Genetics of oil and yield related traits in cotton

Genetics of oil content, plant height, lint index, ginning out turn and lint weight showed that these traits were controlled by dominant genes and dominance x dominance interactions. The traits viz., monopodial branches, seed cotton yield, boll weight, sympodial branches per plant and number of bolls per plant were controlled by dominance gene effects and additive x additive gene interactions. For seed index, additive gene interaction was prominent.

Illustrations explain the paper; they do not show study data.

AI-generated illustration

↗ Enlarge

Comparison of Bt vs Non-Bt Oil ProfilesClose ×

AI-generated conceptual illustration (cotton seeds and cotton seed oil). It explains the idea and does not show study data.

Open full-size image ↗

Introduction

Cotton has a proud place among the cash crops from the earliest times. It finds mention in the Rigveda the oldest scripture of the Hindus. Manu, the law giver also referred to it in his Dharma Shastra. It was the excellence of Indian cotton fibres famed as webs of woven wind which compelled European countries to seek new trade routes with India. Since the discovery of the Mohen-jodaro relics the history of cotton and cotton manufacture came to be treated as beginning from times of the ancient Indus valley civilization, which flourished in India about 5000 years ago. Despite the advent of a multitude of other fibres, cotton, white gold rules the world of textile. Even today, it is unchallenged as a natural textile fibre. It is an important fibre and food crop of nearly 100 countries with China, India, United States, Pakistan and Brazil being five of the largest producers of cotton. Gossypium hirsutum, G. barbadense, G. arboreum and G. herbaceum are grown in varied agroclimate in northern, central and southern Indiafrom 9 0 N latitude to 31 0 N latitude. G. hirsutum is the principal cultivated cotton and accounts for about 90 per cent of the world cotton production. G. barbadense accounts for about eight per cent of the world cotton production. The estimated area under cotton during 2008 to 2009 in India was 93.73 lakh ha with production and productivity of 290 lakh bales and 526 kg lint per ha, respectively (Sharma et al., 2009) [67] . In the year of 2008 in Karnataka, cotton was grown over an area of 3.5 lakh ha with 8.9 lakh bales production. India has 20 per cent of total cotton growing area of the world and contributes over 12 per cent to total production. Cotton plays a key role in the national economy interms of generation of direct and indirect employment of about 60 million people in the agricultural and industrial sectors of cotton production and processing, textile and related exports which accounts for nearly 33 per cent (76,000 crores) of total foreign exchange earnings of our country. Cotton though mainly grown for fibre is also ranked as major oilseed crop in the international market. Out of the four major products i.e., meal, hull, oil and lint, oil is most important. Besides commercial importance in the leather industry and as a lubricant, cotton seed oil can also be used for edible purpose after refining. Cotton seed oil is premium quality oil and there are several reasons for this. India produces about 3.6 million tonnes of cotton seed annually, from which 0.545 MT of oil can be obtained by proper processing of seeds. Cotton harbours about 14 to 26 per cent oil. However, hardly 5 per cent of the cotton seed is processed scientifically resulting in losses of cotton byproducts worth Rs. 2500 crores annually. Recently, India imported 17 per cent of palm oil and 6 per cent of soybean oil. Nearly half of our vegetable oil requirement is met through import.

Genetic research in cotton in India began at the dawn of the last century founded by the provincial governments of the states growing cotton. Later the Indian Central Committee, the Indian Council of Agricultural Research and State Agricultural Universities sponsored genetic research in cotton. The improvement in seed, oil and seed cotton yield related traits requires the presence of genetic variability. The range of variability in cultivated Gossypium species is very wide for cotton seed oil. Kohel et al. (1985) [27] reported the existence of germplasm lines with as high as 32 per cent protein and 30 per cent of oil. Murthi and Appu Rao (1963) [41] have given an account of genetic variability for fuzz, kernel and hull percentage. Now research efforts for yield and quality improvement of oil, seed and yield related traits are in progress in India, USA, Pakistan, China and Egypt. Cotton seed oil is one of the healthy domestic vegetable oils that is grown and processed in India and other countries of the world. Cotton seed oil has taste and flavour. It is also blended and doesn't mask the flavour of food cooked in it.

Little is known about the nature of gene action for oil content, yield and its related traits in G. hirsutum despite an early awareness of the good qualities of cotton seed oil. Systematic investigation on the inheritance was initiated by Kohel (1980) [26] and Khan et al. (1992) [23] . However, these studies are essential in knowing genetic trend, which would allow the development of more efficient selection method and genetic populations. The present investigation aims to obtain such information with reference to oil and its related traits and yield and its related traits. The cotton seed oil is transfree, has neutral flavour, is highly stable, with low flavour reversion, strong shelf life, low fryer turnover, extremely versatile, heart healthy in moderation, according to the American heart Association. It is a good source of essential fatty acids (70% unsaturated, 26% saturated and good source of vitamin E). Cotton oil has a fatty acid profile that makes it acceptable as healthful oil. This profile gives it characteristics that make it very useful as cooking and frying oil (Boghara et al., 1985) [7] . It is one of the healthful domestic vegetable oil that is grown and processed in India. It is blended oil and does not mask the flavour of food cooked in it. Evaluation of different cotton genotypes is mainly focused on cotton yield, fibre yield, fibre quality and resistance to various pest and diseases. The byproducts were being ignored until recently; since many other oil seed like mustard, soybean, groundnut and palm oil dominate the oil seed scenario. In particular a number of genotypes having high oleic acid have been identified in cotton in order to provide high stability cooking oils. These oils provide the opportunity to replace the current wide spread use of saturated fats and hydrogenated oil that contribute significantly to increased risk of cardiovascular diseases due to the effect of saturated and trans fatty acids on elevating LDL cholesterol in the blood stream. Similarly, oil with increased stearic acid content are being developed to enable the production of solid fats without the need for hydrogenation (Sharma et al., 2009) [67] . Yields were low compared to world average partly because of heavy losses from insect pests particularly cotton bollworm complex. Yield losses due to insect pests are estimated to be around 10 to 14 per cent every year. India is addressing the need for increased cotton Bt cultivars. These insect protected cotton varieties contain a naturally occurring substance, Bacillus thuringiensis (Bt) protein which has been used as an ingredient in safe and effective biological sprays for more than 50 years. Bt trait has been successfully transferred into several Indian lines. Extensive and fully replicated field traits of Bt cotton were conducted from 1998 to 2001 cropping seasons, meeting the government requirements for commercialization. Three Bt cotton cultivars have been approved for planting in India in 2002-03. Since, the introduction of Bt cotton hybrid around 44,500 ha were planted with three hybrids of Bt cotton in central and southern zones in 2002-03 season. This increased to some 1, 00, 000 ha in 2003-04. In 2004-05 around four Bt cotton hybrids were planted over 5, 00, 000 ha by three lakh resource poor farmers. With approval of 16 new hybrids of half a dozen companies including six Bt cotton hybrids for northern region, Bt cotton planting for 2005-06 season has experienced the highest yearly percentage growth rate increasing its area by 160 per cent (13 lakh ha). Around 10 lakh farmers elected to plant Bt cotton hybrids in northern, central and southern cotton growing zones of India as compared to 3 lakh farmers in the previous year (Anon., 2006) [6] . In India, cotton is cultivated in diverse situations ranging from irrigated situation characterized by intensive management to assured rainfed situation where moisture is not a constraint to dry rainfed situation characterized by low rainfall (less than 500 mm) accompanied by high temperature and high intensity. To achieve the target of increasing cotton production at national level, it is essential that productivity of cotton increases in all these situations. In this regard the adaptability of genotypes in different environment is very important. Thereby the present investigation to identify Bt and non Bt hybrids in South and Central India was taken up. In a landmark decision, the Genetic Engineering Approval Committee (GEAC) of the Ministry of Environment and Forest (MOEF) has approved 43 new hybrids of Bt cotton varieties to be sold in 2006 season in addition to the 20 Bt cotton hybrids approved for sale in 2005. This brings the total of Bt cotton hybrids to 59 (63 approvals) giving farmers of India's three cotton growing zones more choices about which varieties to cultivate in 2006. This includes 14 hybrids containing three events to be sold by thirteen companies in Southern Zone of India. Of the total available Bt cotton hybrids in 2006, four different events containing Cry1Ac gene (MON 531 event) by Mahyco sourced from Monsanto, stacked Cry X (Cry1Ac and Cry2Ab) gene event (MON 15985) developed by Mahyco sourced from Monsanto, Cry1Ac gene (Event 1) by JK-seeds sourced from IIT Kharagpur and fusion gene Cry1Ab and Cry1Ac (GF Mervent) by Nath seeds sourced from China have received commercial approval for sale in Northern, Central and Southern cotton growing zones. Till now there is not much information regarding expression of Bt gene in different genotypes at different stages of crop life and different parts of the plant.

Genetics of oil, seed cotton yield and its related traits Generation mean

Little is known about the nature of gene action for oil content and protein percentage in G. hirsutum L, despite an early awareness of the good qualities of cotton seed oil, systematic investigations on the inheritance were initiated by Kohel [1980] [26] , and in isolated instances Boghara et al. [1985] [7] and Khan et al. [1992] [23] . However, these studies are essential in knowing genetic trends, which would allow the development of more efficient selection methods and genetic populations. The present investigation aims to obtain such information with reference to oil content and protein percentage, which may ultimately make a sound basis for further improvement work in upland cotton in India.

Oil percentage

Dominance components (H 1 ) was generally higher than the additive (D) component. Kohel (1978) [25] worked out the possibility of breeding for higher yield of oil per hactare through the combination of high yeild of seed and oil per cent. In view of range of variability in the cultivars and genetic resources, the scope for selection for high oil content in future was found to be quite high. Kohel (1980) [27] reported that the oil content in cotton seeds may be under the control of additive genes. He found a heritability of 0.35 based on F 2 /F 3 standard unit regression analysis. Detected epistasis of additive and to a lesser extent that of non-additive type heritability estimates for oil content which ranged between 0.42 and 0.53. He concluded that analysis of gene effects by generation means is relatively more effective in providing genetic information for seed oil content. Wang and Li (1991) [66] observed frequencies and directions of dominant and recessive effects and closely approached complete dominance. The frequencies and direction of dominance and recessive genes, the ratio of positive and negative genes and the narrow sense heritability of oil content trait differed within studied parents. They also studied inheritance of upland cotton parents grown in Xinjiang indicated additive dominance inheritance without epistasis inheritance. Pundhan Singh and Narayanan (1991) [50] studied pedigree selection which was carried out on G 67 × American nectariless and Khandwa-2 × Reba B-50 MB during 1982-88. Reddy and Satyanarayan (2005) [52] reported that low heritability for oil content. Naveed Murtaz et al. (2002) [42] reported that oil content was controlled by both additive and non-additive gene effects.

Al-Raqi and Kohel (1969) [2] The investigation of showed that seed indexwas controlled by partial dominance and polygenically inherited dominance or epistatsis or both. Chahal and Singh (1974) [9] reported seed index influenced by epistatic gene action. Mathapati et al. (1978) [34] made six intrabarbadense crosses. Their studies revealed that seed index was controlled by non-additive gene action. Similar results were observed by Waldia et al. (1980) [64] , Ahmad et al. (2003) [2] , Reddy and Satyanarayan (2005) [52] , and Rokaya et al. (2005) [54] . Singh and Sandhu (1979) [59] observed both additive and nonadditive gene action for seed index and they observed additive x additive type of interaction. Same type of gene action was observed by Reddy et al. (1999) [53] . Silva and Alves (1983) [58] , their studies revealed that seed index was cotrolled by additive gene action observed. Similar results were observed by Abo-El-Zahab (1986) [1] , Pavasia et al. (1999) [48] , Iyanar et al. (2005) [18] and Subramanian et al. (2005) [62] . Reddy et al. (1999) [53] reported additive and additive x additive type of gene action were greater importance for seed index Wang and Li (1991) [66] observed frequencies and directions of dominant and recessive effects revealed closely approached complete dominance. They suggested that to determine the genetic conditioning of yield per plant and additive dominance and epistatic effect involved in inheritance of seed index. Mohamed et al. (2001) [37] showed dominance × dominance gene interaction for seed index. Mohammad et al. (2003) [38] observed non-additive gene action i.e., dominance or epistasis for seed index. Patel et al. (1996) [46] . Found additive type of gene action for yield. Gad et al. (1974) [15] conducted on experiment on interspecific hybrids of G. hirsutum × G. barbadense. They observed dominance × dominance gene effect for seed cotton yield. Jain (1980) [20] found dominance gene effect for yield. Kumar et al. (1974) [28] reported non-additive type of gene action for seed cotton yield. Similar results were observed by Mathapati et al. (1978) [34] in G. barbadense, Waldia et al. (1980) [64] . Desai et al. (1980) [10] reported both additive and non-additive gene action in control of seed cotton yield. Similar results observed by Lertprasertrat et al. (1987) [35] and Jagtap and Kohel (1987) [19] , who made graphical analysis applied to 5 × 5 crosses of Gossypium hirsutum indicated over dominance for seed cotton yield, Reddy et al. (1999) [53] , Kumaresan et al. (1999) [29] , Subramanian et al. (2005) [62] , observed additive type of gene action in kapas weight. Kashif -Nadeem and Azhar (2004) [21] realized that additive type with partially dominance type of gene action for seed cotton yield. Milan et al. (1989) [36] took six G. hirsutum strains used in their experiment and based on generation mean analysis, seed cotton yield controlled by dominant gene effects. Sambamurthy and Ranganadhacharyalu (1999) [56] conducted experiment on G. hirsutum strains. Their studies exhibited seed cotton yield controlled by epistasis type of gene action. Gad et al.(1974) [15] studied gene effects for boll weight in the cross G. barabadense × G. hirsutum. Their study revealed additive × additive epistasis as being significant for boll weight. Singh et al. (1976) [60] conducted on experiment involving J34 and H14 which were crossed as female parents with 20 other varieties. Based on their study, they noticed that additive gene action was important for boll weight. Similar results were observed by Pathak and Kumar (1975) [47] , Iyanar et al. (2005) [18] . Waldia et al. (1980) [64] reported that non-additive gene action was predominant for boll weight. Same effects were observed by Aba-El-Zahab (1986) [1] , Waldia et al. (1984) [65] , Valarmathi and Jehangir (1998) [63] , Subramanian et al. (2005) [62] and Saeed Rou et al. (2005) [55] . Desai et al. (1980) [10] observed both additive and non-additive gene action for boll weight. Milan et al. (1989) [36] , Silva and Alves (1983) [58] , Reddy et al. (1999) [53] , Kumaresan et al.(1999) [29] , Nimbalkar et al. (2004) [44] , Rokaya et al. (2005) [54] also reported similar gene effects. Jagtap and Kohel (1987) [19] applied graphical analysis to 5 × 5 cross of Gossypium hirsutum indicated that over dominance for boll weight. Khan et al. (1999) [24] and Naveed-Murtaza (2002) [42] also observed the same results. Iftkhar-Ahmad et al. (2001) [75] showed that the gene action controlling the inheritance of boll weight studied a 4 × 4 diallel cross experiment in cotton. Additive type of gene action with partial dominance was observed for boll weight. Ahmad and Mehra (2003) [2] studied generation means analysis of data from the intra hirsutum cross Pusa 45-3-6 × Pusa 19-27 which indicated the presence of dominance and epistatic interactions in the genetic control of boll weight. Kashif Nadeem and Azhar (2004) [21] found that additive type of gene action with partial dominance. Mohamed et al. (2001) [37] found additive × additive type of gene interaction for boll weight. Similar results were observed by Reddy et al. (1999) [53] .

Seed cotton yield

Seed cotton yield

Boll weight

Boll weight

Number of bolls per plant

Aguado et al. (2008) [4] reported that number of bolls per plant were controlled by both additive and dominance type of gene action. Do Thi Ha An et al. (2008) [12] observed non-additive (Dominant) gene action for number of bolls per plant. Neelima and Reddy (2008) [43] noticed that number of bolls per plant was controlled by both additive and non-additive gene action in G. hirsutum. Similar results were observed by Pale et al. (2008) [45] , Zhang Yong Shan (2008) [74] observed additive type of gene action for number of bolls per plant in Gossypium hirsutum L. Similar results observed by Elangaimannan et al.( 2007) [13] and Mei-Yong Jan et al. (2006) [35] .

Fruiting node

Soliman et al. (2007) [61] conducted on experiment on biparental mating system. They found fruiting node was controlled by additive type of gene action. Similar results were observed by Zhang-Young Shan et al. (2008) [74] . They found both additive and dominance gene effects.

Plant height

Al-Raqi and Kohel (1969) [2] selected nine varieties, whose all possible F 1 hybrid combinations and their corresponding F 2 s were investigated. This revealed that plant height was controlled by epistasis. Silva and Alves (1983) [58] reported plant height was affected significantly by epistasis. Waldia and Yadava`s (1984) [65] data on yield and six yield related characters from a cross between 10 (Female) lines and 3 (Male) testers indicated gene action was predominantly non-additive for plant height. Lertprasertrat et al. (1987) [35] , generation mean analysis of revealed that plant height controlled by additive × additive interaction. Kumaresan et al. (1999) [29] reported that both additive and nonadditive gene effects were important for these traits. Similar observations were made by Malek and Shamsuddin (1999) [33] . The height of main stem was controlled by additive type ofgene action with parallel dominance by Khan et al. (1999) [24] . Ahmed and Mehra (2000) [3] conducted on experiment on generation mean analysis the data was generated from the intrahirsutum cross Pusa 45-36 × Pusa 1927. The investigation indicated the presence of dominance and epistatic interactions in the genetic control of plant height. The studies of Kashif Nadeem and Azhar (2004) [21] revealed maximum number of dominance genes for plant height. Iyanar et al. (2005) [18] noticed predominance of additive gene action for plant height. Similar results were reported by Subramanian et al. (2005) [62] , Pavasia et al. (1998) [49] . However, plant height was controlled by non-additive type of gene action (Waldia et al., 1980) [64] .

Monopodial branches

Ahmed and Mehra (2000) [3] conducted on experiment on generation means analysis of intra hirsutum crosses. Pusa 45-3-6 × Pusa 19-27 indicated the presence of dominance and epistatic interaction for number of monopodia per plant. Nimbalkar et al. (2004) [44] studied eight genetically diverse genotypes of desi cotton (G. arboreum and G. herbaceum) for genetics of monopodia, their study revealed additive type of gene effect controlled the number of monopodia per plant. Kaushik and Kapoor (2007) [22] also observed similar results.

Lint index

El-Zahab (1973) [14] studied 15 selected lines of diverse geographical origin with two commercial local varieties. There was predominantly non-additive gene action for lint index. Chahal and Singh`s (1974) [9] studies exhibited epistasis type of gene action for lint index. Kumar et al. (1974) [28] observed that lint index was controlled by non-additive gene action. Similar type of results was observed by Mahammed et al. (2003) [40] .

Number of sympodia per plant

Neelima and Reddy (2008) [43] reported additive as well as nonadditive gene effects for number of sympodial branches per plant. Punitha et al. (2008) [51] observed non-additive type of gene action for number of sympodial branches per plant. Similar results were observed by Muhammad et al. (2007) [39] and Pale et al. (2008) [45] . Do Thi Ha An et al. (2008) [12] found that number of sympodia controlled by additive gene effect. Similar gene effects were observed by Zhang Yong Shan (2008) [74] .

Lint yield per plant

Lack of previous studied of lint weight for gene action.

Singh et al. (1968a) [64] in crosses involving 15 selected lines of diverse geographical origin which carried local two varieties, reported gene action for ginning outturn percentage as being predominantly non-additive. El-Zahab (1973) [14] reported that ginning outturn percentage controlled was by additive type of gene action in G. hirsutum. Similar results were observed by Singh et al. (1976) [60] , Chahal and Singh (1974) [9] , Pathak and Kumar (1975) [47] , Gad et al. (1974) [15] , Deshmukh et al. (1980) [11] , Abo-El-Zahab (1986) [1] , Milan et al. (1989) [36] , Pavasia et al. (1999) [48] , Za-Ul-Islam et al. (2001) [73] , Reddy and Satyanarayana (2005) [52] in G. hirsutum. Singh and Sandhu (1979) [59] reported that dominance effects were more important than additive effects in the inheritance of ginning outturn. Similar results were observed by Silva and Alves (1983) [58] . However, for ginning outturn (%), non-additive type of gene action was observed by Waldia et al. (1980) [64] , Larik et al. (1997) [34] , Valarmathi and Jehangir (1998) [63] , Subramanian et al. (2005) [62] and Ahmad et al. (2003) [2] in G, hirsutum gonotypes. Nimbalkar et al. (2004) [44] reported both additive and nonadditive gene action were important for ginning outturn percentage. Similar results were observed by Rokaya et al. (2005) [54] . Chanbunmee and Sriwarnat (1987) [31] found additive × additive gene interaction for ginning outturn percentage. Similar results were observed by Pathak and Kumar (1975) [47] also observed same results. Jagtap and Kohel (1987) [19] observed over dominance gene action in G. hirsutum crosses for ginning outturn. Amarturdiev (1989) [5] observed the inheritance of ginning outturn in F 1's nine hybrids were intermediate between their parents and one showed over dominance of low outturn found Mohamed et al. (2001) [37] reported that additive gene effects were significantly positive for lint percentage and also significant for additive × additive gene action. Kashif Nadeem and Azhar (2004) [21] carried out studies in order to determine type of gene action controlling lint percentage in G. hirsutum. Their studies revealed additive type of gene action with partial dominance.

a) Component variance

The components viz., phenotypic (σ 2 p), genotypic (σ 2 g) and environmental (σ 2 e) variances were used for estimation of phenotypic and genotypic coefficient variations as per the method suggested by Burten and Devane (1953) [8] . (1956) [18] .

1) Genotypic coefficient variability

Where, σ 2 g = Genotypic variance and σ 2 p = phenotypic variance c) Genetic advance The genetic gain was predicted using the formula given by Lush (1949) [32] and Johnson et al. (1955) [22] and it is as follows.

Where, GA = Genetic advance H = Heritability = σ 2 g/σ 2 p Σp = Phenotypic standard deviation K = Standard selection differential which is 2.06 at 0.05 level

Genetics of oil, seed cotton yield and related traits

In the present study two cultivars viz., B58-1290 and VCH were used to produce the F 1 and three other generations viz., F 2 , BC 1 and BC 2 . The study thus had six generations viz., P 1 , P 2 , F 1 , F 2 , BC 1 and BC 2 .Quantitative analysis based on additive and nonadditive gene effects using six generation means was considered appropriate to describe inheritance of characters related to seed and seed cotton yield. Significance of different parameters in scaling tests (ABC) revealed possibility of non-allelic interaction effects. Six parameters model of Hayman (1958) [16] was used to estimate gene effects (m, d, h, i, j and l) in different characters.

Oil content

Dominance (h) gene effect was found prominent. Among the interaction effects, dominance  dominance gene interaction played major role in the control of oil content. Significant and opposite signs were observed in (h) and (l) parameters revealing the presence of duplicate nature of epistasis. Wang and Li (1991) [66] , Mohammad et al. (2001) [37] and Mohammad et al. (2003) [38] obtained similar results.

Seed index is one of the important parameters contributing to the seed cotton yield and oil percentage. All the other genetic parameters viz., d, i, j and l were found to be significantly important except dominance effect (h). There was higher magnitude of additive gene (d) action than the dominance (h) gene action. Similar reports of prominence of additive gene action were made by Silves and Alves (1983) [58] Abo-El-Zahab (1986) [1] , Pavasia et al. (1999) [48] , Iyanar et al. (2005) [18] and Subramanian et al. (2005) [62] . Simple selection for higher seed index is thus recommended.

Based on estimation of gene effects in the present study, dominance component (h) was found more prominent than additive component. Desai et al. (1980) [10] , Lertprasertrate et al. (1987) [35] , Reddy et al. (1999) [53] , Kumaresan et al.(1999) [29] and Subramanian et al. (2005) [62] also reported the importance of dominance gene action. Based on the average value of F 1 which exceeded both parents P 1 and P 2 , over-dominance is implicated. Jagtap and Kohel (1987) [19] made similar conclusions. Presence of dominance and over dominance in these hybrids produce their high heterotic performance for seed cotton yield.

The estimation of gene effects depicted that the dominance component (h) and additive × additive gene interaction were significant in the inheritance of boll weight. Similar results were observed by Ahmed and Mehra (2000) [3] . Considering interaction effects, additive × additive effects played a major role in the genetic control of the trait. Significance of epistatic effects for the trait was also narrated by Gad et al. (1974) [15] , Reddy et al. (1999) [53] and Mohamed et al. (2001) [37] . Considering the complex inheritance patterns noticed in all the traits studied, it is advisable to go for development of hybrids to realize higher seed cotton yield.

Number of bolls per plant

With respect to boll number, all genetic parameters were found to be significant except additive  additive interaction (i) effect. The magnitude of dominance effect (h) with highest magnitude (7.33) was found to play a major role in the inheritance of the trait. Among interactions, dominance  dominance effect was found prominent. The opposite sign of 'h' and 'l' revealed duplicate type of epistasis for this trait. Similar results were observed by Do Thi Ha An et al. (2008) [12] , Aguado et al. (2008) [4] , Nilima and Reddy (2008), Pale et al. (2008) [45] and Zhang Yong Shan et al. (2008) [74] .

Fruiting node Dominance effect (h) with highest magnitude (3.66) was found to play major role in the inheritance of fruiting node. Components h and l were in similar direction, revealing complementary gene epistasis. Similar kind of results was observed by Zhang Yong Shan et al. (2008) [74] .

The mean of F 1 generation was higher than both the parents for this trait. Though all the genetic parameters were significant, based on relative magnitude, dominance effect (h) and dominance × dominance gene effects (l) were found important in the genetic control of plant height. Similar results were observed by Waldia et al. (1980) [64] , Silva and Alves (1983) [58] , Waldia and Yadava (1984) [64] , Ahmed and Mehra (2000) [3] , Malek and Shamsuddin (1999) [33] , Kashif Nadeem and Azhar (2004) [21] . But, additive and additive × additive gene control for plant height was observed by Lestprasertrat et al.(1987) [35] , Pavasia et al. (1998) [49] , Kumaresan et al. (1999) [29] , Khan et al. (1999) [24] , Iyanar et al. (2005) [18] and Subramanian et al. (2005) [62] .

Monopodial branches

Perusal of gene effects revealed higher magnitude of dominance component (h) than additive component (d) revealing its prominence in the genetic control of number of monopodial branches. Among the interaction effects, additive × additive (d) and dominance × dominance effects (l) were found most important indicating duplicate epistasis in the inheritance of this trait. The reports of Ahmed and Mehra (2000) [3] supported this observation. But, Nimbalkar et al. (2004) [44] observed additive gene effects for this trait.

Persual of data indicated that mean value of F 1 generation exceeded the better parental mean of P 1 indicating overdominance for lint index. Similar result was observed by Chahal and Singh (1974) [9] considering gene interaction effects all of which were noticed to be significant at 1 per cent level. For lint index, higher magnitude of interaction observed was dominance × dominance type. Chahal and Singh (1974) [9] obtained similar kind of interaction.

Sympodial branches per plant: Dominance (h) gene effect was highest, revealing dominance gene effect in the control of this trait. Among the interaction effects, additive  additive effect (i) was found prominent followed by additive  dominance (j) effect. Further, estimates of 'h' and 'l' were significant with positive and negative values, respectively. This indicated that duplicate nature of gene interaction was operating for the trait sympodial branches per plant. Punita et al. (2008) [55] also observed similar kind of results.

The mean of F 1 was better than both parents P 1 and P 2 revealing over dominance for lint weight. Among the interaction effects, additive × additive type of interaction was found prominent among the interactions to indicate presence of complementary gene action was operating for this trait.

The ginning outturn percentage is very important in cotton crop, because cotton is mainly grown for its fibre. Jain (1980) [20] reported over -dominance for the trait. Singh and Sandhu (1979) [59] and Silvas and Alvas (1983) [58] reported that dominance gene action was predominant to additive effects in the inheritance of ginning outturn. In the present study perusal of data indicated that ginning outturn controlled by dominance gene effects. Similar results were observed by Jagtap and Kohel (1987) [19] . Both the additive (d) and dominance (h) gene effects were significant. However, magnitude of dominance (h) was higher than additive (d) component. Thus revealing its importance in the inheritance of the trait. The interaction effect, dominance × dominance also showed a major role in the expression of this trait. These heterosis breeding thus will help in increasing the ginning outturn.

Summary

Studies on these Bt hybrids had become important so as to find out the best hybrid suited to a situation. In addition, the oil content and its components in Bt genotypes has not been studied. With these in view a comprehensive study was initiated across two different cotton growing zones of India. The conclusions drawn from the present study has been summarized. At Nagpur, bolls per plant, 20-boll weight, ginning outturn percentage, seed index (g), fuzz percentage, hull percentage and kernel percentage showed high genetic variability. The seed cotton yield, plant height, number of monopodial branches and number of sympodial branches per plant were noticed to have higher genetic variability at Bagalkot. Of the three locations, higher heritability was observed for yield (92.40%), 20-boll weight (59.70%), ginning outturn (74.10%), seed index (92.50%), fuzz percentage (97.80%) and kernel percentage (43.50%) at Nagpur. The sympodial branches (39.40%) had higher heritability at Dharwad. The plant height (78.50%), monopodial branches (92.50%), number of bolls per plant (77.30%) and oil percentage (87.80%) expressed higher heritability at Bagalkot. The higher genetic gain for yield (79.20%), bolls per plant (38.16%), 20-boll weight (23.87%), ginning outturn (15.15%), seed index (35.68%), hull percentage (14.63%) and kernel percentage (9.86%) was observed at Nagpur. The plant height (24.65%), monopodial branches (69.83%), sympodial branches (13.62%), oil percentage (20.67%) and fuzz percentage (23.15%) showed higher genetic gain at Bagalkot. Cotton seed oil had fatty acid composition of 2:1 ratio of poly unsaturated to saturated fatty acid. Generally, 70 per cent of it was unsaturated fatty acids including18 per cent monounsaturated (oleic) and 52 per cent poly unsaturated (linoleic) and 26 per cent saturated (palmitic and stearic) fatty acids. On an average, oil content of Bt hybrids (17.43%) was on par with non-Bt hybrids (17.40%). Unsaturated fatty acid composition of Bt (59.71%) and non-Bt hybrids (59.15%) was similar. It can be safely concluded that the presence of the Cry1Ac gene did not affect the oil content or its profile in the Bt hybrids. The myristic acid of Bt hybrids (1.42%) was numerically superior over non-Bt (1.28%) hybrids which helps to increase the keeping quality of oil.

1. Abo-El-Zahab H, El-Zahab AAA. Diallel analysis of quantitatively inherited traits in Gossypium barbadense L. Ann Agric Sci Ain Shams Univ. 1986;28(3):1337-1358.
2. Al-Raqi KM, Kohel RJ. Generation inheritance of fibre properties in international diallel crosses of upland cotton, G. hirsutum L. Crop Sci. 1969;10:82-85.
3. Ahmad RD, Malik AJ, Chang MA, Gul-Hossan, Muhammad-Subhan, Khan MU. Comparison of combining ability of yield components of plant land and ridges pattern of agriculture in Gossypium hirsutum L. J Appl. Sci. 2003;3(6):391-399.
4. Aguado A, Santos B, De los, Blanco C, Ramero F. Study of gene effects for cotton yield and verticillium wilt tolerance in cotton plant (Gossypium hirsutum L.). Field Crops Res. 2008;107(1):78-86.
5. Ahmed MA, Mehra RB. Genetic characterization of Gossypium hirsutum cross Pusa 45-3-6 × Pusa 19-27. Indian J Genet Pl Breed. 2000;60(4):503-510.
6. Al-Raqi KM, Kohel RJ. Generation inheritance of fibre properties in international diallel crosses of upland cotton, Gossypium hirsutum L. Crop Sci. 1969;10:82-85.
7. Amanturdiev AB. Nature of inheritance of some economically useful traits following hybridization of short and dwarf forms of cotton. Tezisydokla dov Tashkent and Zbek. 1989. p. 22-23.
8. Boghara DG, Mehta NP, Pethani KV. Genetic studies on oil content in upland cotton Gossypium hirsutum L. Gujrat Agric Univ Res J. 1985;10(2):1-4.
9. Burton GW, Devane EM. Estimating heritability in fall fescue (Festuca cimelinaceae) from replicated clonal method. Agron J. 1953;45:479-481.
10. Chahal GS, Singh TH. Application of different mating designs to action in Gossypium arboreum L. diallel versus simplified triple test cross. Crop Improv. 1974;1(1/2):61-71.
11. Deshmukh NY, Vitkare DG, Atale SB. Combining ability in yield characters in Gossypium arboreum. Indian J Agric. Sci. 1980;50(2):902-906.
12. Do Thi Ha An, Ravikesavan R, Iyanar K. Genetic advance and heritability as a selection index for improvement of yield and quality in cotton. J Cotton Res Dev. 2008;22(1):14-18.
13. Elangaimannan R, Anbuselvan Y, Venkatesan M, Karthikeyan P. Diallel analysis in cotton (Gossypium hirsutum L.). Int J Pl Sci. 2007;2(2):53-56.
14. El-Zahab AAA. Combining ability in diallel crosses of Egyptian cotton, Gossypium barbadense L. Zeitscrift fur Pflanzenzuchtung. 1973;69(1):42-49.
15. Gad AM, El-Fawal MA, Bishr MA, El-Khishen AA. Studies on gene action in an interspecific cross of cotton manifestation of types of gene effect. Egyptian J Genetics and Cytology. 1974;3(1):117-124.
16. Hanson GH, Robinson HF, Comstock RE. Biometrical studies of yield in segregating population of Korean hespedeza. Agron. J. 1956;48:267-282.
17. Hayman BI. Separation of epistatic from additive and dominance variation in generation means. Heredity. 1958;12:371-390.
18. Ahmed I, Ali A, Zubair M, Khan IA. Mode of gene action controlling seed cotton yield and various components in Gossypium hirsutum L. Pakistan J Agric. Sci. 2001;38(3/4):19-21.
19. Iyanar K, Ravikesavan R, Subramanian A, Thangaraj K, Varman PV. Studies on combining abilities status in relation to heterosis in cotton (Gossypium hirsutum L.). Adv. Pl Sci. 2005;18(1):317-322.
20. Johnson HW, Robinson HF, Comstock HF. Estimates of genetic and environmental variability in soybean. Agron. J. 1955;47:314-318.
21. Jagtap DR, Kolhe AK. A study on the inheritance of certain agronomic characters in cotton. Indian J Agric. Sci. 1987;57(7):456-464.
22. Jain DK. Genetics of yield component and fibre characters in desi cotton (G. arboretum L.). Thesis Abst. 1980;6(4):304-305.
23. Nadeem K, Azhar FM. Genetic analysis of seed cotton yield and its components in Gossypium hirsutum L. Int. J Agric. Biol. 2004;6(5):865-868.
24. Kaushik, Kapoor CJ. Genetics of yield and other traits over environments in American cotton (Gossypium hirsutum L.). J Cotton Dev. 2007;17(1):24-26.
25. Khan AA, Khan MA, Sadaqat HA. Genetics of oil and protein percentage in Upland cotton. Pak J Sci. Ind. Res. 1992;35(3):95-98.
26. Khan MAJ, Khan MA, Lodhi TE. Genetic study of yield and yield-related components in cotton (Gossypium hirsutum L.). J Ani Pl Sci. 1999;9(1):73-75.
27. Kohel RJ. Genetic studies of seed oil in cotton. Crop Sci. 1980;20:784-787.
28. Kohel RJ, Glueek J, Rooney LW. Comparison of cotton germplasm collections for seed protein content. Crop Sci. 1985;25:961-963.
29. Kumar P, Pathak RS, Singh RK. Heterosis and combining ability in upland cotton Gossypium hirsutum L. Indian J Agric Sci. 1974;44(3):145-150.
30. Kumaresan D, Senthilakumar P, Ganesan J. Combining ability studies for quantitative traits in cotton (Gossypium hirsutum L.). Crop Res., Hissar. 1999;18(3):430-432.
31. Lerik AS, Ansari SR, Kumbar MB. Heritability analysis of yield and quality components in Gossypium hirsutum L. Pakistan J Bot. 1997;29(1):97-101.
32. Lertprasertrat K, Jan-Orn J, Srinivas P, Punklom M, Chanbunmec S, Sriwarnat. A study on the inheritance of certain agronomic characters in cotton. Agric. Sci. 1987;20(2):70-103.
33. Malek MA, Shamsuddin AKM. Combining ability and heterosis for yield and yield contributing characters in upland cotton (Gossypium hirsutum L.). Ann. Bangladesh Agric. 1999;8(1):49-56.
34. Mathapati SN, Hiremath KG, Kodape SN, Goud JV. Indian J Agric. Sci. 1978;49(12):722-729.
35. Milan MR, Jenkins JN, McCarty JC Jr, Parrott WL. Breeding upland cotton for resistance to the tarnished plant bug. Field Crop Res. 1989;21(3-4):227-238.
36. Mohamed SAS, Hassan Hemaida GM. Genetical studies in yield and some yield components in the Egyptian cotton cross (Giza 80 × Giza 85). Ann. Agric. Sci. 2001;39(2):835845.
37. Mohammed SAS, El-Adly HH, Eissa AEM. Evaluation of some Egyptian cotton genotypes under different environments. Egyptian J Agric. Res. 2003;81(4):17971816.
38. Ilyas M, Naveed M, Khan TM, Khan IA. Combining ability studies in some quantitative and qualitative traits of Gossypium hirsutum L. J Agric. and Social Sci. 2007;3(2):39-42.
39. I, Chang MA, Iqbal MZ, Mahmood Ul-Hassan, Khalid M. Genetic analysis of earliness and agronomic character in upland cotton. Sarhad J Agric. 2003;19(3):353357.
40. Murthy KS, Appu Rao B. Studies on the compositional and varietal characteristics of Indian cotton seed during decennium 1951-52 to 1961-62. Indian Central Oilseeds Committee; c1963, 13.
41. Murtaza N, Khan AA, Qayyum A. Genetic studies of seed oil and protein percentage in Gossypium hirsutum L. J Res. (Sci.), Bahauddin Zakariya University, Multan, Pakistan. 2002;13(1):77-85.
42. Neelima S, Reddy VC. Genetic parameters of yield and fibre quality traits in American cotton (Gossypium hirsutum L.). Indian J Agric. Res. 2008;42(1):67-70.
43. Nimbalkar RD, Jadhav AC, Mahetre SS. Combining ability studies in cotton (G. arboretum and G. herbaceum). J Cotton Res. Dev. 2004;18(2):150-155.
44. Pale SP, Kumble SK, Madrap IA, Sarang DH. Diallel analysis for combining ability for seed cotton yield and its components in upland cotton (Gossypium hirsutum L.). J Cotton Res. Dev. 2008;22(1):19-22.
45. Patel UG, Patel JC, Patel NN, Patel AO. Variability parameters in diploid cotton. Gujarat Agric. Univ. Res. J. 1996;22(1):9-13.
46. Pathak RS, Kumar P. Combining ability studies in upland cotton (Gossypium hirsutum L.). Zeitschrift-furPflanzenzuchtung. 1975;75(4):297-310.
47. Pavasia MJ, Shukla PT, Patel UG. Combining ability analysis over environments for fibre characters in upland cotton. Indian J Genet. Pl. Breed. 1999;59(1):77-81.
48. Pavasia MJ, Shukla PT, Poshiya VK. Combining ability for vegetative characters over environments in upland cotton (G. hirsutum L.). Gujarat Agric. Univ. Res. J. 1998;23(2):28-32.
49. Singh P, Narayanan SS. Seed oil improvement in upland cotton (G. hirsutum) through the pedigree method selection. Indian J Agric. Sci. 1991;61(1):43-44.
50. Punitha D, Raveendran TS, Vindhiyavarman P. Combining ability studies in hybrid cotton (Gossypium hirsutum). Int. J Pl. Sci. 2008;3(1):40-45.
51. Reddy AN, Satyanarayan. Genetic of yield and fibre quality traits in American cotton (Gossypium hirsutum L.). Ann. Agric. Res. 2005;26(2):190-193.
52. Reddy MS, Hendawy FA, Abd El-Hamid AM, Esmail RM. Combining ability of yield and yield component in cotton. Egyptian J. Agron. 1999;21:53-66.
53. Rokaya MH, El-Marakby AM, El-Agroudy MH, Scif MG. Heterosis and combining ability for fibre to seed attachment force earliness, yield and yield components in half diallel cross of cotton. ArabUniv. J Agric. Sci. 2005;13(3):741753.
54. Raut S, Khan TM, Nazir S. Combining ability and heterosis in Gossypium hirsutum L. Int. J Agric. Biol. 2005;7(1):109113.
55. Sambamurthy JSV, Ranganadhacharyulu VN. Diallel cross analysis for combining ability in upland cotton (Gossypium hirsutum). Madras Agric J. 1999;86(7/9):407-414.
56. Bhaskaran S, Ravikesavan R. Combining ability analysis of yield-related traits and fibre quality traits in cotton (Gossypium spp.). J Cotton Res Dev. 2008;22(1):23-27.
57. Silva FP, Da Alves JF. Diallel analysis of quantitatively inherited traits in Gossypium barbadense L. Ann. Agric. Sci. Ain-Shams Univ. 1983;28(3):1337-1358.
58. Singh TH, Sandhu BS. Estimates of gene effects for fibre and seed characters in Gossypium arboreum Linn. Indian J Agric Sci. 1979;49(5):325-329.
59. Singh T.H., Bhardwaj H.L., Dhillon S.S. The combining ability analysis in upland cotton (Gossypium hirsutum L.) using Indian and exotic germplasms. J Res Punjab Agric Univ. 1976;13(1):8-13.
60. Sharma VK, Yngard RA, Lin Y. Silver nanoparticles: green synthesis and their antimicrobial activities. Advances in colloid and interface science. 2009 Jan 30;145(1-2):83-96.
61. Soliman YAM, El-Yazied MAA, El-Bary AMRV. Application of F₂ biparental mating design to produce new cotton recombination. Egyptian J Pl Breed. 2007;11(1):1123.
62. Subramanian A, Ravikesavan R, Iyanar K, Thangaraj K, Vindhyavarman P. Combining ability analysis in upland cotton (Gossypium hirsutum L.). Pl Arch. 2005;5(1):23-28.
63. Valarmathi M, Jehangir KS. Line × tester analysis for combining ability in Gossypium hirsutum cotton. Madras Agric J. 1998;85(2):103-105.
64. Waldia RS, Mor BR, Jotasra DS. Line × tester analysis for yield and economic attributes in Gossypium arboreum. Indian J Agric Sci. 1980;50(10):745-747.
65. Waldia RS, Mor BR, Yadav JS. Combining ability for yield and its components in desi cotton (Gossypium arboretum L.). Haryana Agric Univ J Res. 1984;14(2):151-156.
66. Wang GY, Li ME. Genetic study of cotton seed quality traits. Acta Agriculture Boreali Sinica. 1991;6(2):20-24.
67. Zia-Islam, Sadaquat HA, Khan FA. Combining ability of some hirsute cotton types for economic traits. Int. J Agric. Biol. 2001;3(4):411-412.
68. Zhang Y, Yu S, Guo X, Wang Z, Wang Q, Chu L. Dissection of genetic effects of quantitative trait loci (QTL) in transgenic cotton. Euphytica. 2008 Jan;159:93-102.
69. Ahmad I, Beg AZ. Antimicrobial and phytochemical studies on 45 Indian medicinal plants against multi-drug resistant human pathogens. Journal of ethnopharmacology. 2001 Feb 1;74(2):113-23.