Effect of management practices on rhodes grass and lucerne pastures with special references to developmental stages at cutting and associated changes in nutritional quality

Daniel Keftasa
Institute of Agricultural Research,
Kulumsa Research Centre, Ethiopia.

Abstract
Introduction
Methodology
Results
Discussion
Acknowledgements
References

Abstract

Rhodes grass (Chloris gayana) fertilized at 0 or 138 kg N/ha/yr and lucerne (Medicago saliva) were grown at Kulumsa (8° N. 2200 m altitude) representing medium-highland zones of Ethiopia. Samples were taken every 10 days for about 100 days during two growing seasons (March-May and July-November) in 1986. The samples were analysed for dry matter yield, crude protein (CP), organic matter digestibility (OMD), fibres (NDF, ADF), Ash, lignin and major mineral elements.

Nitrogen fertilization increased yield, rate of growth and improved CP content and OMD during the earlier part of growth but N-fertilized Rhodes grass produced reproductive inflorescence earlier and produced forage of low CP, high fibres, low OMD, P. K, Mg and Na at the advanced stage of growth. The average rate of decline in CP due to advance in maturity was 0.14 and 0.11%/day with and without N-fertilization respectively. The corresponding decline in OMD was 0.28 and 0.19%/day respectively. Lucerne produced forage of high CP, low fibres, high OMD, K, Mg, Ca and high lignin at all stages of growth and the minimum levels of CP & OMD reached due to advance in maturity were significantly higher than those of Rhodes grass.

It was concluded that maturity stage at cutting is the most important factor which determines the quality of Rhodes grass pasture and cutting at 10-50% heading or about 50 days regrowth period can be recommended.

As to the utilisation of these and such research findings at present in Ethiopia, some issues were raised. Lack of appropriate integrated livestock/forage crops research, extension and production systems, poor research-extension linkage, less emphasis on livestock and forage extension, shortage of inputs (e.g. forage seeds) are recognised as some of the problems facing forage development in Ethiopia to date.

Introduction

Livestock production plays an important role in Ethiopian agriculture. It is an integral part of all farming systems and provides milk, meat' draught power' manure, hides and skins. Approximately 60 % (Taylor, 1984) of the total land area is utilised for raising livestock which are thus the largest single users of land resource in the country. Cattle production constitute the main component of the highland mixed farming system with small dairy herds, high traction power demand and an intensive crop and/or vegetable production. The highland is characterized by high human and livestock density leading to overgrazing, land degradation and low agricultural productivity.

The main reason for low productivity of livestock is inadequate feeds both in quantity and quality. There are few areas that can supply sufficiently good quality natural herbage for existing livestock for the whole year owing to the marked seasonality of rainfall distribution. Fodder production from improved grasses and legumes is limited to a few experimental farms due to lack of well planned crop/livestock husbandry systems, low technical expertise, restrictions relating to small farm size due to population pressure and overall low standard of agricultural production.

In recent years, the use of sown pastures has received considerable attention in areas where high producing cross-bred daily animals are owned. Rhodes grass (Chloris gayana) and lucerne (Medicago saliva) mixture is probably the most successful sown perennial grass-legume mixed pasture. Results from various experiments in low medium altitude parts of Ethiopia (Sisay 1975; Haile, 1977; Tsegahun et al, 1986; Jutzi et al 1986; FNE 1986; 1987) indicate that Rhodes grass and lucerne have a high potential as livestock feed in terms of dry matter production and nutritional value.

Rhodes grass has been a popular perennial grass in the tropics and sub-tropics of East & Southern Africa, Australia and Central America. Originating in Eastern and Southern Africa, it is valued for its (1) ability to set seed, (2) relative ease of establishment and ability to cover ground, (3) tolerance for drought, light frost, soil salinity and (4) suitability to be grown in association with many tropical legumes, clovers and lucerne. A detailed review on botanical and agronomic attributes and utilisation of Rhodes grass has been published by Bogdan (1969). Agronomic characteristics and cultivation techniques of Rhodes grass with reference to seed production in Kenya are presented in Boonman (1973) and Keftasa (1985) has presented highlights on Rhodes grass seed production in Ethiopia.

Lucerne is probably the oldest cultivated forage crop in the world and has been called the "Queen of forages". Although it originated in the Mediterranean climate of the Near East and Central Asia it is grown in almost all parts of the world. Some of its merits include tolerance to drought and low temperatures, a vigorous symbiotic relationship with Rhizobium and its high feeding value. Extensive review on lucerne have been published by Hanson (1972) and Leech (1978).

Mixed Rhodes grass-Lucerne pastures have been known to be advantageous for higher yields and quality over their monocultures. A mixture of Rhodes grass benefits from the transfer of fixed nitrogen from lucerne when the nodule and/or the roots (and shoots) decay. Lucerne contains higher digestible protein and major mineral elements (Ca, P. K, Mg) than Rhodes grass but its yield is usually lower, less persistent and more difficult to harvest and cure as hay. So Rhodes grass-lucerne mixture combines the yield and quality aspects of the component monocultures and may reduce harvesting and utilisation problems.

Benefits from mixed stands of such pastures can be efficiently exploited only if proper management strategies such as optimum fertilization and accurate cutting or grazing frequencies are followed. A number of research results and literature reviews indicate that nitrogen fertilization (Howard et al, 1962; Raymond, 1966; Clatworthy, 1967; Miaki, 1968; Whitehead, 1970; Wilson and Haydock, 1971; Henzell, 1971; 1977; Minson, 1973; Binnie, 1974; Hacker & Minson, 1981) and plant maturity at cutting (Blaser, 1964; Butterworth, 1967; Milford and Minson, 1968; Stobbs, 1971; Minson, 1971b; 1972; Soneji et al 1971; 1972; Said, 1974; Rocha and Vera, 1981; Hacker 8 Minson 1981) is the factor which influences the nutritional value of pastures. Nitrogen frequently limits grassland productivity in the tropics. Nitrogen fertilization has been well known to increase dry matter yield and protein content of the herbage but affects herbage digestibility only slightly. Whitehead (1970), Wilson (1982) and Van Soest (1982) have reviewed published reports from different sources and concluded that the change in dry matter digestibility due to fertilizer N has been positive, negative or insignificant.

It is generally recognised that the nutritive value of tropical pasture falls as they mature due to a rise in fibre content with increasing maturity. Minson, (1971a) showed that the rate of decline in digestibility of Rhodes grass was about 0.1%/day when the overall trend of a long period is considered. In another report (Minson, 1972) showed the mean rate of fall in digestibility of Rhodes grass to be 0.25%/day between 28 and 70 days of regrowth but decreased to 0.17%/day during 70-98 days of regrowth.

Minson and Milford (1967) have shown that the digestibility of different varieties of Rhodes grass declined from 0.08 to 0.15 %/day due to advanced maturity. Hacker and Minson (1981) reviewed numerous research results and concluded that after initial growth pasture plants decline in digestibility with time. They reported that the decline is more rapid in grasses than herbaceous legumes in which digestibility remain high. Raymond (1966) reviewed reports of various authors on pattern of herbage digestibility and established a prediction formula for temperate grasses showing a steady decline in digestibility with increasing stages of maturity. A detailed work on this subject by Said (1974) shows that digestibility occurs after forage plants has headed, the reasons being increase in structural constituents (CF. cellulose and lignin) and a decrease in the non-structural constituents, mainly the soluble carbohydrates. The common explanation for the decline has been the fall in leaf: stem ratio and rise in cell wall components coupled with increased lignification.

There are a number of interesting pieces of work done on the nutritional content of lucerne in temperate regions and Rhodes grass in East Africa and Queensland, (Australia). However, full information is lacking on the yield and quality profile of Rhodes grass-Lucerne pastures relating to different cutting stages under Ethiopian conditions. The results depicated in this paper obtained from field experiments carried out to study effects of nitrogen fertilization and maturity stage on yield and quality of Rhodes grass-Lucerne pastures at Kulumsa; representing the medium-highland parts of Ethiopia.

Methodology

The research site is located at an altitude of 2200 m, latitude 8ºN. on a clay soil with a pH of 6.2, a P content of 32 ppm, a long time when total annual rainfall of 850 mm and with mean maximum and mean minimum temperatures of 22° and 10°C respectively.

Rhodes grass and lucerne were sown in rows of 20 cm spacing at sowing rates of 2.5 and 8 kg/ha pure germinating seed were used. Fertilizer was applied at planting and at the rate of 100 kg/ha. DAP (18/46 N/P₂0₅) was 100 kg/ha. TSP (46%, P₂05) was applied every second year. Nitrogen fertilization at the rates of 0 or 138 kg N/ha/yr was superimposed on Rhodes grass at the beginning of each growing season. Herbage samples were taken at 10 days interval to determine dry matter yield, digestibility, contents of neutral detergent fibre (NDF,) acid detergent fibre (ADF), ash, lignin and contents of crude protein (CP), calcium (Ca), phosphorus (P), magnesium (Mg), potassium (K) and sodium (Na). The samples were analysed for dry matter yield and contents of crude protein and minerals according to the specifications by Association of Official Analytical Chemists (AOAC) (1975). NDF, ADF, lignin and ash were determined with a micro fibre apparatus using the Goering and van Soest (1970) procedures. Organic matter digestibility was determined according to Lindgren (1979) and metabolizable energy content was estimated from the regression equations presented by Lindgren (1979).

Rainfall, evapotranspiration, relative sunshine hours and temperatures are shown on figures 1 and 2 respectively.

In this paper the two growing seasons are denoted as short (Fete-May) and main (June-Sept) growing seasons following the traditional terms for Ethiopia highlands.

Results

As shown in Tables 1 through 4, dry-matter yields of Rhodes grass increased steadily upto 72-83 days of regrowth period and then decreased slightly or remained high. Average rates of increases in dry-matter yields during these periods were 121 and 65 kg/ha/day in the short rainy season with and without nitrogen fertilization respectively. The corresponding rates of increase in the main rainy season were 70 and 30 kg/ha/day respectively. Dry-matter yields of lucerne was generally lower than those of Rhodes grass except in few cases in non-nitrogen fertilized Rhodes grass in the earlier part of the regrowth period. Lucerne started growth faster than Rhodes grass as soon as the rains began and also declined earlier in dry-matter yield mainly due to leaf senescence and leaf diseases such as leaf spot (Pseudopeziz medicaginis).

Nitrogen fertilization increased the crude protein content of Rhodes grass by about 15% at the early stage of growth but N fertilized Rhodes grass contained less crude protein content at the advanced growth stage. The crude protein of Rhodes grass declined markedly due to advance in maturity. The rate of decline was about 0.1 and 0.08%/day in the short and the main rainy seasons respectively in non-nitrogen fertilized Rhodes grass and 0.17 and 0.14% in the short rainy seasons and the main rain seasons respectively in nitrogen fertilized Rhodes grass. Lucerne contained high crude protein in the earlier part of regrowth period but declined at the rate of 0.15%/day during 72 days of regrowth and then stagnated around 19 %.

Nitrogen fertilization improved the organic matter digestibility of Rhodes grass in the earlier part of growth period but at the advanced stage of maturity the organic matter digestibility of nitrogen fertilized Rhodes grass was inferior to the non-fertilized ones.

The organic matter digestibility of Rhodes grass increased slightly during 25 days of regrowth in the short rainy season and declined steadily at the rate of 0.36 and 0.28%/day with and without nitrogen fertilization respectively (Table 1). In the main rainy season organic matter digestibility declined steadily at the rates of 0.20 and 0.10%/day with and without nitrogen fertilization respectively. The organic matter digestibility of lucerne was almost always higher than Rhodes grass in both seasons. It declined steadily at the rate of about 0.2%/day in the main rainy season (Table 4) during the short rainy season it increased slightly in the growth period of up to 25 days and declined then after at the rate of 0.17%/day.

Figure 1. Monthly rainfall evapotraspiration and relative sunshine and hours at at Kulumsa, A. 1986.

Figure 2. Mean monthly maximum and minimum temperature at Kulumsa, Ethiopia, 1988.

Table 1. DM yield, crude protein, organic matter digestibility and metabolizable energy contents of Rhodes grass and Lucerne at different date, with and without N-fertilization during the short rainy season, of 1986.

(N- fertilization; O = nil and + =46 kg N/ha)
Species	N-fert.	Harvesting date	Regrowth days	DM yield kg/ha %	CP %	OM dig. %	Estimated ME MJ/kg OM	Botanical stage
Rhodes	0	4/3	17	1340	13.3	75	10.1	100% leafy
Rhodes	+	4/3	17	2200	18.1	76	10.3	100% leafy
Lucerne	0	4/3	17	1240	27.7	78	11.2	100% leafy
Rhodes	0	12/3	25	1740	15.0	79	10.7	95% leafy
Rhodes	+	12/3	25	3210	17.0	81	19.1	95% leafy
Lucerne	0	12/3	25	2150	32.1	81	11.5	100% leafy
Rhodes	0	21/5	34	1880	12.3	74	9.9	90% leafy
Rhodes	+	21/3	34	3790	15.1	79	10.7	90% leafy
Lucerne	0	21/3	34	2470	26.8	81	11.5	100% leafy
Rhodes	0	31/3	44	3000	8.2	69	9.1	75% leafy
Rhodes	+	31/3	44	6680	11.9	73	9.8	75% leafy
Lucerne	0	31/3	44	3600	25.5	77	11.1	100% leafy
Rhodes	0	9/4	53	3760	9.3	70	9.3	50% headed
Rhodes	+	9/4	53	7430	8.8	67	8.8	50% headed
Lucerne	0	9/4	53	3640	22.9	74	10.8	10% flowering
Rhodes	0	28/4	72	4410	6.6	66	8.7	75% headed
Rhodes	+	28/4	72	8900	8.9	62	8.0	75% headed
Lucerne	0	28/4	72	4520	19.1	69	10.2	10% flowering
Rhodes	0	7/5	81	5640	5.3	61	7.9	80% headed
Rhodes	+	7/5	81	8150	4.8	60	7.7	Fuel heading
Lucerne	0	7/5	81	3320	20.3	72	10.6	15% flowering
Rhodes	0	17/5	91	5530	5.1	53	6.6	Full heading
Rhodes	+	17/5	91	7640	4.0	55	6.9	anthesis
Lucerne	0	17/5	91	2600	19.8	71	10.5	20% flowering
Rhodes	0	27/5	101	5670	5.1	57	7.2	Seed maturity
Rhodes	+	27/5	101	7290	4.4	50	6.1	Seed maturity
Lucerne	0	27/5	101	2240	19.9	71	10.5	20% flowering

Mean ± standard deviation (SD)

Rhodes No	3663 ± 1757	8.9 ± 3.8	67 ± 8.7	8.8 ± 1.4
Rhodes N+	6143 ± 2419	10.3 ± 5.5	67 ± 11.0	8.7 ± 1.6
Lucerne No	2864 ± 991	23.8 ± 4.5	75 ± 4.5	10.9 ± 0.5

As shown in Tables 2 and 5 both the NDF increased with advance in maturity. It appears that both NDF and ADF were lower in nitrogen fertilized Rhodes grass if cut early but higher if cut late (advanced maturity). This feature of NDF and ADF contents followed the pattern of crude protein content and organic matter digestibility. Lucerne contained less NDF and ADF than Rhodes grass at all stages of cutting and their contents increased at slower rate than that of Rhodes grass. The ADF ash of Rhodes grass was found to be quite high as compared to lucerne indicating the presence of large amount of silica which could severely interfere with digestibility. It appears higher in nitrogen fertilized Rhodes grass and its trend of increase with maturity is not consistent. The ADF ash content of lucerne was somewhat higher in the earlier part of growth and was not detectable in most of the cases in the latter part of growth. The lignin content of both Rhodes grass and lucerne increased with maturity. Lucerne contained more lignin than Rhodes grass and there were no marked differences in lignin contents of Rhodes grass due to nitrogen fertilization.

As shown in Tables 3 and 6, generally the total ash content of both species declined as the maturity advanced but some increases were also observed in non-nitrogen fertilized Rhodes grass. It was observed that the P. K, Mg and Na contents declined due to advance in maturity in both species but the Ca content was fluctuating and the trend was not consistent. This higher content of Ca in the middle (about 50 days regrowth) and at the latter growth stage might be due to the role of Ca in the plant as a structural element. Lucerne contained more K, Ca, Mg and Na than Rhodes grass in almost all of the cases. The P contents of Rhodes grass and Lucerne were not distinctly different but it appears that Rhodes grass contained higher P than lucerne at early part of the cutting stages but less in the later part. Ca:P ration increased with maturity, the peak being around 50 days growth period and the magnitude is higher in lucerne than Rhodes grass.

The crude protein content, organic matter digestibility and mineral contents of Rhodes grass reported in this paper are comparable to the earlier findings in Ethiopia (Evaldson, 1969), Uganda (Soneji et al, 1971; 1972) and Kenya (Said 1974; Abate et al, 1981). This work indicates that stage of maturity at cutting is the most important aspect to determine feeding value. Both yield and quality varied according to growing season. That the short rainy season produced forage of high yield and quality in this particular study may be due to higher maximum and minimum temperatures (Figure 2) more irradiation and sufficient rainfall condition (Figure 1). Work is going on at Kulumsa (Keftasa, unpublished) to demonstrate the practical application of these findings to adopt more frequent cutting systems in a year than the conventional single-cut in the main growing season.

Table 2. Neutral detergent fibre (NDF), Acid detergent fibre (ADF) ADF ash and Lignin contents of Rhodes grass and Lucerne at different cutting dates, with and without N-fertilization during the short rainy of season 1986.

N-fertilization 0 = nil, +=46 kg/ha

Species	N-fert.	Harvesting date	Regrowth days	NDF	ADF	ADF ash	Lignin
				% DM
Rhodes	0	4/3	17	62.7	33.0	7.6	3.8
Rhodes	+	4/3	17	58.3	30.4	7.8	4.0
Lucerne	0	4/3	17	31.1	27.2	0.6	5.1
Rhodes	0	12/3	25	61.1	30.6	6.7	3.7
Rhodes	+	12/3	25	61.0	30.9	6.0	3.6
Lucerne	0	12/3	25	30.9	28.8	ND	7.0
Rhodes	0	21/3	34	65.5	34.1	6.2	4.9
Rhodes	+	21/3	34	62.4	32.4	5.2	3.6
Lucerne	0	21/3	34	31.8	26.8	0.3	5.3
Rhodes	0	31/3	44	67.0	35.3	5.4	4.5
Rhodes	+	31/3	44	65.3	36.2	5.4	4.1
Lucerne	0	31/3	44	34.9	27.5	ND	5.3
Rhodes	0	9/4	53	70.6	36.7	4.5	4.4
Rhodes	+	9/4	53	69.1	38.1	5.7	5.0
Lucerne	0	9/4	53	38.5	31.0	0.5	6.5
Rhodes	0	28/4	72	72.9	40.5	4.4	4.7
Rhodes	+	28/4	72	68.2	39.0	6.2	5.9
Lucerne	0	28/4	72	41.6	31.9	ND	7.1
Rhodes	0	7/5	81	17.8	41.2	5.0	5.6
Rhodes	+	7/5	81	72.0	40.9	4.4	5.2
Lucerne	0	7/5	81	41.5	31.6	0.5	7.4
Rhodes	0	17/5	91	71.7	41.1	3.9	4.8
Rhodes	+	17/5	91	73.6	4.26	4.0	5.5
Lucerne	0	17/5	91	41.4	31.3	ND	7.4
Rhodes	0	27/5	101	74.3	40.4	5.1	5.9
Rhodes	+	27/5	101	75.9	42.3	5.2	5.6
Lucerne	0	27/5	101	43.4	32.3	0.2	7.7

Table 3. Total ash, Phosphorus, Calsium, Potassium Magnesium and Sodium contents of Rhodes grass and lucerne at different cutting dates, with and without N-fertilizations during the short rainy season of 1986.

N-fertilization 0 = nil +=46 kg N/ha

Species	N-fert	Harvesting date	Regrowth day	Ash	P	Ca	K	Mg	Na ppm
				% of DM
Rhodes	0	4/3	17	18.0	.55	.54	2.82	.20	128.6
Rhodes	+			16.7	.45	.59	3.60	.22	141.1
Lucerne	0			13.2	.44	1.53	4.15	.25	127.0
Rhodes	0	12/3	25	15.8	.48	.55	3.32	.18	102.1
Rhodes	+			14.5	.40	.43	3.72	.19	87.4
Lucerne	0			15.2	.42	1.08	4.71	.25	117.8
Rhodes	0	21/3	44	14.3	.43	.41	3.21	.16	112.6
Rhodes	+			15.2	.28	.60	3.72	.19	93.5
Lucerne	0			12.3	.28	1.19	4.31	.22	110.8
Rhodes	0	31/3	53	12.4	.40	.51	1.38	.13	75.1
Rhodes	+			13.8	.22	.58	3.49	.17	86.1
Lucerne	0			12.4	.25	2.19	3.54	.22	113.0
Rhodes	0	94/4	62	11.6	.23	.50	2.49	.14	93.9
Rhodes	+			13.1	.25	.55	2.54	.19	77.2
Lucerne	0			12.2	.23	1.66	2.90	. 19	103.8
Rhodes	0	28/4	72	10.8	.34	0.40	2.20	.11	72.2
Rhodes	+			12.6	.31	.50	2.06	.12	62.4
Lucerne	0			11.3	.25	2.07	2.74	.20	71.1
Rhodes	0	7/5	81	10.7	.28	.36	2.14	.11	58.8
Rhodes	+			10.6	.28	.31	2.21	.11	56.6
Lucerne	0			11.6	.25	1.52	2.65	.18	99.2
Rhodes	0			9.9	.18	.37	2.09	.10	72.0
Rhodes	+			9.8	.15	.34	2.17	.10	49.5
Lucerne	0			10.9	.21	1.56	2.64	.21	124.5
Rhodes	0	27/5	101	10.4	.21	.34	1.52	.10	70.1
Rhodes	+			9.7	.15	.37	1.55	.10	59.3
Lucerne	0			10.3	.24	1.40	2.56	.19	48.30

Means ± SD

Rhodes No	12.6±2.8	0.35±0.13	0.44±0.08	2.46±0.57	0.14±0.04	87.3±13.3
Rhodes N+	12.9±2.5	0.28±0.10	0.47±0.11	2.78±0.85	0.15±0.04	79.2±27.9
Lucerne No	11.9±0.9	0.29±0.08	1.58±0.36	3.35±0.84	0.21±0.03	101.7±26.0

Table 4. DM yield, crude protein content, organic matter digestibility and metabolizable energy contents of Rhode grass and lucerne at different cutting dates with and without N-fertilization during the, main rainy season, of 1986.

N-fertilization 0 = nil; + = 92 kg N/ha)

Species	N-fert.	Harvesting date	Regrowth days	DM yield kg/ha	CP %	OM dig %	Estimated ME MJ kg Om	Botanical stage
Lucerne	0	29/7	14	1280	30.1	82	11.6	100% leafy
Lucerne	0	8/8	24	1650	30.8	78	11.2	100% leafy
Rhodes	0	18/8	34	1200	8.8	69	9.7	100% leafy
Rhodes	+	18/8	34	2140	10.0	74	9.9	100% leafy
Lucerne	0	18/8	34	2100	22.7	76	11.0	100% leafy
Rhodes	0	6/9	43	1860	6.7	65	8.5	10% heading
Rhodes	+	6/9	43	3460	8.0	68	9.0	10% heading
Lucerne	0	6/9	43	2223	21.3	74	10.8	20% flowering
Rhodes	0	16/9	53	1980	6.5	62	8.0	15% heading
Rhodes	+	16/9	53	3840	6.9	68	9.0	15% heading
Lucerne	0	16/9	53	2471	22.2	73	10.7	30% flowering
Rhodes	0	26/9	63	2150	5.1	65	8.5	50% heading
Rhodes	+	26/9	63	4220	6.6	66	8.7	50% heading
Lucerne	0	26/9	63	2250	22.6	74	10.8	40% flowering
Rhodes	0	6/10	72	2260	5.3	65	8.5	90% heading
Rhodes	+	6/10	72	5680	6.0	68	8.2	90% heading
Lucerne	0	6/10	72	2500	18.8	68	10.1	40% flowering
Rhodes	0	17/10	83	2400	4.5	61	7.9	Anthesis
Rhodes	+	17/10	83	5880	5.9	65	8.5	Anthesis
Lucerne	0	17/10	83	1420	18.8	67	10.0	50% flowering
Rhodes	0	28/10	94	1760	4.1	66	8.7	Seed maturity
Rhodes	+	28/10	94	4900	3.7	57	7.2	Seed maturity
Rhodes	0	10/11	107	1800	4.3	61	7.9	Seed maturity
Rhodes	+	10/11	107	4970	3.0	58	7.4	Seed maturity

Mean ± SD

Rhodes No	1936±368	5.7±1.6	64±2.8	7.4±0.43
Rhodes N+	4386±1239	6.3±2.2	65±5.6	8.5±0.89
Rhodes No	2112±433	23.4±4.6	73±6.3	10.8±0.53

Table 5: Neutral detergent fibre (NDF), Acid detergent fibre (ADF) ADF ash and lignin contents of Rhodes grass and lucerne at different cutting dates, with and without N-fertilization during the main rainy season, of 1986.

Species	N-fert.	Harvesting date	Regrowth days	NDF	ADF	ADF ash	Lignin
							% of DM
Lucerne	0	29/7	14	32.9	28.9	0.24	4.8
Lucerne	0	8/8	24	35.5	28.3	0.22	5.4
Rhodes	0	18/8	34	68.9	39.1	6.6	4.5
Rhodes	+	18/8		65.3	37.4	3.5	4.5
Lucerne	0	18/8		37.2	31.6	ND	5.9
Rhodes	0	6/9	43	69.1	38.5	8.2	4.1
Rhodes	+			68.4	37.5	5.0	4.1
Lucerne	0			36.4	29.3	0.29	5.9
Rhodes	0	16/9	53	69.2	27.6	7.6	4.2
Rhodes	+			70.4	37.7	3.6	3.7
Lucerne	0			39.8	31.2	ND	7.2
Rhodes	0	26/9	63	70.0	37.4	7.0	4.3
Rhodes	+			71.2	37.7	4.3	4.1
Lucerne	0			40.8	31.4	ND	7.0
Rhodes	0	6/10	72	70.4	37.7	7.0	4.2
Rhodes	+			72.7	39.4	5.2	4.9
Lucerne	0			45.2	31.4	ND	7.6
Rhodes	0	17/10	83	68.3	37.7	8.5	4.8
Rhodes	0			59.9	39.7	4.6	5.1
Lucerne	-			46.6	37.9	ND	8.0
Rhodes	0	28/10	94	65.5	36.3	8.6	4.4
Rhodes	+			72.5	43.2	4.9	5.7
Rhodes	0	10/11	107	67.0	37.9	8.6	5.0
Rhodes	+			71.6	41.3	6.0	5.5

Mean ± SD

Rhodes No	68.6±1.6	37.8±0.8	7.8±0.8	4.4±0.3
Rhodes N+	70.3±2.4	39.2±2.1	4.6±0.8	4.7±0.7
Lucerne No	39.3±4.76	31.3±3.0	-	6.5±1.1

Table 6. Total ash phosphorus calcium, potassium, magnesium and sodium contents of Rhodes grass and lucerne at different cutting dates with and without N-fertilization during the main rainy season of 1986.

Species	N-fert.	Harvesting date	Regrowth days	Ash	P	Ca	K	Mg	Na
						% of DM		ppm
Lucerne	0	29/7	14	12.9	.64	1.47	5.16	.27	89.4
Lucerne	0	8/8	24	12.6	.48	1.71	4.80	0.26	98.1
Rhodes	0	18/8	34	13.5	.51	.54	2.02	.13	31.3
Rhodes	+	18/8		12.9	.50	.48	3.26	.19	64.5
Lucerne	0	18/8		11.4	.33	2.25	3.75	.23	136.9
Rhodes	0	6/9	43	12.9	.55	.51	1.88	.13	49.4
Rhodes	+	6/9		10.8	.45	.50	2.78	.14	52.1
Lucerne	0	6/9		11.6	.23	2.75	3.05	0.21	133.0
Rhodes	0	16/9	53	14.2	.54	.51	1.57	.13	34.4
Rhodes	+	16/9		10.6	.45	.44	2.62	.12	50.4
Lucerne	0	16/9		11.3	.28	2.47	2.62	.22	122.3
Rhodes	0	26/9	63	13.2	.52	.51	1.81	.13	48.9
Rhodes	+	26/9		10.6	.35	.47	2.15	.13	49.3
Lucerne	0	26/9		9.9	.32	2.17	2.50	.21	95.0
Rhodes	0	6/10	72	12.4	.42	.51	1.68	.10	8.1
Rhodes	+	6/10		10.9	.33	.50	2.03	.11	31.1
Lucerne	0	6/10		9.0	.26	1.93	2.35	.19	43.3
Rhodes	0	17/10	83	14.1	.45	.61	1.63	.14	13.8
Rhodes	+	17/10		11.0	.30	.54	2.42	.12	36.2
Lucerne	0	17/10		8.5	.28	1.61	2.70	.19	151.9
Rhodes	0	28/10	94	15.4	.47	.58	1.77	.15	13.1
Rhodes	+			10.1	.29	.42	1.84	.12	22.2
Rhodes	0	10/11	107	15.3	.30	.61	1.55	.15	11.9
Rhodes	+	10/11		11.0	.29	.54	1.79	.12	27.0

Means ± SD-Rhodes No	13.88±1.08	0.47±0.08	0.55±0.05	1.73±0.13	0.13±0.02	26.4±16.9
Rhodes N+	10.99±0.83	.37±0.08	0.49±0.04	2.34±0.50	0.13±0.03	43.7±14.5
Lucerne No	10.90±1.61	0.35±0.14	2.05±0.44	3.37±1.09	0.21±0.04	108.7±34.6

Discussion

Various field studies and laboratory analyses concluded that the existing feedstuffs in Ethiopia, native pastures and crop residues, are poor in quality thus providing inadequate protein, energy, vitamins and minerals. In certain areas where improved forage crops are introduced farmers fail to utilise them at the optimum developmental stage when the herbage is high in quality to satisfy livestock needs and support production. One reason for this failure is lack of strong extension support in adopting the recommended practices.

In Ethiopia the Institute of Agricultural Research (IAR) with the began research on crops, livestock and forages began in 1966. IAR's broad objective was the formulation of a national policy for agricultural research which should be within the framework of the country's central plan. This national policy for agricultural research should then be to implement through coordinated programmes of applied research. Soon after the establishment of JAR, the Chilalo Agricultural Development Unit (CADU), the first comprehensive pilot project, was established in Arsi region in 1967 under the Ministry of Agriculture (MOA). CADU undertook various research programmes in cattle crossbreeding of cattle and forage evaluation in addition to the extension programme. As a result of the effort of these two major institutions in forage research, it has been possible to recommend a number of forage species and their cultural practices for different ecological zones of Ethiopia (Lulseged Gebre-Hiwot and Tadesse, 1985).

Agricultural extension was initiated since the 1950s. The Ministry of Agriculture has been the main organisation responsible for organising, giving advice and assistance as well as services to the peasant sub-sector. Within the Ministry today, departments for animal breeding and feed resources development are directly concerned with introduction and utilisation of improved forage crops.

The Animal Breeding and Feed Resources Development Department through its various projects (e.g. Fourth Livestock Development Project, FLDP) promotes forage development, introduction of improved forages crops and management techniques, forage seed production on farmers fields (on contract basis), adaptive research and training of farmers and development agents. The Soil Conservation Development Department multiplies seeds of grasses and multipurpose forage shrubs on nursery sites and undertakes plantations on catchment sites. The two departments work in close co-operation.

At the field level the linkage between research and extension has been weak in general and worse in the fields of livestock and forage development. To promote the research extension linkage certain measures have been taken in crop production, such as IAR/MOA joint research programmes, farming system research (on-farm trials) and research-extension liaison committees of which none is involved in livestock/forage fields. Although there are development agents on over 2000 centres (FAO, 1985), quite a small propotion (in terms of time or resource allocation is working in this field) indicates that there is less emphasis on livestock and forage development. The current research and extension systems focus mainly on improving of major cash crops (cereals, pulses, oil crops, cotton and coffee) with little integration with the livestock subsystem in such areas as forage legume-based crop rotation, alley cropping, etc.

There is an acute shortage of forage seeds in the country. There is still no agency responsible for production and certification of forage seeds. There are some forage seed production programmes within the MOA at the South-Eastern Zonal Office for Agricultural Development (ex CADU), FLDP and soil conservation nurseries. However, production has been much below the requirements. There is considerable demand for forage seeds for planting at state-owned cattle breeding stations dairy development and beef production enterprises as well as dairy cooperatives and individual smallholders who own upgraded dairy cows. Utilisation of improved forage crops at the smallholder level has not been successful due to low return to focal zebu cattle and competition for scarce land and labour. At present the use of fertilizers on pasture lands remains to be questionable due to limited availability, low level of credit services and capital investment. FAO (1985) reported that only 14% of Ethiopia farmers used fertilizers (8.76 kg/farmer) but that only 2% use improved seeds (food crops). In Arsi region where research and extension programmes have been relatively more effective, the corresponding figures were 47% (31.75 kg farmer) and 21% fertilizers and improved seeds respectively.

Experience in Arsi region shows that small-scale farmers and farmers co-operatives who own cross-bred dairy cows adopted favourably forage production. Oats/vetch mixtures, fodder beet and Rhodes grass are planted about 600 ha of land every year with dry matter production/ha comparable to state-owned dairy and cattle breeding farms. Any recommended package related to improved forage crops may need to be accompanied by improvement of the existing low producing animals otherwise the package may not be economical.

The introduction of the training and visit extension system is expected to improve the currently weak research-extension linkages through regular training of development agents who visit farmers frequently and regularly to pass on relevant technical messages and also bring back farmers' problems to researchers. The integration of forage development with soil conservation is expected to reduce competition for arable land and help to develop a good landuse system. Emphasis should be placed on forage legumes for a sound integration of forage production in cereal crop rotation systems requiring minimum inorganic fertilizer input particularly at smallholder level. The success of forage production in Ethiopia would be measured through its incorporation into dairy and beef production programme.

Acknowledgements

I would like to thank Dr. John Tothill and Professor Abdullah N. Said of ILCA for their sincere interest in my work and for their encouragement during the preparation of this paper.

Many thanks also go to Dr. Seme Debela, General Manager of the Ethiopian Institute of Agricultural Research, for his permission and assistance to attend this workshop.

References

Abate, A., Kayongo-Male, H., Karue, C.N. 1981. Dry matter protein, energy and fibre intake by dairy heifers grazing a Rhodes grass Chloris gayana) pasture. Animal Feed Science and Technology. 6:15-26.

AOAC (Association of Official Analytical Chemists). 1975. Official methods of analysis (12th ed) Association of Official Analytical Chemists. Washington. D.C.

Binnie, R.C., Harrington, F.J., Murdoch, J.C. 1974. The effect of cutting height and nitrogen level on the yield, in vitro digestibility and chemical composition of italian rye grass swards. J. Br. Grassld Soc., 29:57-61.

Blaser, R.E. 1964. Symposium on forage utilisation. Effects of fertility levels and stage of maturity on forage nutritive value. J. Anim. Sc. V.23 No. 1:246-253.

Bogdan, A.V. 1969. Rhodes grass. Review article, Herbage Abstracts V. 39. (1).

Boonman, J.G. 1973. On the seed production of tropical grasses in Kenya. Agr. Rep. (Versal land bouwk. onderz) 794 Netherlands.

Butterworth, M.H. 1967. The digestibility of tropical grasses. Nutrition Abstracts and Reviews. 37:349-368.

Clatworthy, J.N. 1967. The response of giant Rhodes grass (Chloris gayana) to heavy dressing of nitrogen fertilizer. Rhodesia, Zambia and Malawi. J. Agri. Res. 5(1):87-95.

Evaldson, O. 1969. An inventory of feeding systems and feed stuffs in Chilalo Awraja, Ethiopia. Minor research tasks at CADU No. 5.

FAO, (Food and Agricultural Organisation of the United Nations). 1985. Proceedings of the research-extension linkage workshop, Nazareth, Ethiopia.

FNE (Forage Network of Ethiopia). 1986. Results of 1985/86 FNE trials, FNE Newsletter No. 11 pp. 19-27. ILCA, Addis Ababa. Ethiopia

FNE (Forage Network of Ethiopia) 1987. Results of 1986/87 FNE trials. FNE Newsletter No. 16, pp. 17-23 ILCA, Addis Ababa, Ethiopia.

Gegre-hiwot, L. and Tadesse, A. 1985. Pasture Research and development in Ethiopia. In: J.A. Kategile (ed) proceedings of a workshop on pasture improvement research in Eastern and Southern Africa Harare, Zimbabwe. pp. 71-91.

Goering, H.K. and van Soest, P.J. 1970. Forage fibre analysis (Apparatus, reagents, procedures and some application) USDA Hand book No. 379, US Govt, printing office, Washington, DC. pp. 1-20.

Hacker, J.B. and Minson, D.J. 1981. The digestibility of Plant parts, herbage abstract Vol. 51 No. 9:459-482 review article.

Haile, A. 1977. Results of experiments in forage crops and pasture management in the highlands of Ethiopia 1971-76 JAR, Ethiopia. Forage and Range Bulletin No. 1 pp. 24.

Hanson, C.H. 1972. Alfalfa science and technology. American Soc. Agron. Monography No. 15.

Henzell, E.F. 1971. Recovery of nitrogen from fertilizers applied on Rhodes grass in small plots. Aust. J. Exp. Agric. anim. husb. vol 11 (51): 420-30.

Henzell, E.F. 1977. Nitrogen nutrition of tropical pastures in P.J. Skerman (ed) Tropical forage resumes F.A.O. Rome 609 pp.

Howard, D.A. Burdin, M.L. and Lampkin, G.H. 1962. Variation in the mineral and crude protein content of pastures at Muguga in the Kenya highlands. J. Agric. Sci. 59:251-256.

Jutzi, S., Haque, I. and Tedla, A. 1986. The production of animal feed in the Ethiopian highlands potentials and limitations. Paper presented at the National livestock conference Dec. 3-5, 1986 Addis Ababa, Ethiopia. pp. 141-142.

Keftasa, Daniel, 1985. Rhodes grass seed production. Forage network in Ethiopia Newsletter No. 7. ILCA, Addis Ababa, Ethiopia. pp. 19-21.

Leech, G.T. 1978. The ecology of lucerne pastures. In: J.R. Wilson (ed). Plant relations in pastures. pp. 290-308.

Lindgren, E. 1979. The nutritional value of roughages determined in vitro and by laboratory methods. Swedish Univ. Agric. Sci. Department of animal nutrition. Report No. 45. Uppsala.

Miaki, T. 1968. Studies on chemical composition and feeding value of forage crops. II. An influence of level of nitrogen fertilization and stage of maturity on chemical composition and feeding value of Rhodes grass (Chloris gayana) hay. Jap. J. Zooetech. Sci. 39(1):48-52.

Milford, R. and Minson, D.J. 1968. The digestibility and intake of six varieties of Rhodes grass (Chloris gayana) Aust. J. exp. Agric. Anim. Husb. 8:413-418.

Minson, D.J. 1971a. The digestibility and voluntary intake of six varieties of panicum. Aust. J. Exp. Agric. Anim. Husb. 11:18-24.

Minson, D.J. 1971b. The nutritive value of tropical pastures J. Aust. Inst. Agric. Sc. 37:255-263.

Minson, D.J. 1972. The digestibility and voluntary intake by sheep of six tropical grasses. Aust. J. Exp. Agric. Anim. Husb. 12(54):21-27.

Minson, D.J. 1973. Effect of fertilizer nitrogen on digestibility and voluntary intake of Chloris gayana, Digitaria decumbens and Pennisetum clandestinum: Aust. J. Expt. Agric. Anim. Husb. 13:153-157.

Raymond, W.F. 1966. The nutritive value of herbage. In: J.T. Abrams (ed) Recent Advances in Animal Nutrition. J & A Churchill Ltd. London.

Rocha, G.P. and Vera, R.R. 1981. Structural carbohydrates, protein and in vitro digestibility of eight tropical grasses. Turrialba (costa rice) Jan-Mar. 1981. Vol 31(1):15-20.

Said, A.N. 1974. The nutritive value of some Kenya feedstuffs and the effect on protein and energy rich concentrate supplementation on the utilisation of Chloris gayana hay by wether sheep. Ph.D. thesis, University of Nairobi, Nairobi.

Sisay, Bekele. 1975. Effect of harvesting interval and seeding rate on percentage alfalfa, dry matter yields in alfalfa and rhodes grass mixtures. Proceedings of 5th Annual research seminar. IAR Holleta Ethiopia. pp. 190-205.

Soneji, S.V., Musangi, R.S. and Olsen, F.J. 1972. Digestibility and feed intake investigations at different stages of growth of Brachiaria ruziziensis, Chloris gayana and Setaria sphaceluta using corriedale wether sheep. I. Digestibility and voluntary intake. East Afr. Forest J. 37(2):125-128.

Soneji, S.V., Musangi, R.S. and Olsen, F.J. 1971. Digestibility and feed intake investigations at different stages of growth of Brachiania ruziziensis, Chloris gayana and Setaria sphaceluta using corriedale wether sheep. II. Chemical composition and yield. East Afr. Agric. Forest. J. 37(4):267-271.

Stobbs, T.H. 1971. Quality of pasture and forage crops for dairy production in the tropical regions of Australia 1. Review of the literature. Tropical grasslands vol 5 No.3.

Taylor, M.S. 1984. Assistance to land use planning in Ethiopia. Livestock production and feed resources F.A.O. Rome, technical report No. 6.

Tsegahun, A., Hagos, G.M. Kandiah, A. 1987. Optimum irrigation frequency and amount of alfalfa/Rhodes grass mixed pasture in the middle Awash region. Proceedings of the First National Livestock Improvement Conference, 1987 Addis Ababa, Institute of Agricultural Research Ethiopia. pp. 167-169.

Van Soest, P.J. 1982. Nutritional ecology of the ruminant. O and B books. Inc. Oregon, USA.

Whitehead, D.C. 1970. The role of nitrogen in grassland productivity. The Grassland Research Institute, Hurley, Bulletin 48.

Wilson, J.R. and Haydock, K.P. 1971. The comparative response of tropical and temperate grasses of varying levels of nitrogen and phosphorus nutrition. Aust. J. Agric. Res. 22:573-583.

Wilson, J.R. 1982. Environmental and nutritional factors affecting herbage quality. In: J.B. Hacker (ed) Nutritional limits to animal production from pastures Commonwealth Agricultural Bureaux, Farnham Royal, U.K. pp. 111-131.