Intercrop Stylosanthes effects on millet yields and animal performance in the Sahel

C Kouamé ¹*, S Hoefs ¹, J M Powell ¹, D Roxas ² and C Renard ³

¹ International Livestock Centre for Africa (ILCA)
ICRISAT Sahelian Center
BP 12404, Niamey, Niger

² University of the Philippines at Los Baños
College Laguna, Philippines

³ International Crops Research Institute for the Semi-Arid Tropics (ICRISAT)
ICRISAT Sahelian Center
BP 12404, Niamey, Niger

* Present address
Institut des savanes
BP 633, Bouaké, Côte d'Ivoire

ABSTRACT

The increasing interest in crop-livestock integration in sub-Saharan Africa has emphasized the beneficial uses of forage legumes in cereal-based cropping systems. Forage legume-cereal cropping systems were evaluated in 1989 and 1990 at the ICRISAT (International Crops Research Institute for the Semi-Arid Tropics) Sahelian Center along with the performance of sheep fed millet residues as a basal diet and supplemented with forage legume hay.

Intercropping Stylosanthes fruticosa (Retz.) Alston and S. hamata (L.) Taub. with millet (Pennisetum glaucum [L.] R. Br.) did not significantly affect grain yields during the legume establishment year. During the second year, when millet was planted into pre-established stylo, total biomass and crude-protein yields increased by 45 and 125%, respectively, yet millet grain yield decreased by more than 30%. Sheep supplemented with 500 g stylo hay per day consistently gained weight during 60-70 day pen-feeding and grazing trials.

Management strategies are needed to fully exploit the agronomic benefits and feeding value of forage legumes introduced into Sahelian mixed farming systems.

RESUME

Effet de l'introduction de Stylosanthes dans les systèmes agraires du Sahel sur la production de mil et les performances animales

L'intérêt croissant que suscite l'intégration de l'agriculture et de l'élevage en Afrique subsaharienne a permis de faire ressortir les avantages de l'introduction des légumineuses fourragères dans les systèmes agraires à base de céréales. Des systèmes agraires associant cultures de légumineuses et de céréales ont été évalués en 1989 et 1990 au Centre sahélien de l'ICRISAT (Institut international de recherche sur les cultures des zones tropicales semi-arides). Ces études avaient été effectuées simultanément avec des travaux visant d'examiner les performances d'ovins soumis à un aliment de base composé de résidus de mil et complémenté avec du foin de légumineuses.

L'introduction de Stylosanthes fruticosa (Retz.) Alston et de S. hamata (L.) Taub entre les rangées de mil (Pennisetum glaucum) n'avait aucun effet significatif sur la production de grains au cours de l'année d'établissement des légumineuses. Au cours de la seconde année, lorsque le mil était semé entre ces légumineuses, les productions de biomasse totale et de protéines brutes avaient augmenté de 45% et de 125% respectivement, alors que le rendement en grains du mil avait baissé de plus de 30%. Des ovins soumis pendant 60 à 70 jours à un régime d'alimentation à l'auge et de pâturage le jour complémenté par 500 g de foin de Stylosanthes enregistraient des gains réguliers de poids.

Des efforts doivent être déployés pour élaborer des méthodes de gestion permettant de tirer pleinement parti des avantages agronomiques et nutritionnels de l'introduction des légumineuses fourragères dans les systèmes agraires mixtes du Sahel.

INTRODUCTION

The introduction of forage legumes in cereal-based cropping systems is a promising strategy for increasing crop and livestock productivity in sub-Saharan Africa (Gryseels and Anderson, 1983; Tothill, 1986). Forage legumes can enhance soil fertility, improve yields and nutritive value of harvested products, sustain food production and combat erosion (Mohamed-Saleem, 1985; le Houérou, 1989; Izaurralde et al, 1990; Garba and Renard, 1991). However, the beneficial effects of legumes vary according to crop species, management and environmental factors (Waghmare and Singh, 1984; Nnadi and Haque, 1988; Varvel and Peterson, 1990). Because forage legumes do not contribute directly to food security, farmers are reluctant to devote land and other resources solely to forage production. The adoption of forage crops by farmers will, therefore, depend on the demonstration of their productivity and subsequent positive impact on cereal and livestock production, The following studies were conducted to determine the effect of forage legume-cereal intercropping on yields, nutrient uptake and fodder quality, and performance of sheep fed cereal stover as a basal diet and supplemented with forage legume hay.

MATERIALS AND METHODS

The studies were conducted during 1989 and 1990 at the International Crops Research Institute for the Semi-Arid Tropic (ICRISAT) Sahelian Center (ISC), at Sadoré (latitude 13° 15' N. longitude 2° 18' E), Niger. The sandy soil (Siliceous Isohyperthermic Psammentic Paleustalf) of ISC is acidic (pH 5.6 in 1:1 soil:water mixture) and low in native fertility (West et al, 1984). Average annual rainfall is 560 mm; rainfall was 623 mm in 1989 and 499.5 mm in 1990.

Agronomic trials

Legumes for the intercropping trials were selected based on their previous superior performance in screening evaluations. Seed of Stylosanthes fruticosa (Retz.) Alston was collected locally and seed of S. hamata (L.) Taub. cv Verano was obtained from the International Livestock Centre for Africa (ILCA), Addis Ababa, Ethiopia. Seed of both species was multiplied at the ISC. The millet (Pennisetum glaucum [L.] R.Br.) cv CIVT, a recommended cultivar in the region, was used. Experimental treatments consisted of two planting strategies:

· millet and stylo sown on the same day
· millet sown into year-old pre-established stylo; and seven cropping patterns:
· pure millet
· pure S. fruticosa
· millet plus S. fruticosa in alternate single rows
· pure S. hamata
· millet plus S. hamata in alternate single rows
· millet plus S. fruticosa in alternate triple rows
· millet plus S. hamata in alternate triple rows.

The treatments were replicated four times in a split plot design with planting date assigned to the main plot and cropping system randomly assigned to 8 x 12 m subplots. Each plot received an annual application of 13 kg P/ha as single superphosphate which was broadcast before planting and worked into ridges with an oxen-drawn plough after the first rains. Nitrogen was applied to millet at the rate of 15 kg N/ha as calcium ammonium nitrate in a single application about 30 days after emergence. Planting occurred on 30 June 1989 and 30 May 1990. Millet was sown in pockets at 1.2-m spacing on ridges 0.75 m apart, and was thinned to three plants per pocket three weeks after sowing. Stylo was sown in continuous lines and was not thinned.

To determine treatment effects on yields, the total biomass from the innermost 3 x 4 m of each subplot was hand harvested and weighed. Millet was fractionated into panicles, leaf (blade and sheath) and stem. Subsamples of millet plant parts were sun-dried to constant weight for dry-matter (DM) determination. Panicles were threshed to determine grain yield. The legumes were harvested twice, approximately 60 and 120 days after emergence. Herbage subsamples were oven-dried (75°C, 48 hours) for DM determination. Seed yields for the legumes were determined at the second harvest. All plant samples were milled to pass a 1-mm screen, acid-digested using a modification of the aluminium block digestion procedure (Gallaher et al, 1975) and nitrogen and phosphorus were determined by semi-automated calorimetry (Hambleton, 1977).

Analysis of variance using the General Linear Model procedures (SAS, 1982) was used to determine treatment effects on yields of millet grain and stover, stylo herbage and seed and fodder crude protein. A least significant difference (LSD) test (Montgomery, 1984) was used to identify possible differences in treatment means.

Sheep feeding trials

A pen-feeding and a grazing trial were conducted in 1989 and 1990 to determine liveweight changes of sheep fed millet stover as a basal diet and supplemented with cowpea (Vigna unguiculata [L.] Walp.) or S. fruticosa hay.

In the 70-day pen-feeding trial, three groups of eight intact male sheep (average liveweight 28 kg) were randomly assigned to one of three rations:

1. millet stover only (control)
2. millet stover plus 300 g cowpea hay
3. millet stover plus 500 g stylo hay.

Millet was fed ad libitum and legume levels were determined to provide a daily nitrogen supplement of 6 g per animal. The hays were offered individually to animals at the end of the day.

In the 60-day grazing trial, 30 intact male sheep (average liveweight 25 kg) were divided into three equal groups. All animals grazed a 3.5-ha millet field during the day and the legume hays were offered individually to animals in the evening at the same level as in the pen-feeding trial.

RESULTS AND DISCUSSION

Agronomy trials

Planting time and pattern highly influenced millet and stylo yields. During the stylo seeding year, yields were highly variable and so no significant differences in grain yields among the cropping systems were detected (Table 1). However, average millet grain yields from the alternate triple row millet/S. hamata treatment decreased by 78% in 1989. Millet stover yields were 10-53% less from intercrops than from sole crops. Reductions were greatest in the millet/S. hamata association. When millet was grown with S. hamata in alternate triple rows, total biomass yield decreased by 40% in 1898 and 49% in 1990. Although stylo intercropping reduced yields in some treatments, crude-protein content of intercrop total biomass was 4-56% higher than that of sole millet crop in all treatments, except in 1990 when stylo herbage yields in intercrops were very low. Yet crude protein concentration of all harvested biomass was below 7%, the minimum maintenance level for ruminants (Humphreys, 1978). This poor feeding quality was due to the low stylo yield, probably caused by intermittent drought and poor germination during the seeding year.

Table 1. Seeding year yields of millet and Stylosanthes (millet and stylo sown on the same date) as affected by cropping system

Cropping system		Millet		Stylo herbage (kg/ha)	Residue a
		Grain (kg/ha)	Stover (kg/ha)		Biomass (kg/ha)	Crude protein (%)
1989
	Pure millet	1080	2706	-	2706	4.3
	Pure S. fruticosa	-	-	213	213	11.9
	Millet/S. fruticosa single rows	805	2405	14	2419	5.6
	Pure S. hamata	-	-	552	552	12.4
	Millet/S. hamata single rows	1020	2428	22	2450	5.5
	Millet/S. fruticosa triple rows	613	2400	327	2727	6.2
	Millet/S. hamata triple rows	246	1276	353	1629	5.9
	LSD (5%)	ns	1181	123	948	1.7
1990
	Pure millet	591	1960	-	1960	4.8
	Pure S. fruticosa	-	-	783	783	12.2
	Millet/S. fruticosa single rows	407	1484	202	1686	5.0
	Pure S. hamata	-	-	126	126	12.2
	Millet/S. hamata single rows	414	1453	51	1504	5.7
	Millet/S. fruticosa triple rows	539	1349	395	1744	7.5
	Millet/S. hamata triple rows	341	939	61	1000	4.1
	LSD (5%)	ns	433	324	476	1.2

^a Residue = millet stover + stylo herbage

When millet was sown into pre-established stylo, grain yield was reduced by 26-83% relative to monocrop millet (Table 2). Total biomass, crude-protein and phosphorus contents of intercrops were, however, higher in both years than those obtained in sole millet crop. Alternate triple row planting of millet with either stylo resulted in lower millet grain yield in 1989 than alternate single row planting. Cropping patterns involving S. hamata reduced millet yields more than those involving S. fruticosa.

Table 2. Yields millet and Stylosanthes (millet sown into pre-established stylo) as affected by cropping system

Cropping system		Millet		Stylo herbage (kg/ha)	Residue a
		Grain (kg/ha)	Stover (kg/ha)		Biomass (kg/ha)	Phosphorus (%)	Crude protein (%)
1989
	Pure millet	642	2662	-	2662	3.4	5.7
	Pure S. fruticosa	-	-	3075	3075	5.1	13.0
	Millet/S. fruticosa single rows	313	1214	2032	3246	4.6	9.3
	Pure S. hamata	-	-	1265	1265	1.7	12.2
	Millet/S. hamata single rows	276	1006	4698	5704	8.0	10.9
	Millet/S. fruticosa triple rows	106	618	2508	3126	5.4	11.1
	Millet/S. hamata triple rows	139	814	4122	4936	6.8	11.2
	LSD (5%)	183	798	889	931	1.1	1.7
1990
	Pure millet	832	3204	-	3204	1.5	2.9
	Pore S. fruticosa	-	-	2889	2889	4.6	12.2
	Millet/S. fruticosa single rows	429	2234	1518	3752	3.0	7.3
	Pure S. hamata	-	-	3936	3936	5.3	12.8
	Millet/S. hamata	217	1286	3064	4350	4.4	9.5
	Millet/S. fruticosa triple rows	610	1845	1154	2999	2.8	6.2
	Millet/S. hamata triple rows	439	1308	2271	3579	3.9	8.5
	LSD (5%)	112	787	429	652	1.0	1.0

^a Residue = millet stover + stylo herbage

Dry-matter yields of both stylos were greatest the second year after establishment. Stylosanthes hamata produced significantly more herbage (P<0.01) and assimilated more nutrients (P<0.01) than S. fruticosa. An exception to this was lower pure S. hamata yields in 1989 due to plants being cut too short the previous year, resulting in poor regeneration. Higher legume yields in the second year versus the first year of establishment contributed to the increase of intercrop dry-matter and crude-protein content but caused a decline of millet yields.

Sheep feeding trials

For sheep grazing millet stover, average daily liveweight gains (ADO) for the control, stylo and cowpea supplemented sheep were 78, 98 and 115 g, respectively, during the first 22 days of the trial (Figure 1a). Unsupplemented animals maintained their weight up to the 42nd day, after which they consistently lost weight. Supplemented sheep continued to gain weight up to the end of the trial. Average daily gains at the end of the trial were 12, 48 and 53 g for the unsupplemented, stylo and cowpea supplemented sheep, respectively. Weight gains of unsupplemented sheep during the first three weeks were probably due to animals grazing selectively the more nutritious portion of the biomass such as leaves, tillers and weeds. A parallel trial involving sheep grazing millet stover and browse wind breaks (ILCA, unpublished) found that crude-protein (CP) content of millet residue dropped from 50 g CP/kg DM at the start of grazing to 39 g CP/kg DM after four weeks. The reason why unsupplemented sheep began to lose weight, and weight gains of supplemented animals decreased after the first three weeks of grazing, was probably a decrease in the nutritive value of millet residues.

Pen-fed sheep consumed mostly millet leaves (blades and sheaths) and refused most stems. Supplemented animals had steady growth rate throughout the feeding period (Figure 1b). Unsupplemented sheep gained weight for a week and then maintained it up to 49 days, after which they lost weight up to the end of the 70-day trial. Resulting ADG for the entire trial were -1.8, 78 and 51 g for the control, stylo and cowpea hay supplemented sheep, respectively. That supplemented animals consistently sustained growth rate throughout the trial period was primarily due to constant nutritive value (53.1 g CP/kg DM) of feed supplied in the basal diet.

CONCLUSIONS

There appear to be distinct tradeoffs in grain and feed output when millet and stylo are intercropped. When millet was sown into year-old established stylo, grain yield decreased but total feed biomass and protein yield increased as compared with sole millet crop. The adoption of such intercropping systems will depend on the relative emphasis placed by small-scale farmers on grain versus feed production. The cut-and-carry system seemed to have an advantage over grazing but this benefit would need to be weighed against other factors such as labour input, storage, nutrient cycling, etc.

Figure 1. Effect of supplementing cowpea or stylo hay on liveweight of sheep grazing millet stover (a) or fed millet stover in pens (b)

In terms of improved feeding systems, the study showed that as little as 6 g N in the form of stylo supplemented daily to sheep fed millet residue is enough to obtain steady animal growth. The average additional 35 kg N/ha produced by intercropping S. fruticosa with millet could, therefore, in theory provide sufficient N supplement to 30 animals for 194 feeding days. Corresponding millet stover yields could, however, be reduced. Appropriate management strategies are therefore required to maximise the benefits of forage legume introduction into mixed crop-livestock farming systems of the Sahel.

REFERENCES

Gallaher R N. Weldon C O and Futral J G. 1975. An aluminum block digester for plant and soil analysis. Soil Science Society of America Proceedings 39:803-806.

Garba M and Renard C. 1991. Biomass production, yields and water use efficiency in same pearl millet/legume cropping systems at Sadore, Niger. In: Sivakumar M V K Wallace l S. Renard C and Giroux C (eds), Soil water balance in the Sudano-Sahelian zone: Proceedings of International Workshop, Niamey, February 1991. IAHS (International Association of Hydrological Science) Publication 199. IAHS Press, Institute of Hydrology, Wallingford, UK pp. 431-439.

Gryseels G and Anderson F M. 1983. Research on farm and livestock productivity in the central Ethiopian highlands: Initial results 1977-1980. ILCA Research Report 4. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. 30 pp.

Hambleton L G. 1977. Semiautomated method for simultaneous determination of phosphorus, calcium and crude protein in animal feeds. Journal of the Association of Official Analytical Chemists 60:845-852.

Humphreys L R. 1978. Tropical pastures and fodder crops. Longman, London, UK 135 pp.

Izaurralde R C, Juma N G and McGill W B. 1990. Plant and nitrogen yield of barley-field pea intercrop in cryoboreal-subhumid central Alberta. Agronomy Journal 82:291-301.

le Houérou N H. 1989. The grazing land ecosystem of African Sahel. Ecological Studies Vol 75. Springer Verlag, Berlin, Heidelberg, New York. 282 pp.

Montgomery D C. 1984. Design and analysis of experiments. 2nd edition. John Wiley and Sons, New York, USA. 418 pp.

Mohamed-Saleem M A. 1985. Effect of sowing time on grain and fodder potential of sorghum undersown with stylo in the subhumid zone of Nigeria. Tropical Agriculture (Trinidad) 62:151- 153.

Nnadi L A and Haque J. 1988. Forage legumes in African crop-livestock production systems. ILCA Bulletin 30: 10-19. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia.

SAS (Statistical Analysis Systems Institute Inc). 1982. SAS user's guide. SAS, Cary, North Carolina, USA. 1028 pp.

Tothill J C. 1986. The role of forage legumes in farming systems of sub-Saharan Africa. In: Haque 1, Jutzi S and Neate P J H (eds), Potentials of forage legumes in farming systems of sub-Saharan Africa. Proceedings of a workshop held at ILCA, Addis Ababa, Ethiopia, 16-19 September 1985. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. pp. 162-185.

Varvel G E and Peterson T A. 1990. Nitrogen fertilizer recovery in monoculture and rotation systems. Agronomy Journal 82:935-938.

Waghmare A B and Singh S P. 1984. Sorghum-legume intercropping and the effects of nitrogen fertilization. 1. Yield and nitrogen uptake by crops. Experimental-Agriculture 20:251-259.

West L T. Wilding L P. Landeck J K and Calhoun F G. 1984. Soil survey of the ICRISAT Sahelian Center, Niger, West Africa. Texas A & M University, College Station, Texas, USA. 66 pp.