Effect of seed proportions of Rhodes grass (Chloris gayana) and white sweet clover (Melilotus alba) at sowing on agronomic characteristics and nutritional quality

Kechero Yisehak

Jimma University, College of Agriculture and Vet. Medicine,P. O. Box 307, Jimma, Ethiopia
yisehak_kechero@yahoo.com

Abstract

An experiment was conducted to assess the effect of different seed proportions of Rhodes grass (Chloris gayana)-white sweet clover (Melilotus alba) on agronomic performance, dry matter yield, chemical composition and in vitro dry matter digestibility. The experimental design employed was a randomised complete block design with 3 replications. Rhodes grass and white sweet clover were mixed at 7 different seed proportions: 1:0, 3:1, 2:1, 1:1, 1:2, 1:3 and 0:1. Agronomic parameters studied were seedling numbers, level of tillering/branching, plot cover and vigour, number of leaves per plant, plant height, leaf length, days to 50% flowering, leaf: stem ratio and dry matter yield (DMY). Biological competition functions computed were land equivalent ratio, aggressivity index, relative crowding coefficient and competition ratio. Forage quality was assessed in terms of total ash, organic matter (OM), crude protein (CP), neutral detergent fibre (NDF), acid detergent fibre (ADF), acid detergent lignin (ADL), in vitro dry matter digestibility (IVDMD) and metabolisable energy (ME).

Seed proportion had a significant effect on agronomic traits, DM yield and forage quality. Pure stand of Rhodes grass and the 3:1 mixture produced the highest DM yields (10.7 t/ha). CP, IVDMD and ME concentrations were higher in mixtures than in pure grass (P < 0.01), while the reverse was true for NDF and ADF.

Overall, results of the study indicated that mixed stands combined the advantages of the higher yield of Rhodes grass and the higher forage quality of white sweet clover. Reasons for the differences in yield and forage quality for the various seed mixture combinations are discussed.

Key words: Agronomic performance, biological competition functions, chemical composition, in vitro dry matter digestibility

Introduction

Feed shortage is a critical problem in livestock farming in most areas of Ethiopia. As grazing lands are gradually brought under cultivation to satisfy the food needs of the increasing human population, livestock are forced to graze on marginal areas, which, combined with the use of low quality crop residues as feeds, has resulted in poor livestock performance. Natural pastures, from such marginal lands are generally high in fibre, low in protein and energy, yet they form the main source of animal feed in Ethiopia (Teshome 1987). These resources are over utilized to the extent that they fail to meet even the maintenance requirements of indigenous animals especially when the dry season persists for long periods. This has resulted in significant decrease in milk production, loss of body weight, reduced draught power, increased susceptibility to diseases, reduced reproductive performance and retarded growth rate and high mortalities of young animals (Alemayehu 1997).

Whereas application of nitrogenous fertilisers with a view to improving crude protein (CP) and energy levels and consequently of enhanced yield and quality of natural pastures is suggested to be one method of improving animal productivity in developing countries (Peyard and Astigarraga 1998). However, the high cost of commercial nitrogenous fertiliser makes it unaffordable for most subsistence farmers in Ethiopia. An alternative strategy to improve the CP concentrations in animal feed resources is the development of grass-legume mixtures (Bogdan 1977), which consequently improves both quality and quantity of available forage (Whiteman 1980). It has long been known that forage legumes are more nutritious as livestock feeds than grasses and other crop residues, and increase total dry matter intake when used as supplements to low CP grass diets (Mero and Uden 1997) but this potential is often under-utilised (Anon 1990; EARO 2001; Alemayehu 2004).

Many grass and legume species have been tested and recommended for the different agro-ecological zones in the country. Among the grass species, Rhodes grass (Chloris gayana) is recommended for low to high elevations. White-flowered sweet clover (Melilotus alba) is a legume species of high potential as fodder crop. However, there is limited information on the management practices that influence yield and quality of these species and their compatibility when grown in mixtures. Therefore, the present study was undertaken to investigate: the agronomic performance, biological potential and dry matter yield of mixtures of Chloris gayana and Melilotus alba; and chemical composition and in vitro dry matter digestibility of the forage produced.

Materials and methods 

The experiment was conducted at Alemaya University Campus Research Station (9⁰ 26^‘ N, 42⁰ 03^’E; 1980 m.a.s.l.) on an alluvial Vertosol (Tamire 1982) during the main rainy season (July-October) of 2003-2004. Mean annual rainfall (1994 -2004) for the area is about 825 mm and mean temperature is 18.2 ^oC (AUA 1998).

Twenty-one plots, each measuring 4 m x 3 m, with 1 m spacing between plots and 25 cm spacing between rows were fertilised with a basal dressing of 40N + 18 P kg/ha (grass plots), 20N + 18P (grass-legume plots) and 10N + 18P (pure legume plots) at the beginning of the experiment, based on local recommendations. Seed of Chloris gayana (3 kg/ha on pure grass plots) and Melilotus alba (4 kg/ha on pure legume plots) was drilled into rows on a well prepared seedbed. Hoeing and hand weeding were conducted during establishment and subsequently, as deemed necessary.

The experiment consisted of 7 treatments arranged in a randomised complete block design with 3 replications. The treatments included Chloris gayana (Rhodes grass) and Melilotus alba (white sweet clover) as pure stands and mixtures of Rhodes and sweet clover as follows: Rhodes grass (100% Rhodes), 3:1 (75% Rhodes + 25% sweet clover), 2:1 (66.7% Rhodes + 33.3% sweet clover), 1:1 (50% Rhodes + 50% sweet clover), 1:2 (33.3% Rhodes + 66.7% sweet clover), 1:3 (25% Rhodes + 75% sweet clover) and sweet clover (100% sweet clover).

Data collection

The following attributes of plant growth, biological potential, quality and yield of forage were recorded: plant population [number of seedlings, tillers (Rhodes) and branches (sweet clover)/m²]; number of leaves per plant; leaf length; plant height; number of tillers/branches per plant (NTPP/NBPP); leaf: stem ratio; plot cover; growth vigour; dry matter yield; land equivalent ratio (LER); crowding coefficient; competitive ratio; aggressivity index; and dry matter, crude protein, organic matter, total ash, acid detergent lignin, neutral detergent fibre and metabolisable energy concentrations plus in vitro dry matter digestibility.

Morphological characteristics were observed during the entire growing period for variations such as leaf colour, growth vigour and symptoms of nutrient deficiencies or diseases, as well as variations in growth and development due to differences in initial seeding rates.

The average number of seedlings was determined from randomly selected sampling areas in the central rows of each plot for each species at 2, 3 and 4 weeks after sowing (Tarawali et al 1995). Stand counts were taken at 6 and 8 weeks after establishment to give an estimate of tillering and branching in Rhodes and sweet clover, respectively. Within each plot, 8 random measurements of seedlings, tillers and branches were taken using 1m by 1m quadrats and averaged. In addition, 3 Rhodes and 10 sweet clover plants from each plot were taken randomly for recording data on number of tillers and branches per plant. Time to 50% flowering for grass was determined by recording the number of days after sowing when half of the plants (or tillers) were flowering (Tarawali et al (1995).

Basal cover was estimated by dividing the 1 m² quadrat into 25 squares, each of 20 cm x 20 cm, using string as described by Tarawali et al (1995). Plot cover and vigour (growth, competitive ability, seedlings) of establishment were recorded on a scale of 1-5: 1 (very poor), 2 (poor), 3 (fair), 4 (good) and 5 (very good).

Mean plant height for each species from each treatment was determined by measuring the height of 10 randomly selected plants for sweet clover and 3 plants for Rhodes from ground level to the tip of the main stolon or stem prior to herbage defoliation.

Dry matter yield was estimated from a 0.25 m² quadrat in each plot. The harvested material was sorted into leaf and stem, and leaf: stem ratio was calculated (DM basis). In grasses, the stem fraction consisted of leaf sheath, stem and inflorescence, and in legumes, the leaf fraction included the lamina and petiole. Flowers were included with stems.

Biological competition (potentials) functions

Equations representing biological competition functions are presented in Table 1.

Sampling procedures for nutritional value and yield parameters

The herbage was cut manually with a sickle to 5 cm above ground when Rhodes achieved 50% flowering. Before cutting the whole plot, proportionate samples were taken from the 1m² quadrat and the harvested green biomass was separated into sown grass and legume components (Rhodes and sweet clover). The fresh material was weighed in the field, sub-sampled and dried in a forced draft oven at 65 ºC for 72 h. The dried composite forage samples from each treatment were ground in a Wiley mill to pass through a 1 mm sieve for all laboratory analyses. All analyses were completed in duplicate. 

The total dry matter (DM) content of the forage was then calculated using the formula described by AOAC (1990). Total ash (TA) concentration was determined by incinerating dried ground samples in a Muffle furnace at 550 ^oC for 6 h to burn/oxidise all organic matter. The inorganic residue was weighed and expressed as TA% (AOAC 1990). Organic matter was calculated as 100 minus TA. N concentration was analysed using the Kjeldhal procedure (AOAC 1990) and converted to CP as N x 6.25. In vitro dry matter digestibility (IVDMD) was determined according to Tilley and Terry (1963) as modified by van Soest and Robertson (1985). Digestible organic matter (DOM) was calculated as 0.95 IVDMD (%) - 2 (AAC 1990). Metabolisable energy (ME) was estimated from DOM using the equation developed for tropical forages: ME (MJ/kg DM) = DOM (g/kg DM) x 18.5 x 0.81. Analyses for NDF, ADF and ADL were carried out using the methods of van Soest et al (1991). Cellulose was determined by subtracting (ADL + ADF-bound nitrogen + ADF ash) from ADF while hemicellulose was calculated as the difference between NDF and ADF. Component yields were calculated by multiplying the respective component concentration by the DMY.

Statistical analysis

 Data on agronomic parameters and quality of forage samples were subjected to ANOVA based on the model designed for a randomised complete block design (RCBD) according to Gomez and Gomez (1984) and using the MSTAT-C computer software package (1989). Duncan’s Multiple Range Test was carried out for subsequent comparison of means as described by Steel and Torrie (1980). The statistical model used for the RCBD design was: Y_ij = µ + t_i + r_j+ e_ij, where Y_ij = output, the dependent variable; µ = the overall mean effect; t_i = the i^th   treatment effect; r_{j =} the j^th replication effect; and e_ij = the random residual error assumed normally and independently distributed. 

Results  

Plant counts

The number of plants among the different treatments varied with the seed proportions used (Table 2).

Seedling counts for grass in the various treatments ranged from 39.0 to 26.3 seedlings/row (P<0.01), with numbers on pure grass plots exceeding those on all other treatments. As with seedling counts, the pure stand of Rhodes showed significantly (P< 0.01) higher numbers of tillers (252 tillers/m²) than other treatments. The pure stand also produced more tillers/plant (81.3), which is a mean of counts at 6 and 8 weeks. Though the trend for sweet clover was similar to that for Rhodes (P<0.01), number of seedlings in pure legume plots was similar to the number of sweet clover seedlings recorded in 1:3 and 1:2 plots. Here also, treatments with a high legume proportion produced more branches/plant than other combinations.

Row cover and vigour scores

Rhodes grass produced large numbers of stolons, which crept over the soil surface, rooting at the nodes, and produced abundant tillers. Plot cover was directly related to the percentage seed of each species in the initial sowing and was higher (P<0.01) for pure stands of Rhodes and sweet clover than for the 1:3 and 3:1 associations, respectively (Table 3).

As for plant cover, vigour score (establishment, growth, competitive ability and seedling numbers) for each species was directly related to the percentage seed of that species in the initial sowing, with significant differences between pure stands and 1:2 and 1:3 (Rhodes) and 3:1 (sweet clover). In the early stages of plant growth, vigour as well as cover in pure legume plots was low, but the situation changed from 45 days after planting, when sweet clover became very vigorous and showed good covering ability.

Plant height

Height of Rhodes grass was inversely related to the percentage Rhodes seed in the original sowing, with the greatest height (121.0 cm) in the 1:3 mixture and the lowest (100.7 cm) in pure Rhodes (P<0.01)(Table 4).

For sweet clover, height was directly related to the percentage sweet clover seed in the initial sowing, with greatest heights (100-109 cm) in 1:2, 1:3 and pure sweet clover stands and lowest (73.0 cm) in the 3:1 mixture (P<0.01).

Number of leaves

Number of leaves per plant for Rhodes in pure stands and mixtures followed a similar trend to plant height, with the highest number for the 1:3 mixture (P<0.01) (Table 4). While the trend for sweet clover was similar to that for plant height, differences were not significant.

Leaf: stem ratio

The proportions of leaf and stem in Rhodes grass varied with treatment, with the highest leaf: stem ratio (1.35) in the 1:3 mixture and the lowest (0.89) (P<0.01) for pure Rhodes (Table 4). Leaf: stem ratio of sweet clover did not differ significantly with treatment (P>0.01). 

Leaf length

The average length of leaves per plant for Rhodes grass varied with treatment with the highest value for the 1:3 mixtures (Table 4).

Days to flowering

Time from planting to 50% flowering of Rhodes was longest in the 1:3 mixtures and shortest in the pure grass stand (P < 0.01) (Table 4).

Dry matter yield

Total DM yield of forage varied with treatment from 10.8 t/ha for the 3:1 mixture to 6.7 t/ha for pure sweet clover (P<0.01) (Table 4).

Nutritional value of Rhodes-sweet clover mixtures

The chemical concentrations of CP, TA, OM, NDF, ADF, ADL, ME, cellulose and hemicellulose and IVDMD of forages are presented in Table 6.

The DM concentration in forages was significantly (P < 0.01) affected by treatment with the highest (26.0%) and lowest (15.0%) values recorded in pure stands of Rhodes and sweet clover, respectively. There were no significant differences between treatments in ash or OM concentrations (DM basis) (Table 6).

The average CP concentration was directly related to the sweet clover percentage in the initial sowing, ranging from 14.1% for pure Rhodes to 22.5% for pure sweet clover (P<0.01) (Table 6). The CP yield (CPY) also varied with treatment with highest values for the 2:1 and 3:1 mixtures.  In general, mixed pastures produced higher CP yields than either pure species.

NDF concentrations in forage were directly proportional to the grass percentage in the initial sowing (Table 6) with 63.1% for pure grass and 37.2% for pure legume (P < 0.01). Acid detergent fibre levels mirrored the NDF concentrations, ranging from 38.9% in pure Rhodes to 33.1% in pure sweet clover (P < 0.05).  In contrast, ADL concentrations in forage were highest in sweet clover (5.87%) and lowest in Rhodes (3.98%). Hemicellulose (NDF-ADF) concentrations in the forages increased with the increase in proportion of Rhodes grass in the mixtures. Cellulose concentration showed a similar trend, with increasing levels as the grass percentage in the mixtures increased.

IVDMD of the forage produced increased as the level of sweet clover in the mixtures increased, ranging from 55.0% for pure grass to 64.1% for pure sweet clover (P < 0.01) (Table 6). The ME values were also directly related to the legume content in the forage, with pure legume having 9.10 MJ/kg DM and pure grass 7.54 MJ/kg DM (P < 0.01).

The digestible DM yield (DDMY) is considered one of the most important criteria in evaluating forage productivity; since it takes into account both DM yield and digestibility of the DM. The DDMY was considerably higher for pure stands of grass (5.79 t/ha) than for pure legume (4.29 t/ha), although the values did not differ significantly. Highest DDMY was obtained with the 3:1 mixture (6.50 t/ha). 

Discussion 

The results of this study have provided valuable information on the yield, composition and quality of first-year stands of a range of mixtures of Rhodes grass and white sweet clover.

The superior dry matter production in the pure grass pasture and the mixtures containing a high percentage of grass reflects the ability of grass to produce high levels of production. Daniel (1990), obtained similar results from an intercropping of Rhodes with alfalfa (Medicago sativa). In the pure grass plots, the application of 40 kg/ha N plus the release of nitrogen following cultivation would have contributed to the high DM produced. In pure grass pastures, in the absence of added fertiliser N, DM production declines with age of the stand as N is accumulated in the below-ground material (Robbins et al 1986). Cultivation is necessary to release this N and boost DM production. The findings of this study with respect to Rhodes grass was also comparable to the findings reported for Tanzania and Zimbabwe (FAO 1981) in which Rhodes grass harvested under different growth stages and in mixtures with forage legumes the DM yield ranged from 20.00 (Fresh, first cutting, early bloom, Tanzania) to 28.20% (fresh pasture, Zimbabwe). The higher DM yield of the grass component under sole stand as well as higher grass seed proportions could be attributed among other factors, to well-established root system that enabled the grass to extract growth resources from the soil.

Among the different intercrop ratios concerning the magnitude difference, the mixture with the highest proportion of the grass component (75%) produced the highest DM yield (10.8 t/ha), which was very close to the yield obtained from sole stand grass (10.6 t/ha). This is due to the fact that the proportion of grass (75%) growing with the legume (25%) produced much more forage through increased production of tillers than that in the 1: 2 and 1:3 associations. Thus, logically the magnitude of DM yield of forages increased due to increased seed proportion of the grass in the different grass-legume mixtures. Similar conclusions on grass (cereal)-legume associations have been reported by Ibrahim et al (1993), Munzur (1993), Soya (1994) and Qamar et al (1999). The comparable DM yields on the 3:1 pasture should be maintained over an extended period as N fixed by the legume stimulates pasture growth. The DM yields of the pure legume pasture (6.7 t/ha) reflects the poor performance of this legume to produce as much yield as rhodes grass. Virtually, all mixed pastures produced LER values in excess of 1.0, indicating that these mixtures produced more DM as mixtures on a plot of land compared to the yield as pure stands on different plots. The highest LER values were the best ones (excess of 1.0), as far as the farmers needs are concerned. This idea was also supported by Willey, (1979). However, there was only limited response in growth of Rhodes grass from its association with sweet clover in the mixtures. This is probably a reflection of the fertiliser applied at planting and N released by cultivation. The enhanced grass growth in the mixtures could be expected to be greater as the pasture consolidated making high quantities of N fixed by the legume component available to the grass.

Data on plant populations indicated that the numbers of Rhodes grass and sweet clover seedlings did not reflect the numbers and proportions of seeds sown. The number of seeds sown in the pure grass plots was 4 times that in the 1:3 grass/legume seed mixture plots, while the number of seedlings was only 50% higher on the pure grass plots. This could be related to the response of the seedlings to light intensity and soil nutrients availability under pure and mixed sward conditions. However no clear cut conclusions could be drawn from this inconsistent growth performance. On the other hand the DM yield of the grass on these two treatments reflected better the seed proportions, with the yield on the pure grass plots being 4.4 times the grass yield on the 1:3 grass/legume mixed stand plots. This was a function of the greater tillering in the pure grass plots than in the mixed pastures. DM production of sweet clover in the mixtures exceeded the yields expected on the basis of the proportion of legume seed sown, reflecting the ability of the sweet clover to effectively compete with the associated grass.

The patterns of growth of the two grass/legume species were of interest. Rhodes grass plants grown in mixtures with sweet clover were taller than those grown in a pure stand, since plants in the mixed stand produced fewer tillers and so devoted their nutrient reserve for the growth of the main culm. By comparison, sweet clover grew taller in pure stands than in mixture with Rhodes since the latter was more competitive and suppressed its companion. This revealed somewhat a strange deviation between yield and seed proportion in mixtures in view of the relatively better yield performance of the legume in the mixtures (on basis of DM production) than it was expected on the basis of seed proportions at sowing.

Clearly, DM production alone is not the sufficient for judging the value of a fodder crop so quality attributes must be considered in pasture selection and development program. As would be expected, the CP levels in the harvested material were higher in sweet clover than in pure Rhodes grass, with the mixtures producing CP levels that reflected the botanical composition of the harvested material. While CP yields on pure grass and pure sweet clover plots were similar, with differences in DM yield cancelling out differences in CP percentage; CP composition in most of the mixed pastures was elevated, with the highest being 30% for the 3:1 treatment. This is expected to be a reflection of the amount of legume in the composition since legumes have generally higher CP contents than grasses at similar groth stage (Whiteman 1980). The levels of CP in the harvested forage exceeded the minimum of 7.5% suggested as necessary for optimum rumen function (van Soest 1994). The CP levels in all of the treatments except the pure Rhodes grass pasture exceeded the 15% required to support lactation and growth of cattle (Norton 1982; McDonald et al 2002). IVDMD was 55% in the pure grass, reflecting the stage of growth. Digestibility increased progressively as the percentage of legume in the harvested material increased, reaching 64% for pure legume. These values would be a reflection of the differences in NDF in the grass and legume (63 vs 37%, respectively). The IVDMD values for the sole grass stands agreed with that reported in Skerman and Riveros (1990) (40 to 60% for sole Rhodes grass). De Gues (1977) also reported that the digestibility of cultivated tropical grasses lies between 50 and 65%, and of temperate grasses between 65 and 80%. When data for DM yield and IVDMD are combined, DDMY values for the pure grass were 35% higher than for the pure legume, and 51% higher than pure legume for the 3:1 mixture. This would result in an increased feed intake as the IVDMD and feed intake are positively correlated (Van Soest 1982). On an energy basis, the MEY for pure grass was 28% higher than for pure legume and 48% higher for the 3:1 mixture.

While management of mono-specific pastures is relatively simple and can be tailored to the specific needs of the particular species, management of mixed grass-legume pastures is more complex, as the requirements of both species must be considered. Therefore, a greater level of skill is essential if mixed pastures are to be utilised to optimal efficiency.

Conclusions 

• This study has highlighted the benefits of growing grasses and legumes as mixed fodder crop to maximize yield and quality in forage production.

  
  

• As observed in this study the planting of sweet clover in association with Rhodes grass has not only enhanced total DM production but also raised the quality of the forage produced, in particular the CP concentration and increased total energy in the fodder produced.

  
  

•  From this study alone it is not possible to determine the most appropriate combination of Rhodes and white sweet clover seed that would optimize fodder yield and quality.

  
  

• More studies need to be undertaken under varying environments and harvesting strategies will be needed to fix seed proportions and develop management practice.

References 

AAC (Australia Agricultural Council) 1990 Feeding standards: Australian livestock. (Ruminants sub-committee, CSIRO: Australia). pp. 50-81

Alemayehu Mengistu 1997 Conservation based forage development for Ethiopia. (Self-help Development International and Institute for Sustainable Development: Addis Abba, Ethiopia)

Alemayehu Mengistu 2004 Pasture and forage resource profiles of Ethiopia. (Alemayehu Mengistu and Associates: Addis Ababa, Ethiopia)

Anon 1990 Forage Extension Manual for the 4^th Livestock Development Project. (Ministry of Agriculture, Animal and Fishery Development Main Department: Addis Ababa, Ethiopia)

AOAC (Association of Official Analytical Chemists) 1990 Official Methods of Analysis. 7^th Edition. (AOAC: Ringtown, VA, USA)

AUA (Alemaya University of Agriculture) 1998 The University Facts and Figures. (Alemaya University of Agriculture: Dire Dawa, Ethiopia)

Bogdan V A 1977 Tropical Pastures and Fodder Plants. pp. 379- 383. (Longman Group Limited: New York, USA)

Daniel Keftasa 1990 Effects of developmental stages at harvest, nitrogen application and moisture availability on the yield and nutritional quality of Rhodes grass (Chloris gayana kunth)-Lucerne (Medicago sativa L.) pastures. PhD Dissertation presented to the Swedish University of Agricultural Sciences, Upsala, and Sweden 96p

DeWit C T 1960 On competition. Verslagen and Landboukundige Onderzoekingen, The Netherlands. 66: 1-83

DeWit C T and Van den Bergh 1965 Competition between herbage plants. Netherlands Journal of Agricultural Science 13: 212-221

De Gues R T 1977 Production Potentialities of Pastures in the Tropics and Subtropics. Zurich: Centre D’Etude De 1’Azote 53P

EARO (Ethiopian Agricultural Research Organization) 2001 National forage and pasture crops seed research and production strategy. (EARO: Addis Ababa, Ethiopia)

FAO (Food and Agriculture Organization of the United Nations) 1981 Tropical Feeds:  Feed information summaries and nutritional values. Bo Gohl (editor). FAO, Rome, Italy 528p http://www.fao.org/ag/AGA/AGAP/FRG/afris/default.htm

Gomez K A and Gomez A A 1984 Statistical Procedures for Agricultural Research. 2^nd Edition. (John Wiley and Sons: New York, USA)

Ibrahim N M, Gofforar M O and D A A Wahab 1993 Intercropping of pioneer sorghum with lablab purpureus L. under irrigation at Sambat. Annuals of Arid Zone.32 (3): 157-159

McDonald P. Edwards R A, Greenhalgh J F D and Morgan C A 2002 Animal nutrition. 5^th Edition. pp. 230-345. (John Wiley and Sons and Longman Scientific and Technical: New York)

McGilchrist C A 1965 Analysis of competition experiments. Biometrics 21:975-985

McGilchrist C A and Trenbath B R 1971 A revised analysis of plant competition experiments. Biometrics 27:659-679

Mero R N and Uden P 1997 Promising tropical grasses and legumes as feed resources in central Tanzania. I. Effect of different cutting patterns on production and nutritive value of six grasses and six legumes. Tropical Grasslands 31: 549-555 http://www.tropicalgrasslands.asn.au/Tropical%20Grasslands%20Journal%20archive/PDFs/Vol_31_1997/Vol_31_06_97_pp549_555.pdf

MSTAT-C 1989 Microcomputer statistical program for experimental design, data management and data analysis. Version 2.10. (Institute of International Agriculture, Michigan State University, Soil and Crop Sciences, and Agricultural Economics: USA)

Munzur M 1993 Herbage yields of barley-vetch at different seeding rates. In: Management of Mediteranean Shrubland and Related Forage Resources. FAO REUX Technical Series.pp.28, 58-61. FAO. Rome

Norton B W 1982 Differences between species in forage quality. pp. 76-84. In: Proceedings of International Symposium held on 24-28 September 1981 at St. Lucia, Queensland, Australia

Peyard J and Astigarraga L 1998 Review of the effect of nitrogen fertilization on the chemical composition and nitrogen balance. Journal of Animal Feed Science and Technology. 72: 235-259

Qamar I A, Keatinge J D H, Ali M A and M A Khan 1999 Introduction and management of vetch/barely mixtures in the rain fed areas of Pakistan. Australian Journal of Agricultural Research. 50 (3): 21-27

Robbins G B, Rickert K G and Humphreys L R 1986 Productivity decline in sown tropical grass pastures with age: the problem and possible solutions. Proceedings of the Australian Society of Animal Production 16: 319-322

Skerman P J and F Riveros 1990 Tropical Grasses. FAO Plant Production and Protection Series No. 23. Rome, Italy. pp. 284-285

Soya H 1994 Effect of the seed yield and yield charachristics of common vetch (Vicia sativa L.) on the percentage of barely (Hordeum vulgare L.) as a nurse crop in mixtures and row spacing.Anadolu 4: 8-18

Steel R G D and Torrie J H 1980 Principles and Procedures of Statistics. pp. 30-56. (McGraw-Hill Book Company: New York, USA)

Tamire Hawando 1982 Summary results of soil research programme during crop Season at Hararghe highlands, eastern Ethiopia. Department of Plant Sciences, College of Agriculture. Addis Ababa University

Tarawali S A, Tarawali G, Garbi G and Hanson J 1995 Methods for the evaluation of legumes, grasses and fodder trees for use as livestock feed. ILRI Manual I. ILRI, Nairobi, Kenya http://www.ilri.org/html/trainingMat/Forage.pdf

Teshome Shenkoru 1987 Nutritional value, dry matter yield and in vitro dry matter digestibility as affected by cutting interval and fertilizer application on native natural pasture grown on the different soil types at Alemaya University of Agriculture Campus. M. Sc. Thesis. School of graduate studies of Alemaya University of Agriculture, Dire Dawa, Ethiopia

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: 104-111

Van Soest P J 1982 Nutritional Ecology of the Ruminant. O. and B books (editors), Inc Cornell University. USA. 373p

Van Soest P J 1994 Nutritional Ecology of Ruminants. 2^nd Edition. pp. 50-65, 478 (Cornel University Press: Ithaca, New York, USA)

Van Soest P J and Robertson J B 1985 Analysis of Forages and Fibrous Foods. A Laboratory Manual for Animal Science, 613, Cornell University, USA

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

Whiteman P C 1980 Tropical pasture science. (Oxford University Press: New York, USA)

Willey R W 1979 Intercropping-its importance and research needs. Part I. Competition and yield advantages. Field Crop Abstracts, 32: 72-79

Willey R W and Rao M R 1980 A competitive ratio for quantifying competition between intercrops. Experimental Agriculture. 16:117-125

Received 21 October 2007; Accepted 11 November 2007; Published 1 February 2008

Go to top