Response of Cowpea [Vigna unguiculata (L.) Walp] Genotypes under Residual Moisture Condition in Rice-Fallow Area of Goa State, India

Cowpea [Vigna unguiculata (L.) Walp] is an important food and fodder legume cultivated in the tropical and sub-tropical regions in more than 65 countries (Singh et al., 1997). It is a major source of inexpensive protein in human diets with dry seeds containing 23-25% protein (Bressani, 1985; Gupta, 1988), 1.8% fat and 60.3% carbohydrates and is also a rich source of calcium and iron (Majnoon Hoseni, 2008).  Cowpea has multiple uses as source of protein for human beings, as feed for livestock and for enriching soil fertility through biological nitrogen fixation (Nhamo and Mpagngua 2003), thus it has become very valuable in areas where land use has intensified (Kamara et al., 2017).
        
In Goa state, cowpea is a major pulse crop cultivated during rabi season in rice-fallow areas under residual moisture condition. Locally known by Alsando (red bold varieties) or Chowli (cream colour or white colour varieties) used in many of the culinary preparations owing to its unique taste, bold size seeds and better cooking quality. Its leaves and immature pods are consumed as green vegetable.  Majority of the farmers in Goa prefer local varieties (Alsando) which are semi-spreading types with tolerance to moisture limiting situation. Farmers are getting premium price for the seeds of these local type cowpea varieties, with each kilo gram of seeds fetching 160 - 180 rupees. However, the yield of these local types is generally low with average seed yield of 0.8-1.0 t/ha. This in turn demands suitable cowpea genotypes with higher seed yield and better seed quality traits. The present study thus aimed at identification of cowpea genotypes suitable for cultivation in rice-fallow areas prevailing in the Goa state. The study also aimed at understanding the genetic variability in the germplasm set for their use in future breeding programme.

Twenty three cowpea genotypes received from All India Co-ordinated Research Project on Arid Legumes, Kanpur, as part of Initial Variety Trial (IVT) were included in the present study along with local check variety Goa cowpea-3. Goa cowpea-3 is a released variety for Goa state. The test genotypes include advanced breeding lines developed at different cowpea breeding institutes in India and also the ruling cowpea varieties like RC 101, GC-3, Pant lobia-3 and Pant lobia-4. All these genotypes were grown in a randomized complete block design (RCBD) with two replications at ICAR-Central Coastal Agricultural Research Institute farm (15° 33’ N latitude, 73° 53’ E longitude and +3.0 m MSL) during the rabi season 2018-19. Seeds were sown at a spacing of 45 cm between the rows and 10 cm between plants. Five rows of 4 m length were maintained for each of the genotypes. Timely management practices and plant protection measures as per the standard recommended practice was followed to raise the good crop.  The observations on growth and yield contributing characters were recorded on ten randomly selected plants in each of the genotypes. Days to 50 per cent flowering, days to maturity, grain yield and straw yield were recorded on plot basis. The variability was estimated as per procedure for analysis of variance suggested by Panse and Sukhatme (1985), genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV) by Burton and De Vane (1953) and heritability and genetic advance by Johnson et al., (1955). The genotypic and phenotypic correlation coefficients were calculated using the method given by Johnson et al., (1955), while genotypic and phenotypic path coefficient was worked out as suggested by Wright (1992) and as described by Dewey and Lu (1959). Analysis was done using the windostat software version 9.1.

A wide range of variation was observed for all the characters studied. Days to 50 per cent flowering varied from 49 days to 68.5 days with an overall mean of 57.3 days. The variation is the resultant of inherent genetic variation and also due to prevailing environmental factors, such as temperature and photoperiod (Hadley et al., 1983). Genotype PCP-1118 was the earliest with 49 days followed by Pant lobia 4 and PCP-1131 with 50 and 52 days respectively.  Genotypes with early maturity are preferred over the late maturing ones when the crop is entirely dependent on the existing moisture in the soil. This in turn help such genotypes to escape from the drought like situation during the end of the cropping season.
 
Plant height ranged from 24.7 cm to 37.2 cm with a mean of 31.2 cm whereas number of primary branches per plant ranged from 4.4 to 8.9 with an average of 5.9 per plant.  Highest number of pods per plant was reported in the genotype SKAU-C-407 with 13.78 per plant followed by RC 101 and MC-17-1 with 12.69 and 10.59 respectively. Goa cowpea-3 recorded lowest number of pods per plant but had longest pod with mean pod length of 23.77 cm.  Bidoli local had the lowest pod length of 11.13 cm. Seeds per pod was highest in the genotype PTBCP-5 with average seeds of 16.6 followed by Goa cowpea-3 and MC-17-1 with 15.5 and 15.4 seeds respectively. With respect to 100 seed weight, Goa cowpea-3 was very unique with 100 seed weighing 22.9 g where as in rest of the genotypes it is less than 16 gm. The lowest number of pods in Goa cowpea-3 was compensated through long pods, more number of seeds per pod and with high 100 seed weight. Highest dry straw yield was recorded in the genotype Goa cowpea-3 followed by PCP-1131 and MC 17-1 where as Pant lobia 3 recorded highest harvest index (0.48) and Goa cowpea 3 with lowest harvest index (0.22).
 
Yield performance
 
Performance of cowpea genotypes under rice fallow situationin terms of seed yield (kg/ha) and ranking based on mean seed yield is presented in Table 1. Highest seed yield of 2204 kg/ha was recorded in the genotype PCP-1131 followed by MC 17-1 (2119 kg/ha) and SKAU-C-407 (2036 kg/ha). Regional check variety RC 101 was fourth among the ranking whereas GC-3 was 11^th among the ranking.  Local check variety Goa cowpea-3 recorded seed yield of 1443 kg/ha. Out of the 23 genotypes tested, only four genotypes viz., PCP-1131 (2204 kg/ha), MC 17-1 (2119 kg/ha), SKAU-C-407 (2036 kg/ha) and RC 101 (1917 kg/ha) recorded significantly higher seed yield compared to the local check variety Goa cowpea-3 (1443 kg/ha). Genotypes PCP-1131, SKAU-C-407 and RC 101 were hence found to be ideal genotype for rice-fallow situation as they produced higher seed yield and straw yield along with100 seed weight of more than 11 g. The genotype MC 17-1 though recorded higher seed yield but has very small seeds with average 100 seed weight of 9.75 g. Such varieties are not preferred by the people in Goa state. 
 

 
Variability for yield and its attributing traits
 
The extent of variability present in the cowpea germplasm set was measured in terms of genetic parameters viz., mean, range, genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV), heritability (broad sense) and genetic advance as percent of mean and are presented in Table 2. In the present study, high PCV and GCV (>20%) values were recorded for seed yield, straw yield, pods per plant and 100 seed weight. Moderate PCV (10-20%) was observed for pod length, seeds per pod, plant height, harvest index and primary branches per plant where as low PCV (<10%) values were recorded in days to 50 per cent flowering and days to maturity. Higher values of PCV and GCV indicate the presence of substantial variability for the traits which in turn offering scope for selection of these traits. Similar observations of high PCV and GCV for seed yield and pods per plant were earlier reported in the studies on variability by Havaraddi and Deshpande (2018), Sarath and Reshma (2017), Olayiwola and Soremi (2014) and Suganthi and Murugan (2008). Similarly high PCV and GCV for straw yield was observed by Olayiwola and Soremi (2014) and for 100 seed weight by Sarath and Reshma (2017).  Characters pod length, plant height, primary branches per plant, seeds per pod and harvest index recorded moderate PCV and GCV whereas days to maturity and days to 50 per cent flowering recorded low PCV and GCV values. In general PCV estimates were more than the GCV estimates in all the characters indicating that apparent variation was not only due to genotype, but also due to influence of environment. Further, the traits plant height and primary branches per plant showed considerable differences between PCV and GCV indicating the influence of environment in the expression of these traits and hence, selection based on these traits may not be effective.
 

        
Heritability estimates ranged from 53 per cent in plant height to 98 per cent in 100 seed weight. High h² (broad sense) values indicate the predominance of additive gene action in the trait expression, which in turn implies that the traits can be improved through single plant selection.  Genetic advance as per cent of mean was ranged from 10.45 per cent to 63.71 per cent in days to maturity and grain yield, respectively. In general most of the characters except days to 50 per cent flowering, days to maturity and plant height showed high heritability and genetic advance as per cent of mean suggesting that they provide good base for selection. This was in accordance with the findings of Olayiwola and Soremi (2014) and Malarvizhi et al., (2005) for pods per plant, 100 seed weight, seed yield and straw yield.  Similar trend for pod length and seeds per pod was observed by Suganthi and Murugan (2008) and Sarath and Reshma (2017). This indicates that the expression of these characters is mainly due to additive gene effect (Panse 1957). Anbu et al., (2000) on the contrary reported higher values of heritability and genetic advance as per cent mean for plant height and days to 50 per cent flowering. All this information on genotypic coefficient of variation, heritability and genetic advance as per cent mean would help in developing reliable selection indices.   

Genotypic and phenotypic correlation coefficient analysis
 
The genotypic and phenotypic correlation for the eleven quantitative yield related traits are presented in Table 3. Genotypic correlation coefficients (rg) were generally found to be of higher magnitude than the corresponding phenotypic correlation coefficients (rp) indicating the strong association between the characters. Seed yield exhibited significant positive correlation with plant height, pods per plant, straw yield and harvest index at both genotypic and phenotypic level whereas it was significant and positive with days to maturity only at genotypic level. Similar findings of significant and positive association of plant height, pods per plant, biological yield and harvest index with seed yield was earlier reported by Mahesh et al., (2016) in their study on assessing the diversity of cowpea genotypes. The characters days to 50 per cent flowering, pod length, seeds per pod, 100 seed weight recorded positive non-significant association with grain yield. The positive association between a pair of traits indicates that selection of a desirable quantitative trait (s) will have concurrent positive effects on the other traits, which would help breeders to improve both characters at the same time (Gerrano et al., 2015). Selection of highly associated traits with seed yield such as plant height (r = 0.65), pods per plant (r = 0.68), straw yield (r = 0.68) and harvest index (r = 0.46) would help in improving the seed yield.
 

 
Path coefficient analysis
 
Path coefficient analysis at phenotypic and genotypic levels was carried out to find the direct and indirect contribution of different quantitative traits to seed yield (Table 4). It is evident from the values of residual effect (0.14) that most of the yield and yield contributing traits were studied in the present investigation. The results indicated that straw yield (1.068) had highest positive direct effect on seed yield followed by harvest index (0.741). Similar observations of high direct effect of harvest index and pods per plant on seed yield was observed by Walle et al., (2018) in their study on assessing the correlation and path coefficient in cowpea landraces from Ethiopia. However positive direct effect of straw yield on grain yield in the current study was contrary to some reports (Peksen and Artik, 2004) which had a non-significant negative effect on grain yield. Hence, direct selection through these traits will be effective for seed yield improvement. Characters viz., plant height and pods per plant (r = 0.65 and 0.68, respectively) can also be considered for selection as both straw yield and harvest index contributed to the seed yield via these two characters.
 

Evaluation of cowpea genotypes under residual moisture condition revealed the positive and direct effect of characters such as straw yield and harvest index on seed yield.  Therefore, maximum importance should be given to these traits during the selection for achieving the higher seed yield under rice-fallow situation. The study identified three promising cowpea genotypes viz., PCP-1131, SKAU-C- 407 and RC 101 which would serve as dual-purpose cowpea for cultivation under rice-fallow areas in Goa state due to their high seed yield and straw yield. 

The authors are thankful to ICAR-Central Coastal Agricultural Research Institute, Goa, India, for providing facilities and support to carry out this research.  Authors are also thankful to Project Co-ordinator, National Network Project on Arid Legumes, Kanpur, for sharing the materials for the study.