In vivo digestibility and performance of growing goats fed maize stover supplemented with browse leaf meals and cotton seed cake based concentrates

E E Ndemanisho* B N Kimoro, E J Mtengeti* and V R M Muhikambele*

Department of Livestock Production, Ministry of Agriculture and Rural Development, P.O. Box 34188, Nairobi, Kenya.
*Department of Animal Science and Production,Sokoine University of Agriculture, P.O. Box 3004, Morogoro. Tanzania
ndema@suanet.ac.tz

Abstract

Rumen degradability, in vivo digestibility and intake studies were undertaken to evaluate the potential of Leucaena leucocephala, Albizia lebbeck, Moringa oleifera and Gliricidia sepium leaf meals compared with cotton seed cake as protein sources for goats. In a rumen dry matter (DM) and crude protein (CP) degradability study, hominy meal (HM), maize stover (MS), and isonitrogenous concentrates (220g CP/kg DM) based on Leucaena (LBC), Albizia (ABC), Moringa (MBC), Gliricidia (GBC) or cotton seed cake (CSC) were used in nylon bags, were inserted into the rumens of 4 fistulated heifers and removed after 2 to 96 hours. In an in vivo digestibility study, 20 goats were fed ad libitum maize stover, supplemented with either of the LBC, ABC, MBC, GBC or CSC concentrates.

There were differences (P<0.05) in DM and CP degradability constants, both between energy feedstuffs (MS and HM) and between protein feedstuffs (leaf meals and CSC). The digestibility study showed differences in DM, organic matter (OM), CP, acid detergent fibre (ADF) and neutral detergent fibre (NDF) intake between treatments (P<0.05), but the corresponding in vivo digestibility coefficients were similar across treatments (P>0.05), except for CP. Total intake varied from 302 ± 4.5 in ABC to 316 ± 0.7 g per day in CSC.

Average growth rate for the goats across treatments was 26.5 ± 2.9 g/d and did not differ among treatments (P>0.05). These weights were rather lower than expected for crossbred dairy goats. This could be attributed to the short period involved in trial, that is not being long enough to allow for tangible growth rate but also due to the maize stover used as basal diet being of too low quality. However, the study suggests that the browses used could replace cotton seed cake as supplement for protein.

Keywords: browse leaf meals, cotton seed cake, growing goats, In vivo digestibility, maize stover

Introduction

Forage trees and shrubs are an integral part of farming system and generally, have multiple uses in all less-developed countries in the tropics (Preston 1995, Khamseekhiew et al 2001). Leucaena leucocephala, Gliricidia sepium, Albizia lebbeckand Moringa oleifera are browse shrubs widely available and fed to animals by farmers in the tropics. Their use as supplements has been shown to enhance intake of poor quality roughages, improve growth rates and increase reproduction efficiency in ruminants (Karachi and Zengo 1997; Alayon et al 1998; Orden et al 2000). The high dry matter degradability values of the leaf meals make them appropriate as supplements with basal diets of poor quality. The high dry matter degradability implies high fermentation and passage rates, thus allowing higher feed intakes by animals. In addition, this feature also facilitates faster release of nutrients to the microbial population in the rumen thus optimising microbial synthesis. In spite of the apparently high dry matter values of these browse plants, there is evidence (Kamatali et al 1992; Azim et al 2002) that animal response is not always predictable and there are differences in their nutritive values. Further, comparison of these browse plants species has not been undertaken effectively with goats. The purpose of this study was therefore to assess the degradability and in vivo digestibility of concentrates based on leaf meals of Leucaena leucocephala, Gliricidia sepium, Albizialebbeck and Moringa oleifera compared to cotton seed cake (CSC) in growing goats.

Materials and methods

Degradability study

Animals and their management

Maize stover, which formed a basal diet, was collected from the University farms in early August 2001 and chopped into 2 - 4 cm lengths using a tractor driven forage chopper. The stover were then stored in a dry concrete pit during the entire period of the experiment. The basal diet was supplemented with concentrates based on Albizia lebbeck (ABC), Gliricidia sepium (GBC), Leucaena leucocephala (LBC), Moringa oleifera (MBC) and cotton seed cake (CSC) as shown in Table 1. The shrubs were harvested at flowering stages around the University farm. They were each dried under the sun for two days while hand turning (twelve times per day) to avoid scorching. The dried leaves were preserved in dried sacks and stored ready for feeding. Cotton seed cake, hominy meal, mineral mixture, salt and vitamins were purchased from various milling enterprises and agro-shops in Morogoro town. All the diets were formulated to be iso-nitrogenous and contained on average 220 g crude protein (CP) per kg dry matter (DM). This was achieved by using different proportions of browse and cotton seed cake. Degradability was estimated by the nylon bag technique, as described by Ørskov (2000). Four heifers with permanent rumen fistulas were used in the study. Prior to commencement of the study, the animals were subjected to a preliminary period of 10 days. The ration during this period, consisting of cut grass (Briacharia/Panicum grass mixture), supplemented with 2.0 kg of a concentrate mixture, consisting of all the four leaf meals, CSC, hominy meal and mineral supplements. Daily feed intakes were recorded and corresponded to roughage to concentrate ratio of 70:30 and this feeding regime was maintained until the removal of the last nylon bags from the rumens.

The energy (ME) of the feedstuffs was calculated according to the following formulae:

OMD = 0.98 DMD - 4.8 (MAFF 1975),

Where OMD = organic matter digestibility and DMD 48 - hour degradability. .

The energy was then estimated as follows:

ME (MJ/kg DM) = 0.15 or 0.16 OMD (Beever and Mould 2000).

The effective degradability was calculated, using a fractional passage rate of 0.04 h ^{- 1} , according to Mgheni et al (2001)

Sample preparation and data collection

All samples were air-dried and ground to pass a 2.5-mm screen of a Christy and Norris hammer mill. The nylon bags used were approximately 7.5 x 10 cm in size with a pore size of about 40 to 50 µm and were filled with approximately 1 g of the respective samples. Rumen incubation times were 0, 2, 4, 8, 16, 24, 48 and 96 h. After removal, the bags were subsequently placed into plastic bags and stored in a deep freezer at a temperature of between -15 ⁰C and -20 ⁰ C to prevent further microbial activity.

After removal from the deep freezer, all bags were thawed, placed into a washing machine and washed for 30 minutes. For the 0 - hour washing loss determination, duplicates of the samples, without incubation in the rumen, were similarly washed in a washing machine for 30 minutes.

In vivo digestibility study

Animals and their management

In vivo digestibility of DM, OM, N, NDF (neutral detergent fibre) and ADF (acid detergent fibre) determination were carried out on 20 cross-bred (local Tanzania x Norwegian Landrace,) male goats aged 6 to 7 months (originating from the intake study). This experiment was conducted three weeks after the feed intake experiment (See intake data in Table 4) where 40 growing goats (20 males and 20 females) had been used. Amounts of feed offered and refusals were recorded daily and samples were taken for subsequent analyses.

Feeding and feed allowance

Maize stover was provided on ad libitum basis (that is to allow for about 10 - 25% refusal) throughout the experimental period. The refusals from the previous day's offer were collected from each animal, weighed, recorded, sub sampled and finally bulked for chemical analysis at the end of the study (Table 1). The supplementary feeds were split into two portions and fed in the morning (0900 h) and afternoon (1400 h). The basal diet was given in the morning (0900 h) after removal of the previous day's refusals. All animals had free access to both water and minerals (mineral blocks) during the entire experiment. The total duration of the study was 17 days (10 days adjustment period and 7 days collection period).

Refusals and faeces collection

During the collection period, samples from the maize stover offered and refusals from each goat were collected, weighed, recorded and bulked. At the end of the study, the bulked refusals from individual animal were thoroughly mixed, re-sampled, oven dried (60^o C for 48 h), ground through a 1-mm sieve and then analysed.

The faeces were thoroughly mixed, weighed and dried to constant weight at 103 - 105 C in a Unitherm forced draught oven for chemical analysis.

In vitro digestibility

Dry matter digestibility and OMD of feed samples offered to experimental animals were determined in vitro using the acid pepsin two-stage method (Tilley and Terry 1963).

Laboratory analyses

Dried samples were milled through a 1 mm screen before being analysed. Feeds and faeces were analysed for DM, OM, N and ash in accordance with AOAC (1990). The ADF, ADL and NDF values were analysed according to Van Soest et al (1991). Wet faeces and urine were analysed for N using the Kjedahl method (AOAC 1990).

Biometrical analyses

The disappearance of DM from the nylon bags was described by the model of Ørskov and McDonald (1979):

P = a + b (1- e^-ct)

Where:

P = percentage degradability at time t.

a = Soluble fraction assumed to disappear from the bag instantly (g/kg)

b = Insoluble but rumen potentially degradable material at time t (g/kg)

c = Rate constant at which b is degraded (1/h).

t = Length of incubation (h).

Effective rumen degradability (EPD) was then calculated using the equation:

EPD = a + bc/(c + k) (Ørskov and McDonald 1979)

where EPD were used to describe the digestion kinetics and to calculate the effective rumen degradability and k = passage rate constant (0.04 h ^{- 1} ).

The 48-hour values were analysed using the General Linear Model (GLM) procedures of the Statistical Analysis System (SAS 1998) using the model:

Y_ij = μ + T_i + e_ij

where

Y_ij = Record of the j^th value belonging to the i^th sample

μ = Overall mean

T_i = Effect of the ith incubation period

e_ij = Error term

A similar model was used to analyse the in vivo digestibility data.

The least square means (LS Means) were computed and tested for significance.
 

Results

Chemical composition of feedstuffs and supplements

The composition of protein supplements and feed ingredients used in the study are shown in Tables 1 and 2.

Maize stover had higher NDF and ADF values and hominy meal had a higher hemicellulose value compared to the other feedstuffs. However, their CP levels were lower compared to values of leaf meals and CSC. Among the leaf meals AL had the highest NDF value and MO the highest CP level. The chemical compositions of the treatment rations were as shown in Table 3.

The supplements were isonitrogenous. However, ABC and CSC had a slightly higher OM values compared to the rest of the concentrates. The ADL of ABC was highest among the concentrates used in the study.

Intake

Intake results are presented in Table 4. The leaf meal supplemented diets had similar trend of intake to maize stover. Goats on LBC had slightly lower DM intakes of maize stover compared to those on other treatments. There were hardly any leftovers in the feed troughs from the supplements.

Growth performance

The results on growth performance are presented in Table 5. There were generally small differences in growth rates between treatments. Goats fed CSC had higher mean gains compared to the other treatments (P<0.05) except GBC.

Rumen Degradability

Degradability of DM and CP at 48 hours, and fitted model parameters are presented in Tables 6 and 7 and the corresponding degradation curves are shown in Figures 1 and 2.

Dry matter

Maize stover had lower while hominy meal (HM) had higher DM washing losses among the feedstuffs (P<0.05). Among the protein concentrates, CSC had the highest DM washing losses. The a values for MB and GB differed (P<0.05) from that of AB and LB. However, differences (P<0.05) for these two constants were evident for the rest of the feedstuffs. Similarly, the potentially degradable (a + b) portion and 48-hour degradability values for Moringa and hominy meal did not differ.

Crude protein

Maize stover had relatively lower disappearance values for a, a + b and 48-hour degradability among the feedstuffs. However, individual values of MO, HM, GB and CSC were higher (P<0.05) than those of AB and MB.

In vivo digestibility study

Intake and apparent digestibility coefficients of the rations are presented in Table 8.

Generally, goats fed GBC had lowest intake values for all parameters except NDF and ADF values where goats on CSC were lowest (P<0.05). The CP digestibility coefficient by goats on GBC was above the other treatments (P<0.05).

Discussion

DM degradability

Maize stover used in the present study had a low 48h DM degradability, a low a-value and low degradation rate. The values are supported by other experiments carried out by Kimbi (1997); Nherera et al (1998) and Mgheni et al (2001). The high NDF and ADF values were probably due to high cellulose, hemicelluloses and lignin contents. Mgheni et al (2001) evidenced this in another experiment, where the low water soluble fraction (a), high insoluble but potentially degradable fraction (b) and low rate of degradation (c) for DM, Predicted Dry Matter Intake (PFDM) and N are common for maize stover. On the other hand, hominy meal had higher washing loss and rate constant values due to its relatively low NDF and ADF contents.

The DM degradability values for the protein feedstuffs in the present study were similar to those of Kaitho et al (1998). However, Albizia had slightly lower values compared to results from other studies. Larbi et al (1996) for example, reported a potentially degradable portion and rate constant of 622 g/kg DM and 0.081 h^-1, respectively, while Ramana et al (2000) reported a value of 623 g/kg DM for the 48-hour degradability value. The 48-hour degradability and rate constant values for Gliricidia obtained in the present study were higher compared to results by Khamseekhiew et al (2001) who obtained 775 g/kg DM and 0.180 h^-1 respectively. Other authors (El Hassan et al 2000) reported 860 g/kg DM (48-hour DM degradability), 613 g/kg DM (potentially degradable portion) and 0.036 h^-1 (rate constant) for Leucaena leucocephala. However, no comparable values for Moringa oleifera were found.

The washing losses for Moringa and Gliricidia did not differ significantly possibly due to relatively similar NDF and ADF values. The relatively high NDF content of Albizia meant that most of the DM was in form of cell walls and therefore not easily available to the rumen microbes. The relatively low 48-hour degradability value of Albizia showed highest NDF value of all the other leaf meal based concentrate. It is obthrr hand, the low 48-hour degradability value in Leucaena was possibly due to the presence of secondary anti nutritive compounds (e.g. tannins). A number of authors (Wheeler et al 1994 and Stewart et al 1998) have reported the presence of varying amounts of secondary compounds (alkaloids, tannins, mimosines and phenols) in Leucaena leucocephala and implicated them in its low degradability values within the rumen.

CP degradability

The low washing losses (a), slowly degradable (b) and 48-hour degradability of CP maize stover were expected. The values obtained in the present study are in line with results by Abdulrazak et al (1997) who reported 810 g/kg DM (potential degradability) and 0.020 h^-1 (rate constant) for maize stover.

The stover was generally high in NDF and ADF. Most of the nitrogen in maize stover was therefore mainly bound to the cell walls and thus mostly unavailable to microbial enzymatic attack. This is supported by the findings of Nherera et al (1998) who noted the level of N bound to NDF maize stover, as NDIN of about 870 g/kg DM while that bound to ADF, as ADIN of about 560 g/kg DM. This implies that such nitrogen would be available to the rumen microbes only after long periods of exposure of stovers to microbial enzymes in the rumen. Conversely, hominy meal showed high values for the washing losses, 48-hour degradability and rate constant because of its relatively low ADF value. Hence, most of the nitrogen was readily available and very little was bound to ADF.

The CP degradability of the protein sources, except for Albizia, were high (>940 g/kg DM) at 48-hours. Values reported from previous studies, close to results in the present study, include those by Kamatali et al (1992), who reported 846 g/kg DM and 0.035 g/h, for potentially degradable portion and rate constant, respectively of Leucaena. Kaitho et al (1998) and El Hassan et al (2000) also reported values close to those in the present study for Leucaena. Khamseekhiew et al (2001) reported 897 g/kg DM and 0.10 g/h for the potentially degradable portion and rate constant, respectively of Gliricidia. Kombe (1999) reported slightly higher values in CSC of 594 g/kg DM and 989 g/kg DM for both the slowly and potentially degradable portions, respectively.

Moringa, Gliricidia and CSC showed relatively high washing losses and 48-hour degradability values compared to Leucaena, indicating that most of the available N was readily available to microbial attack. This was especially true for CSC, which in spite of its slightly higher NDF value had relatively high rate constant (Table 7). The possible effect of secondary compounds on release of CP in Leucaena is again manifested by the slightly lower value for 48-hour degradability value and rate constant.

Nutrient digestibility

In the digestibility trial, supplementation with the leaf meals had significant effect (P<0.05) on apparent digestibility of DM, OM, NDF and ADF. The OM digestibility coefficients of leaf meal based supplements were lower compared to the CSC and this is supported by its highest value of the same in its chemical composition. In addition, the supplements based on protein sources other than CSC had lower ADF values and thus were more digestible. Forages with low cell wall constituents have been reported to be of high digestibility (Tolera and Sundstøl 2000). The slightly lower fibre contents in supplements fed to animals in other treatments therefore favoured the high OM digestibility which was however not statistically different from that of goats fed the control diet.

The results from these experiments proved that when poor quality crop residues such as maize stover is not treated chemically or otherwise prior to supplementation response of animals' performance is poorly expressed. In this study the average daily gain (ADG) of 27g per day was lower than where similar study using cross bred growing dairy goats with urea treated maize stover supplemented with leucaena leaf meal and cotton seed cake realized ADG of 40 g per day (Ndemanisho et al 1998). Various types of supplements (MPTS, fish meal and seed cakes) has been used as supplements in goats fed high fibre basal roughage, such as maize stover (Mgheni et al 1993). Since the feeding value of the stover in this study was very low, in order to increase its production potential improvement method strategies like chemical treatment together with the browse supplementation are inevitable. According to Mgheni et al (1993) where same cross bred growing goats were used, a control diet of urea sprayed rice straw was used, but ADG was 3.3 g per day which shows that potential for these animals to gain lower or better weights is there when they are not or are effectively supplemented. For instance, goats that were eating urea treated rice straw supplemented with fish meal grew up to ADG of 49 g per day (Mgheni et al (1993). Other findings using wood ash to treat rice straw (Mtamakaya 2002; Kimario 2003) have been effective in improving the nutrients digestibility of the latter and would be cheaper to afford by small scale farmers than other conventional treatment methods

Conclusions

• .Based on the findings of this study it may be concluded that Leucaena leucocephala, Albizia lebbeck, Gliricidia sepium and Moringa oleifera leaves contain high concentrations of highly digestible crude protein and low cell wall constituents.

• The feeding potential of these browse leaf-meals was comparable to cotton seed cake and thus could replace its utilisation without drastic reduction in animal performance; however with the need to improve the poor quality of the straw as recommended earlier.

•  Based on differences in efficiency of protein utilisation among the leaf-meals, it is also recommended that investigation of the combinations of various browse plants as opposed to sole protein sources should be undertaken.
   

Acknowledgement

The authors wish to thank The International Foundation for Science (IFS) and the government of Kenya for their financial support and Sokoine University of Agriculture (SUA) for their facility to carry out the study.

References

Abdulrazak S A, Muinga R W, Thorpe W and Ørskov E R 1997 Supplementation with Gliricidia sepium and Leucaena leucocephala on voluntary feed intake, rumen fermentation and liveweight of crossbred steers offered Zea mays stover. Livestock Production Science 49: 53-62.

Alayon J A, Ramirez-Aviles L and Ku-Vera J C 1998 Intake, rumen digestion, digestibility and microbial nitrogen supply in sheep fed Cynodon nlemfuensis supplemented with Gliricidia sepium. Agroforestry Systems 41: 115-126.

AOAC 1990 Official method of analysis. Association of Official Agricultural Chemists. 15^th edition (K Helrick, editor).Virginia pp 69-88.

Azim A, Khan A G, Ahmad J, Ayaz M, and Mirza H 2002 Nutrition evaluation of fodder tree leaves with goats. Asian-Australasian Journal of Animal Sciences 15(1): 34-37.

Beever D E and Mould F L 2000 Forage evaluation for efficient ruminant livestock production. Forage Evaluation in Ruminant Nutrition (Givens D I E, Owen R F Axford E and H M Omed, editors). CAB International, Wallingford, pp. 15-42

El Hassan SM, Lahlou Kassi A, Newbold C J, and Wallace C J 2000 Chemical composition and degradation characteristics of some African multipurpose trees. Animal Feed Science and Technology 86: 27-37.

Kaitho R J, Nsahlai I V, Williams B A, Umunna N N, Tamminga S and Van Bruchem J 1998 Relationships between preference, rumen degradability, gas production and chemical composition of browses. Agroforestry Systems 39: 129-144.

Kamatali P, Teller E, Vanbelle M, Collignon G and Foulon M 1992 Insitu degradability of organic matter, crude protein and cell walls of various tree forages. Animal Production 55: 29-34.

Karachi M and Zengo M 1997 Legume forages from pigeon pea, Leucaena and Sesbania as supplements to natural pastures for goat production in Western Tanzania. Agroforestry Systems 39: 13-21.

Khamseekhiew B, Liang J B, Wong C C and Jalan Z A 2001 Ruminal and intestinal digestibility of some tropical legume forages. Asian-Australasian Journal of Animal Sciences14 (3): 321-325.

Kimario F M 2003 Effect of methods of treating rice straws using wood ash on dry matter intake, nutrient digestibility and performance of lambs. MSc Thesis - Sokoine University of Agriculture, Morogoro, Tanzania, pp 107

Kimbi E F C 1997 The effect of substituting Leucaena leucocephala forage for cotton seed cake as protein supplement for urea treated maize stover on performance of goat weaners. MSc Thesis - Sokoine University of Agriculture, Morogoro, Tanzania, pp 151.

Kombe R A 1999 Effect of supplementing low-quality roughage with different levels of proteins of different degradability and soluble carbohydrate on nitrogen recycling capacity in cattle. In: Proceedings of SUA-MU ENRECA Project workshop (Kimambo A E, Kurwijila L R and Mgheni D M, editors). 1 August 1999, Arusha, Tanzania, pp. 66-74.

Larbi A, Smith J W, Adekunle I O and Kurdi I O 1996 Studies on multipurpose fodder trees and shrubs in West Africa: Variation in determinants of forage quality in Albizia and Paraserianthes species. Agroforestry Systems 33: 29-39.

MAFF 1975 Energy allowances and feeding systems for ruminants. MAFF Technical BulletinNo. 33 HMSO, London. pp 64-88

Mgheni D M, Kimambo A E, Sundstøl F. and Madsen J 1993 Influence of urea treatment or supplementation on degradation, intake and growth performance of goats fed rice straw diets. Animal Feed Science and Technology, 44: 209 - 220 Elsevier Science Publishers B.V., Amsterdam

Mgheni D M, Ndemanisho E E, Hvelplund T and Weisbjerg M R 2001 Evaluation of the feeding value of two tropical cereal straws, maize stover, rice straw and their botanical fractions by nylon and mobile bag technique. African Journal of Science and Technology (AJST). Science and Engineering Series 2 (1): 65-71

Mtamakaya J D 2002 Evaluation of the Effectiveness of Ash extract in improving the feeding value of rice straw to ruminant animals. MSc Thesis - Sokoine University of Agriculture, Morogoro, Tanzania, pp 107

Ndemanisho E E, Mtenga L A, Kimbi E F C, Kimambo A E and Mtengeti E J 1998 Substitution of dry Leucaena leucocephala (DLL) leaves for cotton seed cake (CSC) as a protein supplement to urea treated maize stover fed to dairy weaner goats. Animal Feed Science and Technology 73: 365-375.

Nherera F V, Ndlovu L R and Dzowela B H 1998 Utilization of Leucaena diversfolia, Leucaena esculenta, Leucaena pallida and Calliandra calothyrsus as nitrogen supplements for growing goats fed maize stover. Animal Feed Science and Technology 74:15-28

Orden E A, Abdulrazak S A, Cruz E M, Orden M E M, Ichinohe T and Fujihara T 2000 Leucaena leucocephala and Gliricidia sepium supplementation in sheep fed with ammonia treated rice straw: Effect on intake, digestibility, microbial protein yield and live-weight changes. Asian-Australasian Journal of Animal Sciences 13 (12): 1659-1666.

Ørskov E R and McDonald I 1979 The estimation of protein degradability in the rumen from incubation measurements weighted according to rate of passage. Journal of Agricultural Science Cambridge 92: 499-503.

Ørskov E R 2000 The In situ technique for estimation of forage degradability in ruminants. Forage Evaluation in Ruminant Nutrition (Edited by D I Givens, E Owen, Axford, R F E and Omed H M) CAB International, Wallingford pp 175-188

Preston T R 1995 Tropical animal feeding. A manual for research workers. Animal Production and Health Paper No. 126. FAO, Rome, pp 41-67. http://www.fao.org/DOCREP/003/V9327E/V9327E00.HTM

Ramana D B V, Singh S, Solanki K R and Negi A S 2000 Nutritive value of some nitrogen and non-nitrogen fixing multipurpose tree species. Animal Feed Science and Technology 88: 103-111.

SAS 1998 Statistical Analysis System. SAS/STAT - Users guide. Statistical Analysis InstituteInc, Cary, NC. USA.

Stewart J L, Dunsdon A J, Kass M, López Ortíz S, Larbi A, Premaratne S, Wina E and Vargas J E 1998 Genetic variation in the nutritive value of Gliricidia sepium. 1. Acceptability , intake, digestibility and liveweight gain in small ruminants. Animal Feed Science and Technology 75: 111-124.

Tilley J M A and Terry R A 1963 A two-stage technique for the in vitro digestion of forage crops. Journal of British Grassland Society. 18(21): 104-111.

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

Van Soest P J, Robertson J B and Lewis B A 1991 Methods for dietary fiber, neutral detergent fiber and non-starch polysaccharides in relation to animal nutrition. Journal of Dairy Science 74 (10): 3583-3597. http://jds.fass.org/cgi/reprint/74/10/3583.pdf

Wheeler R A, Chaney W R, Butler L G and Brewbaker J L 1994 Condensed tannins in Leucaena and their relation to psyllid resistance. Agroforestry Systems 26: 139-146.

Received 10 March 2006; Accepted 19 April 2007; Published 3 August 2007

Go to top