Rice is one of the most important crops in the world and plays a vital role in human feeding. It provides up to 50-80% of Asia's residents' calories, protein, and minerals (Anders et al. 2021). Currently, more than three million tons of paddy are produced annually in 19 provinces of Iran (Ahmadi et al. 2021), which shows the high importance of rice cultivation in the food, financial, and job security of people and Iranian farmers.

Iranian local rice varieties usually have favorable characteristics such as high quality of the cooked grain, early maturity, and stable production, However, they have low grain yield, sensitivity to lodging (due to tall plants and weak stems), and high vulnerability to rice blast disease. In Iran, methods based on crossing foreign cultivars with local populations and selecting the best progenies have a history of 50 years (Modarresi 2023). This path started with the crossbreeding of Mesbah, and Taichung imported cultivars with local cultivars such as Dom Sefid, Tarom Mahali, and Musa Tarom (Nematzadeh et al. 2000) and continues until now.

Crossing mating designs such as Line in Tester, North Carolina (1, 2, and 3) and Diallel can be very beneficial for selecting the best parents. Also, with the help of statistical methods, it is possible to know the additive or non-additive effects of the inheritance of different traits (Fasahat et al. 2016). The Line in Tester (Kempthorne 1957) is a widely used and robust method for estimating hybridization and selecting the best parents and their progeny for breeding using the pedigree method. By using information about the effect of genes (additive and non-additive), it is possible to select the appropriate traits and strategy to select the best progeny. Simultaneous use of general and specific combining ability as well as the degree of heterosis, will be helpful in estimating the response to selection in segregating generations (Fellahi et al. 2013).

In plant breeding, general combining ability (GCA) and specific combining ability (SCA) are used to measure the effects of additive and non-additive genes on the inheritance of traits (Fasahat et al. 2016). GCA is attributed to additive gene effects, while SCA is attributable to non-additive gene action, which may be due to dominance (Abera Muluneh et al. 2020; Karaağaç 2021). Heterosis, also known as hybrid vigor, is the phenomenon in which the offspring of two genetically different parents exhibit superior traits compared to their parents. The degree of heterosis is influenced by additive and non-additive gene effects (Sunny et al. 2022).

This research was carried out to form breeding populations, estimate the gene effect, evaluate the combinability of several rice genotypes, and estimate the heritability of traits in the form of a line×tester mating design.

This experiment was conducted at the Rice Research Institute of Iran, Guilan province (longitude 49°36~ and latitude 37°16~ with 7 m above seq level), in 2019-2020. In this study, we have used five high-quality and low-yielding local Iranian genotypes, Deylamani, Hasani, Hashemi, Binam, and TH1, as maternal parents (lines). In addition, we have used three improved genotypes with suitable yield and low-cooking quality, named Gohar, Saleh, and IR75479-199-3-3, as paternal parents (testers) (Table 1).

Table 1 . Characteristics of parental genotypes used in this research.

Genotypes	Type	Origin	Characteristics
Saleh (Khazar (IR2071-625-152×TNAU7456) ×IR39385-20-1-2-1-2)	Improved	Iran (parents from IRRI)	Suitable yield, high amylose content, low gelatinization temperature, low cooking quality, and non-fragrant
IR75479-199-3-3	Improved	Philippines (IRRI)	Suitable yield, lodging resistant, Intermediate amylose content, low cooking quality, and non-fragrant
Gohar (pusa1238-1 ×pusa 1238-81-6)	Improved	Iran (parents from India)	High yield, lodging resistant, blast disease resistant, late-maturing, intermediate to high amylose content, low gelatinization temperature, and fragrant
Hashemi	Landrace	Iran	Low yield, Grains with excellent market value (because of grain shape and taste quality), sensitive to lodging, early-maturing, intermediate amylose content, intermediate gelatinization temperature, and fragrant
Hasani	Landrace	Iran	Low yield, intermediate amylose content, intermediate gelatinization temperature, sensitive to lodging, and fragrant
TH1	Landrace	Iran	Low yield, intermediate amylose content, intermediate gelatinization temperature, sensitive to lodging, and fragrant
Deylamani	Landrace	Iran	Low yield, Grains with good market value (because of grain shape and taste quality), early-maturing, intermediate amylose content, intermediate gelatinization temperature, sensitive to lodging, and fragrant
Binam	Landrace	Iran	Low yield, intermediate amylose content, intermediate gelatinization temperature, sensitive to lodging, and fragrant

The parents were selected based on existing molecular, morphological, and physicochemical data (Allahgholipour 2014; Jahani 2013; Modarresi 2022). Therefore, improved genotypes with different genetic origins, Philippines (Saleh and IR75479-199-3-3) and India's Basmati (Gohar), were used alongside lines with Iranian rice genetic origin. In spring 2019, eight genotypes were sown at three sowing dates with 15-day intervals. Then seedlings were cultured as single plants at the 3-4 leaf stage. At the heading date, improved cultivars were crossed with local cultivars through line×tester mating design, and the hybrid seeds were harvested.

In the spring of 2020, the seeds of fifteen F1 progenies and eight parents (23 genotypes) were sown in the greenhouse and transferred to the field at the 3-4 leaf stage. Studied genotypes were cultivated in a randomized complete block design with three replications. Each genotype grown in a four-row plot had a row length of 2 m and a distance of 25 x 25 cm. The distance between plots and replications was 0.5 and 1 meter, respectively. All evaluations were done on 10 plants from each plot that was randomly selected (after removing border plants). Finally, an evaluation of traits (Plant height, Panicle length, Number of panicles per plant, Days to heading, Grains per panicle, hundred grain weight and Grain yield per plant) was done. All these measurements were performed based on the Standard Evaluation System for Rice (S.E.S. 2002). Meteorological data related to the rice growing season is given in Table 2.

Table 2 . Meteorological data related to rice growing season which has been recorded in Rasht agricultural meteorological station (near the research farm).

Date	Average temperature (°C)	Precipitation (mm)	Total sunny hours (h)
	Max	Min
April	16.8	9.9	131	101
May	24.6	13.8	65	223
June	30.5	18.8	10	306
July	31.0	21.4	171	253
August	31.4	20.4	25	227
Mean	26.9	16.9	80	222

After analysis of the variance of the traits among the parents, the estimation of the combining ability for traits with significant differences was done. In addition, their genetic effects were estimated using Line×Tester analysis by the Kempthorne method(Kempthorne 1957). Equations (1) and (2) were used to estimate general combining abilities (GCA) through lines and testers, respectively. Meanwhile, equation (3) was used to estimate specific combining abilities (SCA). In addition, the variance of general (σgca2) and specific combining abilities (σsca2) were calculated. Since in the present study, the parents of the crosses were pure lines, so their coefficient of inbreeding (F) was considered one. The values of additive (σA2) and dominant (σD2) variances were determined through relations 4 and 5 respectively. The t-test was used to investigate the significance of the effects of general and specific combining of lines and testers. Also, the relative contribution of lines (relationship 6), testers (relationship 7) and the lines in testers (relationship 8) were also calculated. In addition, broad-sense heritability (relationship 9) and narrow-sense heritability (relationship 10) were also calculated (Ghorbani et al. 2013). All calculations were done by writing formulas using R language (de Mendiburu 2021) or using Microsoft Excel 2019 software.

(1) GCALines=gi=Xi…tr−X…ltr

(2) GCATesters=gt=X.j..lr−X…ltr

(3) SCA=Sij=Xij.r−Xi..tr−X.j.lr+X…ltr,∑iSij=∑jSij=∑i∑jSij=0

(4) σgca2=covHS=1+F4σA2

(5) σsca2=MSL×T−MSer=1+F22σD2

(6) Lines  contribute=SSlSScross×100

(7) Testers  contribute=SStSScross×100

(8) Lines  ×  Testers  contribute=SStlSScross×100

(9) H2=vA+vDvA+vD+vE×100

(10) h2=vArA+vD+vE×100

The analysis of the variance of traits among parental genotypes (Table 3) showed that the genotype effect, which indicates the difference among lines, testers, and hybrids, was significant at the 1% level for all studied traits. It indicates the presence of suitable genetic diversity between parents and the results of crosses. Therefore, it became possible to analyze the line in tester and separate the genetic variance into its components, general and specific combining abilities. So, the mean square of the genotypes was broken into its components (mean square of parents, crosses, and parents vs. crosses). The mean square of crosses was also broken into its components, including the mean square of lines, testers and lines in testers. The variation range of different traits is presented in Table 4, which indicates the existence of high diversity in the experimental genotypes.

Table 3 . Analysis of variance from line×testers mating design for the investigated characters.

S.O.V	df	Grain yield plant-1	Hundred grain weight	Grains per panicle	Days to heading	Number of panicles plant-1	Panicle length	Plant height
Replication	2	1.925*	0.003ns	64.701**	1.206ns	2.453**	8.858ns	86.543*
Genotypes	22	265.312**	0.171**	539.367**	59.178**	76.417**	28.021**	1086.759**
Parents	7	96.640**	0.319**	492.928**	92.137**	27.279**	12.994**	1121.708**
Pa. vs Crosses	1	823.692**	0.059**	35.353ns	90.145**	84.324**	180.928**	9397.234**
Crosses	14	309.764**	0.104**	598.587**	40.487**	100.421**	24.613**	475.679**
Lines	4	300.887ns	0.037ns	625.490ns	65.210ns	57.823ns	26.743ns	384.551*
Testers	2	913.172*	0.318ns	2023.466**	63.301ns	148.556ns	56.409ns	2303.081**
Lines×Testers	8	163.351**	0.085**	228.915**	22.422**	109.687**	15.599**	64.393**
Error	44	1.660	0.007	55.525	1.907	4.819	3.535	20.901
C.V. (percent)		4.022	3.123	6.600	1.576	11.038	6.301	3.017

ns, * and ** are not significant, significant at 0.05 and 0.01 probability, respectively.

Table 4 . Descriptive statistics of evaluated traits in considered genotypes.

Traits	Range	Minimum	Maximum	Mean	Standard error
Plant height (cm)	72	113	185	151.522	2.308
Panicle length (cm)	17	22	39	29.842	0.410
Number of panicles plant-1	23	12	35	19.889	0.636
Days to heading)	18	77	95	87.607	0.544
Grains per panicle	64	85	149	112.897	1.754
Hundred grain weight (g)	1.15	2.2	3.35	2.702	0.030
Grain yield plant-1 (g)	48	19	67	32.031	1

The difference of parents' traits and the results (Table 5) showed that, in general, hybrid plants were 2.8 days later flowering. They had 24.5 cm more plant height, 3.5 cm longer panicle length, four more fertile tillers, 1.5 more seeds per panicle, 0.06 grams more weight of one hundred seeds and 7.25 grams per plant grain yield. Therefore, hybrids had numerical superiority in these traits.

Table 5 . Mean performance for yield and yield components of the parents and their crosses.

Genotypes	Grain yield plant-1 (g)	Hundred grain weight (g)	Grains per panicle	Days to heading	Number of panicles plant-1	Panicle length (cm)	Plant height (cm)
Hashemi×Saleh	40.867	2.767	117	91	13.333	33.333	162
Hasani×Saleh	34	2.683	104.667	89.333	24.667	27	142.667
TH1×Saleh	32.800	2.587	124.667	89.667	17	31.567	157.333
Deylamani×Saleh	32.667	2.233	126	85	13.333	28	149
Binam×Saleh	34.361	2.523	118.028	87.583	17	29.325	152.250
Hasani×Gohar	40.333	2.791	110.962	89.453	16.943	31.893	169.433
TH1×Gohar	31.787	2.848	128.637	89.920	25.164	33.989	176.648
Hashemi×Gohar	64	2.700	114.667	91	35	31.333	185
Deylamani×Gohar	33	3.033	144.333	90.667	16.667	39	170.667
Binam×Gohar	41.667	2.800	111.333	93	18	32	170
Hashemi×IR75479-199-3-3	25	2.900	112.667	89	19.333	30	162.333
Hasani×IR75479-199-3-3	25.633	2.823	95.667	80.333	27.333	26	142
TH1×IR75479-199-3-3	28.467	2.767	91.667	85	23.333	33.333	146.667
Deylamani×IR75479-199-3-3	19.400	2.600	106.667	90.667	21.333	29.667	158
Binam×IR75479-199-3-3	34.333	2.800	94.333	92	22	30.833	156.667
Saleh	24.667	2.300	112.667	88	19.333	30	117.333
IR75479-199-3-3	26	2.343	115	91	16	26.667	118
Gohar	40.667	2.920	139	94.333	20.667	27	121.667
Hashemi	26.667	2.583	109.333	85.333	16.333	27.333	151.333
Hasani	26.200	3.283	109	77.333	16	23.667	124.667
TH1	21.867	2.567	113.333	80.667	12	28	141.667
Deylamani	24.667	2.500	102.333	83.333	13	28	137
Binam	27.667	2.800	94.667	88.333	20	30.333	172.667
LSD (5%)	2.120	0.139	12.262	2.273	3.612	3.094	7.523

Green and red show the highest and lowest values, respectively. Meanwhile, intermediate values are indicated with yellow and orange.

Results of this research showed the presence of a positive, and significant general combining ability to increase seed yield per plant, and Deylamani, Hasani and IR75479-199-3-3 parents had negative GCA for this trait. In addition to Gohar, Hashemi, and IR75479-199-3-3 genotypes had positive, and significant GCA for one hundred-grain weight. Gohar, Saleh and Deylamani genotypes had a positive, and significant combining ability for the number of grains per panicle. Meanwhile, Hasani, Hashemi, Gohar, and IR75479-199-3-3 parents had positive, and significant combining ability for the number of grains per panicle trait as well as Gohar, and TH1 parents for the panicle length trait.

In this study (Table 6), the hybrids of Binam×IR75479-199-3-3, Deylmani×Saleh, Hashemi×Gohar, and TH1×IR75479-199-3-3 had the positive and significant specific combining ability for grain yield traits. In addition, the hybrids of Binam×Gohar, Binam×Saleh, Hashemi×IR75479-199-3-3, Hashemi×Saleh, TH1×Gohar and Deylmani×Gohar had significant SCA to reduce grain yield. Moreover, Deylamani×Gohar and Hashemi×Saleh hybrids had positive and significant SCA for hundred-grain weight, and Deylamani×Saleh and Hashemi×Gohar hybrids had negative and significant SCA for the hundred-grain weight. In addition, Hashemi×Gohar, Hashemi× IR75479-199-3-3 and Deylamani×Gohar hybrids had positive and significant SCA for the number of grains per panicle trait, and TH1×IR75479-199-3-3 hybrid had negative and significant SCA for this trait. The Hasani× IR75479-199-3-3 hybrid was identified as the best promising combination for producing precocious lines, while having an early maturity and a significant combination of SCA and parents' GCA to reduce the days to heading. Hybrid TH1×IR75479-199-3-3 was identified as the best genetic combination with desirable alleles for reducing plant height due to the negative and significant SCA of the hybrid for plant height.

Table 6 . Specific combining ability values of crosses for studied traits.

Genotypes	Grain yield plant-1 (g)	Hundred-grain weight (g)	Grains per panicle	Days to heading	Number of panicles plant-1	Panicle length (cm)	Plant height (cm)
Binam×Gohar	-2.723**	-0.018	-5.132	-0.228	-2.659*	-0.958	-3.944
Binam×IR75479-199-3-3	5.534**	0.038	0.345	2.180**	1.029	1.172	3.939
Binam×Saleh	-2.811**	-0.020	5.477	-1.953*	1.629	-0.215	0.005
Deylamani×Gohar	-2.959**	0.300**	10.100*	-0.478	-2.103	3.273**	-2.861
Deylamani×IR75479-199-3-3	-0.968	-0.077	-5.780	2.930**	2.252	-0.864	5.689*
Deylamani×Saleh	3.927**	-0.224**	-4.319	-2.453**	-0.148	-2.409*	-2.828
Hasani×Gohar	-0.592	-0.086	-1.370	3.047**	-7.697	1.357	3.761
Hasani×IR75479-199-3-3	0.299	0.003	5.121	-2.665**	2.382	-1.240	-2.456
Hasani×Saleh	0.293	0.083	-3.751	-0.382	5.315**	-0.118	-1.306
Hashemi×Gohar	13.108**	-0.200**	8.678*	-1.700*	10.786**	-2.460*	0.917
Hashemi×IR75479-199-3-3	-10.301**	0.057	11.108*	-0.292	-5.193**	-0.497	-0.533
Hashemi×Saleh	-2.807**	0.143**	-2.430	1.992*	-5.593**	2.958**	-0.383
TH1×Gohar	-6.834**	0.003	5.080	-0.642	1.673	-1.212	2.127
TH1×IR75479-199-3-3	5.437**	-0.021	-10.104*	-2.154**	-0.470	1.429	-6.638*
TH1×Saleh	1.398	0.181	5.024	2.796**	-1.203	-0.216	4.512
S.E. (SCA)	0.744	0.0487	4.302	0.797	1.267	1.086	2.640

* and ** are significant at 0.05 and 0.01 probability, respectively

Examining the relative contribution of lines, testers and lines×testers interactions effect for the total variance in the studied genotypes showed that testers have the greatest effect on plant height, grains number per panicle and yield with 69.167%, 48.291% and 42.114%, respectively (Table 7).

Table 7 . The relative contributions of lines, testers, and their interaction with the total variance for the investigated characters.

Contribution	Grain yield plant-1 (g)	Hundred-grain weight (g)	Grains per panicle (no)	Days to heading (day)	Number of panicles plant-1	Panicle length (cm)	Plant height (cm)
Lines	27.753	10.227	29.856	38.805	16.452	31.044	23.098
Testers	42.114	43.407	48.291	15.095	21.133	32.741	69.167
Lines×Testers	30.134	46.366	21.853	46.101	62.415	36.216	7.735

In this research, the mean square of the parents was significant for all traits, which was expected considering the selection of these genotypes based on the variety of traits. The presence of high diversity among the offspring resulting from crossing is desirable and will increase the power of selecting superior plants for the breeder (Modarresi 2022). The mean square of the parents versus the cross was also significant for all traits except the number of grains per panicle. This significance means the presence of positive or negative heterosis in the progeny and the difference between the mean of offspring and parents for these traits (Ghorbani et al. 2013).

The significance of the mean square of the general combining ability of lines and testers (Table 8) indicates the more outstanding contribution of additive variance, and the significance of the mean square of the specific combining ability (Table 6) of lines in testers indicates the more outstanding contribution of non-additive variance (such as dominance or epistasis) (Karademir et al. 2009).

Table 8 . General combining ability values of parents for studied traits.

Parents	Grain yield plant-1	Hundred-grain weight	Grains per panicle	Days to heading	Number of panicles plant-1	Panicle length	Plant height
Lines
Binam	2.233**	-0.016ns	-5.521**	2.420**	-1.696*	0.305ns	0.405ns
Deylamani	-6.199**	-0.101**	12.247**	0.336ns	-3.585**	0.564ns	-0.822ns
Hasani	-1.232**	0.042ns	-9.654**	-4.402**	2.285**	-2.727**	-8.678**
Hashemi	8.735**	0.065*	1.358ns	1.892**	1.859*	0.531ns	9.733**
TH1	-3.537**	0.010ns	1.570ns	-0.246ns	1.137ns	1.938**	0.172ns
S.E. (Lines)	0.429	0.028	2.484	0.460	0.731	0.626	1.523
S.E. (d) gi – gj (lines)	0.607	0.039	3.512	0.651	1.034	0.886	2.155
Testers
Gohar	7.603**	0.111**	8.567**	2.366**	1.659**	2.238**	14.305**
IR75479-199-3-3	-7.988**	0.054*	-13.220**	-1.042**	1.971**	-1.058*	-6.911**
Saleh	0.385ns	-0.165**	4.653*	-1.325**	-3.629**	-1.180*	-7.394**
S.E. (Testers)	0.332	0.021	1.924	0.356	0.566	0.485	1.180
S.E. (d) gi - gj (testers)	0.470	0.031	2.721	0.504	0.801	0.686	1.669

ns, *, and ** are not significant, significant at 0.05 and 0.01 probability, respectively.

The parents used in the crosses had a diverse range of the desired characteristics, but this does not mean the definite transfer of these traits to the progeny. Parents may not be able to transfer the desired traits to their progeny. Therefore, concepts such as estimating the effects of general (GCA) and specific (SCA) combining ability are proposed (Fellahi et al. 2013). The success of a breeding program depends on the parents' combining ability, amount of heritability and their ability to transfer of traits characteristics to the progeny (Suvi et al. 2020). Results of this research showed the presence of significant GCA (Table 8) and SCA (Table 6) effects for all traits, which indicates the role of additive and non-additive gene action in the inheritance of the examined traits.

In this research, Gohar had a positive GCA for all investigated traits. The existence of a parent with a positive general combining ability for yield and its components has been reported before (Gramaje et al. 2020). However, in many studies, such a parent is not observed (Huang et al. 2015; Yuga 2018).

Due to the limitation of irrigation water, and the prevention of pesticide use, early to mid-maturing plants that escape from drought, and the third generation of stemborers are preferred. Meanwhile, for the days-to-heading trait, parents with negative combining ability are superior (Azad et al. 2022). In this study, the parents Hasani, IR75479-199-3-3, and Saleh were identified as parents with the negative and significant general combining ability for the day-to-heading trait. Also, Gohar, Hashemi, and Binam were not favorable for this trait and had positive and significant GCA. Hasani, IR75479-199-3-3, and Saleh's parents had significant, and negative GCA for plant height trait. Due to the desirability of semi-dwarf plants compared to tall plants, and the possibility of these plants being more resistant to the lodging problem at the end of the season, these parents are preferred (Aamer et al. 2020). For this trait, Hashemi, and Gohar had positive, and significant general combining ability (Table 8).

General combining ability is directly related to the breeding value of parents and additive genetic effects, however specific combining ability is related to non-additive effects such as dominance, epistasis, and genotype-by-environment interaction (Rukundo et al. 2017). The specific combining ability is usually not given much attention for self-pollinated plants, unless the goal is to create the best hybrids and use the ability of the heterosis phenomenon. High values of SCA of a cross indicate the more significant influence of the non-additive gene effect in controlling the trait (Suvi et al. 2020). However, the best hybrids are expected to have the potential to produce proper homozygous lines due to transgressive segregation (Rajput et al. 2018). It has been suggested that genetic combinations with a high trait average and desirable SCA estimate, where at least one of their parents has a high GCA estimate, can improve the target trait by increasing the frequency of desirable alleles (Fellahi et al. 2013). Therefore, this research identifies the Gohar×Hashemi genotype as the best-expected hybrid for improving seed yield. Our observations in the fourth generation of these crosses also show that Gohar×Hashemi progenies have a high yield (unpublished).

Examining the relative contribution of lines, testers and lines×testers interactions effect for the total variance in the studied genotypes presented in Table 7. A similar result regarding the greater effect of the male parent (tester) on the height of the plant (Bagheri et al. 2010; Ghorbani et al. 2013; Montazeri et al. 2014), grains number per panicle (Fellahi et al. 2013) and yield (Baloch Zehi et al. 2015; Ghorbani et al. 2013) have been reported by other researchers. However, some researchers reported the greater effect of the maternal parent (line) on plant height (Faiz et al. 2006) and the lines×testers interactions on plant yield (Devi et al. 2015; Gaballah et al. 2022). The reason for this is the difference in the chosen parents and the complexity of the heritability of some traits, such as yield.

Due to the significance of the mean square of lines, testers and lines×testers, the plant height trait is simultaneously affected by additive and non-additive effects (Table 3). The ratio of the variance of GCA to the variance of SCA for the plant height trait was calculated as 1.637 (Table 9). Therefore, this trait is more affected by additive effects and the selection among the diverging generations will be successful for the height of the plant. The low degree of dominance, the higher values of additive variance compared to the variance of dominance and environment, the high narrow-sense heritability and the low difference between broad-sense and narrow-sense heritability also confirm this. Similar results have been reported by other researchers (Chakraborty et al. 2009; Zewdu 2020). However, regarding the other studied traits (panicle length, number of panicles per plant, days to heading, number of grains per panicles, hundred-grain weight, and grain yield), the ratio of the variance of general combining ability to the variance of specific combining ability was calculated to be less than one (Table 9). Therefore, these traits influenced mainly by non-additive effects and the selection among the diverging generations will not be successful. The lower values of the additive variance compared to the dominance variance also confirm this. High broad-sense heritability and low narrow-sense heritability indicate that these traits are under the control of genetic effects, but additive effects play a smaller role in controlling these traits compared to non-additive effects (such as dominance and epistasis). Therefore, methods such as the use of hybrid cultivars or selection in advanced generations can be used to improve these traits. Similar results have been reported by other researchers (Ambikabathy et al. 2019; Saidaiah et al. 2010; Srijan et al. 2016).

Table 9 . Estimates of genetic components and heritability of traits in the studied genotypes.

Parameters	Grain yield plant-1 (g)	Hundred-grain weight (g)	Grains per panicle (no)	Days to heading (day)	Number of panicles plant-1	Panicle length (cm)	Plant height (cm)
σGCA2	31.057	0.004	78.415	3.832	0.100	1.912	87.243
σA2	124.229	0.017	313.660	15.328	0.200	7.648	348.970
σD2	215.738	0.102	231.638	26.379	140.943	16.256	53.308
σE2	0.516	0.003	18.396	0.879	1.327	1.136	8.137
hbs2	0.998	0.977	0.967	0.979	0.991	0.955	0.980
hns2	0.365	0.138	0.556	0.360	0.043	0.305	0.850
σSCA2	215.738	0.102	231.638	26.379	140.943	16.256	53.308
σGCA2/σSCA2	0.144	0.041	0.339	0.145	0.001	0.118	1.637
σD2/σA21/2	1.318	2.470	0.859	1.312	37.542	1.458	0.076

σGCA2 : General combining ability variance, σA2 : Additive genetic variance, σD2 : Dominance genetic variance, σE2 : Environment variance, hbs2 : Broad sense heritability, hns2 : Narrow sense heritability, σSCA2 : Specific combining ability variance, σGCA2/σSCA2 : Ratio of variance of general to specific combining ability, and σD2/σA21/2 : Degree of dominance.

Correlation between different traits (Fig. 1) shows a positive and significant correlation between plant height and panicle length, number of panicles per plant, days to heading, hundred grain weight and plant yield. Other researchers have also reported a positive and significant correlation between plant height and panicle length (Sari et al. 2019), yield (Bhutta et al. 2019), thousand-grain weight and panicle length (Islam et al. 2020). Also, the results show a positive and significant correlation of yield with plant height, panicle length, number of grains per panicle and days to heading. Other researchers have also reported positive and significant correlation of yield with plant height and days to heading (Nath et al. 2021) and panicle length (Reddy et al. 2008).

Figure 1. Correlation coefficients between traits in studied genotypes. Blue colour indicates positive correlation, red colour indicates negative correlation, Multiplication indicates non-significance at the 5% probability level.

The findings of this research showed that the dominance variance was higher for most of the traits compared to the additive variance. The results showed that the contribution of the additive effect was greater than the non-additive effects only in the plant height trait, so for this trait, selection in the diverging generations will be useful. Based on the findings of this research, Hasani and Saleh parents were the best general combining ability for early heading. In addition, Gohar variety hybrids had high SCA for yield components, and Gohar× Hashemi genotype can be identified as the best expected hybrid for improving grain yield. Considering the predominance of non-additive effects in traits related to performance and its components, it is suggested to postpone the selection of desirable plants to advanced generations (Gramaje et al. 2020).

This work was supported by Rice research institute of Iran. The authors thank Mr. Arman Babakhani, Mr. Mohammad Mohammadi and Mr. Mahmoud Sayyadi-Fatmehsari for their assistance and companionship.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.