and its attributes  in maize (Zea mays L.) Indian J. Genet., 50: 57-61.
8. Dodiya. N.S and V. N. Joshi. 2003. Heterosis and combining ability for quality and yield in early maturing single cross hybrids of maize. (Zea mays. L.). Crop Res., 26(1): 114-118.
9. Dodiya. N.S and V. N. Joshi. 2003. Heterosis and combining ability for quality and yield in early maturing single cross hybrids of maize. (Zea mays. L.). Crop Res., 26(1): 114-118.
10. Egesel. C.O., J.C. Wang, R.J. Lambert, T.R. Rocheford. 2003. Combining ability of maize inbreds for  carotenoids and tocopherols. Crop Sci., 43(2): 818-823.
11. FAO year book, 2005. FAO corporate document repository. In: Maize in Human Nutrition. FAO Food and Nutriton  Series, N0.25
12. Fasoulas.A(1980). Principles and methods of Plant Breeding Pub.No.11. Dept. Of Genetics and plant Breeding. Aristotellian Univ. of Tessaioniki.Greece
13. Gautam. A.S.  2003. Combing ability for grain yield and other agronomic characters in inbred lines of maize (Zea mays L.) Crop Res. 26(3): 482-485.
14. Gilbert, N.E.G.1958. Diallel cross in plant breeding. Heredity,12: 477-492.
15. Hedaiyuan, Wu Guangwei, Long Dexiang, Lu Junjie, Liu Qiang. 2003. Analysis of combining  ability and hereditary parameters of main quantitative characters of ten  maize inbred lines. J. of Maize Sciences. 11(1): 26-29.
16. Humpries, E.C.1956 Mineral components and ash analysis. Modern methods of plant analysis. Springer and Verlag, Berlin. Pp. 468-502.
17. Johnson, L.P.V. 1963. Applications of the diallel cross techniques in plant breeding. In: Statistical Genetics and Plant Breeding. Nat. Acad. Sci., Nat. Res. Council Washinton, D.C. Pp: 561-570.
18. Kadambavanasundram, M. 1980. Combining ability as related to gene action in cotton (Gossypium hirsutum L.). Ph.D. Thesis (Unpubl.) TNAU, Coimbatore.
19. Kumar, M.V.N., S.S.  Kumar and M. Gansech. 1999. Combining ability studies for oil improvement in maize. Crop Res. (Hisar),18: 93-99.
20. Mahajan,  V. and A.S. Khehra. 1991. Inheritance of quantitative traits in maize (Zea mays L.) in winter and monsoon seasons. Indian J. Genet., 51: 292-300.
21. Mahto. R.N. and  D.K.  Ganguli. 2003. Combining ability analysis in inter crosses of maize. Madras Agric. J. 90(1-3): 29-33.
22. Mathur, A., P. K. Chunilal, S.K. Bhatnagar and V.Singh. 1998. Combining ability for yield, phonological and ear traits in white seeded maize. Indian J. Geneti., 58: 177-182.
23. Nadarajan, N.  1986.  Genetic analysis of fibre characters in cotton (Gossypium hirstum L.) Ph.D. Thesis (unpubl.).TNAU, Coimbatore.
24. Pooran chand (1999). Comibining ability of maize grown in winter region. Madras agric.J.83: 573-75.
25. Prodhan, H.S. and R. Rai. 1999. Combining ability for popping quality and grain yield along with other associated attributes in popcorn. Environment and Ecology, 17(4): 827-830.
26. Ranganna, S., 1995. Manual  analysis of fruits and vegetables products. 2^nd edition. Tata Mc Graw Hill Publishing Co., Ltd., New Delhi. Pp; 25-27.
27. Rodrigues, L., L.C. Ray Mundo, L.C. Lee, L. Simpson and C.D. Cheester. 1976. Carotenoid pigment changes in ripening Momardica charantia fruits. Ann. Bot., 40: 625-630.
28. Sankarapandian, R. and Subbaraman.N. (1994) Combning ability infodder maize. Madras agric.J. 81:117-20.
29. Sharma, R.L. and  B.P.S. Chauhan. 1985. Combining ability in sesame. Indian J. Genet., 45(1): 45-59.
30. Soliman. F.H.S. and S.E. Sadik 1999. Combining ability of new maize inbred lines and its utilization in the Egptian hybrid program. Bulletin  – of Faculty – of – Agriculture- University of Cairo, 50(1): 1-20: 30 ref.
31. Subba Rao.G. and Aruna, C. (1997). Line (Mutants) x tester studies in sorghum .J.Res.APAU 23: 5-7.
32. Suneetha,Y., J.R. Patil and J.R. Srinivas. 2000. Studies on combing ability for forage characters in  maize. Crop Res.,19: 266-270.
33. Tiwari  V.K. and S.S. Verma. 1999. Correlation and path coefficient analysis in baby corn. Agric. Sci Digest 19(4): 230-234.
34. Xuelu Wang and Brian Larkins. 2001. Genetic analysis of amino acid accumulation in opaque 2 maize endosperm. Plant physiology, 125: 1766-1777.
35. Yousif. D.P., H.C. Ali and R.H. Baker. 2003. Estimation of heterosis and combining ability in local maize inbred lines. Agrl. Science 30(2): 246-259.

Table 1. Details of parents

Table 3.  Estimates  of  components of variance of yield  and quality traits in QPM and normal maize lines

Table 4.  General combining ability effects of parents for yield traits

Table 5.  General combining ability effects of parents for quality traits

Table 6.  Three top ranking hybrids in the basis of gca effects along with their respective gca status and per se performance for grain yield and qualitative traits QPM lines and hybrids

 Table 7 : Sumup of per cent heterosis or grain yield and other qualitative traits in QPM maize lines