Combining ability and gene action for yield improvement in kenaf (Hibiscus cannabinus L.) under tropical conditions through diallel mating design

Abstract

Nine morphologically distinct kenaf genotypes were hybridized to produce 36 hybrids following a half diallel mating design. The combining ability and gene action of 15 yield and yield components were assessed in hybrids and their parents across two environments. Except for the mid diameter and plant height traits, there were highly significant differences (p ≤ 0.01) between the environments and the interaction of genotype and environment. Additive gene effects were considerable for the inheritance of these traits, and the expression of these additive genes was heavily influenced by the environment. Significant differences were found for all studied traits for GCA except top diameter and SCA except plant height and top diameter, implying the presence of both additive and non-additive gene action for the inheritance of the concerned characters. For all features except top diameter and number of nodes, the magnitude of GCA variation was significantly higher than that of SCA variance, indicating the additive gene's predominance. The parental lines P₁, P₃ and P₄ were outstanding general combiners for fiber yield and yield-related parameters. Considering combining ability and genetic analysis study, the crosses P₁ × P₄, P₁ × P₉, P₂ × P₃, P₂ × P₅, P₄ × P₆, P₄ × P₇, P₄ × P₉, P₅ × P₈, and P₇ × P₉ were found promising for their heterotic response to higher fiber yield, stick yield, seed yield and could be for future improvement in kenaf breeding programmes.

Introduction

Fiber crops have been a part of human society from the beginning of time. Humans have been known to gather raw materials from the wild for use as ropes or textiles throughout history. Societies later learnt how to cultivate these types of crops. Natural fibre crops are among the earliest known cultivated plants, and humans have continued to domesticate and improve upon them throughout human history. Fiber crop types have been extensively developed through effective selection and breeding in response to the requirements and values of various communities around the world⁵⁰. Kenaf (Hibiscus cannabinus L., Malvaceae) is a high-value fibre crop with significant economic value (Keshk et al. 2006). It is widely cultivated for food and fibre, and it has been extensively utilized as a cordage crop to produce twine, rope, sackcloth, coarse, burlap, and fiberboard⁵⁰. Kenaf is grown commercially for a variety of purposes, including pulping and paper production, oil spill bioremediation, livestock feed, and the production of biodegradable packaging materials²⁵. Kenaf seed is also commercially important because it is a good oil source (16–22%) lubrication, soap production, cosmetics, linoleum, paints, and varnishes^8,33.

In China, and India, hybrid kenaf cultivars accounted for most cultivated accessions²⁴. Hybrid kenaf has received much attention due to its enhanced fiber quality and resistance to force². Due to the ability of kenaf flowers to both self-pollinate and cross-pollinate, the crossover can range from 2 to 24%, depending on insect activity⁴. Dempsey¹¹ reported that the productivity of F₁ in kenaf ranges from 14 to 43% more than the parents. Compared to alternative mating designs, the diallel analysis is an effective method for screening parents for hybrid production. Different varieties of diallel crosses exist, however, in terms of the number of reciprocal crossings, half diallel crosses are more manageable for breeders than full diallel analysis⁷. Combining ability analysis is an effective tool for identifying superior hybrid parents with high general combining ability (GCA) and progenies with improved specific combining ability (SCA)⁴⁵. It is also useful for measuring the genetic worth of parents and crossings in terms of gene activity in quantitative character inheritance and exploitation and breeding¹³.

General combining ability (GCA) refers to a parent's average performance in a series of crossings. In contrast, SCA refers to a hybrid combination that is better or worse than expected based on the average performance of the parental inbred lines involved⁶. Parents with a high GCA effect have additive gene activity, but they do not always have a favorable SCA in their combination³⁸. Meanwhile, determining the sort of gene action that affects the phenotypes of interest using SCA estimation is useful in genetic research. A high SCA identifies non-additive gene action⁴⁶. When measured in terms of average effects (components), SCA effects were bigger than GCA effects, indicating the importance of non-additive gene activity in influencing yield component expression³². SCA and GCA data aid in selecting hybrids and parents for successful breeding³⁴. Therefore, strong hybrids are produced by parents who have good GCA⁴¹.

Heterosis is a genetic phenomenon caused by heterozygosity and is an important plant improvement measurement. Heterosis for fiber yield is well-known, and kenaf hybrid cultivars have been generated and used commercially in China²⁵. The accumulation of dominant genes from both parents, according to Bruce⁵, is linked to the development of dominance heterosis. Yang et al.⁴⁸ described epistasis as interactions among non-alleles on a genome, whereas Cordell⁹ defined it as interactions between genes. Jianmin et al.²⁰ claimed that F₁ heterosis can survive 1.4–1.7 generations on average, but that favorable hybrids could last 3–4 generations. By selecting genotypes with distinct genetic backgrounds, it will be highly useful for improving the variety of kenaf in the Malaysian tropical environment¹². The primary goal of this research was to identify genotypes (parents and offspring) with good combining ability that could assist future kenaf improvement with high fiber yield.

Materials and methods

Planting materials

Nine kenaf genotype parents and 36 F₁ hybrids were employed in this study (Table 1). Among the nine genotypes, one was a commercial variety from Bangladesh and eight mutant lines were developed from V-36 through acute and chronic gamma irradiation by the Malaysian Nuclear Agency in Bangi, Selangor. The parents were mated in all possible combinations, barring reciprocals (half diallel) at Field 10 in University Putra Malaysia to produce 36 F₁ hybrids. The International Union for Conservation of Nature's (IUNC) statement on research involving species at risk of extinction and the convention on the trade in endangered species of wild flora and fauna were followed during the collection of plant materials and execution of the experiment. Table 1 contains a list of all of the accession numbers, along with their IDs codes.

Table 1 Origin and salient features of nine selected kenaf genotypes used as parents for diallel cross.

Hybridization techniques and raising of F₁ seeds

Nine genotypes were chosen as parental materials and mated in diallel fashion omitting reciprocals, considering group distances, genetic distances, and other agronomic performance. To achieve flowering synchronization was observed during the crossing period where pure and healthy seeds were sown in the experiment field three times at 10-day intervals. The F₁ seeds were collected after the fruits had ripened, using standard methods for emasculation and pollination.

Experimental location

The experiment plots were conducted in two seasons in a humid tropical climate between latitude 2° 59′ N and longitude 101° 42′ E, at 48 m above sea level at Field 10, Universiti Putra Malaysia (UPM), Serdang, Selangor Darul Ehsan. The first season was conducted from June 2020 to September 2020, while the second was from March 2021 to June 2021.

Experimental design and field layout

The field was mechanically ploughed and laddered for cultivation. Kenaf seeds from 45 entries, consisting of nine parents and 36 F₁s were planted in peat moss soil in germination trays at a glasshouse for two weeks before being transplanted into the field plots of 59 m × 9 m. The experiment was laid out in Randomized Complete Block Design (RCBD) with three replications using a table of random numbers¹⁵. NPK Green (15:15:15) and NPK Blue (12:12:17) were applied at the prescribed dose of 450 kg per hectare shortly after seeding and after 40 days of transplanting. Intercultural operations like weeding, thinning, supplemental irrigations, and plant protection measures were performed appropriately throughout the cropping season.

Data collection

The genetic diversity was assessed using 22 features (seven qualitative and 15 quantitative) related to the plant, flower, fiber yield, pod, and seed across two seasons. The observation was recorded from 10 randomly selected plants for each genotype per replication for each trait. The qualitative traits, including stem color and leaf shape were visually recorded at seedling and growth stages in the field. Leaf color, petiole color, pod shape, seed shape and seed coat color were visually recorded at the pre-bud and mature stages. Quantitative data collected include plant height, base diameter, core diameter, middle diameter, top diameter, nodes number, days to first flowering, days to 50% flowering, fresh stem weight with leaves and pods, fresh stem weight without leaves and pods, dry stick weight, dry fiber weight, pods number per plant, seeds number per pod and 1000 seeds weight (Table 2). Days to first flowering, days to 50% flowering, and pods number per plant were all verified in the field, and residual traits were measured in the lab 90 days after transplanting³.

Table 2 Quantitative characters studied from nine parents and their crosses.

Statistical analysis

Combined analysis of variance (ANOVA) was performed on all data to assess the amount of variability present among parents and their offspring using SAS (Statistical Analysis Software) version 9.4 (SAS Institute Inc., Cary, NC, USA). The mean performances of hybrids and paternal inbred lines were compared using the Least Significant Difference (LSD) method at a 5% level of significance. The general combining ability of parents and specific combining ability of hybrids were determined following Griffing’s method 2 model 1 using SAS software according to Zhang and Kang⁵¹.

Estimation of heterosis

The relative heterosis (MP) and heterobeltiosis (BP) were determined and expressed as percentages²⁶. The amount of heterosis was calculated using the differences between the mean of F₁ hybrids and the mid parental value for a given characteristic³⁷.

• i. Mid-parent heterosis = [(F₁ − MP)/MP] × 100 (relative heterosis).

  Where, F₁ is the mean value of the F₁, MP is the mean value of two parental involves in F₁ i.e. (P₁ + P₂)/2.

  The significance of relative heterosis was tested using t-test⁴⁷.

  $${\text{t }} = {\text{ F}}_{{1}} {\text{ij}}{-}{\text{MPij}}/({3}/{8}\sigma {\text{e}})^{{{1}/{2}}} ,$$

  where, F₁ij is the mean of the ijth F₁ cross, MPij is the mid-parent value for ijth cross, σ²e is the estimate of error variance.

• ii. Better parent heterosis = [(F₁ − BP)/BP] × 100 (heterobeltiosis).

Where, F₁ is the mean value of the F₁, BP is the mean value of better parent.

The significance of relative heterosis was tested using t-test⁴⁷.

$${\text{t }} = {\text{ F}}_{{1}} {\text{ij}}{-}{\text{BPij}}/({1}/{2}\sigma^{{2}}_{{\text{e}}} )^{{{1}/{2}}} ,$$

where, F₁ij is the mean of the ijth F₁ cross, BPij is the Better value for ijth cross, σ²_e is the estimate of error variance.

Result

Qualitative variation

Qualitative characteristics assessment provides information on highly diverse or uniform character, which can be quite different or very consistent (Table 3). Figures 1, 2 and 3 show the variation observed among the seven qualitative traits, revealing that plant, stem and leaf characteristics differed significantly.

Table 3 Characteristics of selected parents' growth stages and the F₁ kenaf population.

Figure 1

Photographs of the parents' and F₁ population for stem coloration (70 DAS).

Figure 2

Photographs of the parents' and F₁ population for leaf shapes and coloration.

Figure 3

Photographs of the parents' and F₁ population for pod shapes and colors.

Parents

The selected parents' stem color, leaf shape, leaf color (lamina), and petiole color varied greatly. Two of the nine parents, P₁ and P₂, were green stem types with green petioles. Parent P₇ had a light reddish upper surface but a green petiole on the lower surface with a purple stem. Otherwise, five parents (P₃, P₄, P₅, P₆, and P₈) had green petioles with cordate leaves and green stems with reddish patches. Palmate leaf morphologies with green leaf colors were seen in parents P₁, P₇, and P₉ (Table 3). Parents P₂, P₄, P₆, and P₈ had deep green leaves, whereas P₃ had pale green leaves. The genotypes are further divided into three groups depending on pod shape: globular (round), elongated (pointed), and oval (egg-shaped). Genotypes differed in seed shape and seed coat colors. The parent P₄ seed coat color were brownish, while the P₆ seeds were black with a few brownish patches and the rest were ash gray.

F₁ generation

The stem color showed a lot of variances with standard error (Sem, ± 5.17) (Fig. 1). The color variance for the green with reddish patches, green, reddish above greenish below, reddish, purple, and red stem colors was 41.67%, 16.67%, 13.89%, 11.11%, 8.33%, and 8.33%, respectively (Fig. 4a), from the 36 described kenaf F₁ hybrids. Palmate leaves were found in 55.56% of F₁ hybrids, whereas cordate leaves were found in 44.44% (Fig. 4b). The 36 F₁ hybrids' leaf color (lamina) was diverse, falling into three categories: green, deep green, and pale green, which accounted for 72.22%, 25.00%, and 2.78% genotypes, respectively (Figs. 2 and 4c). Petiole color varied greatly among the hybrids investigated, with 72.22%, 22.22%, 2.78%, and 2.78% for green, upper surface light reddish but lower green, reddish, and purple, respectively (Fig. 4d). The form of the kenaf pod also showed a wide range of variation with standard error (Sem, ± 14.26). The genotypes are divided into three groups based on pod shape: globular (round) 52.78%; ovoid (egg-shaped) 41.67%; and elongated (pointed) 5.55% (Figs. 3 and 4e). Seed shape and seed coat color differed across the 36 F₁ hybrids. Triangular seed shape was found in 80.56% of F₁ hybrids, whereas sub-reniform seed shape was found in 19.44% (Fig. 4f). Blackish seeds were found in up to 41.67% of hybrids, ash gray seeds in 38.89%, brownish seeds in 13.89%, and black with few brownish color seeds in 5.55% (Fig. 4g).

Figure 4

Qualitative variation in the kenaf F₁ population (a) Stem color, (b) Leaf shape, (c) Leaf color (lamina), (d) Petiole color, (e) Pod shape, (f) Seed shape, and (g) Seed coat color.

Variation among all genotypes for quantitative traits in pooled environments

Across the two environments, the combined analysis of variance for the 15 quantitative features among the nine parents and 36 crosses revealed significant differences (Table 4). There were significant differences (p ≤ 0.01) in environments and genotypes (parents and offspring) for all the variables studied, with exception of middle and top diameter, respectively. There were significant variations (p ≤ 0.01 or 0.05) for genotype by environment (G × E) except for plant height. Base diameter, core diameter, middle diameter, top diameter, number of nodes, days to 50% flowering, fresh stem weight with leaves and pod, fresh stem weight without leaves and pod, dry stick weight, dry fiber weight, number of pods per plant, number of seeds per pod, and 1000 seed weight were all shown to be highly significant differences (p ≤ 0.01) for G × E interactions. The CV% for yield and yield-related components ranges from 9.40 (days to 1st flowering) to 69.05 (nodes number), showing that the evaluated traits have a wide variability range.

Table 4 Combined analysis of variance was performed for 15 quantitative attributes of nine parents and their crosses over two environments.

Variation among all genotypes due to combining ability effects in pooled environments

Analysis of variance for combining ability using Griffing's (1956) technique was utilized for all analyzed qualities in the F₁ combined data (Table 5) for both environments. General combining ability (GCA) is generally understood to be a consequence of additive gene effects and additive epistatic variance components. Specific combining ability (SCA), on the other hand, is a result of non-additive gene effects and the remaining epistatic variation (Matzinger et al., 1959). The mean squares analysis of variance results for the effects of combining ability are shown in Table 5. Significant differences (p ≤ 0.01 or 0.05) were observed for all studied traits for GCA except for top diameter and SCA except for plant height and top diameter, indicating the presence of both additive and non-additive gene action for the inheritance of the concerned characters. Except for plant height, and fresh stem weight without leaves and pod, significant differences (p ≤ 0.01 or 0.05) were reported for the interaction between GCA and environment. Except for plant height, top diameter, and days to first flowering, all traits showed significant variations (p ≤ 0.01 or 0.05) when SCA and environment interacted. Thus, the effects of non- additive genes in the traits interacted more with the environment. GCA and SCA ratios ranged from 0.54 (nodes number) to 8.29 (days to first flowering). For top diameter, nodes number, and fresh stem weight with leaves and pod, the ratio of GCA and SCA variances was found to be smaller than unity, indicating that non-additive gene action predominated. The estimated GCA/SCA was greater than unity for the other characters, indicating that additive gene effects predominated in their expression. In kenaf, Jianmin et al.²⁰ and Heliyanto et al.¹⁸ found similar results.

Table 5 Mean squares of analysis of variance for combining ability of the 15 traits in pooled environments.

Mean performance of genotypes over two environments

The mean comparison for all genotypes (parents and offspring) was presented in Table 6. Plant heights ranged from 219.58 to 297.85 cm, with hybrid P₃ × P₇ having higher mean value and P₂ × P₉ recorded the lowest. The base diameter ranged from 19.95 mm (P₂ × P₉) to 31.68 mm (P₂ × P₃), but the base diameters of P₁ × P₉ and P₇ × P₈ were comparable. The P₂ × P₃ had the largest core diameter (28.43 mm), while P₂ × P₉ had the smallest (16.83 mm). The middle diameter was 9.69 mm (P₂) to 15.74 mm (P₅ × P₈), with the maximum and lowest top diameters of P₆ × P₇ and P₁ × P₇, respectively. The hybrid P₆ × P₉ had the most nodes (9.68), while P₂ had the least (4.15). Days to first flowering ranged from 43 to 56.33 days in P₇ × P₉ and P₂ × P₅, respectively. Days to 50% flowering ranged from 51 to 69.17 days. The hybrid P₇ × P₉ was the first to mature, while parent P₂ was the last to mature (Table 6).

Table 6 Mean performance of nine parents and their hybrids for 15 yield and yield contributing characters in kenaf.

Fresh stem weight with leaves and pods ranged from 584.30 g (P₂ × P₉) to 1786.50 g (P₄ × P₆), with the highest and lowest fresh stem weight without leaves and pods being 449.96 g (P₄ × P₉) and 222.41 g (P₆), respectively. The dry stick weight ranged from 71.35 to 152.93 g, with P₁ × P₄ having the highest mean value (152.93 g), followed by P₄ × P₉ (147.69 g), and P₆ having the lowest mean value (71.35 g). The hybrid P₂ × P₃ yielded its most dry fiber weight per plant (38.74 g), whereas P₇ yielded the least (16.3 g). Aside from that, the hybrid P₇ × P₉ produced the most pods per plant (217.15), whereas P₂ had the least (50.82). The number of seeds per pod ranged from 15.08 to 28.76. Parent P₅ had the most seeds per pod, while P₁ × P₂ and P₂ × P₈ contained the fewest. The weight of 1000 seeds varied from 26.20 to 35.05 g. The hybrid P₂ × P₈ had the highest 1000 seed weight (35.05 g), while hybrid P₁ × P₅ had the lowest 1000 seed weight (26.20 g) (Table 6).

Plant height, base diameter, core diameter, middle diameter, top diameter, number of nodes, fresh stem weight with leaves and pod, fresh stem weight without leaves and pod, dry stick weight, dry fiber weight, and pods number per plant are all higher than parental mean. In comparison to the hybrid mean, the parental mean has somewhat longer days to first flowering and days to 50% flowering. Similarly, the parental mean had more seeds per pod and 1000 seed weight than the hybrid means. The fact that the hybrid mean has a lower weight per 1000 seeds than the parental mean is advantageous because it allows for smaller seed sizes. The largest standard deviation (SD) value was observed for fresh stem weight with leaves and pod (518.40) with standard error (Sem, ± 31.55), while the lowest was for top diameter (SD, 1.92; Sem, ± 0.12) (Table 6). The standard error (SE) indicates consistency of the average values, lower SE values suggest that the sample mean is a more precise depiction of the true population mean.

Combining ability effects on genotypes (parents and offspring)

General combining ability effects on genotypes (parents and offspring) in pooled environment

Estimates of GCA effects of individual parental genotypes in the F₁ generation were found to be statistically significant or highly significant for most traits studied. Plant height in a taller stature combination should have a positive GCA effect, while the nodes number should have a negative GCA effect. In the pooled data, the parent P₇ had the maximum plant height (11.99 cm) and the lowest negative GCA values for nodes number (− 0.39), indicating that they were good general combiners for quality fiber yield and might be used in future breeding efforts. In terms of base diameter (2.32), core diameter (2.12), middle diameter (0.80), fresh stem weight with leaves and pod (176.17), fresh stem weight without leaves and pod (59.86), dry stick weight (16.57), and dry fiber weight (3.53), parent P₄ had the greatest positively significant GCA effect. Excluding the P₄, parent P₁ (2.00) and P₃ (1.89) were also shown to be positive and significant general combiners for dry fiber weight, with parent P₃ having a positive and highly significant GCA effect (15.09) for dry stick weight.

The parents P₁ (9.02), P₆ (12.76), and P₇ (17.57) had much higher seed output, which is a critical element in determining the pods number per plant. The parent P₂ had the most positively significant GCA effect, with values of 2.32 and 3.17 for days to 1st flowering and days to 50% flowering, respectively. With − 3.74 and − 3.71, respectively, for days to first flowering and days to 50% flowering, parent P₇ had the smallest GCA effect (Table 7). The parent P₂ showed the lowest negative and highly significant GCA effect, with values of − 17.73, − 140.47, − 18.64, and − 4.50 for plant height, fresh stem weight with leaves and pod, pods number per plant, and seeds number per pod, respectively. For base diameter (− 2.34), core diameter (− 2.49), and fresh stem weight without leaves and pods (− 64.18), the parent P₉ exhibited the lowest negative GCA values. A negative GCA effect is desired for 1000 seed weight to give a smaller seed size combination. The parent P₅ had the most desirable 1000 seed weight with a negative GCA effect (− 1.63) and the highest positive (1.36) for seeds number per pod, both of which are highly significant, indicating that they were good general combiners for reduced seed sizes and the highest seeds number per pod in future breeding programs.

Table 7 Estimates of general combining ability effect (GCA) for 15 morphological characters of kenaf.

Specific combining ability effects on hybrids across the environment

The effects of specific combining ability on hybrids in various contexts in pooled environments are shown in Table 8. SCA effects were detected in all 36 hybrids studied, with positive (desired direction) SCA influences on plant height in 19 of them. The hybrid with significant and beneficial SCA effects were produced by the cross P₅ × P₈ (24.44), which was evaluated as a good specific combiner for tallness. The hybrids P₁ × P₅ (− 23.51) and P₂ × P₉ (− 37.62) were identified as the worst specific combiners for this trait, as shown by the significant negative SCA effects for plant height. The SCA effects on the base diameter ranged from − 6.33 to 5.47. Three of the 16 positive SCA effects for base diameter on the were found to be best specific combiners with highly significant (P₂ × P₃, P₄ × P₆ and P₅ × P₈), while two were found to be good specific combiners with significant (P₁ × P₇ and P₇ × P₉). Other 20 crosses had negative SCA effects, including one (P₂ × P₉) with a highly significant SCA value of − 6.33.

Out of 36 cross combinations, five showed significant positive SCA values for kenaf core diameter, indicating heterotic performance over the mean of their parents. The cross P₂ × P₃ had the highest positive SCA effect (5.81), followed by P₅ × P₈, P₇ × P₉ and P₄ × P₆, indicating that they were the best specific combiners for the trait. The cross P₂ × P₉ had the most negative and highly significant SCA effect (− 5.86) and was considered the worst specific combiner for core diameter. The SCA effects for mid-diameter stem varied from − 1.60 to 2.88. The hybrid P₅ × P₈ had the best SCA effects with highly significant, while another 18 crosses had positive but insignificant SCA values, implying that they could be used as average specific combiners. Other 17 crossings had negative SCA effects, with one (P₁ × P₅) being highly significant (− 1.60) and another (P₆ × P₈) being significant (− 1.31). These are considered the worst specific combiners for mid-diameter (Table 8).

The SCA effects on top diameter ranged from − 0.75 to 1.10. Since four crossings revealed significant positive SCA effects, one of which was highly significant, P₂ × P₉, P₄ × P₇ and P₅ × P₈ and P₆ × P₇ were chosen as good specific combiners for this trait. The SCA values of another 15 crosses were positive but not significant, indicating that they may be regarded as average specific combiners for top diameter. SCA impacts ranged from − 4.15 to 2.39 for a given number of nodes. Hybrid P₂ × P₉ (− 4.15) had the highest negative and highly significant SCA estimate, followed by P₃ × P₉ (− 2.86), P₄ × P₉ (− 2.71), and P₁ × P₉ (− 2.38), with one negative significant P₇ × P₈ (− 1.13), showing that these hybrids had good SCA for lower branch stem and improved fiber yield. Three of the 21 positive SCA effects, on the other hand, were found to be highly significant, and three were found to be significant, therefore they were classified as poor specific combiners for nodes number.

The SCA effects ranged from − 3.32 to 4.10 for days to first flowering (Table 8). The cross (P₂ × P₅) was positively highly significant, with two more significant hybrids (P₄ × P₈ and P₁ × P₉), showing that these hybrids had good SCA for this trait. In contrast, the offspring P₄ × P₆ exhibited the highest negative and significant SCA estimate, followed by P₂ × P₈, all of whom were considered poor specific combiners for the days to first flowering trait. SCA impacts ranged from − 5.71 to 7.44 for days to 50% flowering. The cross P₆ × P₉ of the 15 positive SCA effects, which displayed a good specific combining capacity for this characteristic, had the most significant SCA effects followed by P₂ × P₅. Instead, the crosses P₂ × P₆, P₂ × P₇ and P₄ × P₆ had the most negative and significant SCA estimates, indicating that they were poor specific combiners for days to 50% flowering.

Four crosses viz. P₂ × P₃, P₄ × P₅, P₄ × P₆ and P₅ × P₈ were the best specific combiners for fresh stem weight with leaves and pod trait, with highly significant positive SCA effects. Five more crossings of the 20 positive SCA effects viz. P₁ × P₃, P₁ × P₇, P₂ × P₇, P₃ × P₅ and P₇ × P₉ produced positive and significant SCA effects, making them good specific combiners. Four of the remaining 16 crosses had significant SCA values, indicating negative SCA effects. As a result, they were regarded as poor specific combiners for fresh stem weight with leaves and pod characteristics.

Two crosses, P₅ × P₈ (98.91) and P₂ × P₃ (73.40) had highly significant positive SCA effects, whereas one cross, P₄ × P₆ (61.71), had significant positive SCA effects, indicating that these hybrids had good specific combining ability for fresh stem weight without leaves and pod trait. Another 18 crossings had positive, but minor, SCA impacts and might be classified as average specific combiners. The remaining 15 crosses had negative SCA effects, with three (P₁ × P₅, P₂ × P₉, and P₄ × P₇) having significant SCA values and so being classified as poor specific combiners for fresh stem weight without leaves and pod characteristic.

In kenaf, 18 cross pairings had positive SCA effects, while the other 18 had negative for the stick weight per plant, SCA impacts ranging from − 15.27 to 36.25. Hybrid P₇ × P₉ had the largest significant positive SCA impact (36.25), followed by P₁ × P₄ (28.89) and P₅ × P₈ (27.97), implying that it was a good specific combiner for the trait of stick weight per plant. Three crosses (P₁ × P₉, P₂ × P₉ and P₅ × P₇) having negatively significant SCA values, making them poor specific combiners for dry stick weight.

Two crosses, P₂ × P₃ (11.24) and P₅ × P₈ (7.47) showed highly significant positive SCA effects, while one cross, P₂ × P₅ (3.60) showed significant positive SCA effects, implying better fiber yielding hybrids than their parents' mean and considered best specific combiners for increased fiber weight per plant of kenaf. The remaining 17 crosses, on the other hand, had positive SCA findings, but 16 had negative SCA outcomes. Crosses P₁ × P₅ (− 5.28) and P₂ × P₉ (− 10.37) were the worst specific combiners for the trait in question, with highly significant negative values.

The SCA effects on the number of pods per plant ranged from − 49.49 to 63.49. (Table 8). The cross combinations produced 21 positive SCA effects, with 11 of them having significant positive SCA values for the number of pods per plant. The crosses P₂ × P₃ (7.20), P₃ × P₄ (4.70), and P₂ × P₄ (3.32) had highly significant positive SCA effects, but P₄ × P₆ (− 3.43), P₁ × P₃ (− 2.18), and P₁ × P₅ (− 1.05) had highly significant negative SCA effects.

SCA impacts ranged from − 4.01 to 4.87 for the quantity of seeds per pod (Table 8). Sixteen of the cross combinations had positive SCA effects, with six (P₁ × P₃, P₁ × P₈, P₂ × P₆, P₂ × P₇, P₄ × P₈ and P₅ × P₇) having highly significant positive SCA values and two (P₁ × P₄ and P₆ × P₈) having significant positive SCA values considered as good specific combiners for the number of seeds per pod characteristic. The remaining 20 crosses had negative SCA effects, with 12 of them having highly significant negative SCA effects and were considered the poorest specific combiners for the trait under consideration.

For the character 1000 seeds weight, the SCA effects ranged from − 4.20 to 2.64 (Table 8). Hybrids P₁ × P₅, P₁ × P₆, P₂ × P₇, P₃ × P₈, P₄ × P₇, P₅ × P₈, and P₇ × P₈, all exhibited highly significant negative SCA effects, whilst P₃ × P₉, P₃ × P₄, and P₂ × P₆ had significant negative SCA effects, indicating that these hybrids had good specific combining capacity for smaller seed size. Seven more crossings exhibited negative but minor SCA effects, indicating that they are average specific combiners. The remaining 19 crosses had positive SCA effects, with five having highly significant SCA values and three having significant SCA values that were considered poor specific combiners for the 1000 seed weight trait.

Table 8 Estimates of specific combining ability (SCA) effects for 15 different morphological characters of kenaf.

Estimation of heterosis effect of kenaf hybrids for yield and yield components and morphological traits

Heterosis was calculated using the relative performance of hybrids with mid-parent (MPH) and better-parent (BPH) values. Parent heterosis is favored over mid-parent heterosis in hybrid formation. High positive heterosis values are preferable for yield production traits, whereas negative values are favored for node number and 1000 seed weight.

Relative heterosis response in pooled environments

Percentage heterosis relative to mid-parents (MPs) was significantly positive in 13 out of 36 crosses for dry fiber weight (25.74–91.94), five crosses for dry stick weight (49.3–71.65), and 20 crosses for number of pods per plant (49.22–108.72). However, mid-parents of 18 crosses for fresh stem weight with leaves and pod (32.75–91.43) and five crosses for fresh stem weights without leaves and pod (46.16–56.76) were significantly positive (Table 9). For 1000 seed weight, 29 crosses out of 36 had negative heterosis (in the desired direction). This indicates that they had smaller-sized seeds. Over dominance was observed in cross P₅ × P₈ for base diameter, core diameter, middle diameter, fresh stem weight with leaves and pods, and fresh stem weight without leaves and pods, as well as cross P₄ × P₆ for base diameter, core diameter, fresh stem weight with leaves and pods, fresh stem weight without leaves and pods, and pods number per plant and the cross P₂ × P₃ for base diameter and core diameter, with a high potency’s ratio. Furthermore, the crosses P₂ × P₃ and P₂ × P₅ for fiber yield, and the crosses P₁ × P₄ and P₄ × P₉ for stick yield features had the highest mid parent heterosis. Due to the existence of over dominance, crosses P₁ × P₄, P₂ × P₈ and P₇ × P₉ were chosen for pods number per plant for seed yield. Negative heterosis estimates for the nodes number and 1000 seed weight are desirable, where small values for these traits indicate good fiber quality and smaller seed size, therefore, negative magnitude implies high heterosis. Cross P₂ × P₉ had the lowest negative nodes number value (− 11.02) for quality fiber yield, and cross P₁ × P₅ with the lowest negative heterosis for 1000 seed weight (− 18.76) would produce well where smaller seed size kenaf accessions perform well.

Heterobeltiosis response exposed to a pooled environment

Overdominance for heterobeltiosis was seen in the most outstanding crosses P₂ × P₃ and P₂ × P₅ for fiber yield and the crosses P₁ × P₄ and P₄ × P₉ for stick yield, as shown in Table 9, since their heterobeltiosis values were very significant positive with high potence ratio. Due to the presence of over dominance, the seed yield crosses P₁ × P₄, P₁ × P₈ and P₇ × P₉ were chosen for pods number per plant. Overdominance for heterobeltiosis, as indicated by the potence ratio, was found in the promising crosses P₂ × P₃, P₅ × P₈ and P₇ × P₉ for base diameter and core diameter, cross P₆ × P₇ for top diameter, cross P₅ × P₈ for middle diameter, fresh stem weight with leaves and pods, and fresh stem weight without leaves and pods, and cross P₄ × P₆ for base diameter, core diameter, fresh stem weight with leaves and pods, and pods number per plant. In terms of node number and seed weight, negative heterosis is preferable. Meanwhile, in terms of percentage F₁ heterosis above high parent in node number and 1000 seed weight, crosses P₂ × P₉ and P₁ × P₅ had the lowest negative heterosis values, respectively.

Table 9 Estimates of mid-parent and better parent heterosis for 15 characteristics of 36 crosses in kenaf over two environments.

Selection of best parents and offspring

The yield and yield contributing components of fifteen parents and crosses were investigated in this study. Table 10 lists the best general and specific combiners based on trait performance. It's always possible that some crosses will work better in some characters than others. To determine the most favorable, rate each on a character-by-character basis and see which ones perform better across the range. When the SCA results of 36 crosses were reviewed, the cross P₅ × P₈ was shown to be the best in ranking for all the characters. P₇ × P₉, P₁ × P₃, P₂ × P₃, P₁ × P₇, P₄ × P₈, and P₂ × P₄ were the next crosses (Table 11). This was possibly due to their greater distance from the others. The performance of the cross P₅ × P₉ was determined to be extremely low. This was because these two genotypes are the most similar. However, based on their cumulative rating, ten superior crossings were selected for inclusion in a future breeding effort. Thirty-six kenaf F₁s were arrayed in order of merit, with the rank total against the crosses indicating the lot's position.

Table 10 Best general and specific combiner for 15 traits.

Table 11 Rank positions of the F_1’s based on 15 morpho-physiological characters.

Discussion

The wide variation in the crop's botanical and agro morphological properties suggests that the genotypes are genetically diverse (Ogunniyan, 2016). Segregation was seen in 36 crosses for stem color, leaf shape, leaf color (lamina), petiole color, pod shape, seed shape, and seed coat color in the F₁ generation (Supplementary Fig. 1). In addition, some of the crosses indicated significant differences in leaf shape and pigmentation pattern on the stem and petiole (Table 3). Indeed, the green morphotype kenaf leaves are more valuable for their food value, and their fibers are better for stringing²². The parents had green, green with reddish patches, and purple stem color, however the cross combinations had green, green with reddish patches, purple, reddish above greenish below, reddish, and red stem color. When the upper surface light reddish but lower surface green pigmented parent P₇ was crossed with the rest of the green parents, the F₁ showed various pigmentation patterns on the petiole, including green, upper surface light reddish but lower surface green, reddish, and purple. When the ash gray, brownish, and black with few brownish seed coat color parents crossed, F₁ offspring with blackish, ash gray, brownish, and black with few brownish seeds were produced, as were those observed by Osman et al. (2011). In addition, the leaf shape, leaf color (lamina), pod shape, and seed shape of selected parents and their F₁ hybrids showed a wide range of diversity.

Table 4 shows the results of the mean squares analysis of variance on the combined data of the two environments, which revealed that, except for middle diameter, the effects of the environments were highly significant (p ≤ 0.01). The genotypes performed differently across the environments studied, based on the analysis results on the combined data. The effects of genotype by environment interaction were highly significant (p ≤ 0.01) for all traits measured except for plant height, and significant (p ≤ 0.05) for days to first flowering. These results indicate that the effects of genes controlling these traits were expressed differently at different environments. The mean squares of GCA were bigger than the mean squares of SCA for all traits except top diameter, nodes number, and fresh stem weight with leaves and pods in our investigation indicating that the parental materials studied had a lot of genetic variability. Due to the presence of large additive gene effects, this result usually favors the selection strategy of breeding^39,40. Days to first flowering, dry fiber weight and 1000-seed weight were found to indicate the prevalence of additive gene effects in the development of the features by Mostofa et al.²⁹. For seed yield per plant and number of pods per plant in kenaf, Mukewar et al.³¹ found that additive gene action predominated. For yield components such as plant height, fresh and dry weight of bark, and usable stick in kenaf, Pace et al.³² found additive gene action was more relevant. All the analyzed variables (except top diameter and node number) had large mean square GCA values, indicating that the parental materials studied had a lot of genetic variability. All the examined parameters (excluding plant height, days to 50% flowering, fresh stem weight without leaves and pod) showed a very significant effect on GCA environment interaction, indicating that environmental variation influenced additive gene action. Furthermore, environments had a significant influence on how these characters changed¹. Using combining ability investigations, Mostofa et al.³⁰ and Youcai et al.⁴⁹ found similar results in Hibiscus cannabinus, Sobhan⁴³ in Hibiscus sabdariffa, and Khatun²³ in Corchorus capsularis.

Table 4 shows that except for plant height and seeds number per pod, genotypes and environment interaction were highly significant in the pooled quantitative data showed that the environment significantly impacted both parents and offspring for all attributes. We employed nine kenaf genotypes with a range of morpho-physiological and yield features. The mean squares of GCA were bigger than the mean squares of SCA for all traits except top diameter, nodes number, and fresh stem weight with leaves and pods in our investigation indicating that the parental materials studied had a lot of genetic variability. Due to the presence of large additive gene effects, this result usually favors the selection strategy of breeding^39,40. Days to first flowering, fibre weight per plant, and 1000-seed weight were found to indicate the prevalence of additive gene effects in the development of the features by Mostofa et al.²⁹. Days to first flowering and plant height were controlled by one dominant gene pair, while raw fibre yield was controlled by three (Thombre and Patil, 1985). For seed yield per plant and number of pods per plant in kenaf, Mukewar et al.³¹ found that additive gene action predominated. For yield components such as plant height, fresh and dry weight of bark, and usable stick in kenaf, Pace et al.³² found additive gene action was more relevant. When both general and specific combining ability effects are essential, Comstock et al. (1949) suggested using a reciprocal recurrent selection strategy to generate high yielding varieties. All the examined parameters (excluding plant height, days to 50% flowering, fresh stem weight without leaves and pod) showed a very significant effect on GCA environments, indicating that environmental variation influenced additive gene action. Furthermore, environments had a big influence on how these characters changed¹.

The findings revealed that environmental influences controlled the expression of kenaf phenotypic features. The mean of hybrids is higher than the parental mean for plant height, base diameter, core diameter, middle diameter, top diameter, number of nodes, fresh stem weight with leaves and pod, fresh stem weight without leaves and pod, dry stick weight, dry fiber weight, and pods number per plant. In comparison to their parents, hybrids have higher fiber and stick yields and a higher seed yield. Days to first flowering and days to 50% flowering of hybrids, on the other hand, were lower than the parental mean. However, the mean hybrid of 1000 seeds weight had a lower mean value than the parental mean, indicating that smaller seed size hybrids were preferable to parents, but that when the nodes number was larger than the parental mean, the fiber production per plant was reduced. Parent P₁ had the highest fiber weight per plant (26.71 g) among the parental lines, followed by P₄ (24.78 g), and P₈ (23.59 g). Similarly, the hybrids P₂ × P₃, P₄ × P₆, and P₅ × P₈ had much greater fiber yield (38.74 g, 28.87 g, and 32.81 g, respectively) than their respective parents. Parent P₃ produced the meanest stick weight per plant (123.24 g), followed by P₇ (96.27 g), and P₈ (95.06 g). The hybrids P₁ × P₄, P₅ × P₈, and P₇ × P₉ generated the highest mean stick weight per plant (152.93 g, 137.78 g, and 132.68 g, respectively), which was significantly higher than both parents. The cross combinations P₁ × P₄, P₄ × P₆, and P₇ × P₉ produced the most pods per plant (157, 172 and 217, respectively), far exceeding any of the parents, demonstrating the presence of transgressive segregation in the cross.

The results of combining ability analysis are used to select parents and crosses. However, the nature of gene action revealed that both additive and non-additive gene effects were crucial in governing the diverse characters of kenaf in this investigation, but non-additive gene action was shown to control most of the characters. Gupta and Singh¹⁶ state that non-additive gene activity affects plant height, basal stem diameter, fiber weight per plant, and stick weight per plant (H. sabdariffa). Non-additive gene action predominated for all other parameters except days to first flowering, dry fiber weight, and 1000 seed weight. Through pedigree and single seed descent methods of breeding, it is possible to improve traits with a predominance of additive genetic effects. When non-additive gene effects predominated, however, bi-parental mating and a recurrent selective breeding system would be an effective way to obtain hybrid diversity.

Parents' breeding potential is usually linked to their GCA effects. Parents having high GCA effects for certain qualities could be employed as donor parents in hybridization programs to increase these features. A low or negative combining ability effect reflects a plant's inability to pass on its genetic superiority to hybrids¹⁰. Positive values with the highest magnitudes have the most impact. On the other hand, the greatest negative values have the least impact⁴⁴. Additive genetic variance is a major contributor to the GCA component. As a result, each parent's GCA variation has a major impact on the parents' decisions. A good general combiner is a parent with a higher positive significant GCA effect³⁶. In this study, GCA's high value for the traits of interest was distributed across genotypes, showing that none of the genotypes used had the optimal combination of GCA values for the several characters of interest.

Parent P₄ (ML36-24) was the best parental line in terms of base diameter, core diameter, middle diameter, fresh stem weight with leaves and pod, fresh stem weight without leaves and pods, dry stick weight, and dry fiber weight content, showing the accumulation of favorable additive genes for these traits in the hybrid. The parent P₃ had positive effects on plant height, fresh stem weight without leaves and pods, dry stick weight, and dry fiber weight, conversely the parent P₁ was the good general combiner for days to first flowering and dry fiber weight, and it contributed positively to the hybrid for these traits. The major goal of this breeding program is to develop high-yielding hybrids with high potential fiber production as compared to existing cultivars or on par with them. As a result, for fiber yield per plant, parent P₄ might be well combined. Furthermore, the parent P₃ had positive effects on plant height, fresh stem weight without leaves and pods, dry stick weight, and dry fiber weight, and the parent P₁ was the best general combiner for days to first flowering, dry fiber weight, and number of pods per plant. Other parents are specialized in one or two characteristics. To put it briefly, the parental lines P₁, P₃, and P₄ were determined to be outstanding general combiners for fiber yield and yield-related parameters. Parents P₃ and P₄ were good general combiners in terms of stick yield. The parents P₁, P₆ and P₇, on the other hand, were good general combiners for seed yield (pods number per plant). In contrast, parent P₂ was chosen as a good general combiner for the first flowering and days to 50% flowering features, which are important for late maturity cultivars. Golam et al.¹⁴ found that 50% flowering and days to maturity, in addition to other morpho-agronomic features, may be the two most important variables in classifying kenaf accessions.

Parents with a high GCA for a certain trait and high adaptability indicate additive gene action. Because additive variance may be fixed, selecting for qualities regulated by additive variance is a very effective strategy³⁵. The parents P₃ and P₄ were found to be the finest general combiners, with extremely significant values for fresh stem weight without leaves and pod, dry stick weight, and dry fiber weight, all of which are more suitable for desired kenaf features such as bast fiber and core fiber production. The parents with the highest GCA values (strong GCA effects) could be used to improve the kenaf population in Malaysia through varietal development based on desirable features.

In general, crosses with high × low general combiners for yield components outperform others. According to an investigation of combining ability impacts, high specific combiners involved high × high, high × low, high × average, average × average, average × low, and low × low combining parents. In crosses with high × low and low × low general combiners, Jinks²¹ described severe SCA effects caused by over-dominance and epistasis. In crosses involving high vs. low general combiners for yield components, mutual cancellation of heterosis components, especially dominance and its interaction, resulting in unfavorable SCA effects¹⁷. Crossing two parents with low general combiners produces high performance, attributable to complementary gene activity²⁷. This study, SCA effects were shown to be significant for most yield characteristics. For base diameter, core diameter, fresh stem weight with leaves and pod, and fresh stem weight without leaves and pod, the best crossings suggested by SCA effects (Table 8) were P₂ × P₃, P₄ × P₆, and P₅ × P₈.

In contrast, the hybrids P₂ × P₃ and P₅ × P₈ were shown to be the best specific combiners for fiber yield, with high base diameter, core diameter, fresh stem weight with leaves and pod, and fresh stem weight without leaves and pod features. Furthermore, for stick yield and seed yield (plant number of pods), the hybrids P₁ × P₄ and P₇ × were chosen as the best specific combiners. The hybrid P₂ × P₅ will be the best choice for 1st flowering and days to 50% flowering attributes in late maturity cultivars. In terms of another feature, the rest of the hybrids performed better. The strongest positive estimations of mid-parent and better parent heterosis for the qualities examined, indicating the accumulation of favorable genes inherited from their parental inbred lines. Among the hybrids produced, P₂ (ML9) × P₃ (ML36-10), P₅ (ML36-25) × P₈ (MLRing4P2), P₄ (ML36-24) × P₆ (ML36-27), P₁ (ML5) × P₄ (ML36-24), P₇ (BJRI Kenaf4) × P₉ [ML36-21(2)], P₂ (ML9) × P₅ (ML36-25), P₂ (ML9) × P₉ [ML36-21(2)] and P₁ (ML5) × P₅ (ML36-25) had higher yield and yield component qualities indicated by hybrids evaluated based on pooled data from the two environments.

The hybrids P₁ × P₄, P₄ × P₆ and P₇ × P₉ were chosen as the best specific combiners for dry stick weight and pods number per plant. Days to first flowering and days to 50% flowering for the hybrid P₂ × P₅ were also chosen as unique combiners. The best specific combiners for dry fiber weight and seeds number per pod were hybrid P₁ × P₃. For pods per plant and seeds per pod, the hybrids P₁ × P₄ and P₅ × P₇ were selected as good specific combiners. For another required attribute 1000 seed weight, the hybrids P₁ × P₅, P₃ × P₈, P₄ × P₇, and P₇ × P₈ were chosen as good specific combiners (smaller seed size). The data revealed that the GCA effects of parents were linked to the SCA impacts of their crossings, which had the greatest significant positive intensity. Complementing gene effects could explain the strong SCA effects of these crosses. Hybrid vigor can be induced by dominant, over-dominant, or epistatic gene action in any combination of parents, according to Moll and Stuber²⁸. In this study, both additive and non-additive genetic components influenced morphological and yield-related traits, with non-additive gene activity dominating most of the characters.

Table 9 shows estimates of mid-parent and better parent heterosis for yield and yield component qualities indicated by hybrids evaluated based on pooled data from the two environments. Liu²⁵ reported heterosis in yield characteristics based on the mid-parent and better parent. Heterosis estimates based on mid parental values were generally high, with stalk dry weight and bast percentage ranging from 10 to 55%⁴. The strongest positive estimations of mid-parent and better parent heterosis for the qualities examined were derived from P₄ × P₆ based on the combined data of the two environments (108.72% and 76.64%, respectively, for the pods number per plant). Hybrid P₂ × P₃ showed the greatest mid-parent and better parent heterosis for 4 of the 15 phenotypic parameters measured, including base diameter, core diameter, fresh stem weight with leaves and pods, and dry fiber weight, based on the combined data from the two environments.

In addition, hybrid P₅ × P₈ had the second highest mid-parent and better parent heterosis for base diameter, core diameter, middle diameter, fresh stem weight with leaves and pods, fresh stem weight without leaves and pods, and dry fiber weight, as well as P₄ × P₆ for base diameter, core diameter, fresh stem weight with leaves and pods, and pods number per plant. In terms of the features indicated, these hybrids outperformed their inbred parents. Hybrid P₇ × P₉ showed the better mid-parent and better parent heterosis for base diameter, core diameter, fresh stem weight with leaves and pods, and pods number per plant based on the combined data. Furthermore, the hybrid P₁ × P₄ was chosen as the best combiners for stick yield and pods number per plant, whereas hybrid P₄ × P₉ for stick yield and fiber yield showed the greatest mid-parent and better parent heterosis. In late maturity cultivars, the hybrid P₁ × P₉ will be the best choice for 1st flowering and days to 50% flowering features.

In contrast, hybrid P₂ × P₉ for Plant height, base diameter, core diameter, and fresh stem weight with leaves and pods showed low mid-parent and better parent heterosis estimates, followed by P₁ × P₅ for middle diameter, fresh stem weight without leaves and pods, and dry fiber weight, where the high values were unfavorable. Higher heterosis values compared to the better parent and the mid-parent suggested the absence of epistasis and the frequency of partial or total dominance of genes for fiber and seed yield. Crosses that produce superior transgressive segregants with higher fiber production could be found by looking at the percent F₁ heterosis over the high parent.

The strongest positive estimations of mid-parent and better parent heterosis for the qualities examined, indicating the accumulation of favorable genes inherited from their parental inbred lines. Among the hybrids produced, P₂ (ML9) × P₃ (ML36-10), P₅ (ML36-25) × P₈ (MLRing4P2), P₄ (ML36-24) × P₆ (ML36-27), P₁ (ML5) × P₄ (ML36-24), P₇ (BJRI Kenaf4) × P₉ [ML36-21(2)], P₂ (ML9) × P₅ (ML36-25), P₂ (ML9) × P₉ [ML36-21(2)] and P₁ (ML5) × P₅ (ML36-25) had higher yield and yield component qualities indicated by hybrids evaluated based on pooled data from the two environments. Both additive and non-additive genetic components were essential in the control of many morphological and yield-related characters, with non-additive gene activities predominating for most of the characters.

Conclusion

Additive and non-additive variations played a role in the genetic control of all variables in this study, including fiber yield and yield-related traits. New high-yielding kenaf hybrid types could be released via diallel selective mating or mass selection with simultaneous random mating. GCA effects were higher than SCA effects, except for top diameter, node number, and fresh stem weight with leaves and pod, as demonstrated by mean squares, showing that additive gene action predominates for these traits. Fiber yield and seed yield (number of pods per plant) are two important characteristics of kenaf production. Parent P₄ (ML36-24) was the best general combiner for fiber yield (bast fiber) based on GCA performance, followed by P₁ (ML5) and P₃ (ML36-10). Conversely, P₄ (ML36-24) was the best general combiner for stick yield (core fiber), followed by P₃ (ML36-10). For the number of pods per plant (seed yield), P₇ (BJRI Kenaf4) was shown to be the best general combiner, followed by P₆ (ML36-27) and P₁ (ML5). For days to 1st flowering and days to 50% flowering qualities, P₂ (ML9) was chosen as an excellent general combiner. In a breeding programme, the hybrids P₂ (ML9) × P₃ (ML36-10), P₄ (ML36-24) × P₆ (ML36-27), and P₅ (ML36-25) × P₈ (MLRing4P2) will be the best for bast fiber and P₁ (ML5) × P₄ (ML36-24), P₅ (ML36-25) × P₈ (MLRing4P2), and P₇ (BJRI Kenaf4) × P₉ [ML36-21(2)] will be the best for core fiber if fiber yield is the most significant selection factor. If seed yield is important in the breeding programme, the hybrids P₁ (ML5) × P₄ (ML36-24), P₄ (ML36-24) × P₆ (ML36-27), and P₇ (BJRI Kenaf4) × P₉ [ML36-21(2)] will be the best choices for number of pods per plant, and the hybrid P₂ (ML9) × P₅ (ML36-25) will be the best choice for days to first flowering and days to 50% flowering traits for late maturity cultivars. Crossings of P₁ (ML5) × P₄ (ML36-24), P₁ (ML5) × P₉ [ML36-21(2)], P₂ (ML9) × P₃ (ML36-10), P₂ (ML9) × P₅ (ML36-25), P₄ (ML36-24) × P₆ (ML36-27), P₄ (ML36-24) × P₇ (BJRI Kenaf4), P₄ (ML36-24) × P₉ [ML36-21(2)], P₅ (ML36-25) × P₈ (MLRing4P2), and P₇ (BJRI Kenaf4) × P₉ [ML36-21(2)] showed promising heterotic responses and could benefit future breeding programs.