Previous Page Table of Contents Next Page

Potential for by-product feeding in tropical areas

by P.B. O'Donovan

Efficient livestock production depends a great deal on sufficient feed of the right quality being available throughout the year. During the rainy season in tropical areas there is usually enough pasture for feeding ruminant livestock; the success or failure of livestock enterprises depends on feed supplies being available during the dry season when little or no grass grows.

Large quantities of various by-products are usually available in the topics during the dry season. In some areas they are discarded or only partially used, and valuable sources of potential feed are lost. In others there may be a total dependence on by-products for dry-season feeding; if well planned this system can be satisfactory, but it is usually difficult to maintain a continuous supply for the entire period. In still other areas, by-products may be used in addition to feed conserved as silage and hay; if some herbage growth occurs animals may be grazed for a limited member of hours a day and their additional requirements met with conserved feed or by-products. Where there is little or no herbage growth, feeding systems may be based on conserved feed or by-products, or a combination of the two. Local conditions will determine which system is the most practical and economic. Some of the more important byproduct feeds available in tropical areas are discussed in this article.

P.B. O'Donovan is an FAO Animal Nutrition Officer at the Facultad de Agronomia, Universidad de la República, Montevideo, Uruguay.

Sugarcane is widely grown and is an important source of income for many countries. The processing of sugarcane yields a number of important by-products for animal feeding.

Sugarcane by-products


Molasses is a valuable source of energy, and is often either underutilized or exported at a low price. When used for livestock feeding, frequently much less than its maximum value is obtained because it is fed with unsuitable ingredients and in unbalanced diets. It has long been known that the inclusion of a small percentage of molasses improves the palatability of the diet and it is particularly valuable when incorporated with coarse unpalatable roughages. Lofgreen and Otagaki (1960a, 1960b) found less net energy in mixed diets containing more than 10 to 15 percent molasses, although this varied with total feed ration composition. Hatch and Beeson (1972) reported that replacing 5 percent rolled maize with molasses in rations had no apparent effect, but 10 and 15 percent replacement increased nitrogen retention; both energy and dry-matter digestibility and butyric acid in the rumen were significantly increased by the higher molasses percentages. Because of the large quantities of cane molasses produced in tropical areas, it is important to explore the possibility of using higher than conventional levels. Reports indicate (Preston et al., 1967; Elias et al., 1968) that molasses may comprise up to 80 percent of the metabolizable energy for beef cattle under specific feeding conditions. In these studies the fresh forage allowance was restricted and fish meal supplied the additional protein. The animals had free access to molasses/urea in troughs and daily liveweight gains ranged from about 700 to 900 g.

O'Donovan and Chen (1972) fed dairy heifers at different stages of growth on diets in which cane molasses accounted for 25, 33 and 45 percent of the total weight of feed (Table 1). The animals gained weight satisfactorily on the first two levels, but the 45 percent molasses diet, which contained only 10 percent soybean meal, significantly depressed rate of gain. Results indicated that growing heifers could initially be fed 25 percent molasses and the level could later be increased to 33 percent. In the latter diet two by-products, rice straw and molasses, accounted for 68 percent of the total weight of feed. Liveweight gain can be regulated by increasing or decreasing the percentage of straw fed.


Conventional bagasse contains two fractions: an outer portion called the rind and the finer inner part known as the pith. Millions of tons of bagasse are produced in sugarcane factories, and a significant percentage is burned as fuel; thus a potential source of animal feed is not fully exploited.

Bagasse per se is a low-quality feed, principally because of its high percentage of lignocellulose. When mixed with protein and a source of energy such as cane molasses, moderate levels of bagasse can promote satisfactory liveweight gains. Levels higher than 30 percent by weight tend to depress rate of gain.

Various by-products suitable for livestock feed are usually available in large quantities in the tropics during the dry season — in some areas these are discarded or only partially used

During the sugarcane harvest the leaves are traditionally fed to draught animals. Here, buffaloes are taking a midday break from helping with the harvest and are feeding on the leaves. However, the use of this by-product can extend well beyond such traditional practice.

The treatment of high-fibre materials (cereal straws, bagasse and wood pulp) with sodium hydroxide or steam significantly increases digestibility. These possibilities have been discussed by Pigden (1971). Even if chemical treatment does not become a commercial reality, much can be achieved by feeding untreated bagasse at optimum levels in mixed rations. Since bagasse and molasses are usually available in combination, there is a great potential for feeding these in the correct ratio with supplementary protein and nonprotein nitrogen. Where less than maximum gains are required, by-product mixtures are useful in maintaining livestock or promoting a low rate of gain among beef animals during the dry season.

Sugarcane leaves

At harvest, sugarcane leaves are burned or otherwise separated from the cane. If they are not burned they may be collected and made into bundles to facilitate transport from the field. To maintain a high feeding value, they should preferably be collected on the day of harvest when they are still green. The feasibility and economics of collecting the leaves vary from one region to another. Where labour costs are low, they provide a cheap source of feed. The leaves should be chopped to facilitate feeding and to avoid rejection of the fibrous parts.

O'Donovan (1970) fed chopped sugarcane leaves, containing 5–6 percent crude protein, to a herd of commercial dairy cattle and to beef cattle. This feed was found to be sufficient to meet the maintenance requirements of dairy cattle and to provide for the production of about 2 kg of milk per cow — a production level which could scarcely be realized with average quality pangola grass silage. With beef cattle, maintenance requirements were met and an average daily liveweight gain of 0.25 kg was obtained. It was found necessary to feed more protein and energy in order to achieve higher rates of gain.

If mechanized harvesting becomes more widespread, the bulk collection of the leaves may be possible. There is evidence that satisfactory silage can be made from the chopped leaves; this is a convenient method of conserving the surplus for subsequent feeding.

Sugar factory residue

In certain regions there is a residue of sugar extraction consisting of a mixture of the first froth of boiled cane juice and fine bagasse or pith, mixed in a ratio of about 5 to 1. Its approximate composition, on a dry-matter basis, is 11–12 percent protein, 26–35 percent fibre, 1.4 percent Ca and 1 percent P. This residue, called cachaza in Cuba, is used mainly as organic matter for soils. Its rather high phosphorus content could be an advantage if it is fed to cattle when herbage P contents are low. This was tested in Cuba, where the fresh residue was incorported in mixtures (with molasses) and fed to beef cattle on pasture for part of the dry season. The mixtures were consumed without difficulty and the inclusion of the residue did not significantly depress the rate of liveweight gain. Animals fed the mixtures had higher levels of serum phosphorus.

Because the residue has a low dry-matter content (about 25 percent), moulds grow rapidly on it in hot humid conditions. Economic considerations are likely to rule out drying, so that it is necessary to utilize the material in fresh form. It must therefore be fed in close proximity to sugar factories. There is little information as yet concerning its real value as a livestock feed. The extent to which it will be fed to livestock will also depend on alternative sources of better quality feed.

Rice straw

The value of rice straw as a roughage feed for ruminants has never been fully exploited, although traditionally it has been used for feeding buffaloes and zebu cattle in many areas. Its true potential has not been realized for two main reasons: (a) only a fraction of the total rice straw produced is fed to animals, the remainder being used for the manufacture of paper or ploughed into the soil or burned; (b) the rice straw used for feeding is offered with little or no supplements (energy, protein and minerals) and this results in very poor utilization.

Rice straw is nutritionally comparable to barley and wheat straws, and in some cases may surpass these. Its nutritive value is influenced greatly by the stage of harvesting; the protein content is higher if it is harvested when it still retains some of its green colour, when it compares favourably with poor to average hay made from tropical pasture species. Straws are consumed in small amounts when they constitute the only feed. There is ample scope for feeding rice straw in tropical areas, and its potential is enhanced by the availability of cane molasses as a readily available source of energy. A starchy feed ingredient and protein are also needed. The latter may be fed in part as true protein and in part as urea.

O'Donovan and Chen (1972) fed growing dairy heifers with rations in which chopped rice straw comprised 25 and 35 percent of the total mixture, the remainder consisting of cane molasses, sweet-potato chips, soybean meal and urea (Table 1). Daily liveweight gains ranged from 460 to 820 g, the former level reflecting a low soybean content in the ration and the depressing effect of 45 percent cane molasses. Satisfactory gains were obtained when all the feed ingredients, except soybean meal, were home grown and about two thirds of the entire ration consisted of rice straw and molasses.

Pineapple bran (pulp)

Pineapple bran (more aptly described as pulp) consists of the skin and often the core of the pineapple, and accounts for an estimated 40–50 percent of the total pineapple weight. Because of its rather high fibre content, the bran is more suitable for ruminants than for monogastric animals. The net energy values of pineapple bran and pineapple hay are reported to be 118.8 and 85.8 Mcal/100 kg (Otogaki et al., 1961). Fresh pineapple bran usually contains only about 10 percent dry matter, and because of this it is more convenient to feed it in the vicinity of the factory during the canning season. Long-distance transport is both difficult and expensive. However, if pineapple bran is needed elsewhere for feeding, it is possible to employ such techniques as drying (either in the sun or using a conventional dryer) and conserving as silage, in which case the fresh bran is mixed with other ingredients with higher dry-matter content.

O'Donovan, Chen and Lee (1972) investigated a number of pineapple bran silage mixtures (Table 2). Those incorporating molasses or molasses and a source of starch (maize and sweet potatoes) fermented satisfactorily, and it was possible to formulate a wide variety of suitable mixtures, bearing in mind the importance of silage-making and fermentation principles. Partial or complete sun drying is possible, but it is hazardous during the rainy season.

Citrus pulp

Citrus pulp is a by-product of citrus canning, and it contains about 25 percent dry matter. Many of the considerations relating to pineapple bran apply also to citrus pulp. Surplus quantities of pulp may be conserved as silage when mixed with other ingredients such as molasses, urea and bagasse. A variety of these ingredients may be used, the aim being an end-product sufficiently high in dry matter; molasses supplies readily available energy for fermentation. There should be little difficulty in formulating a number of suitable mixtures for animal feeding.

Citrus pulp can also be sun-dried or drum-dried, permitting storage for subsequent feeding. The latter method, however, is expensive when large quantities of liquid have to be removed and may increase the cost beyond that of other more valuable ingredients. The aim should be to utilize the material without undue processing costs. For this reason the use of fresh or ensiled material seems to offer the best possibilities. Citrus pulp has an important future in livestock feeding, as evidenced by a number of experiments. A review of experimental work at the University of Florida (Chapman et al., 1972) indicated that dried citrus pulp can comprise up to 40 percent of the concentrate ration with excellent results. This is supported by the findings of Bhattacharya and Harb (1973) in studies with Awassi lambs; the digestion coeffcient for energy was highest when citrus pulp was incorporated at a level of 40 percent in mixed diets. Satisfactory gains were obtained with fattening cattle when 70 percent of the concentrate portion of the diet consisted of citrus pulp (Carnevali et al., 1972). A feeding trial was conducted in Cuba (unpublished data) where growing steers were fed 3 kg per day of one of two concentrate supplements; where 25 percent cane molasses was replaced by dried citrus pulp, significantly increased daily gains resulted.

There appears to be little doubt as to the value of citrus pulp in ruminant livestock rations. Finding the most practical and economic means of utilizing it in areas where there are large supplies deserves priority.

Sweet potato haulms

Sweet potatoes are an important crop in many areas of the tropics and are a source of animal feed as well as human food. The haulms have not received deserved attention as a feed for ruminants, although they are fed in some countries to buffaloes and zebu cattle. They are available during the dry season when feed supply is usually critical.

Sweet potato haulms were fed to dairy cows with good results by O'Donovan (1970). They invariably increased milk yields when succeeding a diet of sugarcane leaves or silage, principally because of their higher protein content. The haulms were consumed avidly and there were no noticeable ill effects. They may supply part of the feed for dairy and beef cattle, especially in countries where labour costs for collection and transport are not too high.

Above: High-grade Holstein heifers receive a diet in which rice straw and molassestwo widely available by-products in tropical areas — constitute more than two thirds of the total.
Left: Small quantities of pineaple bran silage mixtures were evaluated in a pilot test. The darker colour of the mixture on the right indicates better preservation.
Below: Santa Gertrudis crossbred cattle. which performed satisfactorily on a diet that included 70 percent dry pineapple bran, are shown in the early stages of the test.

Use of urea in by-product feeding

Lack of protein is often the most important limiting factor in the feeding of by-products. Nonprotein nitrogen in the form of urea is now available in increasing quantities in many developing countries, and could help overcome this limitation.

Research in developed countries provides information on conditions under which urea may be utilized for ruminant feeding. Work by Virtanen (1966) suggested that urea could supply all the dietary nitrogen insemipurified diets for cows and could at the same time support high milk yields. Later work (Virtanen and Ettala, 1969) showed that higher yields resulted when some true protein was present. Recommendations from the United States are that urea may replace a maximum of one third of the dietary nitrogen. Comparisons indicate that the response from urea (in terms of daily liveweight gains and milk production) has approximated but rarely exceeded that from soybean meal.

Research in the arid areas of the world has shown a variable response to urea when the aim is to increase the nitrogen intake of animals consuming low-protein roughages (FAO, 1971). A review of the subject by Loosli and Mc Donald (1968) illustrates the variability in results when experiments are conducted under widely different conditions. While urea is beneficial in alleviating a large weight loss, it has rarely promoted gains without the presence of some energy feed.

The application of urea feeding in the tropics is of considerable importance. Hsu (unpublished data) fed milking cows restricted amounts of green forage in addition to a mixture of cane molasses, sweet-potato chips and urea; the latter supplied 55–65 percent of the total nitrogen. A steady decline in milk yields was partially averted by replacing 2 kg of sweet potatoes in the ration with ground maize — a response which may be attributed to the additional protein. O'Donovan, Liang and Chen (1972) compared urea with soybean meal in concentrates for milking cows under zero-grazing conditions (Table 3). Urea supplied about 50 percent of the total dietary nitrogen. Although 8 percent less milk was produced, the results were encouraging. Economic considerations may take priority in some areas: it could be expedient to accept a small decrease in milk yield by using a cheap local source of urea nitrogen rather than import expensive protein.

Evidence is still lacking regarding the successful use of very high levels of urea for animal feeding under tropical conditions. Urea may supply a maximum of 50 percent (ideally somewhat less) of the total dietary nitrogen.

Table 1. Rations containing different percentages of cane molasses (rice straw as roughage) and daily liveweight gains when fed to growing heifers

Ingredients1Ration A fed to heifers weighing
132–182 kg
Ration B fed to heifers weighing
182–205 kg
Ration C fed to heifers weighing
205–310 kg
Rice straw, chopped253535
Cane molasses253345
Sweet-potato chips, dried301810
Soybean meal201410
 Grams per kilogram
Crude protein16.013.713.5
Liveweight gains:   
Test 1-0.58-
Test 20.710.820.46

SOURCE: O'Donovan and Chen, 1972.
1 All rations fortified with bone meal, salt and vitamin A.

Table 2. Ingredients of six pineapple bran silage mixtures

IngredientsSilage mixtures1
Pineapple bran, wet756575656582.5
Rice straw, chopped10201020207.5
Maize meal151555--
Cane molasses--1010155.0
Sweet-potato chips, dried-----5.0
Urea (additional) 
Percentage dry matter28.836.227.635.234.626.2

Source: O'Donovan, Chen and Lee, 1972.
1 Mixtures 1–5 in small plastic bags; mixture 6 in small silos.

Table 3. Composition of soybean meal and urea mixtures fed with roughages, and milk yields obtained

Ingredients1Soybean meal mixtureUrea mixture
Sweet-potato chips, dried2550
Maize, ground1525
Cane molasses2525
Soybean meal35-
Milk Production2  
Total yield (kg)2 7822 371
Average yield(kg/day)12.410.6
Average decrease(kg)3.24.1

Source: O'Donovan, Liang and Chen, 1972.
1 Bone meal and salt added.
2 From 28 to 224 days of lactation.

Role of by-products

Only a few of the many by-products that can be exploited in tropical areas for animal feeding have been discussed in this article. They should not be regarded as providing a complete solution to livestock feeding problems generally, but rather as filling a gap in meeting overall requirements. Although by-product mixtures can be employed in intensive animal production systems, they are used to best advantage during the dry season when feed is in short supply; thus, they can provide for maintenance and low levels of production until pasture starts growing again. Full utilization of by-products can only be realized when they are appropriately combined to form balanced diets.

In many tropical areas there is competition between livestock producers and industry for certain by-products. This competition may proceed to a point where industry can afford to pay a higher price than the livestock producer, who must calculate the costs according to the nutritive value of the material. An example is the utilization of rice straw for the manufacture of paper; the production of alcohol from molasses is yet another, while the lack of fuel in some areas may lead to the use of sugarcane bagasse for this purpose.

Despite this competition, by-product feeding can be of great importance in livestock systems where the problem is largely one of uneven feed distribution: a rainy season with much herbage growth and a dry season with little or no growth. Where conserved herbage (hay and silage) is not available, by-products are often the only source of feed in the dry season. For each area it is necessary to determine the months of the year when these become available and what mixtures may be employed. Conserving the surplus by drying, ensiling or other means ensures a continuous supply of fresh and conserved feed over a prolonged period. Judicious use should be made not only of by-products but of all other feed resources as well.

It is encouraging that there is a move toward the establishment of regional research stations in different parts of the world. This should lead to more coordinated research on common problems. Results obtained with byproduct mixtures would be similar within a given area, thus obviating the duplication of research effort. There is need to investigate further the feeding value of by-products and by-product mixtures, in particular those of sugarcane, rice, pineapple and citrus.


Bhattacharya, A.N. & Hard, M. 1973. Dried citrus pulp as a grain replacement for Awassi lambs. J. Anim. Sci., 36: 1175–1180.

Carnevali, A.A., Chico, C.F. & Verde, G. 1972. Utilización de altos niveles de pulpa de cítricos y de urea, raciones de engorde para bovinos. Agronomia trop., 22(3): 261–269.

Chapman, H.L. jr., Ammerman, C.B., Baker, F.S. jr., Heutges, J.F., Hayes, B.W. & Cunha, J.J. 1972. Citrus feeds for beef cattle. Gainesville, University of Florida. Bulletin 751.

Elias, A., Preston, T.R., Willis, M.B. & Sutherland, T.M. 1968. Intensive beef production from sugar cane. 4. Molasses/urea as a substitute for grain in low-fibre diets. Rev. cubana Cienc. agríc., 2: 55–63.

FAO. 1971. Report of an ad hoc Consultation on the Value of Non-Protein Nitrogen for Ruminants Consuming Poor Herbages, Kampala, Uganda, 29 June–3 July 1971.

Hatch, C.F. & Beeson, W.M. 1972. Effect of different levels of cane molasses on nitrogen and energy utilization in urea rations for steers. J. Anim. Sci., 35: 854–858.

Lofgreen, G.P. & Otagaki, K.K. 1960a. The net energy of black-strap molasses for lactating dairy cows. J. Dairy Sci., 43: 200.

Lofgreen, G.F & Otagaki, K.K. 1960b. The net energy of black-strap molasses for fattening steers as determined by a comparative slaughter technique. J. Anim. Sci., 19: 392–403.

Loosli, J.K. & McDonald, I.W. 1968. Nonprotein nitrogen in the nutrition of ruminants. Rome, FAO. FAO Agricultural Study No. 75. 94 p.

O'Donovan, P.B. 1970. Livestock production on marginal land in Taiwan. Rome, FAO.

O'Donovan, P.B. & Chen, M.C. 1972. Performance of dairy heifers fed different levels of cane molasses with rice straw as roughage. Trop. Agric., 49: 125–134.

O'Donovan, P.B. Chen, M.C. & Lee, P.K. 1972. Conservation methods and feeding value for ruminants of pineapple bran mixtures. Trop. Agric. 49: 135– 141.

O'Donovan, P.B., Liang, S.P. & Chen, M.C. 1972. Comparison of urea with soybean meal concentrates for milking cows. Trop. Agric., 49: 311–320.

Otagaki, K.K., Lofgreen, G.P., Cobb, E. & DULL, G.G. 1961. Net energy of pineapple bran and pineapple hay when fed to lactating dairy cows. J. Dairy Sci., 44: 491–497.

Pigden, W.J. 1971. Effect of physical and chemical processing and non-protein nitrogen (NPN) supplementation on utilization of lignocellulose. Proc. Ad Hoc. Consultation on the Value of Non-protein Nitrogen for Ruminants Consuming Poor Herbages, Kampala, Uganda, 29 June-3 July 1971, p. 114–131.

Preston, T.R., Willis, M.B. & Elias, A. 1967. Intensive beef production from sugar cane. 1. Different levels of urea in molasses given ad libitum to fattening bulls as supplement to a grain diet. Rev. cubana Cienc. agríc., 1: 33–40.

Virtanen, A.I. 1966. Milk production of cows on protein-free feed. Science, 153: 1603–1614.

Virtanen, A.I. & Ettala, T. 1969. On milk production of cows on feeds with different protein and urea levels. Suom. Kemistil, A42: 241–244.

Previous Page Top of Page Next Page