Evaluation of nutritive value of leaves of tropical tanniferous trees and shrubs

Kechero Yisehak and Geert P J Janssens*

Department of Animal Sciences, College of Agriculture and Veterinary Medicine, Jimma University, P. O. Box 307, Jimma, Ethiopia
yisehakkechero.kebede@ugent.be
* Laboratory of Animal Nutrition, Faculty of Veterinary Medicine, Ghent University, Heidestraat 19, 9820 Merelbeke, Belgium

Abstract

This study was conducted to compare the nutritive value of indigenous fodder trees and shrubs (IFTS) and assess the relationship between farmers' IFTS preference, the perception of their characteristics, and analyzed nutritional value at two distinct altitudes within the same area ("high altitude" and "low altitude"). Results were based on laboratory analyses of plant samples and a diagnostic survey of randomly selected 360 livestock farmers. Fifty IFTS were identified and examined for proximate and fibre components, in vitro digestibility, digestible nutrients, energy and condensed tannins (CT). Farmers scored the identified IFTS on a scale of 1 to 4 on nutritive value, growth rate, biomass, compatibility and multifunctionality.

Nutritive value ranged widely among IFTS from 66 to 242 g CP/kg dry matter (DM), 185 to 502 g neutral detergent fibre (NDF)/kg DM, 0.1 to 228 g CT/kg DM, 478 to 745 g total carbohydrate (CHO)/kg DM, 332 to 963 g total digestible nutrients (TDN)/kg DM and 5 to 15 MJ ME/kg DM. Trees showed higher CP contents than shrubs though CHO was higher for shrubs, especially at high altitude (P<0.05). Farmers' scores for nutritive value were positively correlated with CP content of IFTS (r = 0.36; P<0.05). Even though the association was negative for CHO content (P<0.01; r = -0.32), these scores were higher at high altitude (P<0.05). A negative relationship was observed between CT and TA, CP, DMD, OMD, ME and TDN (P<0.05).

It was concluded that although variation within shrubs and within trees was high – CP was higher in trees than in shrubs and lower CHO in trees than shrubs, therefore warranting further research in the added value for ranging ruminants' nutritional status of providing fodder tree material instead of only access to pasture and shrubs. Farmers' perception of nutritive value of IFTS was partly associated with protein content, but other unidentified factors were contributing to their preference. Geographical differences exert shifts in the perceived and analyzed nutritive value of IFTS, thus care should be taken when developing recommendations for the use of IFTS in an entire region.

Keywords: tannin, fodder trees and shrubs, in vitro digestibility, nutritive value, total digestible nutrients

Introduction

Inadequate feed supply, both in quality and quantity, is a major constraint of ruminant livestock production in many southern parts of the world (De Leeuw 1995; Adugna et al. 2000; IFAD, 2008). In the Gilgel Gibe catchments of Southwest Ethiopia, this is incredibly prevalent and acute in the dry season. Crop by-products, available during dry season, have a low nutritive value due to low protein and fermentable energy, despite their rapid growth during the period of heavy rainfall; in addition, high temperature leads to grass maturing early before the dry season. This obviously adds to the poor performance of ruminant livestock. Improvement of the productive and reproductive performance of smallholders' ruminants warrants methods of extending the availability and quality of local feedstuffs produced on smallholder farms. One potential way (Solomon 2004; Mekoya et al. 2008) for increasing the quality and availability of feeds for smallholder ruminant animals in the dry season may be through the use of various locally available fodder trees and shrubs (IFTS). Fodder trees and shrubs of tropical origin are important in livestock production because they can supply significant amounts of protein. Unfortunately, their content of anti-nutrients like condensed tannins (CTs) vary widely and unpredictably (Babayemi et al. 2004b). Their effect on animals ranges from beneficial to toxicity and death (Makkar et al. 2003). A first step in the targeted use of shrubs and trees as feed resource is the analysis of their nutritive value, and identification of environmental factors that may affect their nutritive value.

As it is described in many studies that plant species, farming systems, soil types, feed quality and availability, and many more characteristics vary between geography (ILCA 1990; Ayana and Barrs, 2000; Holecheck et al. 2005) and feeding season (Zarazaga et al. 2009; Kim et al. 2006). Studying combined effects of geography, season and species diversity could give a clue for farmers that can design a feeding strategy based on locally available feed resources.

A variety of FTS are growing in the Gilgel Ghibe basins of Ethiopia, mainly due to the suitability of the environment and the need to use them as fuel wood, construction, mulch, and shade for cash crops like coffee and spices. They replenish soil fertility, are sources of human and veterinary medicine, and also serve as environmental conservation. There is limited information regarding the effect of species diversity, sites of growing, and their interaction effect on farmers feed preference traits, chemical composition, digestibility and energy densities of IFTS with variable CT contents.

Objectives

The present study was undertaken to

Determine and compare the nutritive value of fodder trees versus fodder shrubs at different sites and
Relate the nutritive value information to farmers' traditional feed resource preferences.

Material and Methods

Location and duration

The survey was carried out in two distinct locations in the Omo-Ghibe river basin of Ethiopia. The climate of the area is characterized from arid to humid tropical with bimodal rainfall. Farmers in the area carry out mixed crop-livestock agriculture. As a consequence of the high population density, as much as 90% of the land is cultivated. Livestock production is characterized by traditional smallholders that are kept mainly in severely overgrazed private and communal rangelands throughout the year. Multipurpose trees and shrubs local to the region as well as many tropical regions are becoming valuable feed supplements to livestock species.

Data collection and analytical techniques

The data were collected between January and February 2011 through a cross-sectional field survey following a series of sampling procedures. A reconnaissance survey was conducted to have a notion of understanding about the study area and to select the representative study sites before getting to participatory rural appraisal (PRA) and structured questionnaire. Thereafter, the study area was systematically stratified into two regions based on altitude variations: low altitude region (LAR, 1600-1800 metres above sea level (m.a.s.l)) and high (HAR, 2001-2200 m.a.s.l). Measurement of boundaries of each altitude stratum was done by geographic positioning system (GPS). A total of 360 knowledgeable elder farmers (324 men and 36 women, 180 farmers from each strata), aging between 40 and 69 were included from 12 selected peasant associations (PA) were interviewed. The knowledgeable elders were selected with the help of workers in the Agricultural Development office (DA) at each PA. Feed value preference scoring was done for all identified IFTS on a point scale from 1 (not preferred) to 4 (highly preferred) (Kuntashula and Mafongoya 2005). In view of these authors, feed value (nutritive value) preference score was based on palatability by animals, improvement of body condition, growth and milk production, improvement of intake of straw diets, improved health of animals whilst preference score for growth and re-growth potential used criteria like growth rate after establishment, re-growth potential after frequent cutting or looping. Farmer’s preferences on compatibility mainly focused on absence of competition with crops on available soil nutrients and moisture improve soil fertility; improve growth of below canopy of annual and perennial crops. Multifunctionality indices were used for timber, poles and other local constructions, fuel wood, fence, medicinal value, shade tree, source of honey, soil stabilization, and farm implements.

Main questions were related to names of IFTS, their season of use, plant parts given to animals, species of animals fed, features of their availability, feeding system and calendar, farmers’ indigenous knowledge of feed value preference, and ecological functions. For ethical purposes, data were collected with permission of the informants and knowledge of the local administration.

Three individual plants per species and location were sampled. Leaf samples of browse species were ground and analyzed for dry matter (DM,105°c for 16 hours), organic matter (% OM,100-% crude ash), crude protein (CP, NÍ6.25), crude ash (CA, 550^oc for 8 hours) and ether extract (EE) according to the standard procedures of AOAC (2005). Neutral detergent fiber (NDF) and acid detergent fiber (ADF) were determined sequentially by the method of Van Soest et al. (1991). NDF and ADF values expressed inclusive of residual ash. Lignin (ADL) was determined by solubilization of cellulose with H₂SO₄(Van Soest and Robertson 1985). Hemicellulose (% HC) was calculated from the difference between % NDF and % ADF. Determination of total condensed tannins (CT) was based on oxidative depolymerization of CTs in butanol-HCl reagent using 2% ferric ammonium sulfate in 2N HCl catalyst (Porter et al.1986).

Metabolizable energy (ME, MJ/kg) value was estimated from the in vitro organic matter digestibility: ME = 0.16Í % OMD according to McDonald et al. (2002). The two stage in vitro technique developed by Tilley and Terry (1963) was used to determine in vitro dry matter digestibility (DMD) and OMD of the feeds with some slight modifications. About 0.5 g of milled sample (1 mm sieve) was weighed into a test tube. A 10 ml rumen liquor and 50 ml buffer solution was added to the sample in the tube. The mixture was incubated at 39^oC for 48 h, ensuring that it was carefully shaken from time to time. Finally, the tubes were centrifuged and the supernatant decanted. The residue was again incubated with 60 ml pepsin-hydrochloric acid solution (to digest protein) for another 48h at 39^oC. This was followed by centrifugation, filtering, drying the residues and ashing. Two blanks (rumen liquor mixed with buffer only) and two standards with known digestibility were included to correct for the indigestible DM from the rumen liquor and to check whether the system was working perfectly.

The gross energy (GE) and digestible energy (DE) of feeds were calculated using equations from Hveplund et al. (1995) as follows:

GE (MJ/kg DM) = CP Í 24.237 + EE Í34.116 + CHO Í17.300

DE (MJ/kg DM) = dCP Í 24.237 + dEE Í 34.116 + dCHO (kg/kg DM) Í 17.300

where

dCP = digestible CP (kg/kg DM) = (0.93 Í % crude protein in DM -3)/100

dEE = digestible EE (kg/kg DM) = (0.96 Í % crude fat in DM -1)/100

dCHO = digestible CHO (kg/kg DM) = (digestible OM/100 Í 100-% crude ash in DM)/100.

The total carbohydrate (% CHO) was calculated as: % total CHO = 100-(% CP + % EE + % Ash + % lignin).

The content of total digestible nutrient (%TDN) per kg and per kg DM of a feedstuff was calculated as follows (Ranjhnan 2001): TDN, kg = kg digestible crude protein (DCP) + 2.25Íkg digestible ether extract (DEE) + kg digestible carbohydrates (DCHO).

Statistical analysis

A two-way variance analysis was carried out to see mean differences between two altitudes, between fodder trees and shrubs and their interaction through a 2Í2 factorial arrangement for farmers feed preference score values, chemical analyses, OMD, DMD, digestible nutrients and energy contents. These analyses were performed using GLM procedures of SAS (SAS 2010 version 9.3) computer software following the model Y_ij= µ + L_i+ P_j + M_ij+ å_ij, with Y_ij: response variable (feed preference and nutritional quality); µ: overall mean effect; L_i: i^th altitude effect; P_i: j^th effect of plant nature (trees and shrubs); M_ij: interaction effect between altitude and plant nature and å_ij is the random error. Duncan’s multiple range tests were used for mean separation. Mean differences were considered significant at P<0.05. To establish the magnitude of relationships that exist, if any, between farmers’ assessment of IFTS feed value scores with the relative assessments derived from laboratory-based indications of feed quality, Spearman’s rank correlation analysis was carried out independently for each district.

Results

Fifty IFTS species were identified by the interviewed farmers: A majority of the IFTS was used in the dry season, especially the leaves (Table 1a and 1b). There was a wide spread in the number of respondents that acknowledged the use of a particular species as feed resource: for example, the shrub Vernonia amegdalina was reported by 85.8% whereas another shrub in the same high altitude region, Tagetes minuta, was only listed by 15.6% of the respondents. In addition to leaves, farmers collected fruits of Ficus species and pods of legume trees and shrubs as a fodder source. This was witnessed by 40.2% and 43.2% of respondents, respectively. However, no grinding or any other physical treatment was reported to be practiced for the purpose of improving the feeding value of the leaves, fruits and pods in both of the farming locations. Reasons given to the question as to why they did not process (not wilt/dry/grind) the plant parts varied. None of the respondents knew if this could be of value in feeding practices. The highest feed preference value (score 4) was given for 12 IFTS species whereby 63.3% of respondents gave the highest mean score, i. e. 63.3% of all the respondents well knew and utilized the species in ruminant feeding. Accordingly, these species were highly preferred and their perceived nutritional value was considerably different from the rest of the 38 IFTS. However, the lowest preference for feed/nutritive value (score 1) was given to 29 IFTS species out of 50. Little or no overlap was observed between IFTS highly ranked for their multiple functions (in addition to fodder source, e.g. for shade of cash crops, soil erosion control, construction, timber, fuel wood, live fence) on the one hand, and compatibility on the other hand.

Table 1a. Fodder trees and shrubs (N pooled to 360; feed preference score 1 to 4) with their characteristics as identified and perceived by farmers in the high altitude region
	Plant species	Parts used	Feeding season	% Respondents	Nutritive value score	Growth rate score	Biomass score	Compatibility score	Multifunctionality score

Shrubs
	Calpurnia subdecandra	Leaf	Dry	54.2	2.81	3.83	3.13	2.81	3.53
	Clausena anisata	Leaf	Dry	24.2	2.10	4.00	2.30	2.55	1.80
	Erythrina brucei	Leaf	Dry	33.3	2.10	3.90	3.78	3.84	2.73
	Ficus sur	Leaf	Dry	26.4	2.42	3.11	3.02	3.20	2.05
	Lippia adoensis	Leaf	Dry	16.7	2.11	2.84	2.41	2.11	2.51
	Myrsine Africana	Leaf	Dry	33.3	2.86	3.87	2.20	3.75	2.45
	Phytolacca dodecandra	Leaf	Dry	39.7	1.80	3.87	3.78	3.94	3.01
	Rungia grandis	Leaf	Dry	27.8	2.30	3.31	3.45	4.00	3.95
	Satureja calmintha/spp	Leaf	Rainy	77.2	3.71	3.33	1.33	2.11	1.50
	Tagetes minuta	Leaf	Rainy	15.6	1.13	2.51	3.55	2.04	3.65
	Vernonia amegdalina	Leaf	Dry+rainy	85.8	3.89	3.12	2.20	3.95	3.91
Trees
	Arundinaria alpine	Leaf	Dry+rainy	41.5	3.50	3.20	2.81	2.70	2.60
	Buddleja polystachya	Leaf	Dry+rainy	52.5	3.10	2.17	2.75	2.12	2.05
	Dodonaea angustifolia	Leaf	Dry	40.3	1.25	2.51	3.13	2.71	2.43
	Dombeya torrida	Leaf	Dry+rainy	33.3	3.83	3.70	2.70	1.70	2.71
	Draceana steudneri	Leaf	Dry+rainy	55.6	3.91	3.31	2.88	2.45	2.61
	Ensete ventricosum	Leaf, stem	Dry+rainy	83.1	3.90	3.83	3.73	3.81	3.55
	Hagenia abyssinica	Leaf	Dry	55.6	3.88	3.43	3.10	4.00	3.46
	Maesa lanceolata	Leaf	Dry+rainy	33.3	2.37	1.51	3.30	2.12	1.81
	Millettia ferruginea	Leaf, pod	Dry+rainy	77.8	3.92	2.11	2.14	2.51	2.82
	Olinia rochetiana	Leaf	Dry	27.2	1.28	4.00	3.41	2.10	1.44

Table 1b. Fodder trees and shrubs (N pooled to 360; feed preference score 1 to 4) with their characteristics as identified and perceived by farmers in the low altitude region
	Parts used	Feeding season	% Respondents	Nutritive value score	Growth rate score	Biomass score	Compatibility score

Shrubs
Carica papaya²	Leaf	Dry	33.3	1.03	3.30	1.30	2.73
Catha edulis	Leaf	Dry+rainy	79.7	2.42	3.55	3.87	2.16
Celtis africana	Leaf	Dry	19.2	1.30	4.00	1.10	3.35
Coffee arabica	Leaf	Dry	33.3	2.20	4.00	3.80	3.80
Coffee Arabica¹	Husk	Dry	33.3	1.13	2.30	3.00	1.30
Ekebergia capensis	Leaf	Dry	24.7	1.13	2.43	2.00	2.05
Maytenus obscura	Leaf	Dry	51.9	1.50	2.42	2.30	2.82
Morus alba	Leaf	Dry+rainy	24.7	3.10	2.35	2.40	1.50
Myrica salicifolia	Leaf	Dry	33.3	1.01	3.31	2.31	2.25
Ocimum lamiifolium	Leaf	Dry	30.3	1.83	3.92	3.33	2.20
Premna schimperi	Leaf	Dry	48.6	2.74	1.51	3.51	3.11
Rhamnus staddo	Leaf	Dry	24.4	1.47	2.30	3.30	2.20
Rhus glutinosa	Leaf	Dry	33.3	2.35	4.00	1.50	1.55
Salix subserrata	Leaf	Dry	54.2	1.53	2.17	3.11	3.25
Sida tenuicarpa	Leaf	Dry+rainy	55.6	2.49	3.52	1.52	3.25
Trees
Acacia abyssynica	Leaf	Dry	21.1	1.21	1.83	2.20	2.55
Albizia gummifera	Leaf, pods	Dry	46.7	2.10	3.92	3.02	2.10
Carissa edulis	Leaf	Dry+rainy	63.9	1.22	3.06	2.88	2.78
Cordia africana	Leaf	Dry	55.3	2.48	3.21	3.00	2.10
Erythrina abyssyinica	Leaf	Dry	25.8	2.11	2.80	3.78	2.21
Euclea divinorum	Leaf	Dry	24.2	2.51	1.74	3.74	3.74
Ficus ovata	Fruit, leaf	Dry	27.8	2.53	3.41	3.64	3.01
Ficus sycomorus	Leaf	Dry	33.3	1.48	2.81	3.13	4.00
Ficus thonningii	Leaf, fruit	Dry+rainy	80.3	3.87	3.46	3.75	3.78
Ficus vasta	Fruit, leaf	Dry	33.3	2.73	1.90	3.81	4.00
Grewia ferruginea	Leaf	Dry+rainy	72.8	3.93	4.00	3.51	3.85
Prunus africana	Leaf	Dry	33.3	1.23	4.00	3.20	1.86
Sapium ellipticum	Leaf	Dry+rainy	76.4	3.75	3.42	3.10	2.81
Syzygium guineense	Leaf	Dry+rainy	53.3	3.85	3.90	2.90	2.56

The number of farmers identifying a species as IFTS was only associated with its nutritive value score (r = 0.54; P<0.01). Correlations with the other preference scores were not significant (Table 5). Farmers at high altitude attributed a higher nutritive value score to IFTS than in the low altitude region (Table 2). At both the low and high altitudes, farmers gave a moderately higher appreciation of nutritive value for trees in comparison with shrubs (P<0.05).

Table 2. Subjective scores of utilization traits of fodder trees versus fodder shrubs by interviewed farmers (N pooled to 360) compared at low and high altitude (score 1 = low; score 4 = high).
Score (1-4):	Low altitude		High altitude		SEM	Low vs. High	P
Score (1-4):	Shrubs	Trees	Shrubs	Trees	SEM	Low vs. High	Shrub vs. Tree	Interaction

Nutritive value	2.10	2.14	2.38	3.30	0.14	**	NS	**
Growth rate	3.07	2.98	3.12	3.29	0.10	NS	NS	NS
Biomass	2.90	2.97	2.91	2.81	0.10	NS	NS	NS
Compatibility	2.47	2.72	2.97	3.12	0.11	*	NS	NS
Multi-functionality	2.47	2.46	2.57	2.82	0.12	NS	NS	NS
SEM, standard error of means; P<0.05; * P<0.01; NS= non significant*

No influence of altitude or plant type could be identified on the farmers' scoring of growth rate, biomass and multifunctionality. Yet, scores for compatibility were higher at high altitude and the benefit of trees over shrubs was most pronounced at low altitude (P<0.05). On the other hand, the interaction effect of altitude and plant type was also found to be significant for nutritive value score (P<0.05).

Table 3a and b displays the wide range in analysed nutritive value, e.g. On DM basis, analytical results ranged between 66 to 242 g CP/kg, 185 to 502 g NDF/kg, 0.1 to 228 g CT/kg DM, 331 to 961 g OMD/kg, 282 to 908 g DMD/kg, 478 to 745 g CHO/kg, 5 to 15 MJ ME/kg and 332 to 963 g TDN/kg. Yet, ADF content also tended to be higher in fodder trees at high altitude. Gross energy (GE) was lowest in shrubs at low altitude, but this difference disappeared when estimating digestible energy. Interaction effects of altitude versus plant types resulted significant variation in GE content (P<0.05). Influence of plant nature or altitude on other nutrients could not be identified.

Table 3a. Potential nutritive value of the edible parts of fodder trees and shrubs in the Gilgel Gibe catchment, southwest Ethiopia at low altitude region
Plant Species	DM	TA	CP	EE	NDF	ADF	ADL	CT	OMD	DMD	ME	CHO	DCHO	DEE	DCP	GE	DE	TDN

Acaccia abyssynica	946	90	178	25	315	272	51	2.4	627	613	10	657	627	2	16	16	11	629
Albizia gummnifera	948	82	206	45	276	248	116	72	398	346	6	552	398	4	19	13	8	400
Cordia africana	946	106	242	42	345	330	36	0.5	455	393	7	574	455	4	22	15	9	458
Croton macrostachyus	948	93	231	32	306	290	111	0.7	606	587	10	534	606	3	21	17	11	609
Ekebergia capensis	948	26	143	15	185	88.4	78	80	961	908	15	738	961	1	13	21	17	963
Erythrina abyssyinica	942	97	240	51	332	302	112	3	421	408	7	500	421	5	22	15	8	424
Euclea divinorum	945	92	140	19	336	275	151	81	670	656	11	597	669	2	13	16	12	671
Ficus ovata	914	123	186	27	444	302	99	191	456	414	7	566	456	3	17	13	9	459
Ficus sycomorus	943	114	172	20	399	348	73	110	464	421	7	621	464	2	16	13	8	466
Ficus vasta	946	99	186	14	346	303	99	8	523	437	8	603	523	1	17	14	10	525
Prunus africana	901	96	137	50	352	331	68	76	649	625	10	679	648	5	10	16	12	650
Sapium ellipticum	908	69	130	19	318	269	54	2	653	610	10	728	653	2	12	15	12	654
Syzygium guineense	911	84	126	38	503	465	94	172	385	350	6	658	385	4	11	11	7	387
Carica papaya²	832	136	92	29	313	268	56	0.7	788	723	13	688	788	3	8	17	14	790
Calpurnia subdecandra	901	77	205	37	286	194	46	1	366	358	6	636	366	4	19	13	7	369
Carissa edulis	905	92	136	41	340	273	68	164	393	356	6	663	393	4	12	12	7	395
Catha edulis	936	69	102	30	417	259	106	114	705	673	11	693	705	3	9	16	13	706
Coffee arabica	929	87	84	28	452	341	107	13	443	405	7	694	443	3	8	11	8	445
Coffee arabica¹	891	41	66	19	465	401	129	1	337	282	5	745	337	2	6	8	6	338
Dodonaea angustifolia	946	85	140	35	314	251	86	89	638	624	10	654	638	3	13	16	12	640
Maytenus obscura	913	68	151	35	408	345	115	228	453	387	7	631	453	3	14	13	8	455
Morus alba	912	185	134	36	310	268	87	1	702	674	11	558	702	3	12	17	13	704
Myrica salicifolia	926	132	138	45	483	362	105	0.9	409	363	7	580	409	4	13	12	8	412
Ocimum lamiifolium	946	104	216	38	352	331	68	0.1	348	339	6	574	348	4	20	13	7	351
Premna schimperi	946	95	183	27	328	285	146	67	672	659	11	549	671	3	17	17	12	674
Rhamnus staddo	915	143	85	35	370	333	76	3	331	325	5	662	331	3	8	9	6	332
Rhus glutinosa	946	83	144	41	313	253	55	188	355	334	6	676	355	4	13	11	7	357
Sida tenuicarpa	881	120	131	21	327	217	56	1	721	666	12	672	721	2	12	16	13	723
Saccharum officinarum	948	93	231	32	306	290	111	0.7	522	461	9	534	522	3	21	16	10	525
CP, crude protein; EE, ether-extract, NDF, neutral detergent fiber; ADF, acid detergent fiber; ADL, acid detergent lignin; Hemi, hemicellulose; CT, total condensed tannins; OMD, in vitro organic matter digestibility; DMD, in vitro dry matter digestibility; ME, metabolisable energy

Table 3b. Potential nutritive value of the edible parts of fodder trees and shrubs in the Gilgel Gibe catchment, southwest Ethiopia at high altitude region
Plant Species	DM	TA	CP	EE	NDF	ADF	ADL	HC	CT	OMD	DMD	ME	CHO	DCHO	DEE	DCP	GE	DE	TDN

Dombeya torrida	947	100	238	25	331	321	112	10	6	593	580	10	525	593	2	22	17	11	596
Draceana steudneri	946	104	222	24	353	328	107	26	53	667	649	11	543	667	2	20	18	12	669
Erythrina brucei	947	103	238	61	329	317	120	12	2	455	412	7	478	455	6	22	16	9	459
Ficus sur	945	94	145	28	351	287	94	64	7	461	459	7	639	461	3	13	12	8	462
Ficus thonningii	892	125	115	42	484	388	101	95	1	628	603	10	617	628	4	10	15	11	630
Grewia ferruginea	947	96	229	29	317	300	93	17	52	529	491	9	553	529	3	21	16	10	532
Hagenia abyssynica	946	95	175	23	336	286	89	50	37	573	553	9	619	573	2	16	15	10	575
Millettia ferruginea	957	88	241	31	415	380	146	35	73	508	489	8	494	508	3	22	16	9	511
Vernonia amegdalina	946	105	228	45	345	317	97	28	2	585	543	9	526	585	4	210	17	11	588
Arundinaria alpine	900	110	130	55	484	363	143	121	0.1	376	368	6	562	375	5	12	12	7	378
Buddleja polystachya	943	114	161	17	396	342	146	55	5	681	653	11	562	681	2	15	16	12	683
Celtis africana	945	107	194	40	370	333	76	37	82	455	393	7	584	455	4	18	14	8	458
Clausena anisata	934	77	164	28	340	273	68	67	7	653	610	10	663	653	3	15	16	12	655
Ensete ventricosum	944	118	190	15	387	350	57	38	0.1	531	433	9	621	531	1	17	14	10	533
Lippia adoensis	946	104	216	31	352	331	58	21	1	667	630	11	591	666	3	20	18	12	669
Maesa lanceolata	923	91	203	72	346	305	65	41	68	428	396	7	569	428	7	19	15	8	431
Myrsine africana	944	99	129	37	353	297	48	56	3	552	513	9	687	552	4	12	14	10	554
Phytolacca dodecandra	935	92	107	27	354	242	134	119	76	647	611	10	640	647	3	10	15	12	649
Rungia grandis	947	96	237	14	314	303	89	11	54	654	634	11	564	654	1	22	18	12	656
Salix subserrata	920	60	151	64	453	355	116	98	66	436	397	7	609	436	6	14	13	8	439
Satureja calmintha/spp	942	114	122	35	418	334	73	84	0.2	682	632	11	656	682	3	11	16	12	684
Tagetes minuta	945	92	143	29	332	269	64	62	0.1	648	625	10	672	648	3	13	16	12	649

Fodder trees showed a significantly higher protein content than fodder shrubs, with a tendency to be more pronounced at high altitude (P<0.05) (Table 4).

Table 4. Nutritive value analysis of fodder trees versus fodder shrubs in the Gilgel Gibe catchment (southwest Ethiopia) at low and high altitude.
*Nutritive value*	Low altitude		High altitude		SEM	P
*Nutritive value*	Shrubs	Trees	Shrubs	Trees	SEM	Low vs. High	Shrub vs. Tree	Interaction

DM, g/kg	923	927	938	938	3	NS	NS	NS
Ash, g/kg	94	93	103	98	3	NS	NS	NS
OM, kg	902	907	906	898	3	NS	NS	NS
CP, g/kg	143	170	166	193	7	NS	*	*
EE, g/kg	33	30	35	35	2	NS	NS	NS
NDF, g/kg	365	341	365	376	9	NS	NS	NS
ADF, g/kg	294	292	306	330	8	NS	NS	NS
ADL, g/kg	91	85	77	110	4	NS	NS	*
Hemi, g/kg	71	49	60	46	5	NS	NS	NS
CT, g/kg	58	57	28	30	8	NS	NS	NS
IVOMD, g/kg	493	575	590	543	19	NS	NS	NS
IVDMD, g/kg	460	535	548	516	19	NS	NS	NS
GE, MJ/kg	13	15	14	15	0.3	NS	NS	*
DE, MJ/kg	9.0	10.4	10.7	9.9	0.3	NS	NS	NS
ME, MJ/kg	7.9	9.2	9.4	8.7	0.3	NS	NS	NS
CHO,g/kg	628	590	632	592	9	*	*	**
DCHO,g/kg	564	533	532	560	19	NS	NS	NS
DEE,g/kg	3.3	3	3.2	3	1.7	NS	NS	NS
DCP,g/kg	14.2	16.5	14	16.5	6.4	*	*	*
TDN, g/kg	495	577	592	545	19	NS	NS	NS
DM, dry matter; TA, total ash; OM, organic matter; CP, crude protein; EE, ether-extract, NDF, neutral detergent fiber; ADF, acid detergent fiber; ADL, acid detergent lignin; HC, hemicellulose; CT, total condensed tannins; IVOMD, in vitro organic matter digestibility; IVDMD, in vitro dry matter digestibility; GE, gross energy; DE, digestible energy; ME, metabolisable energy; TDN, total digestible nutrients; =P<0.05; *=P<0.01;NS= non significant

The nutritive value score showed a mild positive correlation (Table 5) with the analyzed CP concentration (r = 0.36; P< 0.05), but not with other analyzed and derived nutrient concentrations, except for a weak positive correlation with ADF concentration (r = 0.24; P<0.01). None of the other scores showed significant correlations with the analyzed nutrient concentrations.

Discussion

The fairly high number of reported IFTS and the high percentage of respondents listing several IFTS, indicates that the use of IFTS is widespread in the study region, and agrees with other studies in similar areas by Aucha (2004) and Ntakwendela (2003) in Kenya, Samanya (1996) of Uganda, and Backlund and Bellskong (1991) and Komwihangilo et al. (1995) in Tanzania. The wide range in the measured nutrient and energy content among the identified IFTS presents a challenge to formulate simple recommendations on the use of fodder trees and shrubs. For the most part, farmer perceptions were neither correlated with chemical essay, nor with energy content and digestibility value. This could be due to the fact that the population in the study area has been mainly devoted to crop agriculture, and has had less experience with nutritive value parameters devised based on laboratory analyses. Roothaert and Franzel (2001) and Kuntashula and Matongoya (2005) reported that farmer’s preference and use of local multipurpose fodder trees and shrubs based on roles of fodder trees in improving animal productivity, multiple uses of fodder trees and shrubs, ever increasing trends of soil erosion that affects availability of feed resources and unpredictable climate changes that are reversing the performance of livestock in the region.

Other surveys on fodder trees also observed this heterogeneity in nutrient composition (Kaitho et al. 1996; Mekoya et al. 2008). The magnitude of variation within the group of shrubs or within the group of trees likely overruled many of the potential differences between shrubs and trees. Nevertheless, trees were shown to have a higher protein concentration than shrubs, especially at high altitude. Studies on free-ranging cattle in Africa have previously demonstrated that protein is a crucial limiting nutrient, especially in the dry season (Zinash et al. 1995; Abdulrazak et al. 2000; Solomon 2002). Therefore, trees might have a benefit over shrubs as a feed resource in such situations. Whereas shrubs are available to the animals without human intervention, cattle usually have no access to the edible parts of trees, and their owners need to harvest this material in order to be available to the animals. Without knowing the exact relative contribution of protein and other nutrients of IFTS to the total diet of animals, it is difficult to estimate the nutritional importance of the difference in protein between trees and shrubs. The present data therefore warrant further investigation to quantify the added value of harvesting fodder trees on the nutritional status of ranging cattle in regions where fodder shrubs are present. Many studies have documented the importance of fodder trees as protein sources for free-ranging ruminants (Osuji and Odenyo 1997; Thorne et al. 1999; Sumberg 2002), but this is the first study to our knowledge that suggests a practical advantage of actively feeding fodder tree parts over fodder shrubs to optimise the protein provision to free-ranging livestock.

Although many of the seemingly important analysed features, such as energy content, digestibility and tannin load, were not reflected in the perception of the farmers (as determined with the scoring system), their estimation of nutritive value to some extent correlated with the protein content of the IFTS. This suggests that farmers throughout the years have unintendedly identified protein as a limiting factor in the dry season diet of their animals, although it is clear from the fairly low correlation coefficient that other factors still affected their judgment that might be worthwhile being identified. Higher condensed tannin levels (CT<50g/kg DM, total 26 out of 50 IFTS) become highly detrimental (Barry and Manley, 1984) as they reduce digestibility of fiber in the rumen (Reed et al. 1985) by inhibiting the activity of bacteria (Chesson et al. 1982) and anaerobic fungi (Akin and Rigsby 1985), high levels also lead to reduced intake (Merton and Ehle, 1984 cited by Leng 1997). According to Brooker et al. (1999), livestock consuming tannin-rich diets over 50g CT/kg DM usually develop a negative nitrogen balance and lose weight and body condition unless supplemented with non-protein nitrogen, carbohydrate and minerals. It is therefore remarkable that the CT content of the IFTS was not identified as a factor related to the preference of the interviewed farmers (Fig 1 and 2). Perhaps, the huge variability in CT content among the IFTS within altitude and plant nature did not allow to separate such effects, and there might also have been a trade-off between CP and CT, in a way that IFTS can combine high protein levels with high tannin content, as seen in the present study (e.g. Ficus ovata) and others (Makkar 2003).

The altitude factor included in the present study was obviously not just a reflection of differences in meters above sea level as such, but implied differences in soil type, erosion status, management system, and other related aspects. Taken this into account, the results still demonstrate that both the perceived and the analysed nutritive values of IFTS can be affected by geographical differences within the same catchment, and that for instance the added value of using fodder trees will depend on these circumstances. For instance, significant interactions or tendency to interactions between plant nature and altitude were observed for CP, ADL and GE; protein and energy are generally considered as the primary aspects of nutrition, and ADL is commonly judged as a negative factor for digestibility (Preston 1995; McDonald et al. 2002).

Farmers at high altitude not only gave a higher score for nutritive value, but also for compatibility. Although they perceived this difference in compatibility between shrubs and trees within an altitude region, it did not seem to affect their preference, hence apart from nutritive value, other factors not identified in this study will have affected the preference of farmers for particular IFTS. Probably, sociological factors will have been in play as well, such as tradition, but maybe also lack of knowledge on the potential use of certain species, and methods to improve nutritive value through available processing techniques, as has been demonstrated by Thapa et al. (1997) and Thorne et al. (1999). We should also acknowledge that other nutrients than the ones studied here might be of importance as limiting factor for animal performance. Previous work in the Gilgel Gibe catchments demonstrated the wide occurrence of trace element deficiencies in cattle, e.g. copper, that might affect nutrient utilization (Dermauw et al. submitted)

Conclusion

Acknowledgements

The authors gratefully acknowledge the VLIR-IUC Institutional University Cooperation Program for funding the research budget. Animal Nutrition Department of Holeta Agricultural Research Center and Jimma University, Ethiopia are also duly acknowledged for providing different research materials and facilities for the study.

References

Abdulrazak S A, Fujihara T, Ondiek J K and Ørskov E R 2000 Nutritive evaluation of some acacia tree leaves from Kenya. Animal Feed Science and Technology 85: 89–98.

Adugna Tolera, Roger C M, Arthur L G, Tilahun S and Tegene N 2000 Nutritional constraints and future prospects for goat production in east Africa. In: R.C. Merkel, G. Abebe and A.L. Goetsch (eds.). The Opportunities and Challenges of Enhancing Goat Production in East Africa. Proceedings of a conference held at Debub University, Awassa, Ethiopia .

Akin D E and Rigsby L L 1985 Influence of phenolic acids on rumen fungi. Agronomy Journal 77:180–182.

AOAC 2005 Official methods of analysis of the official analytical chemists, 2 vol., 18^th Edition. Horwitz W, Ed. Association of Official Analytical Chemists (AOAC), Arlington, V A, USA.

Aucha J, Kabirizi J, Nyanganga J, Moges Y, Kingamkono M and Kitalyi A 2006 Research and development on indigenous fodder shrubs and trees in eastern Africa. Nairobi, Kenya: RELMA and ICRAF, pp 8.

Babayemi O J, Demeyer D and Fievez V 2004b. Nutritive value and qualitative assessment of secondary compounds in seeds of eight tropical browse, shrub and pulse legumes. Communication in Agriculture Applied Biological Sciences 69 (1):103–110.

Backlund M and Belskog J 1991 The Role of Trees and Shrubs in Livestock Production in Central Tanzania. A survey of their nutritive value during the dry season. Swedish University of Agricultural Sciences. International Development Centre. Uppsala, pp 24.

Barry T N and Manley T R 1984 The role of condensed tannins in the nutritional value of Lotus pedunculatus for sheep. 2. Quantitative digestion of carbohydrate and Protein. British Journal of Nutrition 51:493-504.

Brooker J D, O’Donovan L, Skene I and Sellick G 1999 Proceedings of the 8^th International Symposium on Microbial Ecology Bell CR, Brylinsky M, Johnson-Green P (ed) Atlantic Canada Society for Microbial Ecology, Halifax, Canada.

Chesson A, Stewart C S and Wallace R J 1982 Influence of plant phenolic acids on growth and cellulolytic activity of rumen bacteria. Applied Environmental Microbiology 44:597–603.

De Leeuw P N and Rey B 1995 An analysis of the current trends in the distribution of Ruminant livestock in the Sub- Saharan Africa. World Animal Review 83: 43-59.

Dermauw V, Belay D, Yisehak K, Solomon D, Sileshi B andJanssens G P J (Submitted). Nutritional status of cattle around the Gilgel Gibe dam in Ethiopia: the search for easy-to-establish indicators.

Food and Agriculture Organization of the United Nations 2007 Tropical Crop-Livestock Systems in Conservation Agriculture. The Brazilian Experience. Rome: FAO.

GOR 2006. Socio-economic profile of the Jimma zone. GOR (Government of the Oromia region of Ethiopia), Addis Ababa, Ethiopia.

Holechek J L, Pieper R D and and Herbel C H 2004 Range management: Principles and practices (5^th ed.). Upper Saddle River, NJ: Pearson Education, Inc.

Hvelpund T, Andrieu J, Weisbjerg M R and Vermorel M 1995 Prediction of the energy and protein value of forages for ruminants. In: Recent developments in the Nutrition of Herbivores (Jounet M, Grenet E, Farce H-M, Theriez M, Dermaquilly C, Eds.). Proceedings of the IVth International symposium on the Nutrition of Herbivores. INRA Editions, Paris, 205-227.

IFAD (International Fund for Agricultural Development) 2008 Improving Crop-Livestock Productivity through Efficient Nutrient Management in Mixed Farming Systems of Semi-arid West Africa. http://www.ifad.org/lrkm/tags/384.htm.

ILCA 1991 Livestock Research Manual. Vol. 1. International Livestock Center for Africa (ILCA) working Paper 1. ILCA, Addis Ababa, Ethiopia.

JKim J D, Kim SG, Abuel S J, Kwon C H, Shin C N, Ko K H and Park B G 2006 Effect of Location, Season, and Variety on Yield and Quality of Forage Oat. Asian-Australian Journal Animal Science (19) 7:970 – 977.

Kaitho R J 1997 Nutritive value of browses as protein supplement(s) to poor quality roughages. PhD Theses, Department of Animal Nutrition, Wageningen Agricultural University, Wageningen, The Netherlands.

Komwihangilo D M, Goromela E H and Bwire J M N 1995 Indigenous knowledge in utilization of local trees and shrubs for sustainable livestock production in central Tanzania. Livestock Research for Rural Development 6(3):15.

Kuntashula E and Mafongoya P L 2005 Farmer participatory evaluation of agroforestry trees in Eastern Zambia. Agricultural Systems 84:39–53.

Makkar H P S 2003 Effects and fate of tannins in ruminant animals, adaptation to tannins, and strategies to overcome detrimental effects of feeding tannin-rich feeds. Small ruminant research 49:241-256.

McDonald P, Edwards R A, Greenhalgh J F D and Morgan C A 2002 Systems for expressing the energy value of foods. In Animal nutrition. 6^th edition. PP. 266-283. Longman group Ltd, UK

Mekoya A, Oosting S J, Fernandez-Rivera S and Van der Zijpp A J 2008 Multipurpose fodder trees in the Ethiopian highlands: Farmers’ preference and relationship of indigenous knowledge of feed value with laboratory indicators. Agricultural Systems 96:184–194.

Melaku S, Peters Kurt J P and Azage Tegegne 2004 Supplementation of Menz ewes with dried leaves of Lablab purpureus or graded levels of Leucaena pallida 14203 and Sesbania sesban 1198: effects on feed intake, live weight gain and estrous cycle. Animal Feed Science and Technology 30:39–50.

Ntakwendela L 2003 The effect of substituting sunflower seed cake with Acacia tortilis pods as a protein supplement on the performance of small east African goats. A dissertation submitted in partial fulfillment for the degree of Master of Science in agriculture of Sokoine University of Agriculture.

Osuji O P and Odenyo A A 1997 The role of legume forages as supplements to low quality roughages - ILRI experience. Animal Feed Science Technology 69:27-38.

Porter L J, Hrstich L N and Chan B G 1988 The conversion of procyanidins and prodelphinidins to cyanidin and delphinidin, Phytochemistry 25:223-230.

Preston T R 1995 Tropical animal feeding-a manual for research workers. FAO, animal production and health paper 126. Rome, Italy.

Ranjhnan S K 2001 Animal nutrition in the tropics. 5^th ed. Vikas Publishing house PVT LTD, New Delhi, India.

Reed J D, Horvath P J, Allen M S and Van Soest P J 1985 Gravimetric determination of soluble phenolics including tannins from leaves by precipitation with trivalent ytterbium. Journal of the Science of Food and Agriculture 36:255–61.

Roothaert R L and Franzel S 2001 Farmer s’ preferences and use of local fodder trees and shr ubs in Kenya. Agroforestry Systems 52:239-252.

Samanya J P 1996 Evaluation of feed resources for the zero grazing dairy production system in the fertile Lake Victoria Crescent of Uganda. A case study of Jinja Women heifer project. MSc. Thesis.1996, Mekerere University.

SAS 2008 Statistical analysis system (SAS/STAT program). SAS institute Inc, carry, NC, USA.

Skerman P J and Riveros F 1990 Tropical Grasses. FAO Plant Production and Protection Series No. 23. Rome, Italy.

Solomon M 2002 Evaluation of selected multipurpose trees as feed supplements in teff (Eragrostis teff) straw based feeding of Menz sheep. PhD. Thesis, university of Berlin, Berlin, Germany.

Sumberg J 2002 The logic of fodder legumes in Africa. Food policy 27: 285-300.

Thapa B, Walker D H, and Sinclair F L 1997 Indigenous knowledge of the feeding value of tree fodder. Animal Feed Science and Technology 67: 97-114.

Thorne P J, Subba D B, Walker D H, Thapa D, Wood C D and Sinclair F L 1999 The basis of indigenous knowledge of tree fodder quality and its implications for improving the use of tree fodder in developing countries. Animal feed science and technology 81:119-131.

Tilley J M A and and Terry R R 1963 A two stage technique for in-vitro digestion of forage crops. Journal of British Grassland Society 18:104.

Van Soest P J 1994 Nutritional Ecology of the Ruminant, 2^nd ed. Cornell University Press, New York,USA.

Van Soest P J and J B Robertson 1985 Analysis of forages and fibrous foods. A Laboratory manual for animal science, 613, Cornell University, USA.

Van Soest P J, Robertson J B and Lewis B A 1991 Methods of fiber, neutral detergent fiber and non-starch carbohydrates in relation to animal nutrition. Journal of dairy science 74:3583-3597.

Yisehak K, Becker A and Janssens G P J 2010 Micro mineral status of free-ranging Zebu cattle and their salivary response to dietary tannins. In: Proceedings of 14^th ESVCN Congress, Zurich, Switzerland, pp.121-121.

Yisehak K, Belay D, Taye T and Janssens G P J (xxx) Impacts of soil erosion associated factors on available feed resources for free-ranging cattle at three altitudes: measurements and perceptions. Submitted to journal of Agricultural systems.

Zarazaga L A, Guzmán J L, Domínguez, C, Pérez M C and Prieto R 2009. Effects of season and feeding level on reproductive activity and semen quality in Payoya buck goats. Theriogenology 71:1316–1325.

Zinash S, Seyoum B, Lulseged G and Tadesse T 1995 Effect of harvesting stage on yield and quality of natural pasture in the central highlands of Ethiopia. Addis Ababa, Ethiopia. In: Proceedings of 3^rd National Conference of the Ethiopian Society of Animal Production. 27-29 April 1995. pp. 316-322.