Citation of this paper

Yields and the nutritive value of early harvested common bean (Phaseolus vulgaris L.) crop residues for ruminants

Mesfin Dejene^1,6, Rob M Dixon¹, Alan J Duncan^2,7, Kerry B Walsh³, David McNeill⁴ and Endalkachew Wolde-meskel⁵

¹ Queensland Alliance for Agriculture and Food Innovation (QAAFI), The University of Queensland, Gatton, Qld, 4343, Australia
mesfindegene@yahoo.co.uk
² International Livestock Research Institute (ILRI), PO Box 5689, Addis Ababa, Ethiopia
³ Central Queensland University, Rockhampton, 4701, Qld, Australia
⁴ School of Veterinary Science, The University of Queensland, Gatton, Qld, 4343, Australia
⁵ World Agroforestry (ICRAF), Addis Ababa, Ethiopia
⁶ Etiophian Institute of Agricultural Research (EIAR), Holetta Research Centre, PO Box 2003, Addis Ababa, Ethiopia
⁷ Global Academy of Agriculture and Food Security, The Royal (Dick) School of Veterinary Studies and The Roslin Institute, Easter Bush Campus, Midlothian, EH25 9RG United Kingdom

Abstract

The yields and nutritive value of haulm (stem and leaf) and pod wall (HPW) crop residues of common bean early-harvested at the green pod fill stage were measured in nine genotypes grown in two dissimilar environments (Boricha and Mandura) in Ethiopia. The concentrations of total N, neutral detergent fibre (NDF) and acid detergent fibre (ADF), and the in vitro DM digestibility (IVDMD), in the HPW fractions (stem, leaf and pod wall) were measured. At Mandura the yields of seed and HPW averaged 1.03 and 2.26 t/ha, respectively, and varied (p<0.001) among genotypes (0.31 - 1.57 t/ha and 1.19 - 2.93 t/ha, respectively). However no such differences among genotypes (p>0.05) were observed at Boricha. Stem, leaf and pod wall comprised 52.4, 23.1 and 24.5% of HPW, respectively. The concentrations of N, NDF and ADF, and IVDMD in the HPW, and the HPW fractions, generally varied (p<0.05) among genotypes. The digestible DM (DDM) yields of HPW ranged among genotypes from 0.75 - 1.81 t/ ha at Mandura, and averaged 1.89 t/ha at Boricha. N yield ranged from 35.3 - 70.3 kg N/ha at Boricha and 18.3 - 43.4 kg N/ha at Mandura. Since yields of seed were positively correlated with yields of HPW DM, DDM and N, common bean genotypes can be selected concurrently for seed and HPW yields. Uses of high-yielding genotype can double the N and DDM yields without reducing seed yield. The N and IVDMD of HPW residue showed that early-harvested common bean is a high nutritive value feedstuff for ruminants.

Keywords: feed quality, food legume, forage, morphological fractions, Ethiopia

Introduction

In many developing countries smallholder crop-livestock systems provide the majority of animal products (Randolph et al 2007; Herrero et al 2013) and provide the greatest potential for increased production of animal foods (Wright et al 2012; Herrero et al 2013). In such systems the crop and livestock components are usually closely integrated with ruminant livestock dependant primarily on crop residues as feeds for much of the annual cycle (Williams et al 2000; Moritz 2010). However, the poor nutritive value of most crop residue feeds is a major constraint to livestock productivity (Kabaija and Little 1988; Tolera and Sundstøl 2000; Zerbini and Thomas 2003). Thus locally available high quality feeds are important to improve productivity.

Grain legume food crops are often grown to provide high-protein foods and in rotation with cereal crops in smallholder farming systems. Importantly, the harvest residues of grain legume crops are usually higher in nutritive value, especially in N, than cereal crop residues. In many global regions common bean (Phaseolus vulgaris L.) is grown as a food legume crop (Beebe et al 2013), often grown in smallholder crop-livestock systems in association or rotation with maize, sorghum, bananas, or other crops (Broughton et al 2003). Most of the crop is harvested at seed maturity, but some is early-harvested at the green pod fill growth stage (Early or Eshet harvest) to provide a vegetable when food is scarce and to provide seasonal food variety. These are known as ‘green beans’ or ‘snap beans’ when harvested before seed development, or as ‘shell beans’ if harvested when the seeds are physiologically mature but before desiccation. The haulm (stem and leaf) and pod wall crop residues (collectively HPW), of such early harvested common beans are expected to be of high nutritive value for ruminants. For example it has been demonstrated that they can be used with maize or sorghum stover for fattening cattle (Gebreyohannes and Hailemariam 2011). However, information is lacking on the yield, nutritive value and variation among genotypes of early harvested common bean crop residues.

The objectives of the present study were to: (1) assess the yield and nutritive value for ruminants of the various morphological fractions of the crop residues from a range of genotypes of common bean when early-harvested at green pod fill stage, (2) examine the relationships between yield and HPW crop residue quality parameters, and (3) compare the yields and nutritive values with those observed in common bean crop residues harvested at seed maturity.

Materials and methods

Location, experimental design, and sample collection

The study was undertaken at two trial sites, one in the South-west (at Boricha Wereda; 6°947’N and 38°222’E) and the other in the North-west (at Mandura Wereda; 11°118’N and 36°722’E) regions of Ethiopia during the 2013 cropping season. These sites had been selected to represent smallholder crop-livestock systems in differing environments where common bean is an important crop (Farrow 2014). The nine genotypes grown at each site were chosen to represent those well-adapted and often grown in the regions. At each trial site genotypes were examined in a randomized complete-block design with three replicates. The trial sites descriptions, genotypes, management, planting, sampling and measurements at mature seed harvest at the two sites have been previously reported (Dejene et al 2018). The plants harvested at early maturity at the green pod fill stage for the present study were sampled from a 0.5 m x 2 m area in the plots also harvested at seed maturity and with a 0.5 m border between the harvested areas. Dates of sowing at Boricha and Mandura were on 07 and 24 August 2013, respectively. The early harvest at Boricha was on 14 and 17 October 2013 at the R6 (mid-seed fill) stage when 50% of pods have fully developed seeds (Schwartz and Langham 2016). Due to unavoidable delay the harvest at the Mandura was on 2 and 4 November 2013 when~80% of pods had fully developed seeds (i.e. shortly before physiological maturity). At harvest the numbers of plants were counted, the plants were harvested at the soil surface and then separated with care to avoid leaf loss into seed, pod wall, stem and leaf; the last 2 fractions comprised the haulm. Following measurement of fresh weight, samples (~500 g) were placed into cotton bags, sun-dried and later oven-dried (60 °C for 48 h).

Laboratory analyses

Samples of leaf, stem and pod wall fractions were analysed for in vitro dry matter digestibility (IVDMD), and the concentrations of total N, neutral detergent fibre (assayed with α–amylase and corrected for the ash concentration of the residue, NDF), and acid detergent fibre (corrected for the ash concentration of the residue, ADF) using near infrared spectroscopy (NIRS). The laboratory procedures used to analyse the reference samples, and the calibration and validation procedures, have been described by Dejene et al (2018).

Calculations and statistical analyses

The yields and composition of haulm and pod wall (HPW) were calculated from the measurements for the leaf, stem and pod wall fractions. Analysis of variance was performed for yields, morphological traits and nutritional attributes of the residues as described by Dejene et al (2018) except that the results from the stage of seed maturity at harvest (early or mature) were also included as a main effect. Since there were interactions for genotype (G) and site (i.e. environment (E) (i.e. G x E) and harvest stage in the pooled analyses of treatments across sites (Tables 1 and 2) the results from each E were also analysed and are presented separately. Within each E the effects of genotype, morphological fractions and harvest stage were examined. When averaged across harvest stage at each E, replicate was considered as nested within G and the replication x genotype mean square was used as the error term for testing the significance of G. When averaged across E, replicate was considered as nested within E and the replication x environment mean square was used as the error term to test E. Differences among means were compared using the least significant difference (LSD) test in PROC MIXED with the PDIFF option of the LSMEANS statement. Relationships between yield, residue composition and residue digestibility were analysed using linear regression approach.

Results

Seed and early harvest residue yields of haulm and pod wall (HPW) and proportions of HPW fractions

There were large (p<0.05) effects of genotype on the yields of both seed (range 0.31 - 1.57 t/ha) and of HPW (range 1.19 - 2.93 t/ha) at Mandura, but no differences (p>0.05) were observed at Boricha (Table 1). Genotype sometimes affected the proportions of leaf, stem and pod wall in the HPW within environment, but the effects were not consistent (Table 1). The stem comprised the largest component of the HPW (mean 50.2 and 54.7%), while there were similar proportions of leaf (means 25.2 and 21.0 %) and pod wall (means 24.7 and 24.3%) at the Boricha and Mandura sites, respectively (Table 1).

Table 1. Yields of seed, haulm and pod wall (HPW) dry matter (DM), nitrogen (N) and digestible dry matter (DDM), and proportions of the HPW morphological fractions of nine common bean genotypes harvested at green pod fill stage at Boricha (n = 3) and Mandura (n = 3) sites in 2013
Environment	Genotype	Seed yield (t/ha)	HPW yields			Proportions of HPW fractions (%)
Environment	Genotype	Seed yield (t/ha)	DM (t/ha)	N (kg/ha)	DDM (t/ha)	Pod wall	Leaf	Stem

Boricha	A-Melka¹	1.26	2.95	58.6^ab	1.97	18.3	23.9^bc	57.7^a
	Argene	0.85	2.74	70.3^a	1.87	27.3	27.5^a	45.2^bc
	Awash-1	1.13	2.75	56.8^ab	1.84	22.0	25.1^bc	52.9^ab
	Dimtu	0.82	2.06	38.8^bc	1.42	22.7	24.7^bc	52.7^ab
	Dinknesh	1.43	3.16	69.9^a	2.12	25.8	26.1^ab	48.1^bc
	H-Dume²	1.40	3.17	57.1^ab	2.14	27.8	25.1^bc	47.1^bc
	Ibado	0.83	2.06	35.3^bc	1.36	23.8	23.3^c	52.8^ab
	Nasir	1.53	3.15	58.9^ab	2.15	31.6	24.8^bc	43.6^c
	SARI	1.35	3.15	63.2^a	2.17	22.7	26.0^ab	51.3^abc
	Mean	1.18	2.80	56.5	1.89	24.7	25.2	50.2
	SEM	0.201	0.291	6.94	0.202	2.81	0.76	2.80
	p value	0.12	0.06	0.03	0.07	0.12	0.05	0.05

Mandura	A-Melka¹	0.65^d	1.79^c	28.7^d	1.07^c	20.9^b	21.4	57.7^ab
	Argene	0.31^e	1.19^d	18.3^e	0.75^d	20.2^b	20.7	59.2^a
	Awash-1	0.87^cd	1.88^c	31.7^cd	1.15^c	21.8^b	20.2	58.1^ab
	Dimtu	1.24^b	2.57^b	41.9^a	1.64^ab	25.1^a	20.6	54.4^bc
	Dinknesh	1.21^b	2.52^b	43.4^a	1.62^ab	26.1^a	22.1	51.8^c
	H-Dume²	1.57^a	2.93^a	36.5^bc	1.81^a	26.6^a	20.7	52.7^c
	Ibado	1.26^b	2.68^ab	39.7^ab	1.64^ab	25.2^a	21.4	53.4^c
	Nasir	1.07^bc	2.35^b	31.7^cd	1.44^b	26.9^a	21.0	52.1^c
	SARI	1.09^bc	2.40^b	36.3^bc	1.51^b	26.3^a	21.2	52.6^c
	Mean	1.03	2.26	34.2	1.40	24.3	21.0	54.7
	SEM	0.085	0.112	1.667	0.072	0.78	0.98	1.26
	p value	<0.001	<0.001	<0.001	<0.001	<0.001	0.93	<0.01

Pooled	Mean	1.10	2.53	45.4	1.65	24.5	23.1	52.4
	p value
	Genotype (G)	<0.001	<0.001	0.04	<0.01	<0.01	0.38	<0.01
	Environment (E)	0.23	0.08	0.01	0.04	0.81	<0.001	0.02
	G x E	<0.01	<0.001	<0.001	<0.001	0.21	0.28	0.06
¹Awash-Melka; ² Hawassa Dume; SEM, standard error of means ^abcMeans in the same column without common letter are different at p<0.05

Quality attributes of the haulm and pod wall (HPW) fractions

The N concentrations in the entire HPW varied (p<0.05) among genotypes at both environments, ranging from 17.1 – 25.3 g/kg DM at Boricha and 12.5 – 17.4 g/kg DM at Mandura (Table 2). At Mandura site the IVDMD of leaf, stem, pod wall and HPW differed among genotypes (p <0.05) and the IVDMD of HPW ranged from 602– 642 g/kg DM. There were comparable differences among genotypes within the morphological fractions.

Table 2. Quality attributes (g/kg dry matter) of haulm and pod wall (HPW) morphological fractions (pod wall, leaf and stem) and HPW of nine common bean genotypes harvested at green pod fill stage at Boricha (n = 3) and Mandura (n = 3) in 2013
	Genotype	N				IVDMD
	Genotype	Pod wall	Leaf	Stem	HPW	Pod wall	Leaf	Stem	HPW

Boricha	A-Melka¹	21.3^bc	32.5^b	14.0^bc	19.8^bcd	750^ab	735	614	667
	Argene	28.2^a	36.5^a	16.6^a	25.3^a	695^bcd	720	653	683
	Awash-1	23.3^b	30.7^bc	14.7^ab	20.7^bc	687^cd	732	630	667
	Dimtu	23.0^b	30.7^bc	11.4^d	18.8^cde	779^a	732	628	688
	Dinknesh	28.4^a	31.4^bc	13.3^bcd	22.0^b	739^abc	692	621	670
	H-Dume²	16.9^de	30.7^bc	11.8^cd	18.0^cde	691^bcd	719	636	674
	Ibado	13.9^e	30.2^bc	13.0^bcd	17.1^e	677^d	726	631	662
	Nasir	18.5^cd	30.4^bc	12.1^cd	18.7^cde	708^bcd	737	631	682
	SARI	22.5^b	28.2^c	14.9^ab	20.1^bcd	724^abcd	724	657	689
	Mean	21.8	31.2	13.5	20.1	717	724	634	676
	SEM	1.185	1.234	0.832	0.820	0.02	12.37	9.58	8.82
	p value	<0.001	0.02	<0.01	<0.001	0.02	0.35	0.10	0.31

Mandura	A-Melka	15.1^ab	31.2^b	10.8^a	16.0^abc	715^a	726^a	514^d	602^d
	Argene	14.2^ab	30.3^b	10.9^a	15.5^bc	681^bc	732^a	580^a	632^abc
	Awash-1	14.4^ab	34.9^a	11.5^a	16.8^ab	665^bcd	722^ab	553^abc	611^cd
	Dimtu	15.1^ab	34.1^a	10.1^ab	16.3^abc	694^ab	750^a	576^ab	641^ab
	Dinknesh	15.7^a	34.5^a	10.9^a	17.4^a	694^ab	749^a	571^ab	642^a
	H-Dume	9.6^c	25.9^c	8.6^c	12.5^e	638^d	766^a	550^bc	618^bcd
	Ibado	9.9^c	30.4^b	10.9^a	14.8^cd	642^d	679^b	572^ab	612^cd
	Nasir	10.9^c	28.7^b	8.8^bc	13.5^de	649^cd	731^a	542^c	610^cd
	SARI	13.4^b	28.6^b	10.6^a	15.1^bcd	680^bc	752^a	557^abc	631^abc
	Mean	13.1	30.9	10.3	15.3	673	734	557	622
	SEM	0.582	0.857	0.493	0.575	11.16	15.34	9.23	7.77
	p value	<0.001	<0.001	<0.01	<0.001	<0.01	0.04	<0.01	0.01

Pooled	Mean	17.5	31.1	11.9	17.7	695	729	595	649
	p value
	Genotype	<0.001	<0.001	<0.001	<0.001	<0.001	0.18	<0.001	<0.01
	Environment	<0.001	0.76	<0.01	<0.01	<0.01	0.17	<0.001	<0.001
	G x E	<0.001	<0.001	0.12	<0.001	0.54	0.02	0.06	0.39
¹Awash-Melka; ²Hawassa Dume; SEM, standard error of means; IVDMD, in vitro dry matter digestibility; N,concentrations of total nitrogen ^abcMeans in the same column without common letter are different at p <0.05

Concentrations of N, NDF and ADF, and IVDMD content, differed among the HPW fractions (p<0.001) at both environments (Table 3). When averaged across environments and genotypes, generally the leaf was higher ( p<0.001) in N and IVDMD (31 and 729 g/kg DM, respectively) than in pod wall and stem fractions, while the concentrations of NDF and ADF followed the inverse trends (Table 3).

Table 3. Quality attributes (g/kg DM), and nitrogen (N) (kg/ha) and digestible dry matter (DDM) (t/ha) yields, of residue morphological fractions (pod wall, leaf and stem) of nine common bean genotypes harvested at green pod fill stage at Boricha (n=27) and Mandura (n=27) in 2013
Environment	Morphological fractions	Quality attributes (g/kg DM)				Nutrient yields
Environment	Morphological fractions	N	IVDMD	NDF	ADF	N (kg/ha)	DDM (t/ha)

Boricha	Pod wall	21.8^b	717^a	480^b	357^b	15.0^c	0.49^b
	Leaf	31.2^a	724^a	357^c	300^c	22.2^a	0.51^b
	Stem	13.5^c	634^b	557^a	424^a	19.3^b	0.89^a
	SEM	0.364	4.819	6.836	6.250	0.965	0.030
	p value	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001

Mandura	Pod wall	13.1^b	673^b	527^b	387^b	7.2^c	0.38^b
	Leaf	30.9^a	734^a	384^c	308^c	14.6^a	0.35^b
	Stem	10.3^c	557^c	627^a	484^a	12.5^b	0.68^a
	SEM	0.259	4.187	4.884	4.792	0.255	0.010
	p value	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001

Pooled	Pod wall	17.5^b	695^b	504^b	372^b	11.1^c	0.43^b
	Leaf	31.1^a	729^a	370^c	304^c	18.4^a	0.43^b
	Stem	11.9^c	595^c	592^a	454^a	15.9^b	0.79^a
	SEM	0.223	3.192	4.201	3.938	0.499	0.016
	p value	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001
ADF, acid detergent fibre corrected for the ash concentration of the residue; IVDMD, dry matter digestibility; N, total nitrogen; NDF, neutral detergent fibre assayed with α–amylase and corrected for the ash concentration of the residue; SEM, standard error of means ^abc Means in the same column without common letter are different at p <0.05

Nutrient yields of the haulm and pod wall (HPW) fractions

The N yields in the HPW varied (p<0.05) among genotypes at both environments, ranging from 35.3 – 70.3 kg/ha at Boricha and 18.3 – 43.4 kg/ha at Mandura. Digestible dry matter (DDM) yield of HPW varied among genotypes (p<0.05) (range 0.75 – 1.81 t/ha) at Mandura, but did not vary with genotype (p>0.05) at Boricha (Table 1). Among the morphological fraction of the HPW averaged across environment and genotypes, the leaf had highest (p <0.001) N yield (18.4 kg/ha), followed by stem (15.9 kg/ha), and then by pod wall (11.1 kg/ha) (Table 3). In contrast stem had the highest (p<0.001) DDM yield (0.79 t/ha), followed by similar DDM yields for the leaf and pod wall (both 0.43 t/ha).

Relationships between yield and HPW quality attributes

There were correlations between the yields of seed harvested at green pod fill stage and the yields of HPW (r = 0.79; p<0.001), of DDM in the HPW (r = 0.76; p<0.001), and to a lesser extent with the yield of N in the HPW (r = 0.57; p<0.001) (Figure 1a, 1b and 1c, respectively). The seed yield harvested at green pod fills stage was closely correlated with the seed yield at harvest maturity (r = 0.98; p<0.001) (Figure 1d), and was generally not correlated with the N content or IVDMD attributes of the HPW or the HPW fractions. Also the composition of HPW constituents was generally not related (p >0.05) to HPW yield in either site (data not shown). Total N concentrations of the HPW were strongly correlated with IVDMD of HPW (r = 0.70; p<0.001) and leaf-to-stem-ratio in the haulm (r = 0.73; p<0.001) (Figure 1e and 1f). Nitrogen yields of HPW were strongly correlated with the yields of HPW DM (r = 0.86; p <0.001) and DDM in the HPW (r = 0.90; p<0.001, respectively) (Figure 1g and 1h). There were stronger associations between stem and HPW IVDMD at Boricha and Mandura (r=0.66 and r = 0.82; p<0.001) than for leaf (r=0.45 and r = 0.40; p <0.05) and pod wall (r=0.41 and r = 0.44; p<0.05) fractions, respectively (data not shown).

Figure 1. The regression relationships between the seed yield (SY) at early harvest with the yields of (haulm + pod wall; HPW)
DM, digestible dry matter (DDM) or total N concentration in HPW are shown in Figs (a), (b) and (c). The seed yields
at early harvest and mature seed harvest (Dejene et al 2018) are shown in Fig (d). The relationships between total N
concentration of HPW and the in vitro DM digestibility (IVDMD) or the leaf-to-stem ratio of HPW are shown
in figs(e) and (f). The relationships between yields of total N with yield of HPW DM or DDM are shown
in Figs (g) and (h). Measurements at Boricha site (○) and at Mandura site (●) were pooled

Discussion

The measurements in the present experiment clearly demonstrated that the crop residues of early-harvested common bean crops are of high nutritional value for ruminants. This was evident from the high total N concentration (mean 17.7 g/kg DM) and IVDMD (mean 649 g/kg DM) and low concentrations of NDF and ADF in the HPW (Table 2). These measurements are in accord with the general high nutritive value of forages of legumes in vegetative growth and before seed maturity (Norton 1982; Norton and Poppi 1995). Diets comprised mainly of the HPW of early-harvested common bean would be expected to provide high voluntary intakes of DM and DDM by ruminants and metabolizable energy, protein and other nutrients sufficient for high animal productivity (e.g.> 0.5 kg live weight gain/day in young growing cattle) (CSIRO 2007). Furthermore this good quality forage from common bean residue would usually be even more valuable as a ruminant feedstuff when mixed with low quality forage ssuch as cereal crop residues to achieve much higher ruminant productivity than that usually observed in smallholder crop-livestock systems.

The observations that at early harvest the HPW yields of N and DDM ranged widely among the genotypes, up to 2-fold at Boricha site and up to 5-fold at Mandura site (Table 1), indicated that use of the most appropriate genotypes for the environment could provide large increases in the amounts of high quality forage from early-harvested common bean. Wide variation among common bean cultivars in shoot mass at the green pod fill growth stages was also reported by Araújo and Teixeira (2008). Furthermore the close (r = 0.79) positive correlations between the yields of HPW and seed in early harvested common bean in the present experiment indicated that genotypes can be selected for higher yields of both seed and crop residue. Similar positive correlations occurred in the same genotypes and environments when harvested at seed maturity (Dejene et al 2018). These results clearly showed that higher-yielding genotypes of common bean provided higher yields of both seed and crop residue regardless of the stage of harvest, and genotypes can be selected concurrently for both food and forage irrespective of maturity at harvest.

Since the same sites and genotypes of common bean were harvested early and at seed maturity (Dejene et al 2018) the effects of harvest maturity could be compared directly (Table 4). As expected, at mature harvest the seed yield was much (35%) higher at mature-seed-harvest. However there was a much lower proportion of leaf in the HPW, a much lower yield of leaf, and lower yields of both DDM and N. The leaf in legumes, including in common bean, is clearly much higher in N concentration and IVDMD and lower in fibre constituents than the stem or pod wall, and this was observed in the present experiment. Variation in the leaf-to-stem ratio among genotypes in the residue is thus expected to have an important effect on the nutritive value of their harvest residues. Also the stage of physiological maturity of the plant generally has an important influence on leaf shedding and leaf content of the crop residue. This explains the generally lower leaf content at the Mandura site where the early harvest was unavoidably at a later stage of physiological maturity of plants than at the Boricha site. It also explains the lower leaf content at the harvest at seed physiological maturity at these sites (Dejene et al 2018); leaf yield at the mature harvest was only 21% of that at the early harvest. Such differences emphasise the importance of the leaf to maintain the nutritive value of the common bean residue. Since leaf loss in common bean approaching seed maturity is affected by genotype (Larbi et al 1999) consideration of leaf loss during selection of common bean intended as a dual-purpose crop for both food and forage is likely to be advantageous.

Table 4. Yields of seed, haulm and pod wall (HPW) morphological fractions and HPW (t/ha), nitrogen (N) (kg/ha) and digestible dry matter (DDM) yields of HPW (t/ha), and proportions (%) of HPW morphological fractions of nine common bean genotypes harvested at green pod fill and seed maturity stages at Boricha (n = 27) and Mandura (n = 27) in 2013

Environment

Harvest
stage

Seed yield
(t/ha)

Yields of HPW and fractions (t/ha)

Yields of HPW

Proportions of HPW (%)

Pod
wall

Leaf

Stem

HPW

N
(kg/ha)

DDM
(t/ha)

Pod
wall

Leaf

Stem

Boricha

Green pod fill

1.18^b

0.68^a

0.71^a

1.41

2.80^a

56.5^a

1.89^a

24.7^b

25.2^a

50.2^b

Seed maturity

1.67^a

0.61^b

0.15^b

1.41

2.17^b

16.7^b

1.02^b

28.6^a

6.9^b

64.5^a

SEM

0.017

0.007

0.017

0.015

0.013

1.466

0.024

0.371

0.203

0.313

p value

<0.001

0.95

<0.001

Mandura

Green pod fill

1.03^b

0.56^a

0.47^a

1.22^a

2.26^a

34.2^a

1.40^a

24.3^b

21.0^a

54.7^b

Seed maturity

1.32^a

0.50^b

0.10^b

1.01^b

1.61^b

13.1^b

0.78^b

29.9^a

6.8^b

63.3^a

SEM

0.008

0.004

0.007

0.008

0.453

0.010

0.137

0.235

0.269

p value

<0.001

N, total nitrogen; SEM, standard error of means ^ab Means in the same column without common letter are different at p <0.05

Conclusions

In conclusion the HPW residues of early harvested common bean had high nutritive values for ruminants. They should be particularly valuable as a supplement for N-deficient cereal crop residues. Leaf loss as plants approached seed maturity had a major effect on the nutritive value of HPW as a feedstuff. Use of the most appropriate genotypes can double the N and DDM yields of HPW without reducing seed yield, and since yields of seed and of HPW DM, DDM and N are positively correlated genotypes can be selected concurrently for seed and HPW yields.

Acknowledgements

We greatly appreciate the support from Australian Centre for International Agricultural Research through a John Allwright Fellowship for Mesfin Dejene to study at the University of Queensland. The assistance received through the SIMLESA (Sustainable Intensification of Maize-Legume cropping systems for food security in Eastern and Southern Africa) project, CIMMYT, the Ethiopian Institute of Agricultural Research and ILRI-Addis Ababa for research support through N2Africa project is highly appreciated. We thank the N2Africa project team members in the research centres and experimental sites for assistance during field data collection, and staff at ILRI Addis Ababa for support in sample grinding and processing for lab analysis and staff at Gatton, UQ for laboratory analysis support. We would like to add a sentence under this section. Plant samples were imported to Australia under Australian Quarantine Permit-IP14007043.

References

Araújo A P and Teixeira M G 2008 Relationships between grain yield and accumulation of biomass, nitrogen and phosphorus in common bean cultivars. Revista Brasileira de Ciência do Solo. 32(5), 1977-1986. DOI: 10.1590/S0100-06832008000500019

Beebe S E, Rao I M, Blair M W and Acosta-Gallegos J A 2013 Phenotyping common beansfor adaptation to drought. Drought Phenotyping in Crops. Frontiers in Physiology, 4, 1 - 20. https://doi.org/10.3389/fphys.2013.00035

Broughton W J, Hernández G, Blair M, Beebe S, Gepts P and Vanderleyden J 2003 Beans (Phaseolus spp.) – model food legumes. Plant and Soil 252, 55–128.

CSIRO 2007 Nutrient requirements of domesticated ruminants . CSIRO publishing.

Dejene M, Dixon R M, Duncan A J, Wolde-meskel E, Walsh K B and McNeill D 2018 Variations in seed and post-harvest residue yields and residues quality of common bean (Phaseolus vulgaris L.) as a ruminant feedstuff. Animal Feed Science and Technology, 244, 42-55.

Farrow A 2014 Managing factors that affect the adoption of grain legumes in Ethiopia in the N2Africa project, www.N2Africa.org, 56 pp.

Gebreyohannes G and Hailemariam G 2011 Challenges, opportunities and available good practices related to zero grazing in Tigray and Hararghe, Ethiopia . In Drylands Coordination Group (DCG) Report, 54 pp. https://www.utviklingsfondet.no/dcg/assets/documents/Publications/1062-dcg_report_no.66.pdf

Herrero M, Grace D, Njuki J, Johnson N, Enahoro D, Silvestri S and Rufino M C 2013 The roles of livestock in developing countries. Animal 7n(s1), 3-18.

Kabaija E and Little D 1988 Nutrient quality of forages in Ethiopia with particular reference to mineral elements . In Pasture Network for Eastern and Southern Africa.(PANESA). Proc. of 3rd Workshop, 440-448, ILCA, Addis Ababa. https://hdl.handle.net/10568/49880

Larbi A, Dung D, Olorunju P, Smith J, Tanko R, Muhammad I and Adekunle I 1999 Groundnut (Arachis hypogaea) for food and fodder in crop-livestock systems: forage and seed yields, chemical composition and rumen degradation of leaf and stem fractions of 38 cultivars. Animal Feed Science and Technology, 77(1), 33-47.

Moritz M 2010 Crop–livestock interactions in agricultural and pastoral systems in West Africa. Agriculture and Human Values 27(2), 119-128.

Norton B 1982 Differences between species in forage quality. In Nutritional Limits to Animal Production from Pastures: proceedings of an international symposium, 89-100 (Ed J. Hacker). St. Lucia, Queensland, Australia, August 24-28, 1981: Farnham Royal, UK, Commonwealth Agricultural Bureaux.

Norton B W and Poppi D P 1995 Composition and nutritional attributes of pasture legumes . In Tropical legumes in animal nutrition, 23-47 (Eds J. P. F. D'Mello and C. Devendra). Wallingford, UK, CAB INTERNATIONAL.

Randolph T, Schelling E, Grace D, Nicholson C F, Leroy J, Cole D, Demment M, Omore A, Zinsstag J and Ruel M 2007 Role of livestock in human nutrition and health for poverty reduction in developing countries. Journal of Animal Science, 85(11), 2788-2800.

Schwartz H F and Langham M A C 2016 Growth stages of bean (Phaseolus vulgaris L.). Legume ipmPIPE Diagnostic pocket series. In Colorado State University and South Dakota State University. http://legume.coop/wp-content/uploads/2011/10/Card-Common-Bean-Growth-Stages.pdf

Tolera A and Sundstøl F 2000 Supplementation of graded levels of Desmodiumintortum hay to sheep feeding on maize stover harvested at three stages of maturity: 1. Feed intake, digestibility and body weight change. Animal Feed Science and Technology, 85(3), 239-257.

Williams T O, Hiernaux P and Fernández-Rivera S 2000 Crop-livestock systems in sub-Saharan Africa: Determinants and intensification pathways . In Property rights, risk and livestock development in Africa, 132-151 (Ed N. Mccarthy, Swallow. B., Kirk, M., Hazell, P.). IFPRI. Washington D.C.

Wright I A, Tarawali S, Blummel M, Gerard B, Teufel N and Herrero M 2012 Integrating crops and livestock in subtropical agricultural systems. Journal of the Science of Food and Agriculture, 92, 1010–1015.

Zerbini E and Thomas D 2003 Opportunities for improvement of nutritive value in sorghum and pearl millet residues in South Asia through genetic enhancement. Field Crops Research, 84(1), 3-15.

Yields and the nutritive value of early harvested common bean (Phaseolus vulgaris L.) crop residues for ruminants

Mesfin Dejene1,6, Rob M Dixon1, Alan J Duncan2,7, Kerry B Walsh3, David McNeill4 and Endalkachew Wolde-meskel5

Abstract

Introduction

Materials and methods

Location, experimental design, and sample collection

Laboratory analyses

Calculations and statistical analyses

Results

Seed and early harvest residue yields of haulm and pod wall (HPW) and proportions of HPW fractions

Quality attributes of the haulm and pod wall (HPW) fractions

Nutrient yields of the haulm and pod wall (HPW) fractions

Relationships between yield and HPW quality attributes

Discussion

Conclusions

Acknowledgements

References

Mesfin Dejene^1,6, Rob M Dixon¹, Alan J Duncan^2,7, Kerry B Walsh³, David McNeill⁴ and Endalkachew Wolde-meskel⁵