INTRODUCTION
The role of multi-purpose tree species (MPTS) in crop–livestock farming system cannot be over-emphasized. According to Akkasaeng et al. (Reference Akkasaeng, Gutteridge and Wanapat1989), before a tree can be referred to as MPTS, it must have forage value and be palatable to animals. Thus, trees known as MPTS are potential sources of cheap feed for ruminant animals, especially during dry seasons. Unlike herbaceous pasture grasses and legumes, MPTS are able to retain their green leaves and nutrient content during dry seasons, and thereby bridge the gap normally created by decline in the nutritive potentials of natural pastures during this period (Shelton et al. Reference Shelton, Lowry, Gutteridge, Bray and Wildin1991). The MPTS perform several other functions, which include the improvement of soil fertility through nitrogen fixation, provision of shade, fuel wood, erosion control and purification of the environment through oxygen–carbon IV oxide balance. The litter from trees has long been recognized as a good source of organic carbon and nitrogen for soil fertility, and farmers use it for mulching for this reason (Kang et al. Reference Kang, Salako, Akobundu, Pleysier and Chianu1997).
Many MPTS were introduced into the sub-humid African countries in the early 1970s, but poor adoption by farmers discouraged their full integration into farming systems (Jabbar et al. Reference Jabbar, Larbi and Reynolds1996). The recent calls for tree planting in many countries to arrest the upsurge in global climate warming could be an avenue for introducing trees that farmers could use for feeding their livestock and from which other benefits could be derived. This would make their investment in tree planting worthwhile and adoption feasible. However, efforts should be made to ensure that such trees are indigenous species, which are well known to farmers. Many of these indigenous species are those that evolved through natural selection by the prevailing soil and climatic conditions. Thus, screening trials should take into consideration the peculiarity of different zones and regions, so as to produce MPTS that will meet the specific requirements of an area.
Several attempts have been made in this direction, to determine the chemical composition and nutritive potential of many MPTS indigenous to tropical African countries, for instance in Ghana (Apori et al. Reference Apori, Castro, Shand and Orskov1998), Nigeria (Arigbede & Ekpenyon Reference Arigbede and Ekpenyon2004), Kenya (Lanyasunya et al. Reference Lanyasunya, Wang, Kariuki, Kuria, Chek and Mukisira2007) and several other sub-Saharan African countries, with some degree of success. The results from the above studies have shown that tree species perform differently in different climatic zones and the use of nutrient content alone to judge the nutritive potential of trees may not be sufficient. Information from in vitro gas production will give a better evaluation of the nutritive potential of such trees. The degree of variation in the contents of these nutrients as the tree increases in age should also be monitored to determine the age at which their optimum levels are attained.
The objective of the current study was, therefore, to determine the effects of species and age on agronomy, chemical composition and in vitro gas production characteristics of six tropical MPTS, namely Enterolobium cyclocarpum, Moringa oleifera, Millettia griffoniana, Pterocarpus santalinoides, Treculia africana and Leucaena leucocephala, which acted as the control. These species were selected because of their good agronomic performance records in alley cropping trials, nutrient content, dry season leaf retention and adaptation to acidic soils and climatic conditions prevalent in many African countries. The aim of the current study was to determine the biomass production and nutritive quality of the MPTS, as well as to provide additional information for ruminant producers on the potentials of the MPTS as feed supplements in dry seasons of south western Nigeria.
MATERIALS AND METHODS
Experimental site
The trees were planted at the Teaching and Research Farm, University of Agriculture, Abeokuta (UNAAB), Ogun State, Nigeria (7°0′N, 3°30′E, 75 m asl). The site lies within the derived savannah zone of south western Nigeria. Mean monthly temperatures of the area range from 22·50 to 30·72 °C and mean annual rainfall is c. 1037 mm. The mean relative humidity is 82%, while annual soil temperature ranges from 24·5 to 31·0 °C.
Field layout, tree planting and sampling
A parcel of land measuring c. 1 ha (10 000 m2) was mapped out, ploughed and harrowed on 5 April 2004. The trees were planted on 10 May 2004 in a randomized complete block design with each of the MPTS constituting a treatment as follows: E. cyclocarpum, M. oleifera, M. griffoniana, P. santalinoides, T. africana and L. leucocephala. L. leucocephala was selected as the control because several studies have been carried out on the species by the International Center for Research in Agroforestry (ICRAF), the International Institute for Tropical Agriculture (IITA) and the International Livestock Research Institute (ILRI) in this region. The experimental area was divided into three equal blocks, each of which was sub-divided into six plots measuring 40 × 10 m. The treatments were randomly assigned to plots in each block to give three replicates. Each plot had five rows of plants with intra-row spacing of 2 m and inter-row spacing of 2 m, making a total of 20 stands per species per row and 100 stands per plot.
Collection of data on agronomic performance including biomass production, height (from base of the trunk to the apex) and collar diameter (distance round the base of the trunk) of the trees commenced in July 2004 and performed monthly, except for biomass production which was once a year. Three plants per row from the middle three rows were used, making nine plants per replicate. The biomass production was taken in July 2005, 2006, 2007 and 2008. The tree stands were cut to 0·5 m above ground level with the aid of a sharp cutlass. The total harvests from each species were sorted out into leaves (leaves + fine stem <6 mm diameter) and stem (>6 mm diameter), weighed fresh in the field, sub-sampled then dried at 65 °C for 48 h for the determination of dry matter yield (DMY), which was reported per plot in tonnes per hectare (t/ha).
The height was taken using a measuring tape, while a pair of venier callipers was used to measure the collar diameter every 3 months. The tree stands measured were tagged to ensure that same stand was always measured throughout the experimental period and not cut for biomass estimation. Foliage samples were randomly collected from all plots every 3 months. Samples were taken from six stands per row selected from the middle three rows to avoid border effects, making a total of 18 stands sampled per plot. The samples were weighed fresh and bulked separately per plot. The foliage samples were first air dried on the field and then oven-dried at 65 °C to constant weight to determine dry matter (DM). Samples per plot were later pooled together for each year, making 18 samples per year and 72 samples for 4 years. They were ground in a hammer mill with 2·5 mm sieve and kept for subsequent analysis. The samples were labelled and taken to the Nigerian Immigration and Control Board for certification.
Chemical analysis
This aspect of the work was carried out at the Key Laboratory of Agro-ecological Processes in Subtropical Regions, Institute of Subtropical Agriculture (ISA), the Chinese Academy of Sciences, Changsha, China. A total of 72 samples comprising yearly harvests for each of the six species in three replicates and for 4 years were taken to ISA in 2008 for analysis.
At ISA, the samples were further oven-dried at 65 °C for 72 h to determine the residual moisture content and milled in hammer mill with 1 mm sieve, before the commencement of chemical analysis. The following analyses were carried out: proximate composition viz DM (943·01), crude protein (CP, 984·13), ether extract (EE, 963·15) and ash (942·05)), macro-minerals (calcium (Ca), phosphorus (P), potassium (K), magnesium (Mg) and sodium (Na)) and micro-minerals (copper (Cu), iron (Fe), zinc (Zn) and manganese (Mn)) using the procedures of AOAC (1990). Neutral detergent fibre (NDFom, not assayed with heat stable amylase and expressed exclusive of residual ash), acid detergent fibre (ADFom, expressed exclusive of residual ash) and lignin (determined by solubilization of cellulose with sulphuric acid) were determined according to the procedures of Van Soest et al. (Reference Van Soest, Robertson and Lewis1991).
Total phenolics were estimated with the Folin–Ciocalteau reagent according to the method of Makkar (Reference Makkar2003) and expressed as tannic acid equivalents. Total tannins were determined as the difference in total phenolics after treatment with insoluble polyvinyl polypyrrolidone (Makkar Reference Makkar2003).
In vitro gas production measurement
The in vitro gas production was determined following the modified procedure of Menke & Steingass (Reference Menke and Steingass1988) according to Sun et al. (Reference Sun, Liu, Tayo, Tang, Tan, Lin, He, Hang, Zhou and Wang2009). A sensitive scale was used to measure c. 200 mg of the milled leaf samples into small nylon bags with pore size of 45 μm (Xinchao Company, Guangdong, China). The bags were half-sealed to prevent the samples from adhering to the wall of the syringes and lowered into 100 ml glass syringes fitted with a plunger. Each sample was replicated three times and three blank syringes were included to assist in estimating the net gas production. Macro- and micro-elements, reduction and resazurin dye solutions were mixed together with distilled water under continuous reflux of CO2 and the pH adjusted to 6·9 with buffer solution. Rumen fluid was collected from three fistulated black Chinese Liuyang goats (body weight 16·0 ± 0·5 kg) before the morning feeding. The goats were fed on diets of 0·5 maize stover and 0·5 concentrate (Table 1). The goats were fed at 08·00 h and 19·00 h daily. The rumen fluid was mixed immediately after collection and taken to the laboratory, sieved with four layers cheese cloth, and c. 610 ml added to the anaerobic buffer medium. About 30 ml of the mixture was then added to each syringe. The syringes were placed vertically in a water bath equipped with an electric motor to automatically shake the syringes at 50 rpm (DSHZ-300, Taicang, Jiangsu, China) and temperature regulated at 39 °C. Gas production was recorded at 0, 2, 4, 6, 8, 12, 24, 48, 60, 72 and 96 h of incubation.
Ingredients and chemical composition of the concentrate diet of donor animals (DM basis)
* One kg of premix contained: 571·4 g NaHCO3, 2 g FeSO4·H2O, 1 g CuSO4·5H2O, 0·01 g CoCl2·6H2O, 0·1 g KIO3, 7·5 g MnSO4·H2O, 4 g ZnSO4·H2O, 0·0025 g NaSeO3, 371·7 g carrier, 250 mg/kg VE, 25 000 IU Va and 50 000 IU VD.
pH, ammonia and volatile fatty acids (VFA) measurements
The pH values were measured in the supernatants with the aid of a pH meter (pH-4CT, Shanghai Instrument Co., Shanghai, China) after 96 h of incubation. After the pH measurements, 5 and 10 ml of the supernatants were decanted into separate plastic bottles for determination of ammonia nitrogen (NH3–N) and VFA. The 5 ml samples were centrifuged at 1 × 104 rpm for 20 min at 4 °C and analysed for NH3–N using a phenol hypochlorite assay, according to Makkar et al. (Reference Makkar, Blümmel, Borrowy and Becker1993). The plastic bottles with 10 ml supernatants were centrifuged at 5 × 103 rpm for 10 min at 4 °C, after which 1 ml of 0·05 m metaphosphoric acid and 1 ml of 3·10 m 2-ethyl butyric acid were added and then centrifuged again at 1 × 104 rpm for 15 min at 4 °C to obtain a clear solution which was separated on a packed column (model SP-1200, Supelco, Bellefonte, PA) and quantified by gas chromatography (Shimadiz sp-501, Keihanna, Kanagawa, Japan).
Calculations and statistical analysis
The data obtained from in vitro gas production were fitted to the non-linear regression equation of Larbi et al. (Reference Larbi, Smith, Adekunle and Kurdi1996):
$$V\left( {{\rm ml/}200{\rm mg DM}} \right) = b{\rm} (1 - {\rm e}^{ - ct} )$$
where V is the potential gas production at time t, b is the volume of gas that will evolve with time and c is the fractional rate of gas production.
Organic matter digestibility (OMD) coefficient and metabolizable energy (ME in MJ/kg DM) were estimated according to Menke & Steingass (Reference Menke and Steingass1988) as follows:
$${\rm OMD} = 14{\cdot}88 + 0{\cdot}00889{\rm GV} + 0{\cdot}45{\rm} {\rm CP} + 0{\cdot}651{\rm ash}$$
$${\rm ME} = 2{\cdot}20 + 0{\cdot}136{\rm GV} + 0{\cdot}057{\rm} {\rm CP} + 0{\cdot}029{\rm CP}^{2}$$
The data obtained were subjected to analysis of variance (ANOVA) using the mixed model procedure of SAS (2002) in a 6 × 4 factorial arrangement. The model used was
$$$Y_{ijk} = \mu + t_i + y_j + \left( {ty} \right)_{ij} + e_{ijk} $$$
where Y ijk is the observation, μ is the population mean, t i is the MPT species effect, y j is the year effect (j = 1–4), (ty) ij is the interaction between MPTS and year and e ijk is the residual error. Year effect was analysed for linear and quadratic trends. Means were compared by applying the probability difference (PDIFF) option of the least square means statement in the GLM procedure. Probability values less than 0·001 were expressed as P < 0·001 rather than the actual value. Since year and species × year interaction have no effect on the fermentation kinetics, the data were analysed for species effect alone. Pearson correlation analyses were used to establish relationships between chemical component and in vitro gas production parameters.
RESULTS
Agronomic performance
Table 2 shows the biomass production, height and diameter increase of the MPTS as affected by age after planting. MPTS biomass yield increased at an increasing rate (linear P < 0·05; quadratic P < 0·01) with age; however the height of M. oleifera increased at a decreasing rate. E. cyclocarpum had the greatest biomass production, height and diameter increment throughout the experimental period. Its biomass production increased from 3·29 t/ha in the first year to 7·01 t/ha in the fourth year, compared with the biomass production of L. leucocephala, which increased from 1·66 t/ha in the first year to 4·56 t/ha in the fourth year. The lowest values were recorded for P. santalinoides throughout the 4 years. Although T. africana initially showed the fastest growth rate in height (2·13 m) during the first year, it was overtaken by E. cyclocarpum, L. leucocephala and M. oleifera during the second year and this was sustained throughout the study. The collar diameter was, however, highest in E. cyclocarpum and least in M. griffoniana throughout the 4 years.
Effects of species and year after planting on biomass production (t/ha, DM basis), height (m) and diameter (mm) of six MPTS
L, linear; Q, quadratic.
Chemical composition
As shown in Table 3, year effect was only observed for DM (linear P < 0·05) and ADFom (linear P < 0·05; quadratic P < 0·01). Both variables increased at a decreasing rate. There were differences in the DM content of the MPTS leaves (P<0·001) with P. santalinoides having the highest (399 g/kg) in the fourth year, while M. oleifera recorded the lowest (270 g/kg) in the first year. In contrast, there was no difference in the CP content of the MPTS. The CP content value recorded for M. oleifera in the fourth year (240 g/kg DM) was the highest, but was not significantly different (P>0·05) from that recorded for other MPTS.
Effects of species and year after planting on the proximate, fibre and tannin contents (g/kg DM) of six MPTS
L, linear; Q, quadratic; NS, not significant.
The contents of fibre in the MPTS were moderate, with NDFom ranging from 518 g/kg DM in P. santalinoides during the first year to 619 g/kg DM in M. oleifera during the second year. ADFom content ranged from 332 g/kg DM in T. africana in the first year to 449 g/kg DM in L. leucocephala and M. oleifera in the fourth year, while lignin was highest in P. santalinoides in the second year. Species × year interactions (P ⩽ 0·001) were also observed in the fibre contents of the MPTS.
The tannin content ranged from 15 g/kg DM in M. griffoniana during the second year to 132 g/kg DM in T. africana in the first year. T. africana and E. cyclocarpum had greater tannin contents than the other MPTS (P<0·001) and there was also species × year interaction for tannin content of the MPTS. Only the P content of the MPTS was affected by year (linear P < 0·05; quadratic P<0·01, Table 4). Species and year interaction effects on mineral contents of the MPTS were also not different, with the exception of P (P <0·001). The Ca content of the MPTS ranged from 0·86 mg/kg DM for E. cyclocarpum in the second year to 3·30 mg/kg DM for L. leucocephala in the third year. Calcium was consistently greater in L. leucocephala than in any of the other MPTS. The P contents were low for all the MPTS ranging from 0·47 mg/kg DM for M. griffoniana during the third year to 2·69 mg/kg DM for the T. africana during the first year. The Ca:P ratio was within the range of 2 : 1 and above for all the MPTS except E. cyclocarpum.
Effects of species and year after planting on minerals composition (mg/kg DM) of six MPTS
L, linear; Q, quadratic; NS, not significant.
In vitro gas production
The results of in vitro gas production of the six MPTS are presented in Fig. 1. M. oleifera recorded the highest gas production of 66·67 ml/200 mg DM in the 96 h incubation period, followed by L. leucocephala with 57·0 ml/200 mg DM, while the least (40·17 ml/200 mg DM) was recorded for T. africana. The in vitro characteristics (Table 5), as reflected by the b value, showed that the MPTS evaluated contained very high amounts of soluble nutrients and the lag time was negative in all cases. Year after planting as well as species × year interaction have no effect on in vitro gas production characteristics and were therefore not reported in the current study. There was no year effect on pH, NH3–N and total VFA contents of the supernatant after incubation (Table 6). There were interactions (P<0·05) between species and year for total VFA, acetate, propionate, butyrate, iso-valerate, valerate and acetate–propionate ratio.
In vitro gas production of the MPTS.
Effects of species on in vitro gas production characteristics of the six MPTS.
DM, dry matter.
Effects of species and year after planting on pH, NH3–N (mg/l) and VFA composition (mmol/l) in the in vitro supernatant of six MPTS
TVFA, total VFA; L, linear; Q, quadratic; NS, not significant.
OMD and ME contents
As shown in Table 7, there were differences (P < 0·001) in the effect of species and year after planting on the estimated OMD and ME contents. The highest OMD was recorded by M. griffoniana during the first year, while E. cyclocarpum recorded the highest ME during the fourth year. Year effect was not significant (P>0·05).
Effects of species and year after planting on OMD coefficient) and ME (MJ/kg DM) of six Nigerian MPTS
NS, not significant.
Correlations between chemical composition and in vitro gas production parameters
Correlation coefficients among some chemical components (CP, NDFom, ADFom, lignin and tannins) and in vitro gas production characteristics (b, c and lag) of the six MPTS are presented in Table 8. There were both positive and negative relationships between these variables, but the P value showed that the differences were not significant in most cases.
Correlation coefficients between in vitro gas production parameters and chemical composition of six Nigerian MPTS
DISCUSSION
The performance of the MPTS evaluated in the current study showed that they were potential forage trees that could be depended upon as a source of green fodders for ruminants during the dry season in south western Nigeria and other countries in sub-Saharan Africa. With the exception of M. griffoniana, they grew at a rate equivalent to or more than that of the control plant L. leucocephala. The results recorded for MPTS in the present study were lower than height values of 7–8 m at 36 months after planting reported for Calliandra calothyrsus by Ty (Reference Ty and Evans1996) and 10–12 m for its mature stands by Satjapradja & Sukandi (Reference Satjapradja and Sukandi1981). The leaf biomass yield was comparable with the range of 4·3–5·6 t/ha/yr reported for C. calothyrsus by Ty (Reference Ty and Evans1996), but higher than range of 57–1151 6 kg/ha for 3-year-old stands of L. leucocephala by Malcolmson (Reference Malcolmson1988). The fact that E. cyclocarpum, M. oleifera and T. africana outperformed L. leucocephala throughout the period of the current study shows that they are viable alternatives to this species, which has been criticized by farmers in this area for its weediness and toxicity to animals. Apart from being indigenous to this region and very familiar to farmers, they are more adapted to the acidic soils and environmental conditions in the area. Increases in their biomass production and growth performance are also expected to continue until maturity.
Several studies have documented that variations in chemical composition of browse plants are due to species, season and cutting height. Little has been done to evaluate variations due to age of the trees, probably because of the length of time involved. The current study, however, revealed limited variations due to age of trees and confirmed earlier opinions that trees retained most of their nutrient contents in season and over time. The DM content recorded for MPTS in the current study over 4 years for T. africana and M. griffoniana were slightly higher than what had been reported earlier for P. santalinoides (Anele et al. Reference Anele, Arigbede, Olanite, Adekunle, Jolaosho, Onifade and Oni2008), and this might be due to the age factor. The CP content recorded for MPTS leaves was comparable to earlier reports (Ly et al. Reference Ly, Samkol and Preston2001; Anele et al. Reference Anele, Arigbede, Olanite, Adekunle, Jolaosho, Onifade and Oni2008).
The CP levels in the foliage of these MPTS are adequate for maintenance and growth of ruminant animals (ARC 1984). The EE content of MPTS in the current study was much higher than the 52 g/kg reported by Murro et al. (Reference Murro, Muhikambele and Sarwatt2003) for M. oleifera or the 67·5 g/kg reported by Arigbede et al. (Reference Arigbede, Anele, Jolaosho, Olanite, Onifade and Wahab2008) for T. africana. The ash value was also higher than the mean values of 100, 91 and 88 g/kg reported by Rajaguru (Reference Rajaguru and Gunasena1990) for G. sepium, Erythrina veriegata and L. leucocephala, respectively, but comparable with the findings of Arigbede et al. (Reference Arigbede, Anele, Jolaosho, Olanite, Onifade and Wahab2008) for T. africana.
The NDFom value recorded in the current study was much higher than the values of 362 g/kg reported by Oni et al. (Reference Oni, Onwuka, Oduguwa, Onifade and Arigbede2008) for E. cyclocarpum and 286 g/kg reported by Murro et al. (Reference Murro, Muhikambele and Sarwatt2003) for M. oleifera but comparable with the mean values of 591 and 660 g/kg reported by Ly et al. (Reference Ly, Samkol and Preston2001) for G. sepium and L. leucocephala, respectively. The ADFom content reported for MPTS in the current study was higher than 307, 211, 215 and 166 g/kg reported by Monforte-Briceno et al. (Reference Monforte-Briceno, Sandoval-Castro, Ramırez-Aviles and Capetillo Leal2005) for E. cyclocarpum, Gliricidia sepium, L. leucocephala and Acacia augustissima, respectively. These differences might be due to the time of harvest, which was the dry season in the current study and the rainy season in the Monforte-Briceno et al. (Reference Monforte-Briceno, Sandoval-Castro, Ramırez-Aviles and Capetillo Leal2005) study. The lignin values fell within the range (84–348 g/kg) reported by Khanal & Subba (Reference Khanal and Subba2001) for some fodder trees.
Tannin contents in the current study were found to be high when compared with the findings of Dalzell et al. (Reference Dalzell, Stewart, Tolera, McNeill, Shelton, Gutteridge, Mullen and Bray1998) and the 36·6–116·3 g/kg DM reported for free condensed tannins content of Lysloma acapulcencis by Camacho et al. (Reference Camacho, Rojo, Salem, Provenza, Mendoza, Aviles and Montanez-Valdez2010) during the dry season, as well as 28·9 g/kg DM reported for Acacia saligna by Mahipala et al. (Reference Mahipala, Krebs, McCafferty, Dods and Suriyagoda2009). However, these studies determined total tannin rather than condensed tannin. Tannin content in forages may be in the form of extractable condensed tannin (ECT), protein-bound condensed tannin (PCT) or fibre-bound condensed tannin (FCT). The tannin content may be a major determining factor on the proportion of the protein content in these trees that will be available to animals. According to Makkar et al. (Reference Makkar, Singh and Negi1989), tannins can adversely affect the microbial and enzyme activities, nutrient absorption and utilization by animals, and may cause toxicity and death in severe cases. Moderate quantities of tannins may, however, be beneficial as it may prevent bloating and enhance the supply of by-pass proteins (undegradable protein, UDP) for digestion in the small intestine (Salunkhe et al. Reference Salunkhe, Chavan and Kadam1990; D'Mello Reference D'Mello1992) and improve the utilization of dietary essential amino acids (Waghorn Reference Waghorn1990; McNabb et al. Reference McNabb, Waghorn, Barry and Shelton1993).
The calcium content recorded for MPTS in the current study fell within the recommended range of 1·9–8·2 g/kg reported by McDowell (Reference McDowell1997) for all classes of ruminant animals, except for E. cyclocarpum. The level of P was generally low in all the MPTS and was lower than the normal requirement range of 1·0–2·5 g/kg (McDowell Reference McDowell1997) for ruminants.
The in vitro gas production revealed that the MPTS contained high amounts of soluble nutrients, as judged by their gas production. The negative lag time recorded implied that there was no lag, an indication of high nutrient solubility in the foliage. Since the pH values were regulated with buffer medium, the value for all the samples would be expected to be around neutral, unless there is a serious change in the medium. The values reported for NH3–N in the current study is far below the values reported by Soliva et al. (Reference Soliva, Zeleke, Clement, Hess, Fievez and Kreuzer2008) for various tropical forages and this might be due to lower tannin contents in the MPTS investigated in the current study. The high VFA production was, however, in agreement with earlier reports for VFA content in tropical forages by Odenyo et al. (Reference Odenyo, Osuji, Reed, Smith, Mackie, McSweeney and Hanson2003) and Mbugua et al. (Reference Mbugua, Kiruiro and Pell2008). According to Mbugua et al. (Reference Mbugua, Kiruiro and Pell2008), tannins and alkaloids in forages can adversely affect their fermentation characteristics and thereby their nutritive values. The levels of VFA reported in the current study showed that the level of tannins in the MPTS was within the range that does not reduce the bioavailability and utilization of nutrients presented in them.
The estimated OMD and ME composition of the MPTS evaluated in the current trial were comparable with earlier reports (Sallam Reference Sallam2005; Mahipala et al. Reference Mahipala, Krebs, McCafferty, Dods and Suriyagoda2009). Several studies have established the close relationships between these parameters and the nutritive value of feeds. Mahipala et al. (Reference Mahipala, Krebs, McCafferty, Dods and Suriyagoda2009) used faecal composition to predict the dietary OMD and ME of browse containing diets fed to sheep, which implied a close link with nutrient utilization. It becomes necessary, therefore, for farmers to monitor the levels of these components in their feeds, especially forages.
Correlation coefficients between some chemical components (CP, NDFom, ADFom, lignin and tannins) and in vitro gas production characteristics (b, c and lag) of the six MPTS indicated that several factors determine the nutritive value of feeds.
The current study showed that the indigenous MPTS contained most nutrients at levels above the recommended nutrient requirement levels of ruminant animals, and the effects of species and year after planting did not reduce the nutritive quality of the MPTS to a level that will make them invaluable.
In summary, age had a positive effect on the agronomic performance of the MPTS, with higher yields at the end of the fourth year compared to the first year. The CP content of the MPTS was not affected by age but the NDFom content was, with the least values at the first and fourth year of establishment. The result of the in vitro gas production showed differences among MPTS with M. oleifera being degraded more than other MPTS; this was reflected in greater NH3–N and total VFA for M. oleifera compared to the rest. Additionally, E. cyclocarpum, M. oleifera and T. africana have been found in the current study to be better alternatives to L. leucocephala. These facts depicted them as good sources of cheap protein supplements for ruminant animals, especially during dry seasons.
Thanks go to the Chinese Academy of Sciences (KZCX2-YW-T07 and KZCX2-YW-JS407) for providing the financial support for this study. Dr O.M. Arigbede appreciates the financial support of the Chinese Academy of Sciences and the Academy of Sciences for the Developing World (TWAS).
