Chemical composition and nutritive value of some
tropical by-product feedstuffs for small
ruminants Ð in vivo and in
vitro digestibility
Eroarome Martin Aregheore
*Department of Agricultural Sciences, Animal Production Unit, College of Education, P.M.B. 1251, Warri, Delta State, Nigeria
Received 5 July 1999; received in revised form 4 February 2000; accepted 23 February 2000
Abstract
Chemical composition and nutritive values of by-product feedstuffs (BPF) used for small ruminants were evaluated. Four BPF from groundnut shells (GNS), maize cobs (MC), cassava peels (CaP), and citrus pulp/peels waste (CPPW) were used to prepare complete diets fed to goats and sheep in separate trials. Twelve goats and 16 sheep 16±18 months old weighing on average 12.960.28 and 14.050.18 kg, were divided into three and four groups, respectively, and used in two separate growth and digestibility studies. In vitro digestibility trial was carried out. In vivo and in vitro data were compared. Chemical composition of BPF differed in nutrient contents. MC had the least crude protein content (3.4%). CaP and CPPW had low NDF and ADF. GNS and MC were higher in cell wall constituents. Voluntary DMI was expressed as g/kg live weight (LW) per day and this ranged between 58.0±62.0 g/kg LW per day in goats and 70.0±83.0 g/kg LW per day in sheep. DM and CP digestibility were better (P>0.05) in goats fed CaP diet compared to the other BPF. NDF digestibility was better (P>0.05) in the GNS fed group compared to MC and CaP groups. GE was lower in CaP fed goats than in the others. Nutrient digestibility of the four by-products was better (P>0.05) in sheep that received CPPW diet. Other by-products (GNS, MC, CaP) digestibility followed the trend observed in goats. Net gas production, DM, NDF and OM digestibility and metabolizable energy (ME, MJ/kg of DM) were signi®cantly different among the BPF (P<0.05). GNS had low net gas production than MC however, gas production was highest in CPPW followed by CaP. The low NDF values in CaP and CPWM indicated that they contain more soluble materials which ruminants can bene®t from. In in vivo trial, goats and sheep performed better with rations
85 (2000) 99±109
*Present address: The University of the South Paci®c, School of Agriculture, Alafua Campus, Private Mail Bag, Apia, Samoa, South Paci®c Region.
E-mail address: aregheore_m@samoa.net (E.M. Aregheore)
formulated from CaP and CPPW compared to other BPF such as MC and GNS. However, the high gas production observed shows that care has to be exercised in feeding them to animals at high levels to avoid accumulation of gases and fermentation products which could lead to displaced abomasum and acidosis. The results indicate that estimates of digestibility are of use for BPF evaluation and these estimates can be readily accomplished by the in vitro techniques available. In conclusion, in vitro OM digestibility for GNS, MC, CaP and CPPW were higher compared to the in vivo digestibility.#2000 Elsevier Science B.V. All rights reserved.
Keywords:By-product feedstuffs; Goats; Sheep; In vivo; In vitro digestibility
1. Introduction
Shortages of feed resources often impose major constraints on the development of animal production in the tropics and sub-tropics. Considerable quantities of crop residues and agro-industrial by products (by-product feedstuffs, BPF) are generated every year in most developing countries in the tropics and sub-tropics. These are suitable for feeding livestock, however, because of lack of technical-know-how they are lost or under utilized (Aregheore and Chimwano, 1991).
An intensive feeding system based on locally available BPF is an alternative promising feeding system to rear ruminants economically. Ruminants because of their rumen physiological adaptation can utilize inexpensive BPF to meet their feed requirements for maintenance, growth, reproduction and production. By-product feedstuffs in the diets of ruminants support growth and lactation and result in the production of human edible food (DePeters et al., 1997). Although most have low nitrogen content, more ®bre as well as low nutrient density, effective processing can raise their nutritive value (Reddy and Reddy, 1992; Aregheore, 1994). The concept of matching ruminant livestock production with available feed resources (Preston and Leng, 1987; Dargie, 1989), has therefore intensi®ed research into more use of crop residues and agro-industrial by products in most countries in the tropics and sub-tropics.
The nutritive value of BPF depends not only on their digestibility, but also on the amount of voluntary intake by an animal. Generally, palatability, seasonal variation and availability are some factors that in¯uenced feed intake by an animal. The nutritive value of BPF can be determined by their chemical composition (Gohl, 1981; Aregheore, 1993) or by a combination of chemical constituents and gas released on incubation of feeds in an in vitro medium containing rumen microbes (Menke and Steingass, 1988). Digestibility may be directly determined in vivo or estimated by using in vitro procedures, which are cheaper and more convenient. Subsequently, (Abate et al., 1984) opined that nutritive values derived in vitro should only be treated as estimates, although such data often provide the twin problems of time and expenses when evaluating large number of samples.
Pirie (1987) hypothesized that if a feed nutrient is readily digested by several enzymes in vitro, it is reasonable to expect that it will be digested in the gut. And if it is not digested in vitro, it may still be digested in vivo as a result of simultaneous actions of several enzymes and of possible co-operation from the rumen ¯ora. It is therefore the intention of this paper to compare in vivo and in vitro digestibility of some selected
tropical BPF commonly used in complete mash diets for small ruminant livestock (goats and sheep) in the tropics.
2. Materials and methods
2.1. Selected by-products feedstuffs
BPF from groundnut shells (GNS), maize cobs (MC), cassava peels (CaP), and citrus pulp/peels waste (CPPW) were used to prepare complete diets and fed to goats and sheep in separate trials. The BPF were also used in in vitro digestibility study (Table 1).
2.2. Preparation of experimental diets
The four BPF were processed and ground in a hammer mill. They were then moistened to contain 60% DM, maintained at 218C for at least 10 days and aerated for 24 h before being used in the preparation of experimental rations. The other ingredients were cassava ¯our, urea (46% N), salt and mineral±vitamin premix. Each diet contained one by-product feedstuff. Tables 2 and 3 presents the particulars of the experimental diets offered to goats and sheep, respectively.
Twelve goats and 16 sheep 16±18 months old weighing on average 12.960.28 and
14.050.18, respectively, were divided into three and four groups. They were
respectively used in two different growth and digestibility studies. All animals were housed and fed individually. Each goat or sheep had its diet weighed out for a week. The offer was increased or decreased depending on intake by the animals. Diets offered had enough ®bre, therefore forage was not introduced in all the trials. Diets offered were isonitrogenous through the addition of urea. Individual dry matter intake (DMI) and body weight changes were recorded on a weekly basis. However, only average weights at the
Table 1
Chemical composition of the by-product feedstuffs
Nutrients GNSa MCb CaPc CPPWd
Dry matter (%) 92.3 94.5 94.3 88.5
Proximate composition (% in DM)
crude protein 6.3 3.4 7.2 6.0
ash 2.8 4.2 14.2 7.7
neutral detergent ®bre 63.7 76.1 32.0 37.8
acid detergent ®bre 43.6 49.9 21.0 25.9
acid detergent lignin 9.6 15.9 7.2 6.7
hemicellulose 20.1 26.2 11.0 11.9
cellulose 34.0 23.7 13.8 19.2
aGroundnut shell. bMaize cobs. cCassava peels. dCitrus pulp/peels waste.
beginning and end of experiment were used to express growth rate and voluntary dry matter intake.
2.3. Digestibility studies Ð in vivo (goats and sheep)
At the end of each growth phase, digestibility study was carried out. Animals were allowed 7 day adjustment before a 7 day collection period. Total faecal output for each animal was weighed before a 25% sample was removed for dry matter determination. Faeces were dried in a forced draught oven at 708C for 24 h. Dried faeces for each group over the collection period were bulked, sampled and milled with a Christy and Norris
Table 2
Composition of experimental diets fed to goats (on air dry basis)
Ingredients (%) Dietsc
aCassava ¯our contained: 87.82% DM; 3.28% CP; 6.35% CF; 5.69% ash; 0.86% EE; 19.04 GE (MJ/kg of DM). bSupplied the following: 500 000 IU vitamin A; 1250 000 IU vitamin D
3; 1330 mg cobalt; 250 g calcium; 22.5 g copper; 41.7 g iron; 44.2 g manganese; 38.2 g zinc; and 11.3 g magnesium.
cGNS, groundnut shell; MC, maize cobs; CaP, cassava peels.
Table 3
Composition of experimental diets fed to sheep (on air dry basis)
Ingredients (%) Dietsc
GNS MC CaP CPPW
Cassava ¯oura 60.00 60.00 60.00 60.00
Urea (46% N) 1.55 1.75 1.45 1.35
Groundnut shell (GNS) 36.45 ± ± ±
Maize cob (MC) ± 36.25 ± ±
Cassava peels (CaP) ± ± 36.55 ±
Citrus pulp/peels waste (CPPW) ± ± ± 36.65
Mineral±vitamin premixb 1.50 1.50 1.50 1.50
Salt 0.50 0.50 0.50 0.50
Total 100.00 100.00 100.00 100.00
aCassava ¯our contained: 87.82% DM; 3.28% CP; 6.35% CF; 5.69% ash; 0.86% EE; 19.04 GE (MJ/kg of DM). bSupplied the following: 500 000 IU vitamin A; 1250 000 IU vitamin D3; 1330 mg cobalt; 250 g calcium; 22.5 g copper; 41.7 g iron; 44.2 g manganese; 38.2 g zinc; and 11.3 g magnesium.
cGNS, groundnut shell; MC, maize cobs; CaP, cassava peels; CPPW, citrus pulp/peels waste.
Hammer mill (Process Engineers, Chelmsford, England), passed through a 1.77 mm sieve and stored in air tight bottles until required for chemical analyses.
2.4. Nutritional evaluation using in vitro gas method
Rumen liquor and particulate matter (60:40) were collected from a cannulated dairy
cow (kept exclusively on a medium quality diet) before the morning feeding into
pre-warmed CO2®lled thermos ¯asks and then streamed through a cheese cloth. The samples
(200 mg) were weighed into 100 ml calibrated syringes. In vitro incubation was conducted according to the procedures of Menke and Steingass (1988) using 30 ml buffered rumen inoculum. Three syringes containing 30 ml inoculum only served as blanks. Another three syringes with 200 mg of hay reference standard were also carried out. Incubations were stopped after 24 h incubation and gas volume (Gv) was noted. The organic matter digestibility (OMD, %) and metabolizable energy (ME, MJ/kg of DM)
were calculated using the following equations: OMD14.880.889Gv0.45CP, and
ME2.200.136Gv0.057CP. TheGvis in millilitre and CP in percent in DM. These
equations were derived from digestion experiments (n400) on sheep and have been
tested with 300 other digestion experiments including 15 respiration trials. The residual standard deviation was 4.2% for both equations.
2.5. Rate and potential extent of gas production
The samples (200 mg) were incubated in triplicate in graduated syringes containing 30 ml of the in vitro medium containing rumen liquor (Menke et al., 1979). At 2, 4, 6, 8, 10, 12, 24, 30, 36, 48, 54, 60, 72 and 96 h, gas values were recorded. The potential extent (b) and rate (c) of gas production were determined using a one pool exponential model,
yb(1ÿeÿct), where `y' is the gas produced at time `t'.
2.6. Analytical methods
Proximate analyses of BPF and diets (Tables 1 and 4) and faecal samples were analyzed according to AOAC (1980). Fibre analyses (neutral detergent ®bre, NDF; acid detergent ®bre, ADF; and acid detergent lignin, ADL) were determined by the procedures of Van Soest et al. (1991).
Data obtained were analyzed by ANOVA and signi®cant differences between means were compared by using Duncan's multiple range test with the aid of SAS/STAT program (Statistical Analysis Systems Institute Inc., 1988).
3. Results and discussion
3.1. Nutrient composition of BPF and complete diets
The chemical composition of the BPF is presented in Table 1. Compared to other BPF, MC had the least crude protein content (3.4%). CaP and CPPW had low
neutral detergent ®bre and acid detergent ®bre. However, GNS and MC were high in cell wall constituents. The actual nutrient contents of the BPF may differ from tabular values given due to variation in plant varieties and handling after processing (DePeters et al., 1997). However, the proximate chemical composition values are within values reported for similar tropical BPF (Oyenuga, 1968; Gohl, 1981; Aregheore, 1993). Table 4 presents data on chemical composition of the complete diets fed to goats and sheep. Goats had diets of prepared from GNS, MC and CaP, while sheep had diets from GNS, MC, CaP and CPPW.
3.2. Voluntary dry matter intake (DMI)
Voluntary dry matter intake (DMI) was expressed as g/kg live weight (LW) per day and this ranged between 58.0±62.0 g/kg LW per day in goats and 70.0±83.0 g/kg LW per day in sheep (Tables 5 and 6). In the two trials no feed refusals were observed. Their acceptance by goats and sheep could be due to the method of processing used before their incorporation with other ingredients in the complete mash diets. With this system, it was dif®cult for any selection of a particular ingredient to take place (Reddy and Reddy, 1992). Results on DMI indicated that in con®nement sheep could consume as much and even more ®brous diets than goats if BPF are processed, thus preventing the selective habit of sheep for more palatable foods. The ®ndings in DMI between goats and sheep are at variance with Reddy and Reddy (1992) who reported that goats performed better than sheep in utilizing ®brous residues. The differences observed may be due to any of the following factors Ð breed of animals, source and types of BPF, processing methods used and probably other ingredients incorporated with the residues in diets offered. It was observed that goats and sheep could be maintained on processed BPF diets.
Table 4
Chemical composition of complete diets fed to goats and sheep
Nutrients (%)a Dietsb
GNS MC CaP CPPW
Dry matter 90.08 91.00 90.78 90.54
Crude protein 14.58 14.63 14.45 14.30
Ash 4.72 4.81 3.50 3.60
Ether extract 1.48 1.30 1.40 5.10
NDF 42.79 46.88 48.10 43.68
ADF 24.31 28.18 25.98 24.89
ADL 8.18 12.82 10.87 8.96
Hemicellulose 18.48 18.70 22.12 18.79
Cellulose 16.13 15.36 18.61 15.93
GE (MJ/kg of DM) 12.32 12.38 12.98 12.16
aNDF, neutral detergent ®bre; ADF, acid detergent ®bre; ADL, acid detergent ®bre. bGNS, groundnut shell; MC, maize cobs; CaP, cassava peels; CPPW, citrus pulp/peels waste.
3.3. In vivo digestibility
The values of nutrient digestibility of goats fed on three BPF (GNS, MC, and CaP) diets are presented in Table 7. DM and CP digestibility were better (P>0.05) in goats fed CaP diet compared to the other BPF. NDF digestibility was better (P>0.05) in the GNS fed group compared to MC and CaP groups. However, GE was lower in CaP fed goats than in the other groups. The differences observed in the digestibility of the three BPF may be due to nutrient composition of the original BPF. DM and NDF digestibility varies and this depends on the type of diet and processing methods used for BPF. Aregheore et al. (1991) observed that the higher the cell wall constituents (crude ®bre) of diets and the higher the intake of these constituents, the lower the DM digestibility values. In this trial similar trends were observed. However, the higher NDF digestibility observed in goats on GNS group suggested that it had a longer retention in the digestive system and hence the higher digestibility obtained. The decreased digestibility of dietary energy obtained in goats on GNS and MC could be due to cellulose values present in the by-products and subsequently the diets. Donefer et al. (1963) and Olatunji et al. (1976) showed that increase in the cellulose levels of diets
Table 5
Weight gain, dry matter intake and feed ef®ciency of goats fed complete diets
Parameters Dietsa,b
GNS MC CaP
Initial average live weight (kg) 12.980.26 12.940.28 12.960.30 Final average live weight (kg) 18.930.62 18.080.48 19.920.72 Live weight gain (kg) 5.950.36 ab 5.140.20 b 6.960.42 a
Average daily live weight gain (g) 50 ab 43 c 58 a
Average dry matter intake (kg) 0.630.12 0.580.06 0.620.08
Feed efficiency (feed/gain) 12.64 b 13.48 b 10.68 a
aMeans within each row with different letters differ signi®cantly atP>0.05. bGNS, groundnut shell; MC, maize cobs; CaP, cassava peels.
Table 6
Weight gain, dry matter intake and feed ef®ciency of sheep fed complete diets
Parameters Dietsa,b
GNS MC CaP CPPW
Initial average live weight (kg) 14.000.16 14.100.18 15.050.20 14.400.14 Final average live weight (kg) 20.720.23 21.920.31 22.090.52 23.900.56 Live weight gain (kg) 6.720.07 c 7.820.13 bc 8.040.32 ab 9.500.42 a Average daily live weight gain (g) 56 c 65 bc 67 ab 79 a Average dry matter intake (kg) 0.790.20 ab 0.700.20 b 0.830.18 a 0.800.12 a Feed efficiency (feed/gain) 14.12 c 10.77 a 12.39 b 10.13 a
aMeans within each row with different letters differ signi®cantly atP>0.05.
bGNS, groundnut shell; MC, maize cobs; CaP, cassava peels; CPPW, citrus pulp/peels waste.
decreased digestibility coef®cients of dietary energy. The present study also con®rmed their ®ndings.
Table 8 presents data on sheep nutrient digestibility of GNS, MC, CaP and CPPW
diets. Nutrient digestibility among the four BPF was better (P>0.05) in sheep that
received CPPW diet. The superiority of CPPW over other BPF in the diets of sheep has been reported (Aregheore and Tembo, 1998). Other BPF (GNS, MC, CaP) digestibility in sheep followed the trend observed in goats. The reasons adduced for DM, NDF and GE digestibility in goats for similar BPF also applied to sheep in this trial. However, it was observed that goats were better in the digestion of similar by-products than sheep and this con®rmed earlier ®ndings of Gihad (1976) and Williamson and Payne (1978) that goats digest ®bre better than sheep.
3.4. In vitro digestibility
Net gas production, DM, NDF and organic matter (OM) digestibility and metabolizable energy (ME, MJ/kg of DM) are presented in Table 9. Signi®cant differences were obtained among the byproduct feedstuffs in net gas production (P<0.05).
Table 7
Apparent nutrient digestibility of complete mash diets fed to goats Ð in vivo
Nutrients (%) GNSa,d MCb,d CaPc,d
Dry matter 68.00.72 b 67.80.16 b 78.90.29 a
Crude protein 72.90.42 ab 70.10.68 bc 76.50.94 a
Organic matter 23.80.57 c 50.40.23 b 68.30.26 a
NDF 68.20.03 a 67.00.8 ab 74.90.14 c
Gross energy 64.30.28 b 66.40.38 b 70.60.86 a
aGroundnut shell. bMaize cobs. cCassava peels.
dMeans within each row with different letters differ signi®cantly atP>0.05.
Table 8
Apparent nutrient digestibility of complete mash diets fed to sheep Ð in vivo
Nutrients (%) GNSa,e MCb,e CaPc,e CPPWd,e
Dry matter 69.10.60 c 59.20.43 cd 77.30.28 ab 80.20.54 a Crude protein 70.30.19 b 64.80.24 c 73.20.28 ab 77.90.92 a Organic matter 20.90.52 d 45.20.08 c 68.00.42 b 72.70.82 a NDFf 63.40.23 b 65.40.08 c 70.90.52 b 80.00.83 a Gross energy 62.70.23 bc 63.00.08b c 68.40.28 b 81.80.89 a
aGroundnut shell. bMaize cobs. cCassava peels. dCitrus pulp/peels waste.
eMeans within each row with different letters differ signi®cantly atP>0.05. fNeutral detergent ®bre.
It was observed that goats and sheep could be maintained on processed BPF. GNS had lower net gas production than MC, however, gas production was highest in CPPW followed by CaP. Gas production re¯ects more on the content of digestible energy rather than on protein and fat (Abate, 1980; Krishna and GuÈnther, 1987). The high gas production observed in CPWM and CaP may be due to their high NDF and OM digestibility. The high OM digestibility of CPWM and CaP as estimated by gas test may be due to their low cell wall. Net gas production generally re¯ects the contents of fermentable carbohydrates and also probably available nitrogen in the BPF. While fermentable carbohydrates increase gas production, degradable N compounds have been reported to decrease gas production to some extent because of the binding of carbon dioxide to ammonia (Krishnamoorthy et al., 1995).
Maize cobs digestibility was higher than that of GNS but lower than the values of the other BPF. CaP and CPPW had similar NDF digestibility, but higher (P<0.05) than values in GNS and MC. The low NDF values in CaP and CPWM showed that they contain more soluble materials which ruminants can bene®t from. In the present in vitro report, CaP and CPPW were fermented faster than other residues due to the presence of more soluble materials.
In in vivo trials with goats and sheep, growth rate and nutrient digestibility were better in rations formulated with CaP peels and CPPW compared to BPF such as GNS, MC, sun¯ower heads and cocoa pod husk (Aregheore, 1995, 1996).
Organic matter digestibility was lowest in GNS and highest in CPPW (P<0.05). The BPF have varying effective NDF and OM digestibilities and total gas production. Variation in gas production and OM digestibility has been reported for agro-industrial by-products (Krishna and GuÈnther, 1987). CaP and CPPW had similar ME values and these were higher (P<0.05) than those for MC and GNS (P<0.05).
Data on the potential extent (b) and rate (c) of gas production (Table 10) were
statistically different (P<0.05) among the BPF. They followed the pattern of OM
digestibility. A comparison of gas production and concomitant in vitro OM and NDF digestibility demonstrated that gas production re¯ects substrate fermentation. The conversion rate of true fermented organic matter into gas varied with type of by-product feedstuffs.
Table 9
Net gas produced, neutral detergent ®bre (NDF%) and organic matter (OM) digestibility and metabolizable energy (ME, MJ/kg of DM)
aNDF, neutral detergent ®bre; OM, organic matter; ME, metabolizable energy.
bGNS, groundnut shell; MC, maize cobs; CaP, cassava peels; CPPW, citrus pulp/peels waste. cMeans within each row with different letters differ signi®cantly atP>0.05.
In in vivo trials, goats and sheep had better performance with diets formulated with CaP and CPPW (Aregheore, 1996, 1997) compared to other BPF such as MC and GNS. However, the high gas production observed shows that care has to be exercised in feeding them to animals at high levels to avoid accumulation of gases and fermentation products which could lead to displaced abomasum and acidosis.
The results indicate that estimates of digestibility are of use for BPF evaluation and these estimates can be readily accomplished by the in vitro techniques available. In conclusion, in vitro OM digestibility for GNS, MC, CaP and CPPW were higher compared to the in vivo digestibility data obtained in this trial.
Acknowledgements
EMA is grateful to Alexander von Humboldt Foundation, Bonn for the award of a Research Fellowship. It was at this period that the ®nal part of this project was completed at the Institute for Animal Production in the Tropics and Subtropics, Department of Animal Nutrition and Aquaculture, University of Hohenheim, Stuttgart, Germany.
References
Abate, A., 1980. In vitro estimation of the energy content of some feeds commonly fed to livestock in Kenya. East Afr. Agric. For. J. 45 (4), 255±260.
Abate, A., Kayongo-Male, H., Abate, A.N., Wachira, J.D., 1984. Chemical composition, digestibility and intake of Kenyan feedstuffs by ruminants Ð a review. Nutr. Abstr. 54 (1), 1±13.
AOAC, 1980. Of®cial Methods of Analysis. 13th Edition. Association of Of®cial Analytical Chemists, Washington, DC, pp. 1±1094.
Aregheore, E.M., 1993. Chemical composition of some Zambia crop residues for ruminant nutrition. Zambia J. Agric. Sci. 3, 11±16.
Aregheore, E.M., 1994. Potentials of crop residues in ruminants nutrition in Zambia. Zambia Jour. Agric. Sci. 4, 39±42.
Aregheore, E.M., 1995. Effect of sex on growth rate, voluntary feed intake and nutrient digestibility of the West African dwarf sheep. Small Rum. Res. 15, 217±221.
Aregheore, E.M., 1996. Voluntary intake and nutrient digestibility of crop residues based rations by goats and sheep. Small Rum. Res. 22, 7±12.
Table 10
The rate (c) and the potential extent of gas production (b) of the by-product feedstuffs
Crop residues Gas productiona
Ratec(ml/hÿ1) Potentialb(ml)
Groundnut shell 0.12590.0212 a 10.80.8 a
Maize cobs 0.05660.0067 b 59.52.1 b
Cassava peels 0.05840.0016 b 94.12.3 c
Citrus pulp/peels waste 0.07190.0037 c 104.81.3 c
aMeans within column with different letters differ atP>0.05.
Aregheore, E.M., 1997. Effect of sex on feed intake, growth and nutrients digestibility in Blackhead sheep fed complete mash rations of crop residues. J. Anim. Feed Sci. 6, 71±79.
Aregheore, E.M., Chimwano, A.M., 1991. Agro-industrial by-products and crop residues in Zambia: availability, utilization and potential value in ruminants nutrition. In: Stares, J.E.S., Said, A.N., Kategile, J.A. (Eds.), The Complementarity of Feed Resources for Animal Production in Africa. Proceedings of the Joint Feed Resources Networks Workshop, Garborone (Botswana) African Feed Resources Network. International Livestock Centre for Africa (ILCA), Addis Ababa (Ethiopia), pp. 223±238.
Aregheore, E.M., Job, T.A., Aluyi, H.S.A., 1991. The effect of ®bre source on the potential intake growth and nutrient digestibility coef®cients of West African Dwarf (WAD) ram lambs fed predominantly cassava plus urea diets. World Rev. Anim. Prod. XXVI (3), 92±95.
Aregheore, E.M., Tembo, R., 1998. The utilization of complete rations based on crop residues by the black head sheep. Scientia Agriculturae Bohemica 28 (4), 293±300.
Dargie, J.D., 1989. Helping Small Farmers to Improve their Livestock. International Atomic Energy Agency (IAEA), Yearbook, pp. 35±55.
DePeters, E.J., Fadel, J.D., Arosemena, A., 1997. Digestion kinetics of neutral detergent ®bre and chemical composition within some selected by-product feedstuffs. Anim. Feed Sci. Technol. 67, 127±140. Donefer, E., Lloyd, L.E., Crampton, E.W., 1963. Effect of varying alfalfa:barley ratios on energy intake and
volatile fatty acid production by sheep. J. Anim. Sci. 22, 425±428.
Gihad, E.A., 1976. Intake, digestibility and nitrogen utilization of tropical natural grass hay by goats and sheep. J. Anim. Sci. 43 (4), 883±897.
Gohl, B., 1981. Tropical Feeds. Feed information summaries and nutritive value. FAO, Rome, 529 pp. Krishna, G., GuÈnther, K.D., 1987. The usability of the Hohenheim gas test for evaluating in vitro organic matter
and protein degradability at rumen level of some agro-industrial byproducts. Landwirtsch¯iche Forschung 40 (4), 281±286.
Krishnamoorthy, U., Soller, H., Steingas, H., Menke, K.H., 1995. Energy and protein evaluation of tropical feedstuffs for whole tract and ruminal digestion by chemical analyses and rumen inoculum studies in vitro. Anim. Feed Sci. Technol. 52, 177±188.
Menke, K.H., Steingass, H., 1988. Estimation of the energetic feed value from chemical analysis and in vitro gas production using rumen ¯uid. Anim. Res. Develop. 28, 7±55.
Menke, K.H., Raab, L., Salewski, A., Steingass, H., Fritz, D., Schneider, W., 1979. The estimation of the digestibility and metabolizable energy content of ruminant feedstuffs from the gas production when they are incubated with rumen liquor. J. Agric. Sci. (Cambr.) 93, 217±222.
Olatunji, O., Mba, A.U., Olubajo, F.O., 1976. Dietary effects on digestibilities of protein carbohydrate components and metabolizable energy values for West African Dwarf sheep. East Afr. Agric. For. J. 42 (1), 76±88.
Oyenuga, V.A., 1968. Nigeria's foods and feedingstuffs. Their chemistry and nutritive value. Ibadan University Press, Ibadan, Nigeria, pp. 1±98.
Pirie, N.W., 1987. Leaf protein and its by-products in human and animal nutrition. 2nd Edition. Cambridge University Press, 84 pp (Chapter 8).
Preston, T.R., Leng, R.A., 1987. Matching livestock production systems to available resources. Penambul Press, Annidale, Australia.
Reddy, M.R., Reddy, G.V.N., 1992. Effect of processing on the nutritive value of eight crop residues and two forest grasses in goats and sheep. Asian±Australasian J. Anim. Sci. 5 (2), 295±301.
Statistical Analysis Systems Institute Inc., 1988. SAS/STAT Program, Cary, NC: SAS Institute Inc.
Van Soest, P.J., Robertson, J.B., Lewis, B.A., 1991. Methods for dietary ®bre, neutral detergent ®bre and non-starch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74, 3583±3597.
Williamson, G., Payne, W.J.A., 1978. An Introduction to Animal Husbandry in the Tropics, 3rd Edition. English Language Book Series, Longman, London, pp. 463±484.
