by J. Le Dividich, F. Geoffroy, I. Canope and M. Chenost
As world demand for milk and meat increases, competition in the use of crops for human food and for animal feeding is becoming more intense. World grain production, for example, has declined in recent years, and in consequence stocks have been depleted and prices have risen to such a high level that it is becoming uneconomic to continue feeding grain to livestock as generously as in the past. Thus there is more reason now than ever before, particularly in the tropical regions where animal protein for human consumption is scarce, to make use of every crop or by-product that could serve as animal feed.
J. Le Dividich is a researcher with the Swine Breeding Research Station at the National Centre for Zootechnical Research (CNRZ) of the National Institute for Agricultural Research (INRA) in Jouy-en-Josas, France.
F. Geoffroy and I. Canope are researchers at the INRA Zootechnical Research Station in Antilles-Guyane, Petit-Bourg, Guadeloupe.
M. Chenost is an Animal Nutrition Officer with the Animal Production and Health Division of FAO in Rome, Italy
While direct feeding of sugarcane to cattle is now well established (Leng and Preston, 1976), the feeding of bananas, another important tropical crop in terms of volume and energy value, has been relatively neglected. This is largely because bananas are principally a human food, but is also partly attributable to the fact that their value as animal feed has not been adequately considered (Chenost et al., 1969; FAO, 1969).
It is relevant to note in this context that according to the Institut français de recherches fruitières d'Outre-Mer (IFAC), for every ton of bananas packed and exported to the fruit markets of the world, about 750 kg are rejected as being either unsuitable or in excess of requirements. This discarded fruit represents 10 to 20 percent of the total crop, the variation depending on how culling is done during packing (which depends on the market), and the amount of fruit that is rendered unusable by fortuitous circumstances such as gales and hurricanes.
Of the 36 million-ton world banana production (FAO, 1975), it may be estimated that 7 to 10 million tons per year (20 to 30 Percent) could be recovered as animal feed. But of this, a substantial quantity is either used in a haphazard way or lost to the animal industry.
The purpose of this article is to review what is currently known about the use of waste bananas as animal feed and to consider areas of study warranting further attention.
Composition and ensiling properties
The composition of all banana varieties harvested is determined chiefly by degree of ripeness. Table 1 shows the composition of the green and ripe fruit.
Bananas have a high water content (78 to 80 percent). In the green state in which they are generally picked and packed, the dry matter consists mainly of starch (72 percent), which on ripening changes into simple sugars (saccharose, glucose and fructose). The cellulose content is low (3 to 4 percent) and most of it is found in the skin. The inorganic fraction is poor in the more important minerals, Ca and P, but is rich in Potassium.
Whether green or ripe, the banana has a low protein content (N × 6.25) and is deficient in lysine and in sulphur-containing amino acids (2.3– 2.9 g/16 g N). Moreover, the banana contains tannins which are only slightly polymerized in the green fruit and which therefore inhibit the action of enzymes. In the ripe fruit, however, there is higher polymerization.
Because of its high fermentable sugar content, the banana is easy to ensile. On the basis of work done recently in Guadeloupe (Sève et al., 1972; Le Dividich and Geoffroy, 1973; and Le Dividich et al., 1976) it may be concluded that the ensiling of bananas can be successfully undertaken if the normal rules of silage making are observed, ensuring adequate compaction and drainage; no additives are required. Ripe bananas compact easily, but bananas to be ensiled green should be chopped. Alternatively, green bananas may be ripened either naturally or artificially in an acetylene atmosphere (a procedure developed by IFAC at its Guadeloupe Station) prior to ensiling. In silage made from green bananas the starch keeps well (with losses of only 6 to 7 percent), whereas 84 percent of the simple sugars of the ripe bananas are either degraded or disappear. As a result, weight losses (Table 1) are almost three times less for green banana silage (10 to 15 percent) than for silage from ripe bananas (30 to 35 percent).
The silage stabilizes in three to four days and keeps for at least six months. The characteristics of silage from ripe fruit (e.g. pH and lactic acid content) may be more favourable than those of silage from green bananas. But on balance green banana silage is to be preferred, particularly when sudden peaks occur in banana supplies owing to seasonal factors, market constraints and meteorological conditions.
Use in pig feeding
Fresh Bananas. The digestible energy (DE) value of the green banana (Table 2), measured with pigs of 40 kg average weight; is 3 181 kcal/kg of dry matter (i.e. 80 to 86 percent that of maize). This value may be improved by:
Alpine goats eating fresh green chopped bananas
Table 1. Composition of banana export rejects at different stages of maturity and preservation
|Green 3 to 5 days after picking||Ensiled green||Ripe||Ensiled ripe|
|Physical composition||Percentage of dry matter|
|Dry matter content in fresh feed||21.2||29.0||21.7||23.5|
|Sugars soluble in alcohol at 80° G.L.1||1.8||0||73.6||17.3|
|Lactic acid (g/% DM)||5.3||10.1|
|Volatile acidity (g CH3 COOH/100 g DM)||1.8||3.0|
|Losses as percentage of ensiled dry matter||13.5||33.9|
SOURCE: Le Dividich, Sève and Geoffroy, 1976.
1 On dry matter basis.
Table 2. Apparent digestibility and digestible energy content of fresh bananas for pigs
|Green raw||Whole cooked||Whole green peeled||Whole ripe|
|Per kg dry matter||3 181||3 439||3 440||3 327|
|Per kg fresh weight||674||743||1 021||722|
SOURCE: Le Dividich and Canope, 1975.
NOTE: Averages with the same index letter are not significantly different P < 0.05.
Adding bananas to a pig ration reduces the digestibility of the total nitrogen content of the ration (Le Dividich and Canope, 1970). This means that the digestibility of the protein in bananas (Table 2) is either negative (as in green bananas) or very low (as in ripe bananas).
Thus for pigs fresh bananas are a bulky feed of low energy value (about 700 kcal DE/kg, or 20 percent of the energy value of maize). They also depress the utilization of the nitrogen in the ration. When including them in pig diets it is therefore necessary to provide both energy and nitrogen supplementation.
Fresh green bananas may be fed chopped into thin slices or crushed, and when ripe may be offered whole. Some authors (Viteri et al., 1970; Le Dividich, [unpublished] and Shillingford, 1971) have found little influence of the banana's degree of ripeness on animal growth performance, while others stress the value of the less astringent ripe banana, which is more easily digested than the green fruit and contributes to better growth. Cooking the green fruit improves the growth rate of pigs by 10 percent and improves the nutritional efficiency of diets, but is limited by cost.
In Guadeloupe studies on growing pigs (27 to 54 kg live weight) raised in cages and fed four diéts based on bananas and soybean cake, with protein contents ranging from 12.5 to 20.9 percent, gave the following results (Le Dividich and Canope, 1975):
Increased protein content significantly improved growth performance and feed conversion efficiency.
Daily net nitrogen retention rose significantly (P<0.01) to a level corresponding to a 17.8 percent protein content in the diet; this resulted in the recommendation that for growing pigs, the diets should contain 18 percent crude protein. Because of the quantities of banana ingested, a daily supplement of 270 g protein is necessary. This is in accord with the results of Galles, Clavijo and Maner (1969) and Oliva, Gelleri and Maner (1970) for green and ripe bananas.
|(Above) Ensiling fresh green bananas in Guadeloupe. There have been significant improvements in the milk production and weight gains of ruminants fed bananas in compound feeds or as a supplement to green forages.|
|(Right) Crossbred Creole × Large White pigs eating fresh green chopped bananas. As fresh bananas are a bulky feed of low energy value and depress digestibility of the total nitrogen content of a pig ration, it is necessary to provide both energy and nitrogen supplementation.|
Table 3. Effect of level of concentrates in banana-based rations on the growth performance of pigs (25 to 95 kg live weight)
|Daily feed intake|
|Concentrate for pigs weighing 28 to 60 kg|
|Digestible energy||kcal||2 952||3 996||4 995|
|Concentrate for pigs weighing 60 to 95 kg|
|Digestible energy||kcal||3 646||4 690||5 689|
|Total dry matter intake||a1.90||b2.04||c2.16|
|Average daily weight gain||g||a515||b589||c624|
|Feed conversion ratio|
(kg dry matter/kg weight gain)
|Weight of digestive tract, full||kg||a11.4||ab10.2||b9.5|
|Average backfat thickness||mm||a23.0||a24.8||b28.3|
SOURCE: Le Dividich and Canope [unpublished data].
NOTE: Averages with the same index letter are not significantly different p< 0.05.
Adding fresh bananas (which are high in water content and low in energy) to pig diets results in an energy deficit in the ration that may be overcome in three ways:
In practice, the choice of feeding systems or level of supplementation is determined by economic considerations, i.e. the price of the feed supplement in relation to the added value of the carcass.
Ensiled and dehydrated bananas. The work done in Guadeloupe shows that banana silage is highly acceptable to the growing and fattening pig (Table 4). Despite its depressing effect on the digestibility of the ration protein, green banana silage has nearly the same food value as the ripe fruit. On the other hand, ripe banana silage makes for poor growth performance because of its higher cellulose content. Despite the poor digestibility of the protein in silage-based diets, the carcass is of good quality even if animals are slaughtered at weights of 100 kg or more (Sève, Le Dividich and Canope, 1976).
Fresh or ensiled green bananas can constitute a basic feed for the gestating sow (Le Dividich and Canope, 1975). Farrowing performance (number and average weight of piglets in the litter) is not affected by the inclusion of bananas in the ration. On the other hand, even with a generous supplement (1.0 kg of molasses + 1.5 kg of a feed containing 40 percent protein), bananas do not cover all the energy requirements of the lactating sow. The 5-to 9-kg weight losses that have been recorded indicate that the use of bananas as feed for lactating sows is not advisable. These conclusions agree with those of Clavijo and Maner (1969).
Table 4. Feeding value of diets based on fresh green bananas and banana silage for growing and finishing pigs (30 to 90 kg live weight)
|Composition of feeds offered|
|Fresh green bananas||ad lib.||ad lib.||ad lib.||-|
|Control diet1||-||-||-||ad lib.|
|Raw cane sugar for pigs weighing 30 to 50 kg||g/day||400||400||400||-|
|Concentrate2 for pigs weighing 30 to 50 kg||g/day||600||600||600||-|
|Concentrate2 for pigs weighing 50 to 95 kg||g/day||900||900||900||-|
|Total dry matter intake||1.80||1.88||1.80||2.42|
|Average daily weight gain||g||a432||a434||a400||b631|
|Feed conversion ratio|
(kg dry matter/kg gain)
|Average backfat thickness||mm||a23.6||a25.8||a24.8||b30.6|
SOURCE: Le Dividich and Canope, 1975.
NOTE: Averages with the same index letter are not significantly different p< 0.05.
1 Composition: ground barley 70%; soybean meal (44% C.P.) 20%; alfalfa meal 5%; mineral and vitamin premix 5%.
2 Composition: soybean meal (44% C.P.) 90%, minerals + vitamins 10%.
Banana pulp flour (72.3 percent starch, 1.2 percent crude fibre and 4.5 percent crude protein) was used successfully in weaning feeds for piglets at 5 weeks of age (Le Dividich and Canope, 1974), and even at 3 weeks (Le Dividich [unpublished]). When it comprised 50 percent of the weaning feed, growth performance was comparable to that obtained with cassava flour. It also facilitated keeping the animals in excellent health (as noted by the lack of diarrhoea), which is an important consideration in the choice of weaning feeds.
Thus, the use of fresh and ensiled bananas in pig feeding contributes to considerable savings in concentrate feeds. Such savings may amount to about 50 percent per pig slaughtered at 95 kg live weight (with a good lean carcass) and 75 percent per gestation (without affecting farrowing performance). However, fresh bananas should not be used in early weaning feeds because of the piglet's poor capacity to digest starch from this source. On the other hand, flour made from green bananas is well tolerated by piglets when they weigh 5 kg or more.
Use in ruminant feeding
Only a limited amount of work has been done on the use of bananas for ruminant feeding.
Banana pulp flour. Bananas have hitherto been used after dehydration as a source of starch for the preparation of calf feeds. The flour is used in the manufacture of milk replacers in substitution for lactose. However, its use is restricted to particular situations. It is known that preruminant calves can be fed to high slaughter weights if the lactose and some of the fats are replaced by starches; the substitution promotes protein synthesis (Thivend et al., 1972). This would suggest that at a time when the use of cereals in animal feeds is being increasingly questioned, banana flour should find special favour as an ingredient in calf feeds.
Some work has been done, notably in Ecuador (Spiro, 1973; Rihs et al., 1975), on the introduction of banana flour into ruminant diets. Spiro demonstrated that green banana flour could replace 50 percent of the cereal in feeds for young growing and finishing cattle without changing their intake and growth rates.
Fresh and ensiled bananas. The conversion of bananas into flour is a costly process and may be justified only for preruminant calf feeding for the production of veal, which represents a luxury commodity. For milk production or conventional meat production, however, it is preferable to use bananas with a minimum of processing.
Table 5. Influence of different kinds of energy and nitrogen supplementation on feed intake and milk production in dairy goats
|Energy source||Cereal||Green banana||Ensiled banana|
cake + urea
cake + urea
cake + urea
|Level of urea (N as % total N)||(0)||(32.2)||(0)||(28.9)||(0)||(28.5)|
|Dry matter intake||g/day|
|Pangola (regrowth of 50 days)||679||686||176||321||319||330|
|Bananas||-||-||1 330||1 278||1 099||904|
|Total dry matter intake||a1 507||b1 435||c1 954||d1 788||d1 822||b1 394|
|Dry matter intake per 100 kg live weight||kg||3.14||3.34||4.17||3.26||3.67||3.24|
SOURCE: Geoffroy [unpublished data].
NOTE: Averages with the same index letter are not significantly different p < 0.05.
It was in this context that the work carried out in Guadeloupe, and more recently in Costa Rica, was geared to the use of bananas either fresh or ensiled. In these studies two approaches were followed, one involving the use of bananas as an energy supplement to green forages, and the other using bananas as a basic ingredient of compound feeds that were balanced with fibre, protein, mineral and, where necessary, energy supplements.
1. Use as a supplement to green forages. The first series of digestibility trials conducted with goats in cages (Chenost et al., 1971; Geoffroy and Chenost, 1973; Chenost et al., 1976) provided the following results:
The second series of trials conducted in Guadeloupe (Geoffroy, [unpublished]) with lactating goats demonstrated the feasibility of substituting green and ensiled bananas for all the cereals in concentrate feed. The principal results were as follows (Table 5):
Total substitution of the cereals with bananas resulted in an increase in dry matter intake.
Milk production was significantly higher in animals fed green bananas (whether fresh or ensiled) than in those fed cereals.
The butterfat content remained low and was not affected by the diet.
Weight gains were significantly higher in animals fed fresh or ensiled green bananas than in those receiving cereals.
The third series of trials with growing goats (Table 6), also conducted in Guadeloupe, showed that half of the conventional concentrates could be replaced with bananas if the ration were supplemented with urea to maintain its nitrogen level (Chenost et al., 1971); the non-protein nitrogen could constitute about 30 percent of the total nitrogen in the ration.
For both milk production and finishing purposes, bananas may be regarded as a good substitute for barley. This was demonstrated in Guadeloupe by Poncet (1973), who also observed that banana starch passed into the intestine faster than barley starch, that fermentation in the rumen was less intense and more regular, and that less volatile fatty acids were released with banana rations than with barley rations. The latter observation is consistent with the efficient utilization of urea in both banana-based and barley-based diets.
Table 6. Feed intake and performance of fattening kids receiving either Pangola grass supplemented with cereals or bananas, or complete ensiled ration
|Pangola (ad libitum) diet||Silage diet|
concentrate with 2.5% urea
|Average daily weight gain||g||112||130||155||123||142|
|Dry matter intake||g/day|
|Total dry matter intake||772||814||915||989||776|
|Feed conversion ratio||6.7||6.2||5.9||8.0||5.3|
SOURCE: Chenost and Geoffroy [unpublished data].
Cubillos (1974) also used bananas successfully in Costa Rica as a supplement to green forages. Feeding fresh bananas to steers that were raised on grass made it possible to increase stocking rate, the growth rate of the animals and, in consequence, the weight gain per hectare.
2. Use as a basic ingredient of complete feeds. In order to simplify feed mixing and feeding, a single feed obtained by ensiling all the constituents of the diet (bananas, bagasse, bran and urea) was offered in Guadeloupe to dairy goats and growing kids. The principal results were as follows (Geoffroy and Chenost, 1973; Geoffroy, [unpublished]):
The level of silage intake, which was relatively low (2.0 to 2.5 kg/100 kg live weight), increased by 30 to 40 percent when 5 percent molasses was added on a fresh weight basis.
The performance obtained with this type of feed was promising; milk production was maintained and the growth rate of fattening kids was better than that observed with ordinary cereal-based diets (Table 6).
The results obtained from these studies and applied research suggest that the work begun should be continued in depth. In particular, it would be useful to determine the way in which the different banana preparations affect the digestive and metabolic processes of pigs and ruminants. On the basis of the results obtained so far, bananas may be considered a good energy source for pigs, while in ruminants they are potentially a good substrate for microbial protein synthesis from nonprotein nitrogen. This also appears to be confirmed by the work of Herrera and Ruiz (1976), where the substitution of bananas for 50 percent of the molasses in the supplementary energy provided by sugarcane to finishing bull calves resulted in accelerated growth rates and an improvement in feed conversion efficiency.
More generally, in the humid tropical regions it would be interesting to explore the prospects of using local products or by-products in the compounding of complete rations in which the energy component could, for example, be made up chiefly of bananas, sugarcane and molasses, and the protein component could consist of urea and by-products from the cereal milling industry. Sugarcane could be included in proportions that would vary with socio-economic conditions. Adding bananas to sugarcane-based silage could overcome the problems of feed intake and the unfavourable ratios of volatile fatty acids for glucogenesis that were observed by Leng and Preston (1976).
Surplus and discarded bananas are a potential feed resource of great quantitative and qualitative interest in both pig and cattle production. In practice they can be used to replace the entire cereal component of pig and dairy cattle feeds and probably almost all of the cereals in cattle finishing feeds.
The necessary techniques for feeding bananas to pigs have already been developed; it is only necessary to disseminate them more widely and adapt them to varying local conditions. However, some further work on feeding of bananas to cattle is still necessary. In either case, it should be recognized that because of the difficulty of transporting fresh bananas and the cost of dehydrating them, it would be economical to use them as animal feed only in the vicinity of places where they are grown. This limitation is likely to weaken the possible geographical impact of banana feeding. However, the geographical restriction of use to the producing areas is advantageous in that trade in bananas suitable for livestock feeding would be restricted and would not lead to speculative commercial ventures.
Effective local utilization of bananas in livestock feeds would involve the setting up of rearing and/or finishing establishments in the vicinity of banana-packing facilities; this will have the highly beneficial effect of contributing to greater specialization in livestock breeding and finishing operations.
Beroard-Cerning, J. & Le Dividich, J. 1976. Valeur alimentaire de quelques produits amylacés d'origine tropicale. Etude in vitro et in vivo de la patate douce, de l'igname, du malanga, du fruit à pain et de la banane. Annls Zootech., 25. (In press)
Chenost, M., Le Dividich, J., Candau, M. & Cabanis, J.F. 1969. Perspectives d'utilisation de la banane et de la patate douce en alimentation animale. Proceedings 7th Annual Meeting, Caribbean Food Crops Society, p. 43–49.
Chenost, M., Candau, M., Geoffroy, F. & Bousquet, P. 1971. Utilisation de la banane et de l'urée dans l'alimentation des caprins en zone tropicale humide. Document, Tenth International Congress on Animal Production, Versailles.
Chenost, M., Candau, M., Geoffroy, F. & Bousquet, P. 1976. Possibilities of using bananas for the feeding of ruminants in humid tropical regions. J. Agric. Univ. P. Rico, 43. (In press)
Clavijo, H. & Maner, J.H. 1969. Dieta de lactancia normal comparada con dieta a 40% banano maduro con casava en alimentación de cerdas adultas en lactencia. Quito, Ecuador, Instituto Nacional de Investigaciones Agropecuarias. Experimentos SDGP 2.3.4; 2.3.7; 2.3.8.
Cubillos, G. 1974. Sistemas intensivos de engorde en pastos. Seminario America Tropical: Potencial para Incrementar la Producción de Carne, Cali, Colombia, p. 131–147.
FAO. 1975. Production yearbook 1974. Rome.
FAO. Committee on Commodity Problems. Study Group on Bananas. 1969. Developments in the field of processing and the use of bananas in livestock feeding. Document, Third Session, Study Group on Bananas, Panama City, April 1969.
Galles, A., Clavijo, H. & Maner, J.H. 1969. Banana maduro y suplementos proteinicos a voluntad en engorde y acabado de cerdos. Quito, Ecuador, Instituto Nacional de Investigaciones Agropecuarias. Experimento SDGP 2.2.5.
Geoffroy, F. & Chenost, M. 1973. Utilisation des déchets de banana par les ruminants en zone tropicale humide. Bull. tech. Prod. Anim., (2.3), 67–75. (Mimeographed)
Hernandez, J. & Maner, M.H. 1967. Evaluación del banano verde, maduro y verde cocido en crecimiento y engorde de cerdos. Quito, Ecuador, Instituto Nacional de Investigaciones Agropecuarias. Experimento SDGP 2.2.3.
Herrera, A. & Ruiz, M.E. 1976. Fattening cattle with by-products of sugar cane. 3. Substitution of final molasses by bananas. Trop. Anim. Prod., 1(1): 55.
Leng, R.A. & Preston, T.R. 1976. Sugarcane for cattle production: present constraints, perspectives and research priorities. Trop. Anim. Prod., 1(1): 1–21.
Le Dividich, J. & Canope, I. 1970. Utilisation des déchets de banane dans l'alimentation du porc en croissance. Premiers résultats. Journées de la recherche porcine en France, 2: 131–137.
Le Dividich, J. & Geoffroy, F. 1973. Conservation de la banane. Bull. tech. Prod. Anim., (1). (Mimeographed)
Le Dividich, J. & Canope, I. 1974. Valeur alimentaire de la farine de banane et de manioc dans les régimes de sevrage du porcelet à 5 semaines; influence du taux de protéines de la ration. Annls Zootech., 23: 161–169.
Le Dividich, J. & Canope, I. 1975. Chemical composition, energy value and utilization of banana in swine feeding in tropical areas. Document, Seminar on the Utilization of Local Ingredients in Animal Feedingstuffs, Kingston, Jamaica, April 1975.
Le Dividich, J., Sève, B. & Geoffroy, F. 1976. Préparation et utilisation de l'ensilage de banana en alimentation animale. I. Technologie, composition chimique et bilan des matières nutritives. Annls Zootech., 25. (In press)
Oliva, F., Gelleri, H. & Maner, J.H. 1970. Evaluación de la harina de banano verde con casava en crecimiento y acabado de cerdos confunados. Quito, Ecuador, Instituto Nacional de Investigaciones Agropecuarias. Experimento SDGP 2.2.12.
Oliva, F., Gelleri, H. & Maner, J.H. 1971. Empleo del banano verde y maduro con suplementos proteinicos de 20 y 30% en crecimiento y engorde de cerdos confinados. Quito, Ecuador, Instituto Nacional de Investigaciones Agropecuarias. Experimento SDGP 71-1-1-14.
Poncet, C. 1973. Utilisation digestive comparée de l'orge, de la banane verte et de la banane ensilée chez la chèvre. Journées d'études sur la physiologie et la biochimie de la digestion, Marseille, 10–11 mai 1973. Annls Biol. anim. Biochim. Biophys., 13: 776–777.
Rihs, T. et al., 1975. Harina de banano un substituto del maiz en la alimentación del ganado lechero. Quito, Ecuador, Instituto Nacional de Investigaciones Agropecuarias. Boletín Divulgativo 75.
Sève, B., Le Dividich, J. & Canope, I. 1972. Utilisation des déchets de banane “Poyo” par le porc en croissance aux Antilles françaises. Journées de la recherche porcine en France.
Sève, B., Le Dividich, J. & Canope, I. 1976. Préparation et utilisation de l'ensilage de banane en alimentation animale. II. Incorporation dans la ration du porc en croissance-finition. Annls Zootech., 25. (In press)
Shillingford, J.D. 1971. The economics of feeding bananas and coconut meal for pork production in Dominica, West Indies. Trop. Agric., Trin., 48: 103–111.
Spiro, J.T. 1973. De l'utilisation de la farine de bananes vertes dans l'alimentation du bétail. Equateur, 4. (Mimeographed)
Thivend, P., Vermorel, M. & Toullec, R. 1972. Utilisation des produits amylacés dans les aliments d'allaitement destinés au veau préruminant à l'engrais. Document, Ninth International Congress of Nutrition, Mexico.
Viteri, J., Alvarado, P. & Hervas, E. 1970. Comparación del valor nutritivo del banano verde V.S. banano maduro en iguales cantitades en alimentación de cerdos en crecimiento. Quito, Ecuador. Instituto Nacional de Investigaciones Agropecuarias. Experimento SCEP 126.96.36.199.