The effect of supplements of oilseed by-products on the utilisation of low-nitrogen fibrous diets by sheep

K. Yilala
Co-ordination Office, OAU/STRC/SAFGRAD,
B.P. 1783, Ouagadougou, Burkina Faso

Abstract
Introduction
Materials and methods
Results
Discussion
Conclusion
Acknowledgements
References

Abstract

Four experiments in which native pasture hay, and sorghum stover supplemented with varying levels of cotton seed and cottonseed cake conducted with sheep are described.

Experiment 1 investigated the effects of supplementing native pasture hay (predominantly Pennisetum pedicellatum with two levels of cottonseed cake (0 and 60 g/kg diet DM) and three levels of cotton seed (0,60 and 120 9/diet DM) arranged in 2 × 3 factorial using 6 Bali Bali sheep in a 6 × 6 Latin square design. There were increases in the apparent digestibilities of DM (P<0.001), N (P<0.001) and OM (P<0.001) due to the inclusion of cottonseed cake. Cotton seed and cottonseed cake interacted in their effects on the digestibilities of OM (P<0.05) and N intakes (P<0.01). Cottonseed cake increased both total and grass hay DM intakes (P<0.01 and p<0.05) respectively. Cottonseed resulted in substitution effects as its level was increased (P<0.05).

Experiment 2 investigated the effects of supplements of cottonseed cake without or with (60 and 120 g/g diet DM) cotton seed on the liveweight change of Djallonke sheep fed native pasture hay. Twenty-one rams were grouped into three treatments of 7 animals each in a randomized block design. The animals gained 54, 76 and 74 g/day when 0.60 and 120 g/kg diet DM of cotton seed respectively were combined with the cottonseed cake supplement.

Experiment 3 investigated the effects of supplementing sorghum stover with cottonseed cake at three levels (0,60 and 120 g/kg diet DM) in a double 3 × 3 Latin square design using Bali Bali sheep. The diets containing cottonseed cake increased the apparent digestibilities of DM (P<0.05) and N (P<0.001), and total DM intake (P<0.05). The inclusion of 60 g/kg diet DM resulted in increases in sorghum stover intake (P<0.05).

Experiment 4 investigated the response of Bali Bali sheep in liveweight change to the diets in Exp. 3. Three animals were randomly assigned to each treatment and fed individually. There were progressive increases in liveweight gain (P<0.05) but with diminishing differences as the level of cottonseed cake was increased. Cottonseed cake at 60 g/kg diet DM increased daily intakes of total and stover DM. At 120 g/kg diet DM total intake was increased but stover intake reduced.

It is concluded that cottonseed cake supplement at 60 g/kg diet DM substantially improved the intake of both fibrous diets, and was as effective as the high level (120 g/kg diet DM) in liveweight gain. This level of inclusion may also be within the economic reach of the farmers.

Introduction

There is a heavy concentration of ruminant livestock in the heavily cropped (cereals accounting for 80%, Herman 1983) Soudaninan zone of Burkina Faso. This shift in distribution, particularly during the dry season is mainly due to the relatively abundant cereal crop residues. Although on the decline, there still exists substantial amount of native pasture on fallow land left unutilised by the animals during the long dry season because of its low nutritive components.

Despite the severity of feed deficit in the dry season conservation of herbages is seldom a priority at the smallholder farm level. Conservation of the native pasture in late September, when the requirement of labour for crop production is at its lowest ebb, has enabled harvest of up to 4460 kg/DM/ha of fallow land. This was accompanied with increments in the contents of digestible DM, N and P (by 9.2, 74.5 and 66.7%, respectively), and reduction in NDF and ADF (by 22.3 and 19.5%, respectively) when compared to the standing, dry and mature pasture herbage (Yilala, 1986). However, the N content of the conserved herbage (9.0 g.kg DM) remained below the critical level of 11.2 g/kg DM, below which voluntary intake of DM could be depressed (Whiteman, 1980), to be followed by loss of liveweight.

Several studies have shown increased intake (Egan, 1965; Kempton and Leng, 1979; Yilala, 1987) and positive responses in liveweight gain (Kempton and Leng, 1979; Olayiwole and Olorunju, 1987) and retention of N (Egan, 1965; Nuwanyakpa and Butterworth, 1987) when such low-N fibrous diets were supplemented with protein.

By replacing part of fallow pasture with forage or dual purpose legumes and conservation as hay it was apparent that the smallholder farmer could at least be partially self-reliant in the source of nitrogen supplement (Yilala, 1986). Cotton being the principal export crop representing 4% of agricultural land use in Burkina Faso (Herman, 1983) its by products, cotton seed and cottonseed cake, appear to be the cheapest amongst the purchaseable sources of N and energy, and could be complementary to forage legumes (Yilala, 1988a).

This paper describes the results of a series of experiments, in which native pasture hay and sorghum stover were supplemented with N and energy source. The experiments were carried out to study the effects upon sheep of varying the levels of cottonseed cake with or without cottons seed on the apparent digestibilities of nutrients, voluntary intake and liveweight change. The experiments were carried out on 2 ha of farmers' fallow land on which forage was produced and conserved, and the animal shed constructed. Resources that could readily available and are within the economic reach of the farmers were given due consideration in the overall approach. All the animals for the feeding trials were contributed by the farmers and returned at the end of the experiments.

Materials and methods

Experimental Designs
Feeding and growth period
Digestibility
Statistical analysis

Experimental Designs

Experiment 1

This experiment was conducted to assess the effect of supplementing native pasture hay (predominantly Pennisetum pedicellatum) with two levels of cottonseed cake (0 and 60 g/kg diet DM) and three levels of cottons seed (0, 60 and 120 g/kg diet DM) on the apparent digestibilities of nutrients and voluntary intake. A 6 × 6 Latin square design, using six Bali Bali sheep, with the treatments arranged in 2 × 3 factorial were used. Due to shortage of grass hay only five periods of 19 days each were conducted, and this reduced the number of replications from six to five.

Experiment 2

The effect of supplements of cottonseed cake without or with (60 and 120 g/kg diet DM) cottons seed on liveweight change of Djallonke rams fed native pasture hay was investigated concurrently with Exp. 1. The twenty-one rams contributed by farmers, were grouped into three treatments of 7 animals each in a randomized block design. The trial was conducted for 42 days.

Experiment 3

The effect of supplementing sorghum stover with cottonseed cake at three levels (0,60 and 120 g/kg diet DM) were investigated to assess the apparent digestibilities of nutrients and voluntary intake. A double 3 × 3 Latin square design using six Bali Bali sheep was used.

Experiment 4

A feeding trial was conducted to investigate the response of Bali Bali sheep in liveweight change to the diets in Exp. 3. Three animals were randomly assigned to each treatment. The experiment had two periods: feeding sorghum stover without (21 days), and with cottonseed cake (63 days including the 7 days for adaptation to the supplement).

Feeding and growth period

Djallonke rams of age 10 to 15 months and weighing approximately 12.4 kg (Expt. 2), and Bali Bali rams aged 18 to 21 months and approximately 45.5 kg (Expt. 4) were used for growth trials. All animals were treated against internal and external parasites.

They were fed native pasture hay (Expt. 2) and sorghum stover chopped to 6 cm in length (Expt. 4) ad libitum in groups and individually, respectively. The total amount of diet offered (i.e. in two feeds given at 10.00 and 16.00 h) was 20% above the daily consumption. The animals were given seven days to adapt to the supplements. They had free access to mineral block licks and drinking water.

Sheep were weighed weekly at about the same time on the same day.

Digestibility

Bali Bali sheep were confined in individual pens partitioned with locally made mud bricks. During each experimental period of 19 days the animals were adapted to the diets from day 1 to day 12. The remaining 7 days were used for voluntary intake measurements and collection of faeces. The basal diets, native pasture hay (Expt. 1) and chopped (6 cm) sorghum stover (Expt. 3), were fed ad libitum (20% above the daily consumption) in two feeds given at 10.00 and 16.00 h. Drinking water and mineral blocks were available at all times.

Faeces were collected into canvas bags harnessed to the sheep. The collection was done every 24 h with a one day lag from intake measurements. Faecal samples (10% of daily output) were air and oven-dried for DM determinations and subsequent chemical analysis.

Statistical analysis

The date were subjected to analysis of variance with the variance partitioned into main effects and interactions in Expt. 4 (Cochran and Cox, 1957).

Results

Apparent digestibility
Voluntary intake
Liveweight change

The chemical composition of the ingredients in the diets is give in Table 1.

Table 1. Chemical composition of feed ingredients used in the trials.

Ingredients	Average		Composition of DM (%)
	N	P	NDF	ADF	Lignin
Basic
Native pasture hay	1.04	0.15	57.4	34.5	7.0
Sorghum stover	0.97	0.08	68.1	40.9	7.2
Supplements
Cotton seed	5.00	1.43	-	-	-
Cottonseed cake	7.68	2.85	-	-	-

Apparent digestibility

As shown in Table 2 (Expt. 1) inclusion of cottonseed cake in native pasture hay based diet significantly improved the apparent digestibilities of DM (P<0.001), N (P<0.001) and OM (P<0.001) and DOMD value (P<0.001). Cottonseed also improved the interaction with cottonseed cake (Table 3) and significantly influenced the digestibilities of OM and N. and DOMD value (P<0.05, P<0.01 and P<0.01, respectively). That is, the difference between the observed OM digestibility and DOMD value were largest for the diet containing 60 g/kg diet DM of cottons seed and for N digestibility the diet with 120 g/kg diet DM of cottons seed.

Table 2. Effects of supplements of cottonseed cake and cotton seed on the apparent digestibility of nutrients and voluntary intake of native pasture hay by Bali Bali sheep.

	Main effects
	Cottonseed cake (g/kg diet DM)			Cotton seed (g/kg diet DM)				Significance of effect2
	0	60	s.e.d1	0	60	s.e.d		Cake	Seed
Digestibility (%)
DM	63.7	68.9	1.10	61.9	68.4	68.5	1.30	***	***
OM	64.3	70.6	1.18	64.8	68.8	68.6	1.45	***	*
DOMD	56.3	61.6	1.00	56.2	60.1	60.5	1.20	***	**
N	58.2	74.2	1.15	55.6	66.6	76.6	1.41	***	***
Intake (g/day)
Total	1329	1551	52.28	1375	14.60	1485	60.03	**	ns
Nay	1147	1274	56.82	1309	1261	1130	69.59	ns	ns

¹ Standard error of difference
² ns = not significant, * = P<0.05, ** P<0.01 and *** = P<0.001

Table 3. Effects of interaction of cottonseed cake end cotton seed on the apparent digestibility of organic matter, nitrogen and DOMD (in %).

Cottonseed cake	Cottonseed (g/kg diet DM)
	0	60	120	s.e.d.
0 g/kg diet DM
OM	60.4	63.8	68.7	2.10
DOMD	52.3	55.8	60.7	1.67
N	43.8	56.9	74.2	2.00
60 g/kg diet DM
OM	69.3	73.8	68.5	2.10
DOMD	60.1	64.4	60.3	1.67
N	67.4	76.3	78.3	2.00

Significant effects were also observed in the apparent digestibility of DM (P<0.05) and N (P<0.001) and DOMD value (P<0.001) in animals receiving diets containing both levels of cottonseed cake in the sorghum stover based diets (Table 4, Expt. 3).

Table 4. Effects of supplement of cottonseed cake on apparent digestibilities of nutrients and voluntary intake of sorghum stover fed to Bali Bali sheep.

	Cottonseed cake (g/diet DM)
	0	60	120	s.e.d	Significance of effects
Digestibility (%)
DM	59.2	72.5	71.2	1.83	*
DOMD	50.6	61.4	59.1	0.57	***
N	39.6	71.6	75.7	2.44	***
Intake (g/day)
N	7.2	19.8	28.6	0.82	***
Total DM	964	1410	1265	40.9	*
Stover DM	964	1278	1001	40.9	*

Voluntary intake

Inclusion of cottons seed in Expt. 1 increased total and grass hay DM intakes (P<0.01 and P>0.05, respectively). Cottonseed showed a slight increase in total DM intake (P>0.05) and a progressively decreasing hay intake as its level was increased (P>0.05) as shown in Table 2.

Both levels of cottonseed cake in the sorghum stover based diets (Expts. 3 and 4) also led to significant increases in total DM intake (P<0.05, Table 5 and P<0.05, Table 6, respectively). However, animals receiving the diets containing 60 g/kg diet DM of cottonseed cake showed the greatest intakes in total DM (P<0.05 and P>0.05) and stover DM (P<0.05 and P>0.05) in Expts. 3 and 4 respectively.

Liveweight change

The results of liveweight gain (Tables 5 and 6) showed a high coefficient of variation. This might be associated with the problems of competition in group feeding whereas liveweight measurements were made individually (Expt. 2). The inadequate replications might be responsible for the large variation in Expt. 4.

In Expt. 2 the difference in liveweight gain failed to reach significance at P<0.05. The liveweight gain within treatments varied between 14 and 100, 55 and 95, and 36 and 110 g/day for treatments A, B and C respectively. The greatest liveweight gain was obtained from the diet containing 60 g/kg DM cottonseed cake + 60 g/kg DM cottons seed.

Table 5. Effects of supplementing native pasture hay with cottonseed cake and cottons seed on average daily intake and liveweight change of Djallonke rams.

		Cottonseed cake (g/kg diet DM)
		60				Statistical significance (P=0.05)
		0	60	120	s.e.d.
Intake1
	DM (g/kg/day)	497	483	477	-	-
	Liveweight change:
	Initial weight (kg)	12.4	11.7	11.7
	Final weight (kg)	14.6	14.9	14.8	0.77	ns
	Weight gain (g/day)2	54	76	74	14.5	ns
Supplement cost
	USD/hd/day	0.47	0.81	1.19
	USD/100 g LWG3	0.88	1.07	1.60

¹ Calculated from average group daily intake
² Coefficient of variation = 37.7%
³ Liveweight gain

Although not statistically significant (at P=0.05) there were progressive increases in liveweight gain, but with diminishing differences as the level of cottonseed cake was increased in Expt. 4 (Table 6). This rate of liveweight gains during the first 14 days of the supplement were greater (P>0.05) than in the subsequent period.

Table 6. Effects of supplement of cottonseed cake on voluntary intake and liveweight change of Bali Bali sheep fed on sorghum stover.

		Cottonseed cake (g/kg diet DM)				Significance of effects
		30	60	120	s.e.d
Intake
Total DM
	(g/day)	896	1047	1043	382.6	*
	(g/kg LW)	18.6	21.8	22.3	1.11	*
Stover DM (g/day)		806	915	773	38.3	ns
Nitrogen		13.7	21.8	28.8	0.28	***
Liveweight change
Pre-supplementation
Liveweight gain (g/day)		101	135	108	116	ns
Supplementation
	Initial liveweight (kg)	47.6	46.7	45.3
	Final liveweight (kg)	48.8	49.7	48.7
Liveweight gain(g/day)
	days 1 to 141	33.3	71.8	109.7	45.30	ns
	days 1 to 562	23.8	53.6	60.7	15.30	ns
	Food conversion ratio	37.6	19.4	17.2	6.17	*
Supplement cost
	USD/hd/day	0.61	1.23	1.65
	USD/100g LWG	2.7	2.3	3.0

¹ Coefficient of variation = 36.0%
² Coefficient of variation = 44.5%

Discussion

The low and progressive decline in the intake of sorghum stover prior to supplementation during the first 3 weeks, and the increase during supplementation (Expt. 4) confirms the importance of availability of a source of N for effective utilisation of the stover. This effect might be associated with increased rate of fermentation in the rumen (Mehrez and Orskov, 1978) or improved N status of the animal (Egan, 1965) or both.

The aim of supplementation was to create the conditions that will allow increased intake of the low - N fibrous diet and not substitution by the supplement. Based on the DM intake of the Control group (i.e. without supplements) substitution effects of the supplement were calculated according to the ARC (1980). With the inclusion of cottonseed cake at 60 g/kg diet DM (4.8 g N/kg DM) intake of native pasture hay and sorghum stover increased at the rate of 1.80 and 2.8 units, respectively (Appendix Table I). No substitution effect were observed due to cottons seed at 60 g/kg diet DM without or with cottonseed cake. In another study increased sorghum stover intake was achieved with 200 g cowpea hay/kg diet DM, i.e with 5.2 g N/kg DM (Yilala, 1988a). With the doubling of level of cottonseed cake (except in Expt. 3), cottons seed and cowpea hay substitution effects were observed (Appendix Table I).

These was a progressive fall in liveweight during the 3 weeks prior to supplementation (Expt. 4). However, during the first 14 days of cottonseed cake supplementation, animals receiving 60 and 120 g/kg diet DM gained 52 and 230% more, respectively, than those on 30 g/kg diet DM. This possibly indicates the occurrence of compensatory growth in response to the relatively low N status prior to supplementation.

Due to the high coefficient of variation the author is cautious to draw definitive conclusion from the results obtained. However, considering the evidence from other studies (Orskov et al, 1976; Hovell et al, 1983; Yilala and Bryant, 1985) in which changing animals from low to high protein diets resulted in compensatory growth, the finding in this study does not sound atypical.

There are two reasons for emphasizing this point: (1) Since there appears to be a decline in the rate of liveweight gain with time it may not be necessary to include the same amount of protein supplement in the diet during compensatory growth and subsequent periods. (2) Under the observed farm conditions the poor nutritional status of animals during the long dry season is apparent. Therefore, taking advantage of the relationship between the residual effects of previous low N status and possible greater retention of N during the first few weeks of supplementation is of paramount importance. This will have important implication in the efficiency of utilisation of the limited sources of N that are at the disposal of the smallholder farms.

The effects of cottonseed cake and cottons seed were additive (Expt. 2). The highest response in liveweight gain was obtained from the diet containing 60 g/kg diet DM cottons seed, but with no extra advantage when the level was increased to 120 g/kg diet DM. The diminishing increase in the overall pattern of liveweight gain when the level of cottonseed cake was increased (Expt. 4) might suggest that energy was limiting. The acetate type of fermentation that predominates in the stover based diet might have resulted in low production of the glucogenic propionic acid (Leng, 1982). This inadequacy of energy reducing the effectiveness of high level of N for increased retention may be a possible explanation (Yilala and Bryant, 1985).

Considering the cost of supplement/100 g liveweight gain and rate of increase in the intake of fibrous diets, the supplementary value of cottonseed cake at 60 g/kg DM is more preferred than the other levels. The economic benefits realized from the supplementation of crop residues are mostly assessed based on the balance between the biological outputs (in terms of liveweight gain) and the costs of purchased inputs and labour.

Appendix Table 1: Replacement rates of the fibrous diets by supplements of cotton seed, cottonseed cake and cowpea hay fed to Bali Bali sheep.

Source of data	Nitrogen supplement		Energy supplement		Hay/stover intake	Replacement rate
	Amount		Amount
	Source	(g/kg/diet DM)	Source	(g/kg diet DM)	(g/day)

Experimenta

1		(0		(0	1191	0
	Cottonseed	(0		(60	1248	+0.43
	cake	(0	Cottonseed	(120	1152	-0.15
		(60		(0	1428	+1.80
		(60		(60	1276	+0.32
Experimenta		(60		(120	1118	-0.18
2		(0			964	0
	Cottonseed	(60			1276	+2.78
	cake	(120			1001	+1.14
Yilala (1988)b		(0			1169	0
	Cowpea	(200			1190	+0.10
	hay	(400			1149	-0.25

^a Basal diets in Experiments 1 and 2 were native pasture hay and sorghum stover, respectively.
^b Basal diet was sorghum stover

Table 8. Estimated quantity of dry matter, organic matter, nitrogen, phosphorous ansd potash in feed refusals and faeces of sheep¹ fed sorghum stover supplemented with cottonseed cake and cowpea hay.

Source of data	Supplement	Feed refusals (kg/hd/yr)			Faecal output (kg/hd/yr)
		DM	OM	N	DM	OM	N	P	K2O
Expt. 3	Cottonseed cake (g/kg diet DM)
	0	62	54	0.5	100	111	1.9	0.19	0.59
	60	91	78	1.9	127	109	2.1	0.90	2.26
	120	82	62	2.4	160	137	2.4	1.49	3.00
Yilala (1986)	Cowpea hay (g/kg diet DM)
	0	69	62	0.5	153	130	2.1	0.23	1.2
	200	93	74	0.7	152	112	3.0	0.27	1.3
	400	94	75	0.7	140	119	3.0	0.28	1.5

¹ Bali Bali sheep with average liveweight of 45.2 kg during the experimental periods

This may not be adequate for an integrated crop-animal system where inputs are generated within the system. This holds true in most smallholder farm conditions.

Assessement of economic benefits under such conditions may have to consider aspects of recycling of nutrients in which nutrient outputs of the crop component (through crop residues) serve as nutrient inputs to the animal, and vice versa (through manure). To serve as an example, the quantity of DM and nutrients that could be available for composting (for soil fertilisation) were estimated from the daily feed refusals and daily faecal output of sheep fed sorghum stover with supplements of cottonseed cake and cowpea hay (Appendix Table II). The increase in the contents, for example, of phosphorus (P) in the faeces of sheep fed diets containing cottonseed cake is of paramount importance to conditions of poor soil P status in the Soudanian zone of Burkina Faso. It is, of course, assumed that the cottonseed cake might have adequately supplied P and satisfied the needs for the synthesis of microbial nucleic acids in the rumen (Harrison and McAllan, 1980).

For a technology to be accepted it must be able to fit into the objective conditions of the target farms. Farmers do recognise that creating favourable nutritional conditions will improve the performance of animals. However, the conditions that they could create are limited by their economic stand. The purchase of inputs such as protein or energy supplements in most cases is beyond their economic reach, particularly when one considers the need to supplement during the long dry season as in the case of Burkina Faso. Any research work targeted to smallholder farmers should be deliberately biased towards the utilisation of on-farm or locally available resources that could be readily available to and are within the economic reach of the farmers in the overall approach of this and other studies.

The supplements of cottons seed and cottonseed cake in this study are considered as complementary supplements to forage legumes particularly for the latter part of the dry season feeding. Encouraging results were obtained when Djallonke ewe lambs were fed native pasture hay supplemented with on-site produced cowpea hay for five months followed by cottonseed cake for two months during the dry season. The animals gained up to 19 and 40 g/day liveweight on supplements of cowpea hay and cottonseed cake, respectively (Yilala, unpublished data), during this period when loss of liveweight is a common phenomenon.

The feeding in pens for such long period, a break away from the traditional method, was accompanied with increased collection and improved management of manure and feed refusals to make compost. Organic matter is an indispensable resource that is highly valued by the farmers to increase and maintain soil fertility. With the application of compost (made from sheep faeces, feed refusals and some Burkina rock phosphate) at 10 tons/ha it was possible to increase the yield of cowpea forage DM from 2858 to 7751 kg/ha (Yilala, 1989b). Such benefits cannot, Appendix Table II. Estimated quantity of dry matter, organic matter, nitrogen, phosphorus and potash in feed refusals and faeces of sheep¹ fed sorghum stover supplemented with cottonseed cake and cowpea hay therefore, be ignored when input-output relationships are examined to assess the economic benefits realized from the supplementation of crop residues. The method of analysis might become more complex when the source of supplement is a forage or dual purpose legume grown on-farm.

Conclusion

The following conclusions may be drawn from the study:

1. The low N - fibrous diets, native pasture hay and cereal crop residues, will remain to be the major sources of metabolizable energy (ME) for ruminants in Burkina Faso. The study has undoubtedly confirmed that cottonseed cake, the cheapest among the purchasable sources of N in the country, has improved the utilisation of the above sources of energy through increased digestibility and voluntary intake.

2. The findings suggest that sheep managed in the conventional way prior to supplementation respond positively to protein supplements in liveweight gain consistently. The rapid rate of growth during the first 14 days of supplementation might be as a result of the residual effects of the previous low N status of the animals.

3. The further improvement of liveweight gain when cottons seed was added to the cottonseed cake might indicate the importance of increasing the ME intakes of the animals too, although no extra advantage was realized in liveweight gain by doubling the level of cottons seed.

4. For every unit of inclusion of cottonseed cake at 60 g/kg diet DM (4.6 g N.kg DM) the intakes of native pasture hay and sorghum stover were increased by 1.80 and 2.78 units, respectively. The same level of cottonseed cake in the diet was as effective as the high level (120 g/kg diet DM) in terms of liveweight gain. It is possible that the 60 g/kg diet DM of cottonseed cake may also be within the economic reach of the farmers.

5. The high content of P in cottonseed cake might have possibly helped correct the deficiency of P in both native pasture hay and sorghum stover for rumen microbial protein synthesis. It appears that it can also increase the soil P status. Such indirect contributions of protein supplements may need to be considered in the analysis of economic benefits.

Acknowledgements

The author gratefully acknowledges the Farming Systems Research Programme of OAU/STRC/SAFGRAD-INERA (Institute National D'Etudes et de Recherches Agricoles), Burkina Faso, for making the study possible. Deepest appreciation are extended to Messrs Zoundi Sibiri and Ouedraogo Saidou for their technical assistance. Special thanks are due to PANESA/ARNAB for sponsoring my attendance at the workshop.

References

ARC (Agricultural Research Council) 1980. The Nutrient Requirements of Ruminant Livestock. Commonwealth Agricultural Bureaux, Slough, U.K.

Cochran, W.G. and Cox, G.M. 1987. Experimental Designs. John Wiley & Sons, Inc. London.

Egan, A.R. 1965. Nutritional status and intake regulation in sheep. III. The relationship between improvement in nitrogen status and increase in voluntary intake of low protein roughages by sheep. Aust. J. Agric. Res. 16:163-172.

Harrison, D.G. and McAllan, A.B. 1980. Factors affecting microbial growth yields in the reticulo-rumen. In: Digestive Physiology and Metabolism in Ruminants. eds. Y. Rucker busch and P. Thivend. pp. 205-226. MTP Press Ltd., Lancaster.

Herman, L. 1983. The Livestock and Meat Marketing System in Upper Volta: An evaluation of economic efficiency. Centre for Research on Economic Development. The University of Michigan.

Hovell, F.D.,Oskov, E.R., Macleod, N.A. and McDonald, I. 1983. The effect of changes in the amount of energy infused as volatile fatty acids on the nitrogen retention and creatinine excretion of lambs wholly nourished by intragastric infusion. Br. J. Nutr. 50:331-343.

Kempton, T.J. and Leng. R.A. 1979. Protein nutrition of growing lambs. 2. Effect on nitrogen digestion of supplementing a low-protein-cellulosic diet with either urea, casein or formaldehyde-treated casein. Br. J. Nutr. 42:303-315.

Leng, R.A. 1982. A theoretical discussion on the factors limiting production in cattle fed basal diets of straw. In: Maximum Livestock Production form Minimum Land. eds. T.R. Preston, C.H. Davis, F. Dolberg, M. Hague and M. Saadullah. pp. 79-104. Bangladesh.

Mehrez, A.Z. and Orskov, E.R. 1978. Protein degradation and optimum urea concentration in cereal based diets for sheep. Br. J. Nutr. 40:337-345.

Nuwanyakpa, M. and Butterworth, M. 1987. Effects of urea, molasses, molasses urea, noug cake and legume hay on the intake and digestibility of teff straw by highland sheep. In: Utilisation of Agricultural By-Products as Livestock Feeds in Africa. pp. 87-98. ARNAB/ILCA.

Olayiwole, M.B. and Olorunju, S.A. 1987. Feedlot performance of yearling steers previously maintained on different crop residue/supplementation regimes. In: Utilisation of Agricultural By-Products as Livestock Feeds in Africa. pp. 72-82. ARNAB/ILCA.

Oskov, E.R., McDonald, I., Grubb, D.A. and Pennie, K. 1976. The nutrition of the early weaned lamb. IV. The effect on growth rate, food utilisation and body composition of changing from a low to a high protein diet. J. Agric. Sci. 86:411-423.

Whiteman, P.C. 1980. Tropical Pasture Science. Oxford University Press. Oxford.

Yilala, K. 1986. Observations on the conservation of forage legumes and natural pasture as hay. In: Farming Systems Research Report. pp. 78-82. OAU/STRC/SAFGRAD, Ouagadougou.

Yilala, K. 1988a. Influence of inclusion of forage legume hay in basal diet of sorghum stover upon intake and digestibility by Bali Bali sheep. In: Farming Systems Research Report. pp. 118-122. OAU/STRC/SAFGRAD-INERA.

Yilala, K. 1988b. Effects of phosphates and method of land preparation on cowpea forage yield and nodulation. In: Farming Systems Research Report. pp. 68-76. OAU/STRC/SAFGRAD-INERA. Ouagadougou.

Yilala, K. and Bryant, M.J. 1985. The effects upon the intake and performance of store lambs of supplementing grass silage with barley, fishmeal and rapeseed meal. Anim. Prod. 40:111-121.