Tropical Grasslands (2008) Volume 42, 96–103 96 Forage production and potential nutritive value of 24 shrubby Indigofera accessions under field conditions in South Africa ABUBEKER HASSEN1, N.F.G. RETHMAN2, Z. APOSTOLIDES3 AND W.A. VAN NIEKERK1 1 Department of Animal and Wildlife Sciences, 2 Department of Plant Production and Soil Science, and 3 Department of Biochemistry, University of Pretoria, Pretoria, South Africa Abstract Twenty-four shrubby Indigofera accessions from 7 species were evaluated in terms of their forage production, potential nutritive value and indospicine levels in the forage biomass over two growing seasons. Eighteen seedlings per plot were transplanted into field plots measuring 1.5 × 3 m in January 2003 at spacings of 50 cm between and within rows, with 3 replicates. In both seasons, differences between and within species for plant height, canopy spread diameter, fodder yield and leaf percentage of the biomass were significant (P < 0.05). I. amorphoides 7570, I. cryptantha 7070 and I. arrecta 7709 were superior in terms of forage yield in the first season, while I. amorphoides 7549, I. cryptantha 7067 and I. arrecta 10350 were superior in the subsequent season. Yields as high as 21 t/ha DM (total) and 5 t/ha DM (leaf) were obtained in some accessions in Year 2. Crude protein concentrations were high for I. cryptantha (29.8%) and I. amorphoides (27.7%), while the lowest were recorded for I. coerulea (15.9%) and I. vicioides (20.1%). Phosphorus concentrations in the forage biomass were higher for I. cryptantha (0.37%), I. brevicalyx (0.35%) and I. amorphoides (0.33%) than for I. costata (0.23%). The in vitro organic matter digestibility ranged from 74.8% for I. amorphoides to 63.8% for I. brevicalyx. Correspondence: Abubeker Hassen, Department of Animal and Wildlife Sciences, University of Pretoria, Pretoria, South Africa. E-mail: Abubeker.Hassen@up.ac.za Indospicine levels in the forage biomass varied dramatically both between and within species. Concentrations were insignificant (0–2 mg/kg DM) in I. brevicalyx, and reached 706 mg/kg DM in I. vicioides. Within I. arrecta, levels varied 10-fold (26–289 mg/kg DM). The variation, which exists in this genus in the various parameters, indicates considerable opportunity to select accessions for possible feeding studies with animals to determine acceptability and possible deleterious effects. Introduction Generally, indigenous fodder trees and shrubs are valuable sources of feed during the driest months and drought periods across the semi-arid and arid areas of the tropics and subtropics. However, many are not useful as feed supplements as they contain anti-nutritional compounds, which are toxic to rumen microbes or to the animal, or their metabolic products are toxic (D’Mello 1992; Lowry et al. 1996). Indigofera spp. display excellent adaptation to a range of environments (A. Hassen et al., unpublished data), and possess diverse morphological and agronomic attributes, significant to their use as forage and cover crops (Hassen et al. 2006). Shrubby types generally produce more biomass than prostrate types and there is remarkable variation between and within species (Hassen et al. 2006). However, little is known about their variation in terms of winter survival, forage production in subsequent seasons, potential nutritive value and anti-nutritional compounds, which may limit the feeding value of the forage. Previous studies have indicated that some species, e.g. Indigofera spicata, contain the free amino acid, indospicine, which causes hepato-toxicity when this species is grazed by cattle (Norfeldt et al. 1952) or fed to chicks (Britten et al. 1963), rabbits (Hutton et al. 1958a), mice (Hutton et al. 1958b) or rats (Christie et al. 1975). Among collections Growth of Indigofera accessions evaluated in Australia, genetic variation between and within species was significant (Williams 1981; Strickland et al. 1987), suggesting the need for screening more genetic material before promoting the species widely as a forage crop. Chemical analyses, particularly in combination with in vitro digestibility and the determination of the indospicine levels in the leaf biomass, can help to assess the potential nutritive value of species/accessions at a preliminary stage of evaluation for their use as forage plants. The present study evaluated 24 shrubby accessions of Indigofera, from 7 species, to assess variation in forage biomass production and winter survival at Pretoria, as well as potential nutritive value of the leaf biomass as a forage source for both livestock and game. Materials and methods 97 all accessions were harvested at the same time (March 15–18, 2004). The harvested material was separated into leaf and stem components, which were dried in a forced-draught oven at 70°C for 48 hours to determine moisture concentration in the biomass. Chemical composition and in vitro digestibility Samples of leaf biomass harvested in 2003 were milled to pass through a 1 mm sieve and representative subsamples were stored in airtight containers for subsequent laboratory analyses. The DM, ash, P and N concentrations in the samples were determined following standard procedures (AOAC 2000). Crude protein (CP) was determined as N concentration × 6.25. In vitro organic matter digestibility (IVDOM) was determined by the Tilley and Terry (1963) procedure, as modified by Engels and van der Merwe (1967). Location, field layout and management The field experiment was carried out on the Hatfield Experimental Farm, University of Pretoria (28.11°E, 25.44°S; 1370 m asl), South Africa. The soil at the experimental site is classified as a sandy-loam, with a pH of 4.2, P concentration of 29 mg/kg, K of 73 mg/kg, Ca of 158 mg/kg, Mg of 38 mg/kg and Na of 11 mg/kg. Seeds of 24 shrubby Indigofera accessions were sown in trays in a nursery. After establishment, 54 seedlings of each accession were transplanted into field plots in January 2003. Eighteen seedlings were planted per 1.5 m × 3 m plot with a spacing of 50 cm between rows and plants within rows. There were 3 replications. Spacings of 50 cm and 100 cm were maintained between adjacent plots and blocks, respectively. The plants were irrigated twice per week for 2 hours depending on rainfall events. Plots were kept weed-free by handpulling. A total of 5 middle plants, including 1 border plant, per plot were used to estimate dry matter yield with a harvestable plot area of 1.25 m2. In the first growing season (2002–2003), all accessions were harvested at the 50% flowering stage to a height of 15 cm. The flowering time ranged between 104 and 165 days after planting. Subsequently, all plots were cut to the same height before the commencement of winter (June 2003), and left to grow to determine winter survival and biomass production in the subsequent season. In the 2003–2004 growing season, however, Indospicine determination Indospicine analyses were performed on dried and milled leaf material in triplicate. The analysis involved 3 stages: plant extraction, solid phase extraction and ninhydrin test. The free amino acids were extracted as described by Pollitt et al. (1999). Since indospicine accumulates in the leaves, it comprises the major portion of the free amino acids. To improve method accuracy, the positively charged indospicine and other positive amino acids in minor amounts were purified with solid phase extraction as described in the Strata-X method Manual (Huq et al. 2003). Subsequently, the indospicine was determined with the ninhydrin method (Plumer 1978). The absorbance of 200 µL sample mixture was read in an ELISA plate using Multiskan Ascent V1.24 at 550 nm wavelength. A standard curve of the absorbance against arginine concentration was prepared, from which the unknown indospicine concentration was determined. This method provided a >70% recovery on 2 mg of arginine dissolved in the loading buffer (0.01 M carbonate buffer pH=10). Statistical analyses All studied parameters were subjected to analysis of variance to investigate the effects of replication, species and accession nested within a 98 Abubeker Hassen, N.F.G. Rethman, Z. Apostolides and W.A. van Niekerk species using Proc GLM of SAS (2001). Where the F ratio showed significance for either species or accession nested within a species, differences between least squares means were tested using the LSD test, which computes probabilities for all pair-wise differences. Results Biomass production and winter survival There was wide variation (P<0.05) both between and within Indigofera spp. in plant height, canopy spread diameter, total biomass yield and leaf biomass in both the first (2002–2003) and second (2003–2004) seasons (Tables 1 and 2)). In the first season, average plant height for different species ranged from 17.3 cm (I. brevicalyx) to 91.9 cm (I. arrecta); canopy spread diameters from 19.7 cm (I. costata) to 78.4 cm (I. arrecta); total biomass yield from 97 kg/ha DM (I. brevicalyx) to 2728 kg/ha DM (I. arrecta); and potentially edible biomass (leaf biomass) from 74 kg/ha DM (I. brevicalyx) to 1150 kg/ha DM (I. arrecta). The percentage leaf in the total biomass was highest (P<0.05) for I. vicioides (87.1%) and lowest for I. arrecta (45.8%). Similar trends were observed in the second season, except that higher values were recorded for I. arrecta in terms of agronomic parameters (Table 2). Seedling survival 2 months after transplanting (Figure 1a) and winter survival (after 1 year) (Table 2) varied significantly (P<0.05) between species. Generally, winter survival was highest for I. brevicalyx, I. arrecta and I. cryptantha, followed by I. vicioides and I. amorphoides (Table 2). In 2002–2003, within the accessions of I. amorphoides, Accession 7557 was the tallest, but 7570 was superior in terms of leaf and total biomass yields (Table 3). Similarly, accessions of I. arrecta exhibited remarkable variation in terms of plant height, leaf yield and total biomass yield (Table 3). Seven accessions (7709, 7850, 7570, 10339, 10350, 8644 and 7598) gave total yields in excess of 3000 kg/ha and leaf yields in excess of 1000 kg/ha. In the 2003–2004 season, intra-species variability between accessions of I. amorphoides was significant (P<0.05). I. amorphoides 7549 and 7521 produced above 3500 kg/ha total DM and 1500 kg/ha leaf DM (Table 4). Accession 7557 produced very poorly. Within accessions of Table 1. Inter-species variation in a collection of Indigofera species for plant height, mean canopy spread diameter, leaf percentage and % survival, leaf and total dry matter yields in the first season (2002–2003). Species I. amorphoides I. arrecta I. brevicalyx I. coerulea I. costata I. cryptantha I. vicioides 1 Values Plant height (cm) Canopy spread diameter (cm) Leaf percentage % survival after one year Leaf dry matter yield (kg/ha) Total dry matter yield (kg/ha) 65.9b (±3.20)1 91.9a (±2.07) 17.3d (±5.07) 26.3c (±7.16) 19.2d (±7.16) 37.9c (±5.69) 19d (±8.84) 73.9a (±5.78) 78.4a (±4.38) 30.1bc (±8.24) 20.8bc (±20.6) 19.7c (±11.65) 44.9b (±11.87) 19.6bc (±20.67) 70c (±2.2) 46d (±1.4) 79ab (±3.5) 61cd (±4.9) 64cd (±4.9) 70bc (±3.9) 87a (±6.1) 93b (±3.2) 97b (±2.0) 100a (±5.0) 39d (±7.1) 64cd (±7.1) 99ab (±5.6) 82bc (±8.7) 747b (±112) 1150a (±72) 74d (±177) 127cd (±251) 196bcd (±251) 537bcd (±199) 1120bc (±309) 1165bc (±190) 2728a (±123) 97d (±301) 203d (±425) 314cd (±425) 810ab (±338) 1411bc (±524) within columns followed by different letters differ significantly (P < 0.05). Table 2. Inter-species variation in a collection of Indigofera species for plant height, mean canopy spread diameter, leaf percentage and % survival, leaf and total dry matter yields in the second season (2003–2004). Species Plant height (cm) Canopy spread diameter (cm) Leaf percentage % survival at end of year Leaf dry matter yield (kg/ha) Total dry matter yield (kg/ha) I. amorphoides I. arrecta I. brevicalyx I. coerulea I. costata I. cryptantha I. vicioides 59b (±5.1)1 201a (±3.3) 16c (±8.1) 15c (±11.4) 58b (±11.4) 73b (±9.1) 23c (±14.1) 99.3c (±6.83) 127.3b (±4.41) 53.9e (±10.8) 12.2f (±15.28) 86.2d (±15.28) 154.8a (±12.13) 19.4ef (±18.85) 47.3b (±1.80) 20.5c (±1.10) 63a (±2.57) 66.7a (±3.64) 44b (±3.64) 26.7c (±3.69) 41.5b (±4.49) 52b (±4.3) 97a (±2.8) 100a (±6.9) 26c (±9.7) 23c (±9.7) 90a (±7.7) 57b (±12.0) 1255bc (±295) 3193a (±180) 273c (±420) 29c (±597) 416c (±597) 2559ab (±605) 1420bc (±737) 2481c (±1209) 16620a (±738) 431c (±1230) 43c (±2446) 1013c (±2446) 9740b (±2479) 3981bc (±3020) 1 Values within columns followed by different letters differ (P < 0.05). Growth of Indigofera accessions I. arrecta, 10 produced more than 15 000 kg/ha total yield and 10 produced more than 2500 kg/ha leaf. Nutritive value While there were significant differences between and within species for ash, CP, P, in vitro organic matter digestibility (IVOMD) and indospicine concentrations in the leaves (Tables 5 and 6), the magnitude of differences was much smaller than for agronomic parameters. Ash concentration was, in general, lower in I. cryptantha (9.0%), I. brevicalyx (10.2%) and I. arrecta (10.5%) than in I. amorphoides (12.6%), I. coerulea (12.9%) or I. costata (13.4%) (Table 5). The highest levels of CP were recorded in I. cryptantha (29.9%) and I. amorphoides (27.7%), and the lowest in I. coerulea (15.9%) and I. vicioides (20.1%). Phosphorus concentration ranged from 0.22% in I. vicioides to 0.37% in I. cryptantha. In vitro organic matter digestibility ranged from 74.8% in I. amorphoides to 63.8% in I. brevicalyx. Indospicine concentration in the leaves also varied between species, ranging from undetectable levels in I. brevicalyx (0–2 mg/kg DM) to 706 mg/kg DM in I. vicioides. The levels of indospicine 99 in I. coerulea and I. cryptantha were low (23–35.4 mg/kg DM). While variations between accessions within Indigofera spp. did occur, only indospicine levels showed major differences (Table 6). Indospicine concentrations showed the greatest variation in I. arrecta with a range of 26 mg/kg (7850) to 289 mg/kg (7709) (Table 6). Overall, 8 accessions recorded indospicine concentrations below 50 mg/kg. Discussion A great deal of diversity in forage production potential was demonstrated both within and between the Indigofera species. Among the species included in this study, I. arrecta, I. vicioides, I. amorphoides and I. cryptantha, in decreasing order, demonstrated relatively high forage yield potential in the establishment season, whereas the forage yield potentials of I. costata, I. coerulea and I. brevicalyx were generally inferior. There was significant variation both between and within species in terms of nutritive value of the forage. The leaves contained medium to high levels of CP (15.9–29.9%). NRC (1985; 1989) suggested that the diet for mature beef cattle Table 3. Variation in a collection of Indigofera species in terms of mean plant height, canopy spread diameter, mean leaf and total dry matter yields and leaf percentage in the first season (2002–2003). Indigofera accessions Plant height (cm) Canopy diameter (cm) Leaf yield (kg/ha) Total yield (kg/ha) Leaf percentage I. amorphoides 7069 I. amorphoides 7521 I. amorphoides 7549 I. amorphoides 7557 I. amorphoides 7570 I. arrecta 7524 I. arrecta 7592 I. arrecta 7598 I. arrecta 7709 I. arrecta 7850 I. arrecta 8644 I. arrecta 9045 I. arrecta 10339 I. arrecta 10350 I. arrecta 10355 I. arrecta 10478 I. arrecta 10479 I. brevicalyx 7815 I. brevicalyx 7848 I. coerulea 9004 I. costata 8712 I. cryptantha 7067 I. cryptantha 7070 I. vicioides 10486 LSD (P < 0.05) 64.7 50.6 54.1 109.5 50.8 66.0 74.0 70.9 152.7 100.2 126.3 75.4 92.0 136.2 38.8 87.6 83.2 20.4 14.3 26.3 19.2 19.5 56.4 19.0 19.9 47.3 54.1 53.9 138.0 76.2 55.1 68.1 90.1 68.1 84.9 97.1 82.4 75.6 119.1 74.3 65.5 60.8 34.2 25.9 20.8 19.7 28.1 61.7 19.6 26.6 878 398 294 604 1558 984 961 1201 1770 1680 1347 702 1421 1419 678 725 915 91 57 127 196 278 795 1120 695 1394 554 382 916 2578 2122 2189 2944 5175 3358 3824 2019 3211 4044 1062 1197 1594 121 73 203 314 461 1160 1412 1178 65.6 74.8 77.5 68.9 60.5 45.7 45.3 41.1 36.1 50.2 32.5 35.3 43.8 35.5 64.7 61.5 58.1 77.2 80.5 61.1 63.5 70.3 69.2 87.1 13.7 100 Abubeker Hassen, N.F.G. Rethman, Z. Apostolides and W.A. van Niekerk Table 4. Variation in a collection of Indigofera species in terms of mean plant height, canopy spread diameter, mean leaf and total dry matter yields and leaf percentage in the second season (2003–2004). Indigofera accessions Plant height (cm) Canopy diameter (cm) Leaf yield (kg/ha) Total yield (kg/ha) Leaf percentage I. amorphoides 7069 I. amorphoides 7521 I. amorphoides 7549 I. amorphoides 7557 I. amorphoides 7570 I. arrecta 7524 I. arrecta 7592 I. arrecta 7598 I. arrecta 7709 I. arrecta 7850 I. arrecta 8644 I. arrecta 9045 I. arrecta 10339 I. arrecta 10350 I. arrecta 10355 I. arrecta 10478 I. arrecta 10479 I. brevicalyx 7815 I. brevicalyx 7848 I. coerulea 9004 I. costata 8712 I. cryptantha 7067 I. cryptantha 7070 I. vicioides 10486 LSD (P < 0.05) 81.0 52.2 80.7 31.0 50.9 137.4 186.1 159.2 235.6 227.8 227.8 242.2 210.0 233.4 59.2 251.6 242.8 15.9 16.0 15.0 58.0 67.1 79.7 22.8 31.6 102.2 125.2 107.0 42.3 119.8 131.9 130.7 142.4 87.8 149.3 117.2 135.9 136.2 145.0 133.1 119.0 99.4 55.5 52.2 12.2 86.2 188.7 121.0 19.4 42.4 981 1565 2614 16 1099 3201 2774 3023 2280 5269 2780 3119 2810 4063 1937 3394 3658 260 287 29 416 3358 1760 1420 1654 2390 3516 4894 32 2280 13589 19313 19322 17359 21623 15783 18787 15578 21217 5528 15319 16030 399 464 43 1013 11915 7565 3981 6781 41.5 46.4 46.3 50.2 52.1 23.5 15.5 17.0 13.3 23.4 17.7 16.0 18.7 19.0 36.6 23.0 22.7 64.8 61.1 66.7 43.9 30.9 22.5 41.5 10.1 Table 5. Inter-species variation (± s.e.) in a collection of Indigofera species in terms of nutritive value parameters and indospicine concentration in leaf dry matter. Species I. amorphoides I. arrecta I. brevicalyx I. coerulea I. costata I. cryptantha I. vicioides 1 DM Ash CP (%) P IVOMD Indospicine (mg/kg DM) 89.5a1 (±0.99) 88.9a (±0.64) 89.4a (±1.42) 90.4a (±2.48) 91.4a (±2.01) 91.0a (±2.04) 90.9a (±3.55) 12.6a (±0.35) 10.5b (±0.22) 10.2b (±0.50) 12.9a (±0.88) 13.4a (±0.71) 9.0b (±0.72) 11.1ab (±1.26) 27.7a (±0.80) 24.3b (±0.49) 22.4b (±1.14) 15.9c (±1.99) 22.7b (±1.61) 29.9a (±1.63) 20.1bc (±2.84) 0.33a (±0.021) 0.28ab (±0.013) 0.35a (±0.294) 0.24ab (±0.051) 0.23b (±0.042) 0.37a (±0.042) 0.22ab (±0.074) 74.8a (±1.17) 70.6b (±0.74) 63.8c (±1.67) 69.9bc (±2.93) 65.5c (±2.37) 73.6ab (±2.40) 60.9c (±4.18) 180.8b (±23.9) 126.1bc (±15.54) 2.0c (±48.85) 23.0c (±59.74) 135.9bc (±48.20) 35.4c (±49.07) 705.6a (±85.49) Means within a column followed by different letters differ significantly (P < 0.05). should contain a minimum of 7.0% CP, while high producing dairy cows required 19.0% CP. Almost all Indigofera species tested could potentially provide sufficient nitrogen to supplement low quality roughages for beef animals, while most of the species, except I. coerulea 9004 and I. arrecta 9045, could supply the CP requirements of high producing dairy cows. This assumes that no anti-nutritive compounds are involved. The CP levels of these Indigofera accessions were generally higher than reported levels for other browse species, such as Flemingia macrophylla (Dzowela et al. 1995), Acacia nilotica, Albizia lebbeck, Butea monosperma (Ramana et al. 2000), Vernonia amygdalina (El Hassen et al. 2000), Cassia sturtii (Van Niekerk et al. 2004; Wilcock et al. 2004; Ventura et al. 2004), Rumex linaria, Acacia salicina and Adenocorpus foliosus (Ventura et al. 2004), and were comparable with those for Cajanus cajan, Acacia angustissima, Calliandra calothyrsus, Gliricidia sepium and Sesbania macrantha (Dzowela et al. 1995), Leucaena leucocephala, Pongamia pinnata (Ramana et al. 2000), Medicago sativa, Sesbania sesban (El Hassen et al. 2000), Atriplex nummularia (van Niekerk et al. 2004), Sutherlanda microphylla, Tripteris sinuatum (Wilcock et al. 2004) and Bituminaria bituminosa (Ventura et al. 2004). However, the apparently high CP levels in the Indigofera species, compared with most Growth of Indigofera accessions 101 Table 6. Variation in a collection of Indigofera species in terms of mean ash, CP, P, IVDOM and indospicine concentrations in the leaves in the establishment season (2002–2003). Indigofera accessions Ash CP (%) I. amorphoides 7069 I. amorphoides 7521 I. amorphoides 7549 I. amorphoides 7557 I. amorphoides 7570 I. arrecta 7524 I. arrecta 7592 I. arrecta 7598 I. arrecta 7709 I. arrecta 7850 I. arrecta 8644 I. arrecta 9045 I. arrecta 10339 I. arrecta 10350 I. arrecta 10355 I. arrecta 10478 I. arrecta 10479 I. brevicalyx 7815 I. brevicalyx 7848 I. coerulea 9004 I. costata 8712 I. cryptantha 7067 I. cryptantha 7070 I. vicioides 10486 LSD (P<0.05) 11.8 13.2 11.1 13.3 13.4 10.4 11.9 11.0 8.2 10.2 7.8 10.9 10.1 10.7 11.6 11.4 12.2 10.7 9.6 12.9 13.4 9.0 9.1 11.1 0.20 26.0 27.7 28.7 29.4 26.6 18.5 26.0 25.3 26.7 21.0 22.5 16.3 27.7 22.5 29.6 27.8 27.3 19.4 25.5 15.9 22.6 30.1 29.7 20.1 0.44 browse species, may be misleading, as the high levels of indospicine in some of the Indigofera accessions, along with other plant nitrogenous secondary metabolites, could mean that animals would either not consume the material or not utilise it effectively. Similarly, the P concentration in the forage biomass was higher than that reported for Acacia species (Abdulrazak et al. 2000), and higher than the lowest level (0.20% DM) recommended to meet growth requirements of cattle (ARC 1980). The in vitro OM digestibility of the poorer species was 65.0%, while it was as high as 80.0% for the best. Differences in chemical composition have a strong bearing on the potential use of leguminous multi-purpose fodder trees in feeding systems (Dzowela et al. 1997), as they may affect palatability and intake by livestock both within and between species and provenances. In Indigofera species, secondary plant metabolites could influence palatability and intake, and the level of indospicine is a useful indicator of the potential toxicity of the feed under examination. We have demonstrated both inter- and intra-species variation in indospicine concentration in leaves within these Indigofera accessions (2–750 mg/kg DM). The concentrations of indos- P 0.37 0.27 0.34 0.39 0.26 0.20 0.37 0.29 0.23 0.23 0.27 0.23 0.30 0.29 0.40 0.30 0.28 0.40 0.30 0.24 0.23 0.42 0.33 0.22 0.011 IVDOM (%) 80.0 80.1 72.7 69.7 71.7 69.2 72.0 72.4 65.0 65.4 70.4 72.2 70.6 69.8 75.5 74.8 70.2 62.2 65.5 69.9 65.5 76.6 70.7 60.9 6.6 Indospicine (mg/kg DM) 194 314 126 146 124 29 41 60 289 26 268 46 217 108 56 174 198 8.8 0 23 136 6 65 706 13.4 picine recorded for most accessions (I. brevicalyx, I. coerulea, I. cryptantha, I. arrecta, I. costata and I. amorphoides) were lower than the levels reported for I. volkensii CPI No 33819 (2000 mg/kg DM) and 33 different I. spicata (500–12 000 mg/kg DM) accessions (Aylward et al. 1987). However, the threshold level, detrimental to animals, has not been precisely determined, though in I. nigritana (CPI No. 89268), concentrations as low as 100 mg/kg DM have resulted in incipient liver lesions (Aylward et al. 1987). The same authors reported variability in toxicity of accessions in a rat bioassay study. Of 46 accessions tested, 13 accessions, from 7 species, were considered to be non-toxic, while all accessions of I. spicata depressed liveweight gain and caused varying degrees of liver damage in rats (Aylward et al. 1987). Having a low concentration of indospicine may not ensure the suitability of the species as a forage. In Australia, I. schimperi displayed good chemical composition and no problems in feeding trials with rats, but was poorly accepted in grazing trials with cattle and supported poor weight gains (Clem and Hall 1994; T.J. Hall, personal communication). The data presented on biomass yields, winter survival, CP, in vitro digestibility and indospicine levels have demonstrated that some of the 102 Abubeker Hassen, N.F.G. Rethman, Z. Apostolides and W.A. van Niekerk Indigofera species/accessions under evaluation, have moderate to high biomass yields, high crude protein concentrations, high digestibilities and low indospicine concentrations in the leaves. This makes them potential candidates for use as protein supplements. However, chemical composition alone has limited value in predicting the nutritive value of a new feed, which may contain materials toxic to the animal. The presence of indospicine in some of the species in relatively large quantities may be a major constraint to their efficient utilisation by the animal. Future research needs to address how this may be overcome, if Indigofera species are to be used widely as forage plants. On the other hand, the remarkable variability observed in this study, both between and within species, in terms of CP, IVOMD and indospicine concentration, suggests the possibility of directly selecting accessions with high forage potential and feeding value for subsequent evaluation with target animals. Feeding studies with animals are needed before any recommendations can be made about the potential of any of these accessions as livestock feeds. Acknowledgements The support given by technical assistants in the Departments of Biochemistry (Mrs R. du Toit and Mrs S. van Wyngaardt) and Animal and Wildlife Sciences (Mrs Ferreira) while undertaking the laboratory analyses on the leaf samples is gratefully acknowledged, as are the inputs of the crop section on the Hatfield Experimental Farm. References ABDULRAZAK, S.A., FUJIHARA, T., ONDIEK, J.K. and ØRSKOV, E.R. (2000) Nutritive value evaluation of some Acacia tree leaves from Kenya. Animal Feed Science and Technology, 85, 89–98. AOAC (2000) Official Methods of Analysis. 15th Edn. (Association of Official Analytical Chemists, Inc.: Washington, DC). AGRICULTURAL RESEARCH COUNCIL (ARC) (1980) The Nutrient Requirements of Ruminant Livestock. (Commonwealth Agricultural Bureaux: Farnham Royal). AYLWARD, J.H., COURT, R.D., HAYDOCK, K.P., STRICKLAND, R.W. and HEGARTY, M.P. (1987) Indigofera species with agronomic potential in the tropics. Rat toxicity studies. Australian Journal of Agricultural Research, 38, 177–186. BRITTEN, E.J., PALAFOX, A.L., FRODYMA, M.M. and LYND, F.T. (1963) Level of 3-nitropropanoic acid in relation to toxicity of Indigofera spicata in chicks. Crop Science, 3, 415–416. CHRISTIE, G.S., WILSON, M. and HEGARTY, M.P. (1975) Effects on the liver in the rat of ingestion of Indigofera spicata, a legume containing an inhibitor of arginine metabolism. The Journal of Pathology, 117, 195–205. CLEM, R.L. and HALL, T.J. (1994) Persistence and productivity of tropical pasture legumes on three cracking clay soils (Vertisols) in north-eastern Queensland. Australian Journal of Experimental Agriculture, 34, 161–171. D’MELLO, J.P.F. (1992) Chemical constraints to the use of tropical legumes in animal nutrition. Animal Feed Science and Technology, 38, 237–261. DZOWELA, B.H., HOVE, L., TOPPS, J.H. and MAFONGOYA, P.L. (1995) Nutritional and anit-nutritional characters and rumen degradability of dry matter and nitrogen for some multipurpose tree species with potential for agroforestry in Zimbabwe. Animal Feed Science and Technology, 55, 207–214. DZOWELA, B.H., HOVE, L., MAASDORP, B.V. and MAFONGOYA, P.L. (1997) Recent work on the establishment, production and utilization of multipurpose trees as a feed resource in Zimbabwe. Animal Feed Science and Technology, 69, 1–15. El HASSEN, S.M., LAHLOU KASSI, A., NEWBOLD, C.J. and WALLACE, R.J. (2000) Chemical composition and degradation characteristics of foliage of some African multipurpose trees. Animal Feed Science and Technology, 86, 27–37. ENGELS, E.A.N. and VAN DER MERWE, F.J. (1967) Application of an in vitro technique to South African forages with special reference to the effect of certain factors on the results. South African Journal of Agricultural Science, 10, 983–995. HASSEN, A., RETHMAN N.F.G. and APOSTOLIDES, Z. (2006) Morphological and agronomic characterisation of Indigofera species using multivariate analysis. Tropical Grasslands, 40, 45–59. Huq, S., Dixon, A. and Teuscher, J. (2003) Simplified method development for the extraction of acidic, basic and neutral drugs with a single solid phase extraction sorbent- StrataTM X. Application Note TN-004. (Phenomenex Inc.: Torrance, CA USA) (http://www.sampleprep.com/bin/strata-X__ product_information.pdf) HUTTON, E.M., WINDRUM, G.M. and KRATZING, C.C. (1958a) Studies on the toxicity of Indigofera endecaphylla. I. Toxicity for rabbits. Journal of Nutrition, 64, 321–328. HUTTON, E.M., WINDRUM, G.M. and KRATZING, C.C. (1958b) Studies on the toxicity of Indigofera endecaphylla. II. Toxicity for mice. Journal of Nutrition, 65, 429–440. LOWRY, J.B., MCSWEENEY, C.S. and PALMER, B. (1996) Changing perceptions of the effect of plant phenolics on nutrient supply in the ruminant. Australian Journal of Agricultural Research, 47, 829–842. NATIONAL RESEARCH COUNCIL (1985) Nutrient Requirements of Beef Cattle. 6th Revised Edn. (National Academy of Science: Washington, DC). NATIONAL RESEARCH COUNCIL (1989) Nutrient Requirements of Dairy Cattle. 6th Revised Edn. (National Academy of Science: Washington, DC). NORFELDT, S., HENKE, L.A., MORITA, K., MATSUMOTO, H., TAKAHASHI, M., YOUNGE, O.R., WILLERS, E. and CROSS, R.F. (1952) Feeding tests with Indigofera endecaphylla Jacq (creeping indigo) and some observations on the poisonous effects on domestic animals. University of Hawaii, Agricultural Experiment Station Technical Bulletin, Number 15. PLUMER, D.T. (1978) An Introduction to Practical Biochemistry. 2nd Edn. (McGraw Hill: London, UK). POLLITT, S., HEGARTY, M.P. and PASS, M.A. (1999) Analysis of the amino acid indospicine in biological samples by high performance liquid chromatography. Natural Toxins, 7, 233–240. RAMANA, D.B.V., SINGH, S., SOLANKI, K.R. and NEGI, A.S. (2000) Nutritive evaluation of some nitrogen and nonnitrogen fixing multipurpose tree species. Animal Feed Science and Technology, 88, 103–111. SAS (2001) SAS/STAT User’s Guide. Version 8.2. (Statistical Analysis Systems Institute Inc.: Cary, NC). STRICKLAND, R.W., LAMBOURNE, L.J. and RATCLIFF, D. (1987) A rat bioassay for screening tropical legume forages and seeds for palatability and toxicity. Australian Journal of Experimental Agriculture, 27, 45-53. Growth of Indigofera accessions TILLEY, J.M.A. and TERRY, R.A. (1963) A two stage technique for in vitro digestion of forage crops. Journal of the British Grassland Society, 18, 104–111. VAN NIEKERK, W.A., SPARKS, C.F., RETHMAN, N.F.G and COERTZE, R.J. (2004) Qualitative characteristics of some Atriplex species and Cassia sturtii at two sites in South Africa. South African Journal of Animal Science, 34, 108– 110. VENTURA, M.R., CASTAÑÓN, J.I.R., PIELTAIN, M.C. and FLORES, M.P. (2004) Nutritive value of forage shrubs: Bituminaria 103 bituminosa, Rumex lunaria, Acacia salicina, Cassia sturtii and Adenocorpus foliosus. Small Ruminant Research, 52, 13–18. WILCOCK, T.E., VAN NIEKERK, W.A., RETHMAN, N.F.G. and COERTZE, R.J. (2004) A comparison of Cassia sturtii, Tripteris sinuatum and Sutherlandia microphylla: three fodder shrubs applicable to revegetation of degraded rangeland in the Northern Cape province. South African Journal of Animal Science, 34, 114–116. WILLIAMS, M.C. (1981) Nitro compounds in Indigofera species. Agronomy Journal, 73, 434–436. (Received for publication August 2, 2006; accepted October 1, 2007) NB Editor’s note: Indigofera spp. are considered undesirable plants in Australia. Toxic native Indigofera spp. cause problems with grazing livestock and the introduced African species I. schimperi is a potential serious weed. It was included in Statewide evaluation trials as laboratory analyses indicated good chemical composition and feeding trials with rats produced no liver lesions. In subsequent field trials, it displayed poor palatability and produced poor animal performance. Since it grows from both root suckers and seed and thrives on good clay soils, especially when its deep rooting allows it to thrive when other plants are dying, it poses a serious weed threat. QDPI&F and CSIRO are currently conducting a 6-year study to eradicate this species from old evaluation sites.