SESSION 3a (contd.)

THE USE OF ACACIA KARROO AND ACACIA NILOTICA LEAVES AS DRY SEASON SUPPLEMENTARY FEEDS FOR LIVESTOCK

J. S. Dube
Matopos Research Station, P. Bag K5137, Bulawayo, Zimbabwe

Introduction

In Zimbabwe there is a marked decline in the nutritive value of rangeland grass in the dry season (Plowes, 1957; Elliott and Fokkema, 1960; Sibanda, 1984). However, browse from indigenous trees contains a high concentration of nitrogen (N) and is available during this critical period (August to November) when ruminant livestock are in need of supplementary nourishment (Walker, 1980; Dube and Ndlovu, 1994).

Acacia karroo and Acacia nilotica thorn trees dominate rangeland in the Matopos area and most semi-arid parts of Africa (Barnes et al., 1996). The nutritive value of browse from these trees is complicated by the presence in the browse of anti-nutritive factors which interfere with the availability of nitrogen (N) to the animal.

Strategies are being developed at Matopos Research Station for the use of stored dried Acacia leaves as dry season supplements to rangeland grass.

Materials and methods

Experimental design

Mixtures of high and low condensed tannin (proanthocyanidin) feeds were fed to goats. The following mixtures were fed: (1) A. karroo (Ak) and Securinega virosa (Sv) and, (2) A. karroo and A. nilotica (An). The mixtures were fed according to the following scheme:

The same scheme was repeated with A. karroo and A. nilotica.

Animals and diets

Twenty-four castrated indigenous Matebele goats of nearly the same age and weight were housed individually in metabolism crates. They were allocated randomly to each of the browse mixtures and hay so that there were four animals to each diet. The design was a completely randomised design. The browse (air-dried in a well-ventilated room) was offered at 08.00 hours while hay was offered at 14.00 hours. The adaptation period to the diets and crates was 14 days. Faces and urine were collected in the next seven days, during which time intake of the diets was estimated.

Measurements and analyses

Food eaten was estimated daily from the amount offered minus the amount of food left in the feeding trough. Both food offered and refusals were weighed and their dry matter determined.

Collection of faeces and urine

The daily output of faeces for each animal was weighed and ten per cent of the weight was subsampled and stored at -20° C to await analysis. Ten per cent of the daily output of urine, preserved by collecting the urine into vessels containing 25 per cent (v/v) sulphuric acid, was stored at -20° C to await analysis.

Analyses

All browse and faecal samples were dried at 55° C for 48 hours or until constant weight and then ground to pass a 1-mm screen. The browse and faecal samples were analysed for N, ash, neutral-detergent fibre (NDF), acid-detergent fibre (ADF), neutral-detergent insoluble N (NDIN) and acid-detergent insoluble N (ADIN). Browse samples were, in addition, analysed for total soluble phenolics (YbPPT) and soluble proanthocyanidins (solPAs). Urine was analysed for N only.

Statistical analysis

The data was analysed by Analysis of Variance using a SAS computer program (SAS Institute, 1986) which accounted for diet effect, i.e.,
      Y_i=u+A_i+E_i
where u=mean
          A=diet effect
          E=error.
Regression analysis was used to test if animal response to the diet was dose-dependent.

Results

The chemical composition of the diets is shown in Tables 1 and 2.

Table 1. Chemical composition of the diets in experiment 1

A₅₅₀ - absorbance units of extract at 550 n
YbPPT - total acetone soluble phenolics precipitated by ytterbium metal

Table 2. Chemical composition of the diets in Experiment 2.

Intake

When S. virosa alone or A. karroo and S. virosa mixtures were offered to the goats there was an increase in total dry matter (DM) and browse DM intake (P<0.05) as compared to feeding A. karroo alone (Table 3). There was no change (P<0.05) in hay intake. Intake of total DM or browse was higher (P<0.05) when mixtures of A. karroo and A. nilotica were offered than when either A. karroo or A. nilotica was offered alone (Table 4). Intake of A. karroo was higher than intake of A. nilotica. There was no change in the intake of hay.

Table 3. Intakes of total, browse and hay DM by goats consuming A. karroo (AK) and S. virosa (Sv) mixtures, and hay

DM - g per (kg w^0.75) per animal per day

Table 4. Intakes of total, browse and hay DM by goats consuming A. karroo (Ak) and A. nilotica (An) mixtures, and hay

DM - g per (kg w^0.75) per animal per day

Neutral-detergent insoluble N and Acid-detergent insoluble N

In the A. karroo-S. virosa experiment there was a decrease in NDIN when the animals were fed mixtures but ADIN first increased and then decreased to below the ADIN level recorded when the animals consumed A. karroo alone (Table 5). There was less NDIN and ADIN when A. karroo-A. nilotica mixtures were supplements than when A. karroo was the sole supplement (Table 6).

Table 5. Faecal, urinary and retained N (g per day) in goats consuming A. karroo (Ak) and S. virosa (Sv) mixtures, and hay

Table 6. Faecal, urinary and retained N (g per day) in goats consuming A. karroo (Ak) and A. nilotica (An) mixtures, and hay

N-retention

Animals consuming A. karroo or hay alone had negative N retention. There was a marked increase (P<0.05) in N retention when mixtures of A. karroo and S. virosa were consumed by the goats (Table 7). Furthermore, N retention increased linearly with every additional increase in the amount of S. virosa consumed (R²=0.77).

Table 7. Faecal NDF and ADF N (g per day) of goats consuming Acacia karroo (Ak) and Securinega virosa (Sv) mixtures, and hay

NDIN - neutral - detergent insoluble N
ADIN - acid - detergent insoluble N

Similarly, in the second experiment, animals consuming A. karroo browse as the only supplement had negative N retention. Nitrogen retention became positive with every addition of A. nilotica to A. karroo but N retention rose and then remained constant (Table 8). Nitrogen retention was positive for animals feeding on hay alone in this experiment.

Table 8. Faecal NDF and ADF N (g per day) in goats consuming A. karroo (Ak) and A. nilotica (An) mixtures

NDIN - neutral-detergent insoluble N
ADIN - acid-detergent insoluble N

Discussion

Acacia nilotica has a high content of polyphenolics but low PA content. Acacia karroo, on the other hand, is relatively low in total phenolics but has a high content of PAs. Securinega virosa has a low content of all phenolics. Proanthocyanidins are generally thought to be responsible for decreased availability of N in feeds in which they occur but their effect on intake is controversial (McLeod., 1974; Reed et al., 1985; Woodward and Reed, 1989; Dube and Ndlovu, 1995).

Several methods have been used to alleviate the adverse effects of PAs. Among these are use of concentrates, strategic lopping, use of metal ions, heating, microbial degradation, use of PEG or PVP and simple dilution. Simple dilution or feeding of mixtures is attractive because it is cheap and can easily be handled by the communal farmer.

Nitrogen retention was negative when A. karroo was the sole supplement to grass. It was positive when A. karroo was mixed with S. virosa or A. nilotica. When a low PA browse, for example, S.virosa, is consumed alone, there is rapid ammonia production in the animal rumen and consequent loss of N as urea in urine. When a high PA browse is consumed, most of the N is unavailable to the animal. However, A. nilotica presents a problem because it has a high content of total phenolics which are probably responsible for the low intake of the browse. The hay used in the A.nilotica experiment had a high content of N, thus diminishing the usefulness of the browse species as supplements to hay.

Conclusion and recommendations

The two experiments conducted in this study indicated that A. karroo and A. nilotica could be used together as dry season supplements to grass. The easily available A. karroo can be used either with the less common high protein S. virosa or with the easily obtainable A. nilotica in the Matopos rangeland.

Instead of plucking individual leaves from branches of the plant and then drying the leaves, as was done in these experiments, it is recommended that lopped tree branches be laid down to dry in a well-ventilated room. The branches can then be simply shaken to release the leaves. The regeneration of the lopped trees is rapid. We have been lopping large quantities of browse from trees in the Matopos Research Station area since 1986 without any visible effect on the trees.

Acknowledgements

The author would like to thank Matopos Research Station, Department of Research and Specialist Services, Ministry of Agriculture, Zimbabwe, for the use of the animals and other facilities. Thanks are also extended to Professor L.R. Ndlovu of the Department of Animal Science, University of Zimbabwe, whose advice was invaluable.

References

Barnes, R.D., Filer, D.L. and Milton, S.J. 1996. Acacia karroo. Tropical Forestry Papers No. 32. Oxford Forestry Institute, Oxford University.

Dube, J.S. and Ndlovu, L.R. 1994. A note on seasonal variations in the chemical composition of four browse species in the Matopos redsoil thornveld. Zimbabwe Journal of Agricultural Research. 4: 119–123.

Dube, J.S. and Ndlovu. L. R. 1995. Feed intake, chemical composition of faeces and nitrogen retention in goats consuming single browse species or browse mixtures. Zimbabwe Journal of Agricultural Research. 33: 133–141.

Elliott, R. C. and Fokkema, K. 1960. Digestion trials on Rhodesian feedstuffs. Rhodesia Agricultural Journal. 57: 252–256.

McLeod, M. N. 1974. Plant tannin - their role in forage quality. Nutrition Abstracts and Reviews. 44(11): 803–812.

Plowes, D.C.H. 1957. The seasonal variation of crude protein of twenty common grasses at Matopos, Southern Rhodesia, and related observations. Rhodesia Agricultural Journal. 54: 33–55.

Reed, J.D., Horvath, P. J., Allen, M. S. and Van Soest, P. J. 1985. Gravimetric determination of soluble phenolics including tannins from leaves by precipitation with trivalent ytterbium. Journal of the Science of Food and Agriculture. 36: 255–261.

SAS 1986. SAS User's Guide: Statistics. SAS Institute Inc., Cary, North Carolina.

Sibanda, S. 1984. Composition of diet selected from veld by steers fistulated at the oesophagus and body mass of non-fistulated steers grazing the same paddocks. Zimbabwe Journal of Agricultural Research. 22: 105–107.

Walker, B.H. 1980. A review of browse and its role in livestock production in Southern Africa. In: Browse in Africa, Ed: H.N. Houerou, International Livestock Centre for Africa, Addis Ababa, Ethiopia. pp7–24.

Woodward, A. and Reed, J.D. 1989. The influence of polyphenolics on the nutritive value of browse: a summary of research conducted at ILCA. International Livestock Centre for Africa, Addis Ababa, Ethiopia.

CONSERVATION OF FORAGES FOR DRY SEASON FEEDING OF LIVESTOCK IN THE SEMI ARID AREAS OF THE TROPICS.

Marion Titterton¹, O.Mhere², Barbara Maasdorp³, T. Kipnis⁴, G. Ashbell⁵ and Z. Weinberg^5.
Department of Animal Science, University of Zimbabwe, P.O. Box MP 167, Mt. Pleasant, Harare, Zimbabwe.
Matopos Research Station, P. Bag K 5137, Bulawayo.
Department of Crop Science, University of Zimbabwe, P.O.Box MP 167, Mt. Pleasant, Harare, Zimbabwe.
Agricultural Research Organization, Institute of Field and Garden Crops, The Volcani Centre, Bet Dagan, Israel.
Agricultural Research Organization, Forage Preservation and By-products Research Unit, The Volcani Centre, Bet Dagan, Israel.

Introduction

Livestock is recognized as being an integral component of mixed farming systems which predominate in the tropics, particularly in the developing world. Animal manure and traction make the land more productive than would be the case in their absence. Yet, it has been recognized with equal force that livestock owned in the semi-arid developing world are forced to barely subsist on poor and sparse vegetation in the dry season, leading to severe loss of body condition, productivity and fertility and on the land, the threat of desertification. Technologies aimed at achieving a balance whereby livestock can increase in productivity, so enhancing wealth for the livestock owner, while resource degradation is minimized must be developed (Steinfeld, 1998). One such technology is the conservation of forage produced during the wet season which can be fed to livestock kept in at least partially zero-grazed systems during the dry season. This may, in fact, be the only such technology that would satisfy high demand for nutrients for such livestock production operations as small scale dairy farms in the semi-arid regions of the tropics (Dube, 1995).

Methods of conservation

There are two principal methods of conserving forage for the dry season, that of drying (hay) and ensilage.

Hay

Factors affecting hay quality

• Choice of crop and variety: If a crop or pasture is to be grown with the primary aim of making hay, a suitable variety should be chosen. In the semi-arid areas, there are few species of pasture grasses for hay production which can be grown without irrigation e.g. kikuyu and lucerne or are not frost tolerant, e.g. Buffels grass (Maclaurin and Wood 1987) or require heavy applications of nitrogen Forage legumes such as cowpea (Vigna unguiculata) and dolichos bean (Lablab purpureus) are the preferred choice for dryland production (Walker et al., 1992; Maasdorp and Titterton, 1997). When cut under ideal conditions yield and quality are high (Table 1).

Important considerations are:

The potential yield and quality required: the difficulty is in achieving a balance, as good quality is dependent on cutting early (three months after planting) in the season while good yields are achieved later (four to four and half months into the season) (Maroske et al., 1997).
rapid drying capacity (stem thickness). This is important when related to the need to cut early in the season. At this time, the rains are still prevalent and there may be only as much as one or two days to accomplish this. This can be achieved with the thin-stemmed forage legumes (Cameron, 1988) but bulky crops such as forage sorghums, napier and hybrid pennisetums can take up to three months to dry even when left out on racks (Mhere, 1997) and are more suitable for ensiling (Humphreys, 1980; Maroske et al., 1997).
good crop regrowth if required
flexibility in time of cutting without rapid quality deterioration and leaf loss. Later flowering varieties give such flexibility (Stuart et al., 1997)

• Time of cutting: The quality of forage crops changes over time as they grow older and taller. The time to cut will depend on such factors as the rains and the quantity and quality of hay required. Late cutting (four months after planting) may lead to potentially greater yield but quality will be lower than early cutting (two and a half months after planting). For maximum quality, the crops should be cut when crops are still in the young vegetative stage. Another advantage of early cutting is thinner stems, smaller windrows and therefore faster drying (Stuart et al., 1997) To achieve this , it may be necessary to plant only in early December so that cutting can be done in March, but this risks a short rainy season with the hay crop and in the semi-arid area, maximum use has to be made of the rains, which often come early (Mhere, 1997).
• Degree of weather damage: The longer the hay is drying in the field, the greater the risk of rain damage. Early cutting before stems become thick plus using a mower conditioner (crimper) will reduce the risk (Kaiser et al., 1993).
• Moisture content: The moisture content of forage when it is baled affects hay quality. A high moisture content (above 30%) when baled can lead to overheating in the hay. Heat destroys protein, lowering the quality of the hay and can also lead to a fire risk. On the other hand, allowing the crop to dry for too long (below 25%) can result in significant loss of leaf which is the most nutritious part of the plant. The time of baling or stacking is therefore critical. If hay can be put to dry before moisture falls below 30 percent, most of the field losses from leaf shattering can be avoided (Nobbs and Oliver, 1987).

Making hay

The hay making process involves (Stuart et al., 1993)

• mowing- this can be done with a reciprocating mower, or a forage harvester.
• initial drying- for rapid drying of a crop above 30 percent, a crimper which crushes the grass stem, would be required.
• windrowing
• further drying
• baling or stacking.

Table 1. Yield and quality of legumes when cut under ideal conditions (reasonable rainfall, early cut) in the semi-arid area.

Viability of hay making for small holder livestock production in the semi-arid areas.

In the semi-arid regions of the tropics, the conditions are harsh for conserved forage. High temperatures combine with short rainy seasons on largely poor soils to produce grasses and legumes which, while able to produce high yield under good management, still deteriorate rapidly in nutritional quality after only three months of growth. Protein and digestibility both decline rapidly in tropical grasses after flowering, as lignification proceeds in most tropical grasses and legumes (Skerman, 1988). In order to harvest grass and legumes of high nutritional quality, cutting has to take place at the early stage of growth, in fact while the rains are still prevalent. This leads to leaching and rotting of the lying crop and the hay may often be unpalatable because of mould. There are other problems as well. By April, when it is more feasible to make hay, legumes have mostly podded and at harvesting there is the loss of pod and leaf shattering because the legumes are too dry (Nobbs and Oliver, 1987). Forage sorghums and hybrid pennisetums which grow well in the semi-arid areas are very bulky by this time and take too long to dry. Unless a mower-conditioner is used with the harvested crop and it is then baled and taken for treatment with bulk driers in large hay sheds, it is unlikely a good hay crop can be produced at this time (Kaiser et al., 1993). All this involves machinery that is too expensive for any one small holder to use.

Silage

The process of ensilage (McDonald et al., 1991; Bareeba, 1992; Stuart et al., 1993 and Titterton, 1996).

There are two stages involved in this process.

Stage 1

This stage is one of aerobic respiration and it continues in the plant after it is cut and after it is packed into the silo, until all the plant sugars, or the oxygen, is depleted. This process is carried out by plant enzymes and aerobic organisms on the plant at ensilage and in the process, plant sugars are oxidised to carbon dioxide, water and heat.

Ideally, the process should take no longer than three hours and at the end of it, there should still be plant sugars available for fermentation by the anaerobic bacteria, provided of course that there were sufficient plant sugars in the plant in the first place. Therefore, the quicker the oxygen can be depleted, the better. Prolonged respiration leads to overheating (the silage should heat up to no more than 10°C over ambient temperature, that is, the temperature of the air outside the silo). With the excess production of heat, energy is lost to heat instead of being available for fermentation. Energy is also lost for feeding and proteins are heat damaged and totally indigestible. In addition, as long as oxygen is available, sugars will be oxidised instead of fermented, and lactic acid production will be inadequate.

To ensure rapid respiration requires the correct moisture content, correct length of chop (Table 2), rapid packing, (preferably within 24 hours), correct compaction and complete sealing. Plastic sealing has proved to be the best, which is then packed with sandbags around the edges, and covered with soil.

Table 2. Recommended stage of maturity, moisture content and length of chop of a number of plants for ensiling

Sources: Stuart et al., (1993); Mahanna (1994); Mhere (1997) and Panditharane et al., (1986).

Legumes and grasses can be cut early because they can be wilted to the correct moisture content. Maize and grain sorghum, on the other hand has to be cut at the chosen stage of maturity to obtain the correct moisture content.

Stage 2. Anaerobic phase

Once the oxygen is depleted sufficiently for fermentation, acetic acid bacteria and heterofermentative lactic acid bacteria produce acetic acid, ethanol, lactic acid and carbon dioxide. This takes from twenty four to seventy two hours and the pH is reduced to 5.5. Then, homofermentative bacteria predominate because 5.5 is optimum for their growth and so lactic acid is predominantly produced. Other bacteria continue to produce, in much smaller amounts, acetic, propionic and butyric acid. This brings the pH in maize silage down to 4,0 and the pH in legumes down to 4.5. The silage is essentially ready by this stage. The important thing to achieve in Stage 2, then, is the predominance of lactic acid bacteria who produce the lactic acid so necessary to good quality silage.

This can be ensured by there being sufficient available fermentable sugars after respiration has finished and rapid fermentation of those sugars. This is achieved by ensiling the crop at the correct stage of maturity and dry matter content (Table 2).

The following will occur if the crop is not properly ensiled (Mahanna, 1991):

• If the crop does not have sufficient fermentable sugars, spoilage bacteria will produce unpalatable acids (mainly butyric and acetic acids).
• High protein forage crops and pasture grasses, like lablab or cowpea have insufficient sugar for rapid fermentation, and a high content of protein which is then broken down because there is not enough lactic acid to prevent proteolytic bacteria from acting on the protein. If proteins have been proteolysed the bacteria release amides, which not only affect palatability of the silage but can be toxic. It is important then to put in an additive: maize grain, molasses, commercial additives or chopped sweet sorghum will all assist in providing the sugars for fermentation.
• If the crop is too wet, the sugars and the acids will be leached out of the silage and collect in effluent at the bottom. A rancid milk odour means butyric acid was produced by clostridial bacteria (from too high moisture, low plant sugar) which will make the silage unpalatable.
• If the crop is too dry, there will be poor compaction, allowing yeasts and moulds to spoil the silage and also overheating, with damage to proteins, producing caramelised, dark brown silage with a cooked smell and/or mouldy silage with an alcohol odour.
• If the crop is not well compacted, the silage will overheat, causing caramelisation of the sugars. The silage smells of caramel and the nutritional value of the silage is very low.

Points to remember in making silage (from de Figueiredo, 1994; Titterton, 1996):

• The crop must not be harvested too close to the ground, as soil particles entering the silage can introduce too much organic matter and undesirable bacteria. The crop should be cut at least four cm from the ground.
• The earlier a crop is harvested, the greater the nutritional quality, but it has to be balanced against the yield, dry matter content and the amount of fermentable sugars in the crop which increase with age. Wilting will increase both in a young crop but yield may be too low. Multiple cutting can be implemented with rapid growing crops such as forage sorghum and pennisetums, which increases the total yields. Sweet forage sorghum on the other hand, can be cut later, at flowering, since sugar levels are then high and a good yield can be produced. Maize, which does not grow again, has to be cut at hard dough stage to ensure both yield and good fermentation.
• Dry matter content will depend on the storage method. If the crop is ensiled in bunkers or pits, it should not be lower than 30 percent dry matter, otherwise there is leaching of water and soluble nutrients in effluent. If, however, it is stored in a sealed plastic silo, the effluent will not be lost and the silage can be as low as 25 percent dry matter. If crops are too wet at cutting, they can be wilted in the field for a few hours before being picked up and chopped. Crops should not be higher than 40 percent dry matter, or compaction will be too difficult. If they are too dry (because of, for example, drought), then spraying them with molasses in water will assist with both moisture and readily fermentable sugar when they are ensiled while, also making compaction easier.
• The crop must be chopped to a size that allows easy compaction, especially in large silos. In small silos such as plastic bags, crops have been successfully ensiled with only manual chopping to five cm in length but in the large silos, chopping should be no more than 1.5 cm. Fineness of cut also affects how much silage animals can eat - finer cutting results in higher intake. However, it is important to remember that if silage is the only roughage in the diet, too fine a chop will lead to the problems associated with a deficiency of roughage in the diet in the rumen. If silage is the only roughage, it would be better for rumen health to leave the chop length at a minimum of three cm.
• Air must be excluded, either by good compaction with a tractor in the silage pit or bunker or through extraction from plastic silos.
• A pit or bunker must be sealed with plastic over the top of the silage and which must then be held down with soil and old tyres. Plastic bags must be strong enough to avoid breakage, as this will allow air in and lead to spoilage.

Tests of silage quality (from Mahanna, 1994; McDonald et al., 1987 and Titterton, 1996)

Dry matter (DM)

This is an important factor affecting the success of fermentation, as explained above. The DM gives a good indication of how well fermentation should have taken place, given that all the other methods of ensilage (packing speed, compaction, sealing, etc.) were correct and that there was sufficient sugar for rapid fermentation.

pH

pH is a test of the acidity of silage- it does not tell you whether the acids are good or bad acids- lactic acid is good acid and butyric acid is bad acid-, but it does indicate that fermentation took place and that if it is low enough, it means that sufficient acid was produced to preserve the silage.

pH of maize and cereal silage should be less than or equal to 4.0
pH of legume or grass silage should be 4.3 to 4.5, maximum 5.0.

Fermented acids (% dry matter)

Lactic acid is the desirable acid to be produced because it gives the silage the palatability cows enjoy and indicates that the protein was preserved. The other acids are not desirable in large amounts because they reduce the palatability of the silage and indicate protein breakdown.

Lactic acid should be around six to eight percent of dry matter, acetic acid < two percent and butyric and propionic acid<one percent.

Ammonia to nitrogen ratio (expressed as a percentage)

This reflects the amount of protein broken down to ammonia in relation to total protein content. If there is too much ammonia in relation to total nitrogen, it means that there was not enough sugar or starch in the silage to allow the lactic acid bacteria to produce enough lactic acid and prevent the wrong kind of bacteria from thriving. The “bad” bacteria break down the protein, producing ammonia and amides. It is thought that the amides impart poor palatability to the silage.

In maize and cereals this should be < five percent
In legumes and grasses, this should be no more than 10 to 15 percent.

Water soluble carbohydrates - percent dry matter

In legumes and grasses: four to six percent
In maize and cereals: six to eight percent.

Viability of ensilage of forages for livestock production in the semi-arid areas.

Silage has many advantages over hay as a methods of forage conservation, namely:

• Any forage, whether a forage crop or grass from pasture, can be ensiled, provided there is sufficient fermentable water soluble carbohydrate to ferment by homofermentative lactic acid bacteria (de Figueiredo, 1994). High quality, high yielding forages can be produced from small areas of land for conservation but if cereal crops such as grain sorghum, sweet sorghum or pennisetums are used alone, protein levels are low and inadequate for maintenance of cattle. High protein forages such as legumes need to be incorporated with the cereal forage (Maasdorp and Titterton, 1997).
• Silage, unlike hay, can be made at any time, provided of course the harvesters can get into the fields and rainfall is not too heavy at the time of ensilage. Young crops which may be too wet for ensilage can be wilted adequately to 30 percent dry matter within a few hours, whereas hay needs several days for adequate drying to 95 percent dry matter.
• Silage, unlike hay, can be stored either in pits or bunkers, or plastic bags. This makes it much less vulnerable to the dangers of fire, insect damage and oxidation. For the small holder, however, pit silage is at a high risk of spoilage from poor compression and exposure of the silage. Additionally, the labour involved with daily removal of silage may be a constraint to sustainable forage conservation by this method.
• While a silage chopper is necessary for large scale silage making, it constitutes less machinery than a mower-conditioner and hay baler. For small holders, however any machinery of this nature would be too expensive to use. The use of portable diesel-driven choppers or manual choppers, if shared among a group of farmers is feasible but there are likely to be long delays in the ensilage process from one farm to the next (Musimwa. 1998).
• It is easier to incorporate silage into a total mixed ration than hay, since silage is already chopped to a size that allows efficient mixing, while hay still has to be chopped (de Figueiredo, 1994)
• Silage contributes to water reserves in the animal- one litre of water is required by cattle to digest a kilogram of silage while three litres of water is required to digest a kilogram of hay (Smith and Ncube, 1996).

• It is important, however, that silage be made efficiently to produce a good quality conserved forage. If silage is made badly, a whole pit of, say 200 m/t, of silage can be ruined because poor silage is very unpalatable and may be refused by the cattle. One advantage of hay is that, even if it is of poor quality, digestibility and palatability can be improved through treatment by urea or ammonia and molasses. Poor quality silage on the other hand, cannot be improved since the acids (butyric, acetic) and amides impose a very strong taste and smell to the silage which cannot be covered up.

Overcoming the constraints of ensilage for the small holder

While ensilage as a forage conservation method has distinct advantages over that of drying for hay, there are nevertheless a number of challenges to face and overcome before it can be accepted and adopted as a sustainable practice by small holder farmers. These are:

• The small holder has a very limited amount of land from which sufficient quantities of high quality forage must be produced and ensiled in order to maintain and ensure productivity in livestock during the dry season. Intercropping of cereal and legumes forages which are adapted to the semi-arid area provides a way of achieving this.
• The small holder does not have the farm machinery required to produce, cut, chop and compress crop material to ensure good fermentation and stable silage in pits.
• The small holder has limited labour which often has to be provided by women at feeding out. A low-cost ensilage technology must be found which not only ensures successful ensilage but minimises labour at feeding out.

Studies carried out to address the challenges of producing and ensiling forage crops for dry season feeding.

Intercropping to produce high yields of forages for ensilage

In the semi-arid regions of the tropics, maize, generally accepted as being the ideal silage crop in the high rainfall regions, is very susceptible to moisture stress and is thus questionable as the crop of choice for silage. Alternative crops such as grain sorghums, forage sorghum and forage pennisetums which are drought tolerant yet high yielding have been researched as silage crops and found to be suitable (Havilah and Kaiser, 1992) although it has been concluded after an evaluation of grain and forage sorghums in semi-arid areas that sweet forage sorghums offer better potential than grain sorghums (Cole et al., 1996). However, both forage sorghum and hybrid pennisetums have low protein content (70 and 95 g/kg respectively). Mhere (1999) intercropped forage sorghums (var. Sugargraze) and hybrid pennisetum (var. SDPN3) with cowpea or dolichos bean in three planting patterns over three years. The three planting patterns were those of 1) in-row ; 2) one row cereal, one row legume; 3) one row cereal, two rows legume. The trial was repeated over three seasons. The results showed improved total dry matter yields and crude protein content of the forage mixture and better land usage as shown by Land Equivalent ratios (Tables 3 and 4). However, while forage sorghums showed compatability with the two legumes, the hybrid pennisetum evidently requires modification of harvesting (multiple cutting) to allow better conditions for growth of the legume. Of the three planting patterns, pattern (2), that of one row cereal, one row legume, was found to be a suitable planting pattern. All the cereal and legume crops ensiled well with pH of 3.2 to 4.5, dry matter content of 270 to 350 g/kg and NH₃-N% of total N of 3.1 to 8.5. Crude protein content of the silage ranged from 85 to 140 g per kg DM, with digestibility of 540 to 650 g per kg DM. Varying hours of wilting appear to have had no beneficial effect on quality of silage; however, dry matter content was already optimal at ensilage. Further work is needed on early cut crops such as pennisetums (for multiple cutting technique) when dry matter content may be less than 25 percent at cutting.

Table 3. Quality parameters of hybrid pennisetum and forage sorghum intercrops with cowpeas and dolichos beans from the selected planting pattern (Table 4) during the 1995–1996 to 1997–1998 seasons.

MADF (Clancy and Wilson, 1966; Linn and Martin, 1991)
Digestibility derived from MADF using the formula Digestibility = 99.43 - 1.17 MADF.
Legume content of the total intercrop yield averaged over three years

Table 4. Dry matter yields (m/t per ha) and Land Equivalent Ratios (LER)^Y values for cereal and legume intercrops from three planting patterns averaged over three years (1995–1996, 1996–1997 and 1997–1998).

^* Selected planting pattern for large scale forage production on-station and on-farm during the 1998–1999 season.
^Y Total land are required under sole cropping to give yields obtained in the intercropping mixture. Where the LER = 1, there is no advantage of intercropping. If it is >1, a larger area of land is needed to produce the same yield of sole crop of each component than with an intercropping mixture.

Agroforestry also offers potential for improving protein content of mixed silages (Tjandraatmadja et al., 1994) The addition of wilted Amaranthus hybridus to maize (1:1) at the time of ensiling resulted in good fermentation and raised the crude protein content of the silage from 6.9 percent to 11.6 percent and reduced the crude fibre content (Bareeba, 1977). Maasdorp and Titterton (1999) successfully ensiled, on a fresh mass ratio of 50:50, the leaf material of four varities of forage tree legumes with maize with improvements of crude protein content to 14 percent, 15.5 percent, 17.2 percent and 18.7 percent in maize/Calliandra calothyrsus, maize/Glyricidia sepium, maize/ Leucaena leucocephala and maize/Acacia boliviana silages respectively, (Table 5). Only in the maize/Calliandra silage was organic matter digestibility significantly reduced, while in the other three, it was similar to that of maize silage. A similar trial is planned for forage tree legumes ensiled with forage sorghums and pennisetums.

Table 5. Fermentation and nutritional quality of silages made from maize and legume forage trees.

¹ Acidity of < pH5 needed for good preservation
² Protein-N loss as ammonia of < 15% is reasonable in legumes
³ CP = Crude protein
⁴ OMD-organic matter digestibility.

Development of a low-cost, minimal labour, small scale ensilage technology which can produce high quality silage

The treatments consisted of:

• Five forages (including two mixes) - forage sorghum-dolichos bean; pennisetum-dolichos bean; forage sorghum; pennisetum; dolichos bean.
• Two particle sizes (length of chop) - coarse chop from the use of pangas; fine chop from the use of a motor driven chaffer;
• Two methods of compression - mechanical from the use of a manual tobacco press with a screw press driven down on a metal plate covering the bag followed by twisting the top of the bag for tying with twine; manual with simple leaning on the bag accompanied by pushing out air pockets before twisting the top of the bag for tying with twine.
• Two silos: recycled plastic garbage bags large enough to carry 50 kg (90–100 micron thickness); fertiliser/maize bags large enough to carry 50 kg of + 125 micron thickness. The bags are however, filled to carry eight kg of silage, the amount estimated to comprise the daily total required to supplement an indigenous milking cow on natural pasture grazing.

Summary of results

Fermentation and nutritional quality are generally good, exhibiting similar results of analyses as that of the trials with intercropped and mixed crop silages above, Tables 6 and 7.

Table 6. Fermentation quality of different forage crops ensiled after differing treatments.¹

Table 7. Nutritional quality of silages made from different crops1

¹ Statistical differences are not displayed in these tables but will be shown in a paper to be published shortly. (An. Feed. Sci.).

There are differences in fermentation and nutritional quality of silages due to crop variety but none due to chop length, compression treatment or type of bag. Therefore on-farm silage can be made in plastic bags with coarsely chopped crop material which is compressed by hand.

Discussion

Collaborative studies carried out in Israel (Ashbell et al., 1999) show that the success of ensilage despite the lack of fine chopping and effective compression is due to effective sealing in bags, preventing the loss of effluent containing lactic and volatile fatty acids, compared with pit or bunker silages where loss of effluent is high, necessitating fine chopping and effective compression.

The fertiliser bags, being thicker, could be used over two seasons. This puts their present cost ($9.00) equivalent, when used over two seasons, to the cheaper bags (present cost around $4.50). When no longer usable, the bags can be used for the manufacturing of wax for floor polish, a traditional practice in most small holdings. Some losses (2 to 5%) in the cheaper bags were likely due to poor packing of bags in the silage store room and this requires attention. The technology is found to be ideal for smallholder farmers as losses are minimum, compared with silage pits which have been used in smallholder dairy farms in the higher rainfall areas, where silage losses amount to as much as 30 percent due to poor compression and exposure of the remaining silage in daily removal for feeding. The bag technology has also been found to benefit women and children as the bags can easily be stored and carried to the cows for feeding, thus there is minimum labour in feeding compared with daily digging out from pits. It is environmentally more friendly as there is no effluent discharges into the soil as occurs with pit silage.

Conclusions

• Ensilage of forage crops is the preferred option for production of high quality forage for dry season feeding.
• Intercropping or mixing forage cereal crops with forage legumes or mixing them with forage tree legumes produces a high yield of forage from a small area of land, sufficient to supplement two cows for the dry season in a normal year and up to two months in a drought year in the semi-arid areas.
• Ensilage can be successfully carried out with low-cost small scale technology.
• It is feasible to achieve good productivity in livestock with dry season feeding of conserved high quality forage.

Acknowledgements

The author and collaborators of the studies in producing and conserving forages for small holder livestock owners in the semi-arid areas of Zimbabwe wish to thank the following for their contribution to the research:

• Department for International Development, U.K.-research funding.
• German-Israeli Fund for International Development-research funding.
• Matopos Research Station, Department of Research and Specialist Services-facilities and staff.
• Dairy cooperative of Gulathi communal farming area-participatory research

References

Ashbell, G., Kipnis T., Weinberg, Z.G., Hen, Y., Azriell, A., Katler, S. and Titterton, M. 1999. Developing a technology for ensiling forage crops for small holder cattle owners in developing countries. In: Proc. XIIth Int. Silage Conference on Silage Production in Relation to Animal Performance, Animal Health, Meat and Milk Quality. Uppsala, Sweden. July 5, 1999.

Bareeba, F.B. 1992. Forage conservation and utilization. Nomadic Peoples 31: 91–96.

Bareeba, F.B. 1977. The ensilage characteristics and nutritive value of maize, amaranthus-enriched maize and sorghum silages preserved with either molasses or formaldehyde. M.Sc (Agric) Thesis, Makerere University, Uganda.

Cameron, D.G. 1988. Tropical and subtropical pasture legumes - lablab bean. Queensland Agricultural Journal. 114 (2): 110–113.

Cole, C.A., Kaiser A G, Piltz J W and Harden S. 1996. An evaluation of sorghums for silage production in Northern New South Wales. In: Proceedings 3^rd Australian Sorghum Conference. Australian Institute of Agricultural Science Melbourne Occasional Publications. No. 93. Eds: M.A. Foale, R.G. Henzell and J.F. Kneipp. Tamworth, Australia. Feb. 1996.

Cowan, R.T., Moss R J and Kerr D V. 1993. Northern diary feedbase 2001. 2. Summer feeding systems. Tropical Grasslands. 27: 150–161

Figueiredo de, M. 1994. Silage making and its utilization. Occasional Publications by Cedara Agricultural Development Institute. Pietermaritzburg, South Africa p1.

Havilah, E.J. and Kaiser A.G. 1992. Sorghums for silage: a review. Proceedings 2^nd Australian Sorghum Conference. Feb. 4^th–6^th. Gatton, pp 338–354.

Humphreys, L.R. 1980. A guide to better pastures for the tropics and sub-tropics. Wright Stevenson and Company (Australia) Pvt. Ltd. Pp 27–28.

Kaiser, A.G., Havilah, E.J., Chopping, G.D. and Walker, R.G. 1993. Northern dairy feedbase 2001. 4. Feeding systems during winter and spring. Tropical Grasslanda. 27: 180–211.

Maasdorp, B.V. and Titterton M. 1997. Nutritional improvement of maize silage for dairying: mixed crop silages from sole and intercropped legumes and a long-season variety of maize. 1. Biomass yield and nutritive value. Journal of Animal Feed Science and Technology. 69: 241–261.

Maasdorp, B.V. and Titterton, M. 1999. Ensilage of harvested leaf from legume forage trees with maize. In: (Titterton et al., Forage Production and Conservation for Dry Season Feeding of Dairy Cows in the Semi-Arid Regions of Zimbabwe. In: Proceedings DFID Workshop on Livestock Production in the Semi-Arid Regions of Zimbabwe. Ed: T. Smith, Matopos, Zimbabwe. Feb. 1999 In Publ.

Maclaurin, A.R. and Wood A M. 1987. Veld and pasture management. In: Dairy Handbook Ed. J. Oliver. National Association of Dairy Farmers. pp 79–94.

Mahanna, W.C. 1991. When the cows don't like your silage. Hoards Dairyman. Oct 10. Pp 744–775.

Mahanna, W.C. 1994. Proper management assures high quality silage, grains. Feedstuffs. Jan. 10^th. 1994.

McDonald, P., Henderson, A.R. and Herson, S.J.E. 1991. In the Biochemistry of Silage. 2^nd ed. Chalcombe Publications. Marlow, U.K.

Maroske, M., Lambert, S., Haddow, B., Roth, A., Itzstein, R. and Chambers, G. 1997. In: Forage Agronomy Notes for the QDPI Crop Management Seminars. Ed: S. Gibson, 30^th September – 14 October, 1997. pp 26–28.

Mhere, O. 1997. The effect on hay quality of drying pennisetum on racks. Unpubl.

Mhere, O. 1999. Intercropping forage and legume crops. In (Titterton et al.) Forage Production and Conservation for Dry Season Feeding of Dairy Cows in the Semi-Arid Regions of Zimbabwe. In: Proceedings DFID Workshop on Livestock Production in the Semi-Arid Regions of Zimbabwe. Ed: T. Smith. Matopos, Zimbabwe. Feb. 1999 In Publ.

Musimwa, E. 1999. Personal communication. Dairy Development Programme field day. February, 1999.

Nobbs, C.R. and Oliver, J. 1987. Crops for winter feeding. In: Dairy Handbook. Ed. J. Oliver National Association of Dairy Farmers. pp 95–107

Panditharane, S., Allen, V G, Fontenot, J.P. and Jayasuriya, M.C. 1986. Ensiling characteristics of tropical grasses as influenced by stage of growth, additives and chopping length. Journal Animal Science. 63 (1):197–207.

Rekib, A. and Singh, A.P. 1988. Conservation of forages. In: Pasture and Forage Crops Research: Third International Rangeland Congress. Ed: P.Singh, New Delhi. Pp 379–389.

Skerman, P.J. 1988. Tropical forage legumes. In 2^nd ed. Plant Production and Protection Series, No. 2. FAO.

Stuart, P., Slatter, J., Minson, D., Gibson, S. and Whan, I. 1993. In: The Forage Book. Ed: P.Stuart Pacific Seeds. Toowoomba. P8

Tjandraatmadja, M., Norton, B.W and MacRae, I.C. 1994. Ensilage of tropical grasses mixed with legumes and molasses. World Journal of Microbiology and Biochemistry. 10:82–87.

Titterton, M. 1996. Silage making. In: Ed: C. Levy, Grain Handbook. Zimbabwe Grain Producers Association. Harare. M12–M16 pp

Walker, R.G., Knight, R.I., Lehman, B. and Cowan, R.T. 1992. Pasture based systems for milk production in the tropics. In: Proceedings of the Australian Society for Animal Production. 19:118