Genotypic and Phenotypic Variability for Yield and its Components in Normal and Late Sown Chickpea (Cicer arietinum L.)

Chickpea (Cicer arietinum L.) is an important, semi-arid food legume crop.While in the developed world it represents a valuable crop for export, in the developing world it provides a protein-rich supplement to cereal-based diets. The principal uses of chickpea are as good source of protein (20 to 30%) as compared with cereals (8 to 10%), about 40% carbohydrates and 3 to 6% oil (Jukanti et al., 2012). It is also a good source of calcium, magnesium, potassium, phosphorous, iron, zinc and manganese (Ibrikci et al., 2003). In Turkey, the area under chickpea is 52 thousand hectares with the productivity of 1220 kg ha^-1 and production of 630 thousand tones (TUIK, 2018).
 
The total yield production is quite low in most chickpea growing countries and a wide gap exists between the potential (5 ton ha^-1) and actual (0.96 ton ha^-1) yields (FAOSTAT, 2013). This is due to biotic (wilt, root rot, blight diseases and weeds infestation) and abiotic (drought, high and low temperature) stresses. Drought is one of the most important abiotic stresses, which limits crop production. Chickpea is generally grown without irrigation, planted in marginal regions, and due to lack of rainfall during flowering, podding and seed filling time, terminal drought stress is the major environmental stress, which reducing chickpea production (Yucel, 2018).
 
Estimates of yield losses due to drought range from 15 to 60% depending on geographical region, duration of the crop season and dry spell (Sabaghpour et al., 2006).
 
In the last decade, the main breeding strategy used to cope with terminal drought in chickpea was selecting for drought escape by reducing crop duration and securing seed yield before soil water was depleted. Knowledge of the relative magnitude of various genetic parameters for seed yield and yield components is essential for an efficient breeding program. Genetic variation among traits is important for breeding and selecting desirable types. Direct selection only for higher yield could be misleading because many factors interact to determine crop yield. Separate yield components are less influenced by the environment than yield itself; hence, selection for such yield components can be useful to acquire genotypes with better yield abilities. Consequently, several researchers emphasized the usefulness of estimates of genetic components inpredicting the response of quantitative characters to selection of different traits to gain yield (Mohammed et al., 2019; Alemu et al., 2017). Heritability estimates are used to determine the amount of variation present in the population. Heritability combined with genetic advance will bring out the genetic gain expected from selection.
 
Therefore, the present study was carried out to estimate genetic variability, heritability and genetic advance for yield and yield components in chickpea genotypes under normal and drought stress conditions.

Experimental site
 
The present research was carried out at the Eastern Mediterranean Agricultural Research Institute, Adana, TURKEY. The site of experiment is located at 35° 18’E longitude, 37° 01’ N latitude and 23 m above the sea level. Air temperature, relative humidity and distribution of rainfall from December to July during the years of 2014-2016 are shown in Table 1. The soil is silty-clay; total content of salts was 0.026%, organic matter content 2.0%, pH level 7.72 and P level 87.0 kg and rich in term of lime.
 

 
Experimental materials and design
 
The experimental material consisted of 32 genotypes along with FLIP 87-59 C (Kanouni et al., 2012) as drought tolerant as well as ILC 8617 (Cancý and Toker, 2009) as drought susceptible. Thirty-four chickpea (Cicer arietinum L.) genotypes with their names, types and registration institution or origins were given in Table 2. The experiment was set up during the winter and late sowing conditions in 2015 and 2016, in randomized complete block design (RCBD) with three replications. Each plot was accommodated in two rows of 4 meters’ length, spaced 45 cm between row to row with approximate plant to plant distance of 10 cm. Sowing was performed in December 01, 2014 and December 21, 2015 for winter; March 4, 2015 and February 28, 2016 for late sowing. Drought stress condition was created by delaying sowing time. So, winter sowing was evaluated as normal condition, while late sowing was evaluated as drought stress condition.Fertilizer was applied at a rate of 30 kg nitrogen and 60 kg phosphorus per ha before sowing.
 

 
Traits evaluated
 
Observation was recorded for all traits under normal and drought stress conditions. The data for emergence day, vegetative period, flowering day, podding day, vegetation time and maturity day were recorded in each replication and each genotype was represented by single value. For other morphometric traits viz., plant height and first pod height from five randomly selected plants were taken from each genotype in each replication to record the data. After harvest, hundred seed weight was calculated by weighting of counted 100 seeds in four replications. Seed yields were calculated after harvesting collectively the 3.6 m² area and the data were transformed to kg ha^-1.
 
Statistical analysis
 
The data were analyzed for simple statistics, i.e. mean, min., max., standard deviation using the computer software program of SPSS. (Anonymous, 1983).
 
The mean values of the recorded data were subjected to analysis of variance. The mean squares were used to estimate genotypic and phenotypic variance. The variability present in genotypes were calculated as suggested by Burton (1952) as follows:
 
𝛔GY² = (GY MS-EMS)/r                                                              (1)
𝛔G²= (GMS-EMS)/yr                                                                   (2)
𝛔P²=𝛔G²+ 𝛔GY² /y + 𝛔E^2/ry                                                        (3)
𝛔E² = EMS                                                                                   (4)
 
Where,
GY MS= Genotype*Year mean square, GMS = genotype mean square, EMS= error mean square, y= number of year; r = number of replication.
 
According to Singh and Chaudhary (1999), the phenotypic (PCV) and genotypic coefficients of variances (GCV) are expressed by the following formula:
GCV (%) = (√ G² / X) ×100                                                          (5)
PCV (%) = (√ sP² /X) ×100                                                          (6)
 
Where,
GCV% = Genotypic Coefficient of variation; PCV% = Phenotypic Coefficient of variation; X = Mean value of the trait. 
 
GCV and PCV values were categorized as low (0-10%), moderate (10-20%) and high (20% and above) as indicated by Burton and De Vane (1953) and Sivasubramanian and Madhavamenon (1973).The broad sense heritability (h²b) of the traits was calculated (Hanson et al.,1956). Heritability in broad sense was calculated for each trait by using the following formula (Allard, 1960):
 
                                h²b (%) =(𝛔G²/ sP²) ×100                            (7)
 
 
The heritability estimates were categorized as low (0-30%), moderate (31-60%) and high (60% and above) as suggested by Robinson et al., (1949).
 
The expected Genetic Advance for each trait was calculated as
 
                                GA= K* √sP^2* h²b                      (8)
 
Where, K = 1.40 at 20% selection intensity for trait as suggested by Hanson et al., (1956).

The analysis of variance for investigated traitsindicated highly significant differences (P≤0.01) among the chickpea genotypes in both normal and stress conditions (Table 3). This significant variation suggests the existence of inherent genetic variability among the genotypes. Genetic variation in any given crop population is essential to successfully select and manage yield improvement programs. Similar results have also reported by Meena and Kumar (2015); Kadir et al., (2017); Yucel, (2018).
 

 
The phenotypic variance (PV) of the traits in this research was divided into genotypic, genotype by environment interaction and error variances in both growing conditions (Table 4). The highest genotypic (GV) and phenotypic (PV) variance were obtained from seed yield and hundred seed weight. The genotypic variance (s² g) was lower than the phenotypic variance (s² p) as compared to the variances due to error (s² e) in all of the traits. This showed that the observed phenotypic differences were not alone due to inherent genetic differences among the genotypes but also the substantial portion was due to environmental factor and the interplay between environment and genotype. These findings were agreement with Ene et al., (2016).
 

 
The highest PCV was obtained from seed yield followed by hundred seed weight in both normal and stress conditions. High PCV indicates the existence of a greater scope of selection for the trait being considered, which depents on the amount of variability present. Thus, a greater potential is expected in selecting for the seed yield and hundred seed weight among the investigated chickpea genotypes. The highest GCV was obtained from hundred seed weight in both normal and stress conditions. High GCV indicates the presence of exploitable genetic variability for hundred seed weight which can facilitate selection. However, there was no GCV for seed yield in normal condition because of the negative genotypic variance. These results were in agreement with Yucel et al., (2006). Although, estimates for PCV were higher than those for GCV, they were close, implying that genotype contributed more than environment in the expression of these characters and selection based on phenotypic values is therefore feasible. These results were in agreement with Ewoll and Akromah (2017).
 
These values alone are not helpful in determining the heritable portion of variation. The proportion of genetic variability which is transmitted from parents to all spring is reflected by heritability. Heritability in broad sense is considered to be of importance to breeding programs, because only additive genetic variability is carried on to the next generation. Thus traits with high narrow sense heritability values can be selected more quickly with less intensive evaluation than those with low narrow-sense heritability values and are therefore use­ful in making selection progress estimates in early selection generations (Al-Tabbal and Al-Fraihat, 2012). Heritability estimates are classified as low (5-10%), medium (10-30%) and high (30-60%) (Dabholkar, 1992). Reliant on this suggestion, low value of broad sense heritability was noted for maturity day and moderate heritability was observed traits such as emergence day and seed yield. The highest heritability was found for hundred seed weight followed by podding day, plant height, flowering day and first pod height in normal and stress conditions. These results indicate that there is considerable genetic variation present in these traits to warrant selection for better accessions. These traits can therefore be given special attention for selections aimed at chickpea improvement. High heritability along with high genetic advance are important factors for predicting resultant effects of selecting best individuals. Johnson et al., (1955) suggested that without genetic advance heritability estimation will not render practical values, and they emphasized concurrent use of genetic advance along with heritability. High heritability and high GA for hundred seed weight, podding day, plant height, flowering day and first pod height in both normal and stress growing conditions indicates that genes governing these characters may have an additive effect. The phenotypic expression of these characters may be governed by the genes acting additively and thereby indicating the importance of these characters for selection. A simple selection model will be good enough to do what is necessary and no additional gain is achieved by using sophisticated models.These results indicated that the expected gain from selection would be high if these traits were used as selection criteria in chickpea breeding. The above report is akin with the findings of Meen et al., (2014); Degal et al., (2016); Ton and Anlarsal (2017).

High values of heritability, PCV and GCV obtained for hundred seed weight in normal and stress growing conditions showed that this trait was under control of additive genes and effective selection could be possible for improvement. However, negative heritability coupled with high PCV and high GCV for seed yield may indicate non-additive gene effects governing this trait with limited scope of improvement for this trait which is complex and is mostly influenced by environment.These resultsare in agreement with Ewoll and Akromah, (2017); Mohammed et al., (2019); Alemu et al., (2017).

The range of mean values for most of the traits were large showing the existence of variation among the tested genotypes. Phenotypic coefficients of variation (PCV) were found to be higher than genotypic coefficients of variation (GCV) for all the traits. The two values differed slightly indicating less influence of the environmental factors. The highest heritability was found for hundred seed weight followed by podding day, plant height, flowering day and first pod height in normal and stress conditions. High heritability along with high genetic advance for hundred seed weight, podding day, plant height, flowering day and first pod height in both normal and stress growing conditions indicates that genes governing these characters may have an additive effect. These traits can be improved by giving special attention during selection. Genetic evaluation in these genotypes indicated that there is considerable genetic variation in the studied genotypes to warrant selection for both seasons.