Variation in grain yield, fodder quality and animal intake in two dual purpose legume crops: Mungbean and vegetable soybean grown in semi-arid tropical India

Legume crops grown as sole crop or as part of crop rotation in farming systems provides grain, fodder or green manure. Mungbean [Vigna radiata (L.) R. Wilczek] is a protein rich, short duration (about 60-65 days) grain legume that fits in well into the major cereal cropping systems of South and Southeast Asia (Nair et al., 2013b).  It is also utilised as a forage cum grain crop in countries like Egypt (Abd El-Salam et al., 2013). The use of crop residues of mungbean after grain harvest is also prevalent in some countries. Vegetable soybean [Glycine max (L.) Merr.] is a type of soybean from which the immature pod is harvested at R₆ growth stage (when pods are still green and the beans in the pods are 80% mature) and used as a fresh or frozen vegetable (Nair et al., 2013a). Immature pods are boiled and the seeds extracted as a highly nutritious snack food (Carter and Shanmugasundaram, 1993). Vegetable soybean seeds are larger (over 30 g/100 seeds), have a milder flavor, nuttier texture and are easier to cook in comparison to grain soybean. After the harvest of the fresh pods, the crop residues may be used for feeding livestock.
 
Since the residue after the harvest of grain/pods are used for feeding of livestock particularly ruminants, the current study is aimed to explore the variation for grain yield and forage quality traits in 26 mungbean lines and 26 vegetable soybean lines grown in Hyderabad, India so as to promote cultivars of dual purpose with high grain/pod yield and fodder quality of residue. In addition the study is also supported by in vivo experiment so as to identify the superior cultivar (s) among the mungbean and vegetable soybean lines.

Mungbean trials-2013
 
Twenty six mungbean lines (Table 1) were sown in a randomized block design with three replicates (plot size = 2m x 0.6 m) at the campus of the International Crop Research Institute for the Semi-Arid Tropics (ICRISAT) in Hyderabad, India during June 2013 (kharif season). The soil type was a sandy clay loam. Di-ammonium phosphate fertilizer was applied as basal dose at 100 kg/ha. Irrigation was provided at fortnightly intervals during the duration of the trial. Data on pod yield/plant (PY), grain yield/plant (GY) and haulm yield/plant (HY) were recorded at crop maturity stage. 
 

 
Vegetable soybean trials-2013
 
Twenty six vegetable soybean lines (Table 2) were sown in a randomized block design with four replicates (plot size = 2m x 0.6 m) at the campus of the International Crop Research Institute for the Semi-Arid Tropics (ICRISAT) in Hyderabad, India during June 2013 (kharif season). The soil type was a sandy clay loam. Di-ammonium phosphate fertilizer was applied as basal dose at 100 kg/ha. Irrigation was provided at fortnightly intervals during the duration of the trial. Data on fresh pod yield/plant (PY) and haulm yield/plant (HY) were recorded at R₆ stage. 
 

 
Laboratory haulm fodder quality analysis-2013
 
In addition to the agronomic traits, laboratory quality traits such as nitrogen content (N), neutral detergent fibre (NDF), acid detergent fibre (ADF), acid detergent lignin (ADL), metabolisable energy (ME) and in vitro organic matter digestibility (IVOMD) were determined for samples obtained from the above two experiments by Near-Infrared Reflectance Spectroscopy (NIRS) method. The NIRS instrument used was a FOSS Forage Analyzer 6500 with software package WinISI II. For conventional laboratory analysis nitrogen was determined by auto-analyzer (AOAC, 2005) method, NDF, ADF and ADL by Van Soest et al., (1991) and in vitro digestibility (IVOMD) and metabolisable energy (ME) content were estimated based on sample incubation in rumen microbial inoculum using the in vitro gas production technique and the associated equations described by Menke and Steingass (1988).
 
Mungbean and vegetable soybean trials-2014
 
Four each of mungbean (Table 3) and soybean (Table 4) lines selected based on laboratory quality traits were grown in the field in 10 m x 10.5 m plots per line (30 cm x 10 cm spacing), in a randomized block design with three replicates at the campus of the International Crop Research Institute for the Semi-Arid Tropics (ICRISAT), Hyderabad, India during June 2014 (kharif season). Di-ammonium phosphate fertilizer was applied as basal dose at 100 kg/ha. Irrigation was provided at fortnightly intervals during the period of the trial. Data on pod yield/plant (PY), grain yield/plant (GY) and haulm yield/plant (HY) at crop maturity stage in case of mungbean and  fresh pod yield /plant (PY) and haulm yield/plant (HY) at R₆ stage in case of vegetable soybean were recorded. 
 

 

 
Laboratory haulm fodder quality analysis-2014
 
The fodder quality analysis was conducted for the haulm samples of mungbean and vegetable soybean harvested from the trials of 2014 following the procedures as described for the haulm sample analysis of the year 2013.
 
In- vivo trials-2014
 
Haulms from 4 selected mungbean lines [ML 818 (T1), VC 6510-151-1 (T2), NM 94 (T3) and ML 1628 (T4)] and the 3 selected soybean lines [Swarna Vasundhra (T5), AGS 459 (T6) and GC 84501-32-1 (T7)] were tested as supplements (Table 5) to a sorghum stover based ration using male Brown Nellore sheep with an average starting weight of 24.7 kg kept in metabolic cages. Before the start of the trial the sheep were dewormed and vaccinated against FMD (Foot and Mouth disease), PPR (Peste Des Petits Ruminants),  ET (Enterotoxaemia) and Sheep Pox. Six sheep were allocated to a treatment. Two hundred grams of haulms were offered to the sheep from 8:00 to 10:00 AM and refusals were collected and weighted. Chopped sorghum stover was offered ad libitum at 10.00 am allowing for refusals of 10 to 15%. After 4 weeks of feeding a 7-day fecal and urine collection period was started. Dry matter intake (DMI) and organic matter intake (OMI), dry matter digestibility (DMD), organic matter digestibility (OMD) and nitrogen (N) balance were determined by analysing the samples of experimental feeds, refusals, feces and urine using AOAC (2005) procedures.
 
 

 
Statistical analysis
 
SAS 9.4 (2012) statistical package was used for analysis of variance (ANOVA) by general linear model (PROC GLM). The comparison of means between treatments were carried out using Fisher’s least significance difference (LSD) test at 5% level of significance.
 
The model: Yij=µ+ti +eij was used for the analysis of the data where Y_ij represents the j-th observation (j = 1, 2, ...ni) on the i-th treatment (i = 1, 2, ..., k levels). So, µ overall mean effect, ti represents the i-th treatment effect and eij represents the random error present in the j-th observation on the i-th treatment. The errors eij are assumed to be normally and independently (NID) distributed, with mean zero and variance s²_e.

Mungbean trials
 
No significant differences were recorded for pod yield, seed yield and haulm yield among the mungbean lines tested during 2013. Significant (P< 0.05) variation among lines was found for most fodder quality traits (Table 1). In terms of N content, ML 818 and VC 3890A were the better performers. VC 6510-151-1, ML 818, ML 1628 and PAU 911 recorded high values for IVOMD (Table 1).  Haulm yield was positively associated with pod yield, seed yield, haulm IVOMD (Fig 1a/b/d). Seed yield showed positive association with haulm IVOMD (Fig 1h) but not with haulm nitrogen content (Fig 1g). During 2014, significant differences among the four lines tested were observed only for ME and haulm yield.
 

 
Vegetable soybean trials
 
Significant variation (P< 0.05) was observed for all the traits studied for the vegetable soybean lines (Table 2). Swarna Vasundhra, AGS 406 and GC 84501-32-1 recorded high values for fresh pod yield. In terms of forage yield, Swarna Vasundhra was the best performer. Swarna Vasundhra, GC 84501-32-1, AGS 406 and AGS 292 registered high values for nitrogen content. In the case of in vitro organic matter digestibility, AGS 610 and AGS 406 recorded high values (Table 2). Fresh pod yield showed significant and positive association with haulm yield and IVOMD (Fig 2a/2f) but not with haulm nitrogen content (Fig 2e). Seed yield was positively associated with haulm yield, haulm nitrogen content and IVOMD (Fig 2b/2g/2h). Haulm yield showed positive association with haulm nitrogen content and IVOMD (Fig 2c/2d). During 2014, significant variation (P< 0.05) was observed for all the traits studied (Table 4).
 

 
In-vivo trials
 
Nitrogen content in legume residues is generally twice and more that of minimum microbial N requirements (Table 1 and 2). In the present work supplementation of residues from mungbean variety ML 818 and VC 6510-151-1 resulted in positive N balances, in other words the sheep accreted protein body mass, while sheep with negative N balances lost body weight. Supplementation of ML 818 yielded the best results likely because it had the highest N content of 3.3%. As a general observation residues from both mungbean and vegetable soybean were quite fibrous and had to be chopped into smaller pieces before sheep started consuming them in significant amounts.
 
As pointed out by Sharma et al., (2010) variations in crop residue quality need to be not only statistically significant but livestock nutritionally significant. Among the range of laboratory fodder quality traits investigated, two traits represent fodder quality of legume haulms well: haulms N and IVOMD (Samireddypalle et al., 2017; Prasad et al., 2010). Haulms N content ranged from 2.1 to 3.3% and IVOMD from 57.4 to 63.8%. As shown by Samireddypalle et al., 2017 and Prasad et al., (2010) difference of such magnitude among legume haulms results in high price premiums in fodder markets and substantial difference in livestock performance. It is encouraging to note that variations in haulm fodder traits do not come at the expenses of seed yields. At the most basic level seed yield and haulm yield were strongly positively correlated with variations in the former accounting for about 52% in the variations in the latter (Fig 1). Haulm N was unrelated to pod, seed and haulm yield in mungbean (Fig 1a/b/e). Interestingly IVOMD of mungbean was positively associated (P = 0.05) with seed and haulm yield (Fig 1c/d) but not with pod yield (Fig 1f).
 
Supplementation of legume haulms to cereal residues increase the overall nitrogen (N) content of the animal diet. Cereal crop residues generally contain less than 1 - 1.2% N which is the minimum amount to assure efficient microbial degradation of feed in the rumen of livestock (Van Soest, 1994). Nitrogen content below the minimum microbial requirement result in reduced feed intake and perhaps digestibility. Van Soest (1994) reported data that showed feed intake could double when feed nitrogen content increased from 0.4 to 1.2%. It can be estimated from this data set (nitrogen content range 0.4 to 1.2%) that a 0.1% increase in feed nitrogen resulted in increased intake of more than 10% (Van Soest, 1994).

Core funding to support World Vegetable Center activities worldwide is provided by the Republic of China (ROC), Australian Centre for International Agricultural Research (ACIAR), UK aid, United States Agency for International Development (USAID), Germany, Thailand, Philippines, Korea and Japan. The authors declare no conflicts of interest in this work.