Practical problems of feeding crop residues

E. Owen and A.A.O. Aboud

Department of Agriculture, University of Reading, Early Gate, P.O. Box 236, Reading RG6 2AT, UK

Introduction
Post-harvest and pre-feeding constraints
Background to the reading experiments-the grazing approach
Materials and methods
Results
Discussion
Acknowledgements
References
Discussion

Introduction

The vast amount of crop residues byproducts available as potential ruminant feed-some 2.0 t DM per 500 kg live stock unit annually in developing countries (Kossila, 1984)-is now generally acknowledged. With the world population predicted to double by 2025 (even treble in the developing tropics), cereal production, and hence straw production, will have to increase. With the increased pressure on land for food production, less land will be available to produce animal feed, either from pasture or fodder crops, and crop residues will assume eve n greater importance as animal feed. This will lead to greater integration of crop and animal (mainly ruminant) production (Gartner, 1984).

The importance of small ruminants (especially goats) in developing-country agriculture is now widely recognised (Devendra and Burns, 1983; World Bank and Winrock International, 1983; Timon and Hanrahan, 1986). It is less well recognised, however, that small ruminants are mainly associated with small-scale farmers and that small-scale farmers predominate in developing-country agriculture. It will be these farmers who will need to practice crop-animal integration. A major constraint to crop-livestock integration is the potential damage to food crops from indiscriminate grazing, especially by goats. Owen et al (in press) stressed the need to research and develop stall-feeding systems for small ruminants based on crop byproducts.

Much has been written in recent years about the potential of crop residues as ruminant feed and about ways of overcoming their low nutritive value by upgrading and supplementing (e.g. Sundstøl and Owen, 1984; Doyle et al, 1986a). Much less effort has been put into identifying the factors limiting greater usage of crop residues and new technology, particularly by small-scale farmers in developing countries (Owen, 1985).

With the latter target-group in mind this paper will therefore briefly identify constraints which occur in the post-harvest period, as well as practical problems of feeding crop residues. The results of experiments recently undertaken with goats and sheep at Reading will be used to demonstrate how the amount of barley straw offered can affect the quantity and quality of straw consumed. The implications of this in regard to supplementation, plant breeding and developing strategies for feeding crop residues will be discussed, and areas needing further research will be identified.

Post-harvest and pre-feeding constraints

Decisions on whether or not to conserve crop residues for feed have to be taken soon after harvesting and often long before feeding them. Lack of convincing economic evidence in favour of their greater use as feed is undoubtedly a restraining factor (Edelsten and Lijongwa, 1981; Giaever, 1984; Tambi, in press). Animal scientists are partly to blame in that they have generally emphasized biological rather than economic responses to upgrading and supplementing crop residues. A problem which has a bearing on this is the difficulty of accurately predicting the nutritive (and therefore e economic) value of crop residues from simple laboratory techniques, as evidenced by the recent EEC workshop (Chenost, in press). The problem is likely to be even greater for tropical crop residues containing anti-nutritional phenolic compounds (Mueller-Harvey et al, 1987), especially if feeding systems allow the animals to exercise selective feeding.

Cereal straws/stovers generally either are left in the field or accumulate where the crop is threshed. This is often far from where animals are kept and either the animals must be brought to the field for stubble grazing, or crop residues have to be transported to the animals. The bulk of straws and stovers and lack of transport discourage greater use of straws and stovers as feed. Transporting crop rest even over short distances, may be uneconomic for small farmers (Mlay, 1987).

The handling and storing of crop residues have been discussed by Hilmersen et al (1984). More research and development is required to alleviate problems associated with storage of crop residues. These include risk of loss due to fire, and reduction in nutritive value due to moulding (especially in humid conditions) and damage by vermin and insects. Straws and stovers comprise stem and leaf plus leaf sheath (approximately 1:1 in barley straw), and harvesting, handling and storing systems should minim) se the loss of the more nutritious leaf and leaf sheath. In this regard delayed harvesting, or relay harvesting in an intercropped field, would be expected to cause greater loss of leaf and leaf sheath, with a consequent reduction in nutritive value.

Background to the reading experiments-the grazing approach

Crop residues are characteristically low in metabolisable energy and nitrogen content and thus intake is low. Methods of upgrading straws as feed are well documented (Sundstøl and Owen, 1984) and guidelines on researching the subject are available (Preston et al, 1985). There has been more emphasis on upgrading straws for cattle than for sheep (Greenhalgh, 1984) but goats have received little attention (Owen, 1981; Owen and Kategile, 1984). Upgrading straws for feed is regarded as inappropriate for developing countries, especially for small-scale farmers, because it is expensive and needs technical expertise. Greenhalgh (1984) concluded that in many situations chemical upgrading will be superceded by breeding more nutritious straw, improved harvesting methods and judicious supplementation. The Reading experiments reported here suggest another approach, namely increasing intake of digestible nutrients and therefore animal productivity from crop residues by allowing selective feeding by goats and sheep.

The literature on feeding straws to sheep and goats involves experiments where intake and digestibility have been measured under ad libitum feeding. 'Ad libitum' is defined as offering sufficient feed (usually in chopped form) to ensure that 15 to 20% is left (refused) by the end of the feeding period (Blaxter et al, 1961). This approach is standard and has the advantage (for the experimenter, but not the animal) of minimising selective feeding. We would argue that the latter is a disadvantage. The selective grazing and browsing behaviour of sheep (Gibb and Treacher, 1976) and goats (McCammon-Feldman et al, 1981) is recognised. Indeed, experiments (e.g. Gibb and Treacher, 1976) indicate that maximum intake by grazing sheep is achieved only if the herbage allowed exceeds intake by 400%. We therefore hypothesised that conventional ad libitum straw feeding would restrict - intake by reducing the opportunity for animals to select better quality material.

This hypothesis was tested in the experiments reported. The experiments are also aimed at helping us develop strategies for stall-feeding straw to goats and sheep.

Materials and methods

Seven experiments conducted at Reading during the past 5 years are presented. All measured straw intake and assessed the degree of selective feeding by careful sampling and analysis of feed offered and refused. Except in Experiment 1 (no concentrate fed), the animals were fed a concentrate supplement (sugar-bleat pulp, 600 g kg^-1 ; soya bean meal, 180 g kg^-1; fish meal, 180 g kg^-1; minerals and vitamins, 40g kg^-1) at 15 g DM kg^-1 W^0.75 d^-1 to satisfy nitrogen, mineral and vitamin requirements (ARC, 1980) for maintenance and modest growth in sheep. Numbers of animals used, type and mean weight are shown in Tables 1 to 7. All experiments used housed (16 hours light, 8 hours dark), individually penned castrated animals bedded on sawdust and provided with water. In Experiment 6 goats were in metabolism cages and faeces were collected over 9 days following a preliminary period of 14 days. In all other experiments preliminary periods were of 14 to 21 days and experimental periods lasted 21 days, except for Experiment 4 (42 days). Feeds were offered (in large feed boxes) twice daily and straw refusals carefully collected daily. Representative samples (based on aliquots) of straw offered and refused were taken daily and pooled samples were analyzed (Wahed and Owen, 1986a) for dry matter, ash and nitrogen (AOAC, 1975), acid detergent fibre (Goering and Van Soest, 1970) and in vitro digestibility (Tilley and Terry, 1963).

Results

Experiment 1

Experiment 1 (Table 1) examined whether any of the claimed superiority of goats over sheep, in regard to roughage intake and digestion, could be attributed to differences in feed selected under stall-feeding. The straw fed was treated with aqueous ammonia using a stack method (Sundstøl and Coxworth, 1984). The experiment showed goats to eat more than sheep, but there was no large difference between the quality of straw refused by the two species. It was clear, however, that both species were feeding selectively. Refused straw was of lower nutritive value than that offered.

Experiment 2

Experiment 2 (Table 2) was the first trial to test the hypothesis outlined earlier. Allowing goats to refuse 50% of the straw offered increased DM intake by 31% compared with the more conventional 20% refusal rate. The quality of feed refused indicated that goats allowed the higher refusal rate selected more nutritious straw. Thus the estimated intake of straw digestible OM (based on in vitro digestibility) was 40% higher. The 18 goats used per treatment represented a wide range of liveweight (15 to 65 kg), and small goats tended to be more selective than larger ones.

Table 1. Intake and selection of NH3-treated barley straw by sheep and goats (Experiment L).

	Suffolk cross mule wethers	Saanen castrate goats	SED
Number of animals	8	8
Liveweight (W) (kg)	57.9	50.7	9.0
Straw intake
Offered1 (g DM d -1)	1299	1477
Intake (g DM d -1)	956	1117	152.4
(g DM kg-1 W d-1)	16.4	21.6	1.5

Chemical composition	Straw offered	SE	Straw refused
Nitrogen (g kg-1 DM)	17	0.5	11.6	12.2	0.6
Acid-detergent fibre(ADF) (g kg-1 DM)	567	5.7	612	600	6.4
In. vitro digestibility2 (DOMD) (g OM kg-1 DM)	607	6.0	544	566

1. To allow a refusal rate of 20 to 25% of amount offered.
2. Tilley and Terry (1963). Source: Wahed and Owen (1986a).

Experiment 3

In Experiment 3 (Table 3) increasing the refusal rate allowance increased intake of both long and chopped straw. The trend (no: e-significant) was for greater intake of long straw. Straw-length interacted significantly with refusal rate for refusal digestibility, indicating easier selective feeding with long than with chopped straw. All subsequent experiments were therefore carried out with long straw.

Table 2. Effect of allowing two rates of refusal on intake and selection of barley straw by goats (Experiment 2).

	Straw refusal allowance
	20%	50%	SED
Number of goats1	18	18
Straw intake2 (g DM kg-1 W d-1)	14.4	18.9	0.70
Straw intake (g DM kg-1 W0.75 d-1)	33.1	43.7	1.60
Straw refused (% of offered)	20.5	48.3
Estimated intake of straw digestible OM3 (g kg-1 W d-1)	5.9	8.3

Chemical composition	Straw offered	SE	Straw refused
Nitrogen (g kg-1 DM)	5.1	0.02	4.5	4.6	0.13
ADF (g kg-1 DM)	552	7.0	612	596	4.8
In vitro DOMD4 (g OM kg DM)	412	4.8	320	347	7.7

1. Mean liveweight 32.6 kg.
2. Concentrate supplement also fed at 15 g DM kg-1 W0 75 d-1.
3 Calculated from in vitro digestibility of straw offered and refused.
4. Tilley and Terry (1963).
Source: Wahed and Owen (1986b).

Experiments 4 and 5

Experiment 4 (Table 4) simulated the 'grazing approach' (Gibb and Treacher, 1976) in that the amount of straw offered was based on goat weight and not so as to achieve a target rate of refusal. The results, however, corroborated those of Experiments 2 and 3. They also showed (not unexpectedly) that intake response diminished with increasing allowance rate, particularly for estimated digestible OM intake. Experiment 5 (Table 5) with sheep showed similar results.

Table 3. Effect of chopping the straw on the response of goats to increasing refusal-rate allowance (Experiment 3).

	Refusal rate main effect		Staw length main effect		SED main effect	Refusal rate x straw length interaction
Treatment1	20%	50%	Long	Chopped2
Number of goats3	16	16	16	16
Straw intake (g DM kg-1 W d-1)	13.1	18.0	16 .5	14.7	1.71	NS
Straw refused (% of offered)	19.4	49.1	39.3	40.8
Composition of refused straw
Nitrogen (g kg-1 DM)	4.9	5.0	4.7	5.2	0.22	NS
ADF (g kg-1 DM)	583	582	608	557	5 64	NS
In vitro DOMD4 (g OM kg-1 DM)	343	371	326	388	1.20	P<0.05a

1. Design: 2×2 factorial, 8 replicates.
2. Using a precision-chop forage harvest en
3. Mean liveweight 30.5 kg.
4. Tilley and Terry (1963). a. Difference between long and chopped greater with 20% refusal rate.
Source: Wahed (1987).

Experiment 6

Experiment 6 (Table 6) investigated the feasibility of refeeding 'stall-grazed straw', as such or after treatment with ammonia (stack method; Sundstøl and Coxworth, 1984). Intake of untreated stall-grazed straw (straw-previously-refused) was significantly less than that of the original straw, but intake of digestible OM (measured in vivo) of the treated stall-grazed straw was the same as that with the original untreated straw.

Table 4. Effect of amount offered on intake and selection of barley straw by goats (Experiment 4).

	Straw offered (g DM kg-1 W d-1)
	18	54	90	SED
Number of goats	12	12	12
Initial (day 1) liveweight (kg)	30,2	30,6	30,4	0.56
Final (day 42) liveweight (kg)	30.1	33.1	34.0	0.71
Straw intake1 (g DM kg-1 W d-1)	15.5	22.8	26.2	0,74
Straw intake (g DM kg-1 W0.75 d-1)	36.0	54.2	62.3	1.73
Straw refused (% of offered)	12.5	56.6	70.3
Estimated intake of straw digestible OM2 (g kg-1 W d-1)	7.2	12.8	14.5

Chemical composition	Straw offered	SE	straw refused
Nitrogen (g kg-1 DM)	7.4	0.12	5.5	5.7	6.1	0.11
ADF (g kg-1 DM)	528	2.0	565	583	574	6.9
In vitro DOMD3 (g OM kg-1 DM)	443	4.5	354	370	403	14.5

1. Concentrate supplement also fed at 15 g DM kg^-1 W^0.75 d^-1.
2. Calculated from in vitro digestibility of straw offered and refused.
3. Tilley and Terry (1963).
Source: Wahed and Owen (1986b).

Experiment 7

Experiment 7 (Table 7) was only recently completed and aimed to assess whether intake and selection response to increasing refusal allowance would be affected by treating the straw with sodium hydroxide (dip method; Sundstøl, 1981). The preliminary results are somewhat surprising, indicating no apparent increase in straw DM intake due to increasing the refusal allowance. There was a significant increase in DM intake in response to NaOH treatment. In this experiment samples of straw offered and refused were botanically fractionated (Ramazin et al, 1986), and the results (Table 7) indicate that generous feeding (allowing high refusals) increased intake of leaf plus sheath and reduced intake of stem. As expected with barley straw (Ramazin et al, 1986), the leaf plus sheath fraction was of higher nutritive value than the stem fraction (Table 8).

Table 5. Effect of amount offered on intake and selection of barley straw by sheep (Experiment 5).

	Straw offered (g DM kg-1 W d-1)
	18	54	90	SED
Number of wethers1	10	10	10
Straw intake2 (g DM kg-1 W d-1)	14.1	19.0	22.2	0.81
Estimated digestibility of straw consumed3 (g OM kg-1 DM)	467	562	572
Straw refused (% of offered)	20.8	64.7	75.1

Chemical composition	Straw offered	SE	Straw refused
Nitrogen (g kg-1 DM)	6 4	0.2	4.5	5.1	5.5	0.12
ADF (g kg DM3)	542	7.9	610	581	564	8.4
In vitro DOMD3 (g OM kg-1 DM)	432	0.8	294	361	374	7.2

1. Mean liveweight 52.8 kg.
2. Concentrate supplement also fed at 15 g DM kg^-1 W^0.75 d^-1.
3. Calculated from in vitro digestibility of straw offered and refused.
4. Tilley and Terry (1963).
Source: Naate (1986).

Table 6. Digestible straw intake by goats fed straw of straw-previously-refused, with or without ammonia treatment (Experiment 6).

	Straw		Straw-previously2 refused by goats
	Un - treated	NH3 - treated	Un - treated	NH3 - treated	SED
Number of goats3	6	6	6	6
Straw intake	22.8	24.5	15.8	19.4	1.98
g DM kg-1 W d-1
g DM kg-1 W0.75 d-1	53.9	58.9	37.5	47.3	4.71
g digestible OM4 kg-1 W d-1	9.7	12.6	6.6	9.7	0.90
g digestible OM kg-1 W0.75 d-1	22.9	30.4	15.6	23.7	2.11

1. Barley straw fed in Experiments 2 and 4; fed to allow 50% rate of refusal; straw chopped. Concentrate supplement also fed at 15 g DM kg^-1 W^0.75 d^-1
2. Straw from 50% refusal rates in Experiments 2 and 4; straw chopped Concentrate supplement also fed at 15 g DM kg^-1 W^0.75 d^-1.
3. Mean liveweight 36.0 kg.
4. In vivo digestibility measured; concentrate OMD assumed to be 80%.

Source: Wahed and Owen (1987).

Discussion

The above results clearly support the hypothesis that goats and sheep will consume more barley straw if they are permitted to reject 50% of that offered, rather than the conventional 10 to 20%. Furthermore, the improvement in consumption of digestible straw is even greater because generous feeding allows animals to select the more digestible fractions (leaf rather than stem).

These findings need to be corroborated with in vivo measurements of digestible straw intake and also with measurement of animal productivity. The experiments reported are tedious to execute and offer much scope for arriving at erroneous conclusions. For example, incomplete collection of straw refusals would exaggerate treatment response, as unrecorded refusals would be deemed eaten. Grazing research techniques (e.g. Mayes et al, 1986) might have application for measuring quantity and quality of straw consumed.

Table 7. Effect of refusal rate and NaOH treatment of barley straw on intake and selection by goats (Experiment 7).

Straw refusal allowance (% of offered):		Untreated		NaOH- treated
		20	50	20	50
Number of goats1		9	9	9	9
Straw offered2
	Amount (g DM d-1)	805	1398	1031	1807
	Leaf plus leaf sheath (g kg-1 straw DM)	449	449	451	451
	Stem (g kg-1 straw DM)	477	477	502	502
Straw refused
	Amount (g DM d-1)	166	758	197	933
	Leaf plus leaf sheath (g kg-1 straw DM)	307	359	355	405
	Stem (g kg-1 straw DM)	661	613	612	559
Straw consumed
	Total (g DM kg-1 W d-1)	16.5	16.7	19.8	20.0
	Leaf plus leaf sheath (g DM kg-1 W d-1)	6.6	10.6	7.1	11.2
	Stem (g DM kg-1 W d-1)	7.1	5.3	9.4	8.8

1. Saanen castrates, mean liveweight 41.1 kg.
2. Concentrate supplement also fed at 18 g DM kg^-1 W^0.75 d^-1
Source: E Owen, R Alimon and W El-Naiem (unpublished data).

Table 8. Composition of straw offered in Experiment 7.

	Untreated straw		NaOH - treated straw
	Leaf + leaf sheath	Stem	Leaf + leaf sheath	Stem
Ash(g kg-1 DM)	28.0	22.0	72.0	44.8
Na (g. kg-1 DM)	1.7	1.2	21.1	14.5
ADF (g kg-1 DM)	512	668	501	610
In vitro DOMD1 (g OM kg-1 DM)	515	262	664	367

1. Tilley and Terry (1963).

The extent to which selective feeding by small ruminants occurs with straws other than barley needs researching. Smith et al (in press), in Zimbabwe, recently found that unsupplemented, coarse-milled (14 mm screen) maize stover offered to lambs of 37 kg liveweight, at 15, 20, 25 or 30 g kg^-1 W d^-1, resulted in intakes of 23.2, 21.4, 22.3 and 29.1 g DM kg^-1 W^0.75 d^-1. These rates were associated with refusal rates of 42, 61, 67 and 64%. When supplemented with protein (270 g DM per lamb, daily), maize stover intakes were improved; 26.6, 29.4, 25.3 and 36.1 g DM kg^-1 W^0.73 . The authors found little difference between the chemical composition of the stover offered and refused, but admitted to difficulties in collecting representative samples. Clearly more work is required. The same study involved feeding unsupplemented rotor-slashed maize stover to steers at 15, 20, 25 or 30 g DM kg^-1 W d^-1. Intakes increased with increasing rates of offer; 41.5, 42.9, 49.9 and 49.0 g DM kg^-1 W^0.73. These intakes were associated with refusal rates of 31.8, 48.5, 51.2 and 59.5%. The authors were unable to conclude whether or not the greater intakes by steers were due to selective feeding.

The work of Capper et al (1986), Tuah et al (1986), Ramazin et al (1986), Givens (1987) and Doyle et al (1986b) stresses the magnitude of the differences in feeding value between straws of a given type. Differences in leaf:stem ratio probably account for much of this. Other factors, such as content of soluble phenolics (Reed, 1986) may further contribute to differences in nutritive value between tropical crop residues. Interactions between straw allowance rate and straw type, as affecting selectivity and intake, are therefore likely. Zemmelink (1986) has clearly shown this to be so for tropical forages.

The need to chop residues requires clarification. As indicated earlier, studies on feeding straws/stovers invariably use chopped material to facilitate handling and minimise selection. Experiment 3 showed no clear differences due to physical form, although intake of long straw tended to be greater than that of chopped straw. A recent experiment at Reading with goats allowed refusal rates of 25% showed that intake of barley straw treated with sodium hydroxide using a dip method (straw in long form) was markedly higher than that of straw treated using a commercial onfarm method (JF machine straw shredded) (Wrathall et al, in press). Goats appeared to find shredded straw unpalatable. The case for chopping will vary with type of crop residue. This subject needs more research.

The selective feeding associated with generous feeding of straw, shown by the Reading experiments, is likely to have implications concerning the need for nitrogen supplementation of crop residues. The extent to which interactions occur between crop-residue feeding rates and supplementation needs examining. Type of supplement could also have influence. The homogeneous nature of milled and pelleted concentrates precludes selective feeding but this would not be the case with sun-dried forage legumes. Physical form of the latter therefore needs consideration, along with the extent to which selective feeding is affected by the type of crop residue.

A feeding strategy allowing goats and sheep to reject 50% of the straw offered would be clearly wasteful and could only be justified if the rejected straw could also be used. Experiment 6 demonstrated that rejected straw can be refed and high levels of digestible straw intake achieved if it is treated with ammonia. Feeding untreated straw to allow 50% refusals and refeeding these after ammonia treatment would result in little wastage and high intakes of digestible straw. The economics of such a strategy need investigating as labour costs would be high. A simpler approach would be first to graze the straw in the field and then collect the residues after grazing and refeed it either after upgrading or with generous supplementation. Another strategy would be to feed straw generously to goats and then offer the refusals to less-selective ruminants, such as cattle or buffalo. In future such refusals might well have value for industrial purposes (Hartley et al, 1987) also, especially in developing countries (e.g. paper products, hardboards, egg-trays etc).

Acknowledgements

A.A.O. Aboud acknowledges a postgraduate fellow ship from the Norwegian Agency for Development (NORAD).

References

AOAC (Association of Official Analytical Chemists). 1975. Official methods of analysis of the Association of Official Analytical Chemists. 12th edition. AOAC, Washington DC, USA.

ARC (Agricultural Research Council). 1980. The nutrient requirements of ruminant livestock. Commonwealth Agricultural Bureaux, Slough, UK. 351 pp.

Blaxter K L, Wainman F W and Wilson R S. 1961. The regulation of food intake by sheep. Animal Production 3:51-62.

Capper B S. Thomson E F. Rihawi S. Termanini A and McCrae R. 1986. The feeding value of straw from different genotypes of barley when given to Awassi sheep. Animal Production 42:337-342.

Chenost M (ed.). (in press). Evaluation of straws in ruminant feeding. Proceedings of 3rd COS 84 Bis Workshop, Theix, France. European Economic Community, Brussels.

Devendra C and Burns M. 1983. Goat production in the tropics. Commonwealth Agricultural Bureaux, Slough, UK. 183 pp.

Doyle P T. Pearce G R and Devendra C (eds). 1986a. Rice straw as a feed for ruminants. International Development Programme of Australian Universities and Colleges, Canberra, Australia. 117 pp.

Doyle P T. Pearce G R and Egan A R. 1986b. Potential of cereal straws in tropical and temperate regions. In: M N M Ibrahim and J B Schiere (eds). Rice straw and related feeds in ruminant rations. Agricultural University, Wageningen, the Netherlands. pp. 63-79.

Edelsten P and Ligongwa C. 1981. Utilization of low quality roughages - economic considerations in the Tanzanian context. In: J A Kategile, A N Said and F Sundstøl (eds), Utilization of low quality roughages in Africa. Agricultural University of Norway, Aas, Norway. pp. 189-200.

Gartner J A. 1984. Report of the FAO Expert Consultation on improving the efficiency of small-scale livestock production in Asia: A systems approach. Food and Agriculture Organization of the United Nations, Rome, Italy.

Giaever H. 1984. The economics of using straw as feed. In: F Sundstøl and E Owen (eds), Straws and other fibrous by-products as feed. Elsevier, Amsterdam, the Netherlands. pp. 558-574.

Gibb M J and Treacher T T. 1976. The effect of herbage allowance on herbage intake and performance of lambs grazing perennial ryegrass and red clover swards. Journal of Agricultural Science (Cambridge) 86:355-365.

Givens D I. 1987. The nutritive value of untreated cereal straw in the UK - a short review. Agricultural Progress 62:26-34.

Goering H K and Van Soest P J. 1970. Forage fiber analysis. Agriculture Handbook No. 379. Agricultural Research Service, United States Department of Agriculture, Washington DC, USA.

Greenhalgh J F D. 1984. Upgrading crop and agricultural by-products for animal production. In: F M C Gilchrist and R I Mackie (eds), Herbivore nutrition in the subtropics and tropics. Science Press, Pretoria. pp. 167-181.

Hartley B S, Broda P M A and Senior P J (eds). 1987. Technology in the 1990's: Utilization of lignocellulosic wastes. The Royal Society, London, UK. pp. 405-568.

Hilmersen A, Dolberg F and Kjus O. 1984. Handling and storing of straws. In: F Sundstøl and E Owen (eds), Straw and other fibrous byproducts as feed. Elsevier, Amsterdam, the Netherlands. pp. 25-44.

Kossila V L. 1984. Location and potential feed use. In: F Sundstøl and E Owen (eds). Straw and other fibrous by-products as feed. Elsevier, Amsterdam, the Netherlands. pp. 422.

Mayes R W. Lamb C S and Colgrove Patricia. 1986. The use of dosed and herbage n-alkanes as markers for the determination of herbage intake. Journal of Agricultural Science (Cambridge) 107:161-170.

McCammon-Feldman B. Van Soest P J. Horvath P and McDowell R E. 1981. Feeding strategy of the goat. Cornell International Agricultural Mimeograph 88. Cornell University, Ithaca, NY, USA.

Mlay G I. 1987. The economics of by-products transportation. In: D A Little and A N Said (eds), Utilization of agricultural byproducts as livestock feeds in Africa. ILCA, Addis Ababa, Ethiopia. pp. 111-117.

Mueller-Harvey I, Reed J D and Hartley R D. 1987. Characterisation of phenolic compounds including flavanoids and tannins of ten Ethiopian browse species by high-performance liquid chromatography. Journal of the Science of Food and Agriculture 39:1-14.

Naate N M. 1986. Effect of the level of offer on selection and intake of barley straw by sheep. M.Sc. thesis, University of Reading, Reading, UK.

Owen E. 1981. Use of alkali-treated low quality roughages to sheep and goats. In: J A Kategile, A N Said and F Sundstøl (eds), Utilization of low quality roughages in Africa. Royal Agricultural University of Norway, Aas, Norway. pp. 131-150.

Owen E. 1985. Crop residues as animal feeds in developing countries - use and potential use. In: M Wanapat and C Devendra (eds), Relevance of crop residues in animal feeds in developing countries. Funny Press, Bangkok. pp. 25-42.

Owen E and Kategile J A. 1984. Straw etc. in practical rations for sheep and goats. In: F Sundstøl and E Owen (eds), Straw and other fibrous by-products as feed. Elsevier, Amsterdam, the Netherlands. pp. 454-486.

Owen E, Wahed R A, Alimon R and El-Naiem W. (in press). Strategies for feeding straw to small ruminants - upgrading or generous feeding to allow selective feeding. In: A N Said and D A Little (eds), Overcoming constraints to the efficient utilization of agricultural by-products as animal feed. ILCA, Addis Ababa, Ethiopia.

Preston T R. Kossila Vappu, Goodwin J and Reed Sheila (eds), Better utilization of crop residues and by-products in animal feeding: Research guidelines. 1. State of knowledge. FAO Animal Production and Health Paper 50. Food and Agriculture Organization of the United Nations, Rome, Italy. 213 pp.

Ramazin M, Ørskov E R and Tuah A K. 1986. Degradation of straw. 2. Botanical fractions of straw from two barley cultivars. Animal Production 43:271-278.

Reed J D. 1986. Relationships among soluble phenolics, insoluble proanthocyanidins and fiber in East African browse species. Journal of Range Management 39:5-7.

Smith T. Chakanyuka S. Sibanda S and Manyuchi B. (in press). Maize stover as a feed for ruminants. In: A N Said and D A Little (eds). Overcoming constraints to the efficient utilization of agricultural byproducts as animal feed. ILCA, Addis Ababa, Ethiopia.

Sundstøl F. 1981. Methods for treatment of low quality roughages. In: J A Kategile, A N Said and F Sundstøl (eds), Utilization of low quality roughages in Africa. Agricultural University of Norway, Aas, Norway. pp. 61-80.

Sundstøl F and Coxworth E M. 1984. Ammonia treatment. In: F Sundstøl and E Owen (eds), Straw and other fibrous by-products as feed. Elsevier, Amsterdam, the Netherlands. pp. 196-249.

Sundstøl F and Owen E (eds). 1984. Straw and other fibrous by-products as feed. Elsevier, Amsterdam, the Netherlands. 604 pp.

Tambi E N. (in press). Smallholder production: Constraints on the adoption of improved technologies. In: A N Said and D A Little (eds). Overcoming constraints to the efficient utilization of agricultural byproducts as animal feed. ILCA, Addis Ababa, Ethiopia.

Tilley J M A and Terry R A. 1963. A two-stage technique for the in vitro digestion of forage crops. Journal of the British Grassland Society 18:104-111.

Timon V M and Hanrahan J P (eds). 1986. Small ruminant production in the developing countries. FAO Animal Production and Health Paper 58. Food and Agriculture Organization of the United Nations, Rome, Italy. 234 pp.

Tuah A K, Lufadeju E, Ørskov E R and Blacknett G A. 1986. Rumen degradation of straw. 1. Untreated and ammonia treated barley, oat and wheat straw varieties and triticale straw. Animal Production 43:261-269.

Wahed R A. 1987. Stall-feeding barley straw to goats: The effect of refusal-rate allowance on voluntary intake and selection. Ph.D. thesis, University of Reading, Reading, UK.

Wahed R A and Owen E. 1986a. Comparison of sheep and goats under stall-feeding conditions: Roughage intake and selection. Animal Production 42:89-95.

Wahed R A and Owen E. 1986b. The effect of amount offered on selection and intake of barley straw by goats. Animal Production 42:473 (Abstract).

Wahed R A and Owen E. 1987. Intake and digestibility of barley straw by goats: Effect of ammonia treatment of straw and straw-previously-refused by goats. Animal Production 44:479 (Abstract).

World Bank and Winrock International. (1983). Sheep and goats in developing countries: Their present and potential role. World Bank and Winrock International, Washington, DC, USA.

Wrathall J H M, Owen E and Pike D J. (in press). Upgrading barley straw for goats: The effectiveness of sodium hydroxide and urea dip method. Animal Feed Science and Technology.

Zemmelink G. 1986. Evaluation and utilization of tropical forages. In: M N M Ibrahim and J B Schiere (eds). Rice straw and related feeds in ruminant rations. Agricultural University, Wageningen, the Netherlands. pp. 285-298.

Discussion

Thomson: Would it be feasible to upgrade the stem fraction rather than treat the whole crop?

Aboud: Treating only the stem would reduce the amount of material to be processed, thereby reducing costs of feeding.

Ørskov: Your results suggest that goats eat more than sheep but the opposite may occur with different groups of animals. The extent to which sheep and goats select leaf material depends upon local conditions. I have found that sheep will select leaf blade material in barley straw but cannot separate leaf sheath from the stem. Since, in the United Kingdom, leaf blade may constitute only 15% of the weight of the straw, selection may be less important.

Aboud: We found it difficult to fractionate the refusals. My statement regarding the superiority of goats was qualified as relating to our conditions but is in agreement with the bulk of the scientific literature. So far we have no data on the relative amounts of leaf blade and sheath selected.

Van Soest: Selectivity of feeding is less important in temperate regions than in the tropics because there is less difference in feeding value among plant parts under temperate conditions. In the tropics selection may be vital since the overall value of the feeds available are often sub-maintenance.

Pearce: In Australia we have found that sheep will remove leaf sheath from stems but the response is highly variable. Some animals will exhibit no selection at all whilst others will effectively remove sheath material.

Thomson: The choice of a refusal level in feeding experiments is difficult. At ICARDA we use a level of 20% but follow local farmers' practice of fine chopping.

Little: Experiments have been conducted relating intake of legumes and grasses to the level of refusals allowed. These experiments indicate that a 20% refusal level is the minimum amount that should be allowed. In Cameroon, on-station experiments have shown that sheep perform better than goats, but in the villages goats out-perform sheep.