Assessment of genetic variability , heterosis and heritability for morphological parameters in rice

This study was performed to estimate genetic variability and heterosis in rice genotypes using fourteen parents and their ten F1 hybrids in a randomized complete block design (RCBD) using two replications at The University of Agriculture, Peshawar-Pakistan. Significant differences among the parents and F1 hybrids were observed for all the studied traits. The parental genotypes Sadahayat, Dokri-Bas and Khushboo-95 exhibited maximum mean values for panicle length (36.6 cm), spikelets panicle (264.0) and grains panicle 1 (202.3), respectively while IR-8 displayed maximum mean values for 1000-grain weight (28.7 g) and grain yield plant (50.8 g). Among F1 cross combinations, DR-83/Sugdesi displayed the maximum spikelets panicle (249.1), grains panicle (201.7) and grain yield plant (41.8 g) while Pakhal/Kashmir-Bas displayed maximum 1000-grain weight (27.5 g). Among the F1 hybrids, Pakhal/Kashmir-Bas manifested maximum significantly positive mid and high parent heterosis for spikelets panicle (40.8 and 25.4 %), 1000-grain weight (22.7 and 14.9 %) while DR-83/Sugdesi exhibited significant positive mid and high parent heterosis for grains panicle (66.7 and 18.9 %) and grain yield plant (73.8 and 50.4 %). High phenotypic and genotypic coefficient of variation (PCV and GCV) values were observed for grains panicle 1 and grain yield plant. High heritability values were recorded for all the studied traits. The genetic potential of Sadahayat, Khushboo-95, IR-8 and Dokri-Bas for yield and yield associated traits can be exploited in future rice breeding program. The F1 hybrids viz. DR-83/Sugdesi and Pakhal/Kashmir-Bas on account of their better performance for yield and yield associated traits could be further studied in segregating generations for development of new rice cultivars.


Introduction
Rice (Oryza sativa L.) is the main staple food for about half of the human population and occupies the pivotal place in global food and livelihood security systems.It provides 55% of the protein and 75% of the calories in the average daily diet of the people.It is globally planted on 164.1 million hectares with total production of 722.5 million tones.China (202.7 million tons), India (155.7 million tons), Indonesia (65.7 million tons), Bangladesh (50.6 million tons) and Vietnam (42.3 million tons) are the chief producing countries of rice [1].The world's population is increasing at a faster rate, particularly in rice consuming countries.Almost 810 million tonnes of rice will be required to meet the growing demand of this crop by the year 2025 [2].Rice is one of the major cereals crops of Pakistan.It is grown on 2.5 million hectares area of arable land with total production of 6.1 million tones [1].Rice occupies an important position in the economy of Pakistan.It adds a value of 6.4 % in agriculture and contributes 1.4 % to GDP of Pakistan [3].Heterosis breeding of rice is a phenomenon in which F1 of rice derived from different parents show superiority over their parents for various traits [4].Furthermore, it also enriches the crops like maize, sunflower and most of the vegetables with other desirable qualitative and quantitative traits [5].Heterosis breeding is one of the main tool of plant breeding for increasing productivity of crops [6].In rice heterosis was first observed by Jones in 1926 who observed an increase in culm number and grain yield of the rice hybrids over their parents.Based on the criteria used to judge the hybrid performance, heterosis is expressed as mid parent heterosis, better parent heterosis or heterobeltiosis and high or standard or commercial heterosis [7].Both positive and negative heterosis could be useful depending on the nature of a particular trait.Positive heterosis is considered desirable for yield and yield associated traits, while negative heterosis is desirable for optimum culm length, days to heading and maturity [4].Improvement in yield and yield associated traits mainly depends on magnitude and nature of genetic variability present in a population [8].The variations present among the genotypes of a population results due to diversity either in the genetic make-up of the genotypes of a population or in the environment in which they are grown.The existence of variation among the genotypes for the desired traits is essential for effective selection and wide adaptability.The amount of genetic variation is of paramount importance in a population for initiating a judicious breeding program [9].Genetic parameters such as genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV) are useful in detection the amount of variability present in the germplasm [10].Higher PCV values in comparison with the corresponding GCV values for a particular trait depicts that environmental variation constitutes a major portion of the total phenotypic variance [11].Differences between phenotypic and genotypic variance of low magnitude are indicators of additive gene action for traits expression [12].It is very difficult to judge whether observed variation for traits of concern are highly heritable or not.Moreover, knowledge of heritability is essential for selection based improvement as it indicates the extent of transmissibility of a character into future generations [13].Heritability is a good index of the transmission of traits from parents to offspring.The present study was, therefore, undertaken with the following objectives: a.To assess genetic variability for yield and yield associated traits in F1 rice hybrids and their parents.b.To study heterosis for yield and yield associated traits in F1. c.To estimate heritability for yield and yield associated traits.

Materials and methods
This research work was conducted at Plant Breeding and Genetics Research Farm, The University of Agriculture, Peshawar during rice growing season 2011.The germplasm studied comprised fourteen parents and ten F1 hybrids, derived from the crosses among these parents in the preceding rice crop growing season.The list of genotypes studied is given in (Table1).First nursery was raised and then 30 days old seedlings were transplanted into well puddled field.A randomized complete block design with two replications was used.Each rice genotype was planted in a two-rows plot with row length of 3 m while row to row distance of 30 cm and plant to plant distance of 15 cm was maintained.Recommended dose of fertilizer was applied to the experimental plot and irrigation was applied frequently.Data were recorded on panicle length, spikelets panicle - 1 , grains panicle -1 , 1000-grain weight and grain yield plant -1 using five randomly selected plants of each entry.Mid parent and high parent heterosis for each trait was determined using the following formula [18].

Mid parent heterosis (
Where F1 = Mean of hybrid for a particular trait, MP=Average mean of parents for a particular trait = (P1 + P2)/2, HP = Mean of high parent in the cross for a particular trait, P1 and P2 are the values of a specific trait of the respective parents.The significance of F1 hybrids vs. mid parent and high parent mean was determined by t-test [19] as follows:

Panicle length
Highly significant differences (p≤0.01)among the parental genotypes and F1 hybrids were observed for this trait (Table 2).Mean data among the parental genotypes showed that panicle length ranged from 23.8 to 36.6 cm.Among the parental genotypes, maximum value (36.6 cm) was exhibited by Sadahayat while the lowest value (23.8 4).Moderate PCV, GCV and high broad sense heritability values of 14.9, 13.9 % and 87.2 % were observed for panicle length, respectively (Table 5).

1000-grain weight
Mean squares revealed highly significant (p≤ 0.01) differences among the parents and F1 hybrids for 1000-grain weight (Table 2).Among the parental genotypes, mean values varied between 18.3 and 28.7 g.IR-8 displayed maximum value (28.7 g) while Sugdesi showed minimum value (18.3 g) for this trait.1000-grain weight among the F1 ranged from 19.7 to 27.5 g.The highest value (27.5 g) was observed for cross combination Pakhal/Kashmir while the lowest 1000-grain weight (19.7 g) was observed for Sadahayat/Malhar (Table 3).

Grain yield plant -1
Analysis of variance technique (ANOVA) displayed highly significant (p≤0.01)differences among the parents and F1 hybrids for grain yield plant -1 (Table 2).Mean values among the parental genotypes varied between 23.5 and 50.8 g.IR-8 displayed the highest grain yield plant -1 (50.8 g) while Kashmir-Bas showed the lowest (23.5 g).Among the F1 crosses mean values for grain yield ranged from 22.2 to 41.8 g.Maximum value (41.8 g) was recorded for DR-83/Sugdesi while minimum grain yield plant -1 was observed by Pakhal/Kashmir-Bas (22.2 g) (Table 3).
Data after compilation were subjected to the analysis of the variance technique (ANOVA) as suggested by [14].Least significance difference (LSD) test was used for means separation and comparison.Genotypic variances (Vg), phenotypic variances (Vp), phenotypic coefficient of variability (PCV), genotypic coefficient of variability (GCV) and broad sense heritability (h 2 B) were computed as per the method outlined by [15].Vg = [MSG -MSE / r], Vp = Vg + Ve, of genotypes, MSE= Mean squares of error, r= Number of replications, Vp= Phenotypic variances, Vg= Genotypic variances and X = Grand mean.PCV and GCV were classified either as low (less than 10), moderate (less than 20) and high (greater than 20) [16].Broad sense heritability (h 2 B) was expressed as the percentage of proportion of the genotypic variance (Vg) to the phenotypic variance (Vp).Heritability values were rated either as low (less than 0.30), moderate (0.30-0.60) or high (more than 0.60) [17].

21, 25]. [25]
Reported mid and high parent heterosis values of 34.4 and 15.6 %, respectively for this trait.[21] Observed mid and high parent heterosis values of 29.7 and 28.6 %, respectively for panicle length.Two F1 hybrids exhibited significantly positive mid parent heterosis while cross combination Pakhal/Kashmir-Bas manifested significantly positive high parent heterosis for spikelets panicle -1 .These results are compatible with the findings of [

21, 25]. [25]
Reported significant positive mid and high parent heterosis values of 56.3 and 48.7 %, respectively for spikelets panicle -1 .[21] Reported that spikelets panicle -1 manifested significant positive mid and high parent heterosis of 110.8 and 86.3 %, respectively.Cross combination DR-83/Sugdesi exhibited significantly positive mid and high parent heterosis for the trait grains panicle -1 .These results are in conformity with the findings of [21] who also reported significantly positive mid and high parent heterosis of 111.0 and 87.1 %, respectively.Four F1 hybrids displayed significant positive mid parent heterosis while one cross combination Pakhal/Kashmir-Bas manifested significant positive high parent heterosis for 1000-grain weight.[21, 26] reported similar results for this trait.[21] showed significantly positive mid and high parent heterosis of values 38.3 and 38.4%, respectively for 1000-grain weight.In the present study two cross combinations Sadahayat/Khushboo-95 and DR-83/Sugdesi exhibited significantly positive mid and high parent heterosis for grain yield.These results are supported with the findings of [