The physiological and biochemical basis for feeding pigs and poultry in the tropics (part 1)

The physiological and biochemical basis for feeding pigs and poultry in the tropics (part 1)

Swine Research Institute, Carretera del Guatao km.1, Punta Brava, La Habana, Cuba

Review originally prepared for:
ANIMAL PRODUCTION IN DEVELOPING COUNTRIES. Training Course organized by the Royal Veterinary and Agricultural University Copenhagen, 3-7 April 1989

The sugar cane feeding model, although not perfect, seems to be a suitable strategy for non-ruminant farm animals like pigs and poultry in the evergreen tropical environment. This holds true not only from the point of view of a small scale option, but also because it opens the possibility for developing an intensive pig and poultry production as a profitable new agro-industry, thus providing products of high biological value for human use.

The sui generis characteristics of this model have not previously been discussed so there are no points of reference. This challenge therefore makes generation of information necessary from the point of view of animal physiology and biochemistry. As has proved to be the case with ruminant animals fed on sugar cane derivatives, where new concepts were developed concerning the role of glucose and glucose precursors in ruminant metabolism (Preston and Leng 1987), so it may turn out that a non-conventional feeding strategy for monogastric animals built-up in the tropics, will lead to new knowledge of intermediary metabolism in these species.

Key words: Pigs, poultry, sugar cane, A, B and C molasses, metabolism, digestion, non-conventional feed resources, tropics.

In spite of the great demand for protein of high biological value for human consumption, pig and poultry production in the tropics is scanty and inefficient as a rule (Preston 1984). Although this paradox has been focused in different ways throughout the years, a satisfactory solution remains to be encountered. Meanwhile, the difference between the human needs for foods such as meat and eggs, and the possibility of covering them in this part of the world becomes even more significant.

Tropical countries have the advantage of a great potential of new non-conventional feed resources which can be made available throughout the year, and which can be processed in a way that makes them suitable for non-ruminant animals such as pigs and poultry. In fact the use of these new feed resources could avoid the serious competition between these species and humans, for cereal grains. In this connection, a possibility for establishing and even expanding this type of animal production might render many areas of the tropics self-sufficient in protein and carbohydrate foods, which is in marked contrast with the present situation (Cantner 1987, IVVO 1988).

However, the development of a profitable agro-industrial non-competitive (between non-ruminant animals and humans) feed resource poses certain problems not the least of which is the need to search for new technological solutions so that the proposed feeding strategy can be economically feasible.

The inventory of new feeds as well as their proximal composition started some years ago in several tropical countries with studies such as those of Pandittsekera and Elikewela (1949) in Ceylon, Ynalvez et al (1954) in the Philippines, Oyenuga (1955) in Nigeria, including several from the Latinamerican region such as those of Bressani et al (1968 in Guatemala and Central America, Devendra and Gohl (1970) in the Caribbean and Devendra and Raghavan (1978) in South East Asia. This inventory is still in progress and includes new data from feeds not foreseen at first sight in the catalogue, like amaranth (Bressani et al 1987), Canavalia ensiformis (see Escobar et al 1984) Gliricidia tree foliage (Chadhokar 1982; Adejumo and Ademosun 1985; Riverón et al 1986) or Cashew apple pulp and kila nut waste (Awo lumate 1983).

This sustained effort towards a new feeding strategy has been developed not only in laboratories in the tropical countries (Bressani et al 1968; Devendra and Raghavan 1978; López 1986) but also in temperate countries (Cantner 1987).

This somewhat classical but necessary method for establishing the interrelationship between availability of feed resources and nutrient requirements for non-ruminants cannot be considered the only way for expanding pig and poultry production by an adequate dietary formulation. If it is accepted that non-ruminants need a new feeding strategy in the tropics, based on new locally available feed resources, then the utilization of new feeds by these farm animals must be taken into account considering both their availability and the method of harvesting or collecting these products as well as ways for their distribution among animal producers. As the same time, a profound knowledge about chemical composition and nutritive value must be put forward. This may occur according to different options: (a) if the feed is not a by-product as such, but widely available in the country, and is not used at all, therefore being a pollutant, like animal excreta (see García et al 1985; Morales and Montilla 1985), or if it is destined to other purposes, eg: soil fertilization with azolla (Alcántara and Querubin 1984, 1985; Querubin et al 1986a, 1986b); (b) the feed is employed in a traditional manner, but it is capable of being improved, such as cassava leaf, cocoa husk and other cocoa by-products, coffee pulp or citrus pulp (Jarquín et al 1974; Ashum 1975; Hutagalung and Chang 1978; Jung and Choe 1985; Koh et al 1985; Domínguez 1988); (c) if the feed is a new product, and therefore the design of a new dietary formula and new conservation and distribution technologies is necessary [this cannot be solved in a conventional manner: aquatic biomass could be an example of this kind of feed (Kompiang and Matondang 1985; Lincoln et al 1986)]; or (d) if there is an overproduction of the feedstuff thereby avoiding the competition between farm animals and man [In this category are raw sugar (see Buitrago et al 1977); sugar cane juice (Fermin et al 1983; Mena et al 1981; Donzele et al 1986a,b; Mena et al 1986; Donzele et al 1987; Posso 1987) and high test cane molasses (Preston et al 1968; Marrero and Ly 1977; Figueroa et al 1983; Figueroa 1987; Ly 1987).

Complementary information concerning the size of the agro-industry in question is also needed. Solutions applicable on a large scale are not necessarily appropriate for small scale domestic or household management situations. Different resources are needed varying from more or less sophisticated devices to improve feed utilization efficiency and/or improving the nutritional quality of the new product to be used in the diet [the case of household organic waste (Domínguez 1988)] to procedures in which the animal itself does the processing [the feeding of pieces of whole cane stalk to pigs (Mena A, Personal communication)]..

Therefore, the development of feeding systems for intensively- managed improved breeds of pigs and poultry in the tropics poses a challenge both in identifying new feed resources and also from the viewpoint of how to improved feed efficiency and hence profitability on these non-conventional resources.

In this connection studies at the level of nutritional biochemistry and physiology may contribute by increasing the available knowledge concerning food acceptability, nutrient digestibility or metabolism thus playing, in some cases, a leading role in the development of new pig and poultry feeding systems.

The neutralization of anti-nutritive factors by feed processing or by an efficient diet manipulation requires prior knowledge of the physiological and biochemical basis of the action of such compounds. For example, some bird resistant sorghum varieties need to be treated to lower the tannin content in order to improve their nutritive value if destined for pigs (Fialho et al 1979) or chickens and laying hens (Méndez et al 1986) as has been observed in vitro digestibility tests (Mabbayau and Tripton 1975). On the other hand, cassava (Manihot utilissima) must be free or low in cyanide precursors if the tubers are to be utilized in diets for non-ruminant animals (Mahendranathan 1971; Gómez 1979; Gómez and Valdiviesa 1981). Cassava leaf may be included in these same diets (Acuña and Laffont 1982; Ravindran 1983; Parra 1987; Ravindran et al 1987), but anti-nutritive factors must also be taken into account (Ross and Enríquez 1969; Chew 1972).

In Cuba, animal science researchers have been working in the last years on the design of a sui generis model related to a non- conventional feeding system for non-ruminant species, with special emphasis on pigs and aquatic birds like ducks and geese. This model could be a very interesting example of how to improve pig and poultry production by applying new biochemical and physiological knowledge so as to strengthen the basis of such a strategy.

Different strategies for pig and poultry feeding have been developed in Cuba during the last years. One of the first models for pig feeding in the country used as the basic feed resource the agricultural wastes such as crop residues as well as others originating from fisheries, food industry and institutional and private homes. Food wastes are collected, autoclaved and supplied to the pigs usually mixed with molasses. This technology has evolved to the point where it functions as a complete agro-industry (Domínguez 1988; Pérez et al 1983). All kinds of food wastes and residues are collected daily by trucks through a very well organized network where the inventory of materials that can be feasibly incorporated into the scheme is carefully studied and analyzed. Wastes and residues are processed in small factories (del Río et al 1980) situated near pig units throughout the entire country. All the collected materials are mixed and sterilized by autoclaving in order to mix it later on with sugar cane molasses, and then distributed by pipe lines to the pig troughs as a thick soup (Domínguez 1988). This type of processed swill is usually destined for pig feeding, but it has also been successfully supplied to ducks and geese (Rodríguez and Ocampo 1986). A summary of the structure of food wastes and residues available for collection in Cuba is presented in Table 1.

Table 1: A representative structure of the collection of food wastes and residues in Cuba.

Food waste or residues	---------- Fresh basis ----------		-------- Dry matter basis --------
	Tonnes	%	% in waste	%

Institutional food wastes	256.3	59.8	23.5	60.2
Agricultural food wastes	149.8	34.8	15.0	22.4
Cereal by-products	20.4	4.3	76.0	15.5
Fish residues	2.4	0.6	79.2	1.9
Total	428.9	100.0	-	100.0

The food wastes and residues that are collected and fed to animals represents a major part of the total feed (dry basis) presently destined to pig production in Cuba (Table 2), thus taking advantage of the omnivorous character of this species (Domínguez 1988). In fact the use of this swill for pig feeding has saved a very considerable amount of cereals and grains which, if they had been imported, would have been diverted from human consumption. On the other hand, the value of swill in terms of land area that would otherwise have been employed for growing the equivalent amount of cereals, cannot be neglected (Table 3). Finally, in terms of environmental pollution, the collection of food wastes and other residues represents an invaluable ecological contribution.

Table 2: Estimated volumes of food wastes and residues and their potential for pig fattening.

Year	Processed food wastes and residues, (thousands of tonnes)	Potential of pig fattening, (thousands of animals)

1974	237	202
1977	314	267
1980	508	431
1983	850	710
1986	925	940

Table 3: Estimation of land saved (in thousands of ha) by swill processed from food wastes and other residues (dry basis)

Amount of swill (DM basis) (1)	Maize (ha) (2)

100 000	88 700
200 000	177 400
400 000	354 800
800 000	709 600

(1) Gross energy of swill 16.150 Mjoule/tonne DM
(2) Maize yield, 18 041 Mjoule/ha (FAO 1977)
Source: Pérez et al (1983)

In terms of their nutritive value, processed food wastes and residues, appear to be less variable than would perhaps be expected in view of their heterogeneous origin. This facilitates establishing quality control indices in the different factories throughout the country. As an example of this, the results of an annual survey concerning the quality of swill obtained from a factory which started working in 1980 is shown in Table 4.

It has been argued, that heat treatment causes a deterioration in the protein quality of food, mainly through heat damage of some basic amino acids like lysine and arginine (Buraczewski et al 1967; Eggum 1973, 1977). However, it is also known that heat treatment can neutralize many anti-nutritive factors such as anti-tryptic agents (Borchers and Ackerson 1950), toxic non-protein amino acids like mimosine from Leucaena leucocephala leaf meal (Matsumoto et al 1951) and cyanide precursors (Gondwe 1974). Thus, there may be a compromise between both viewpoints in the case of food wastes and residues turned into processed swill, at least as concerns the nutritive value.

In fact, there have been no reports of negative influence of swill in well balanced diets fed to fattening pigs in the Cuban system (Domínguez et al 1987).

Table 4: Protein and energy contents of swill obtained from food wastes and residues processed in a Cuban factory (mean values and standard error).

	DM	N x 6.25	GE
Month	(g/100 g)	(g/100 g DM)	(Kj/g DM)

January	15.9 ± 0.6	nd	nd
February	15.5 ± 0.5	nd	nd
March	16.2 ± 0.5	nd	17.7 ± 0.4
April	16.5 ± 0.5	19.7 ± 0.9	18.2 ± 0.4
May	16.6 ± 0.5	18.8 ± 0.9	18.8 ± 0.4
June	19.8 ± 0.5	20.3 ± 0.9	19.4 ± 0.4
July	19.0 ± 0.5	20.9 ± 0.9	19.9 ± 0.4
August	15.4 ± 0.6	23.7 ± 1.0	19.5 ± 0.4
September	18.7 ± 0.5	19.9 ± 0.9	19.4 ± 0.4
October	17.0 ± 0.6	19.8 ± 1.0	18.7 ± 0.4
November	17.5 ± 0.6	16.8 ± 1.0	18.7 ± 0.4
December	18.1 ± 0.5	19.3 ± 0.9	17.5 ± 0.4

Sugar cane (Saccharum officinarum) is a herbaceous plant which grows in the tropics. It has an outstanding annual yield per hectare, when compared to other crops either from temperate or tropical countries (see Gohl 1975).

Another very important characteristic of sugar cane is that it is composed almost equally of soluble (NFE and NDS) and fibrous insoluble (NDF) elements (Table 5).

Contrary to cereal grains where starch represents the major NFE fraction, sugar cane feeds can contain fructose as high as almost 50 % in raw sugar in the form of sucrose, and non-carbohydrate, organic substances accounting for 20 to 25 % dry basis in final molasses or 5 to 10 % in high-test molasses and refinery syrup. Sugar cane juice could be considered to be a sucrose solution. Practically all the sugar cane feeds are liquids more or less syrups and in fact constitute true carbohydrate solutions and therefore are hyperosmotic (Ly 1987).

Table 5: Ranges in the chemical composition of sugar cane (1)

Item, %	Mean	Maximum	Minimum

Dry matter (2)	25.75	30.5	17.0
N x 6.25	2.32	3.06	1.06
Ether extract	1.24	1.87	1.70
Ash	4.33	7.12	2.74
Calcium	0.20	0.35	0.06
Phosphorus	0.05	0.09	0.02
Crude fibre	28.12	35.93	22.68
NDF	52.70	67.70	42.56
Cellulose	26.99	31.97	21.89
Lignin	6.31	8.43	4.60
NDS (3)	47.29	57.44	32.30
NFE	53.99

(1) Data from 66 cultivars of sugar cane (all the cultivars are 10 months of age at harvest). Analysis made in whole sugar cane (stalk, tops and adhering leaves or trash).
(2) As percent of fresh weight (all other data reported as per cent of dry matter)
(3) Neutral detergent solubles

Sugar cane is a crop which can support a very efficient agro- industry related to pig and poultry production throughout the world, originating animal feeds mainly from raw sugar, sugar cane molasses and sugar cane juice (see Preston 1987, 1988). As it is well known all these feedstuffs are characterized by their extremely high NFE value, with no fibre and where ether extract and protein are practically absent (Ly 1987). A typical composition of different sugar cane feeds is shown in table 6.

Feeding trials with pigs and poultry fed diets based on sugar cane feeds were carried out at first mainly with raw, unrefined sugar and final sugar cane molasses, both being the two extreme feeding possibilities arising from the usual sucrose extraction process in the sugar mill. In respect to poultry feeding systems, sugar cane final molasses proved to be suitable when included in quite high levels in diets for both broilers and layers with no detrimental effect on performance. However, a serious limitation in total replacement of grains and cereals was due to difficulties formulated with relatively high levels of these molasses, including mixing and handling of the diet, by the laxative effect, as Connor et al (1972) pointed out in Australia, after earlier work performed in Hawaii (Rossenberg 1955; Rosenberg and Palafox 1956a,b; Ross 1960; Kondo and Ross 1962) and later in Cuba (Pérez 1958).

Table 6: Chemical composition of some sugar cane feeds obtained after juice extraction from the plant.

Feed	pH	Dry matter	Ash	Sucrose	Reducing compounds	Source of data

Sugar						2
- Refined	-	99.5	-	100.0	-	Ly 1986
- Raw	-	97.0	1.5	93.2	2.8

Juice						3
- Fresh	5.2	25.7	0.3	19.7	4.7	Alexander 1965
- Dehydrated	-	96.5	3.6	67.5	22.5

Molasses
- High-test	5.6	76.1	2.0	35.0	17.0	Preston et al 1968
- Final	5.4	76.9	5.5	24.0	45.0
- Dehydrated	9.1	93.3	17.5	nr	11.1	Veloso et al 1980

1: All data as percent of fresh weight
2: Unpublished data
3: Quoted by Gohl (1975)
nr: Not reported

A change from a dry feeding system to a liquid system based on sugar cane molasses plus a protein, vitamin and mineral supplement, overcame the obvious barrier presented by a sticky mixture of sugar cane molasses and grains offered to the birds in a meal form (Pérez and Preston 1970). This successful liquid feeding system was extended also to ducks (Pérez and San Sebastián 1970), turkeys (Valarezo and Pérez 1972) and geese (Valdivié and Pérez 1974), either with sugar cane final or high-test molasses. On the other hand, the laxative effect observed when birds were fed on diets with high levels of sugar cane final molasses was only neutralized by mixing these molasses with raw sugar or high-test molasses (Pérez et al 1968; Pérez and Preston 1970). In fact, wet droppings have been associated with a poor bird performance when compared to cereals and grain feeding systems since the first observations made by Rosenberg (1955). Furthermore, the laxative effect of sugar cane final molasses has been claimed to cause a high humidity in bird litters thus favoring the proliferation of coccidia and other pathogenic agents. On the other hand, feathers and egg shell surface become dirty and wet (Rosenberg and Palafox 1955b; Solde villa et al 1970; Sharma and Pallivaal 1973).

Nevertheless, the most important issue in such a feeding system based on sugar cane feeds for poultry could be the selection of the most appropriate species, according to feed intake and feed efficiency, including genotype-nutritional interactions (López et al 1970).

In parallel to poultry, a liquid feeding system for pigs based exclusively on sugar cane feeds have been developed in Cuba, first at laboratory scale with final or high-test molasses (Preston et al 1968) and later in large pig fattening units throughout the country (Figueroa 1987, 1988). Besides, a high-test high-protein sugar cane molasses has been utilized for fattening pigs and breeding sows (Figueroa 1988) by adding torula yeast in cream form to the unextracted molasses in the sugar mill, in order to be distributed by pipe lines to the animals as outlined by Pérez and Ly (1983). On the other hand, the same laxative barrier observed in poultry has been encountered in pigs fed sugar cane final molasses for a long time (see Ly 1987).

Less emphasis has been given, at the farm level, to the search for protein sources for pigs and poultry in tropical countries. Yet, from an economic standpoint, protein shortage is a true barrier to the development of profitable feeding systems for non-ruminant farm animals. In this connection, potential protein sources are to be found in any inventory of tropical feeds, and at the research level efforts have been made to improve not only their nutritional value but also the yield (see Fetuga et al 1974; Fialho and Albino 1983; Flore et al 1988; Ravindran 1988).

In Cuba, two non-conventional, high protein feeds: swill and torula yeast grown on sugar cane final molasses are promising as protein sources for large scale pig and poultry production.

Swill has been used in moderate proportions in diets for geese, hence incorporating moderate proportions of protein to the feed (Rodríguez and Ocampo 1986). Contrary to poultry, the feeding of swill to pigs has been a major activity in Cuba (Domínguez 1988).

Earlier studies related to the protein content and biological value of swill have suggested that this feed can be improved by the conventional method of amino acid balancing (Maylin and Cervantes 1982; Maylin et al 1984) or its combination with other well known protein sources (see Domínguez 1983). In Cuba the major role of swill has been to provide the protein component in diets based on molasses.

In the particular case of broilers, torula yeast has been extensively studied as a potential major protein source in the diet (Valdivié 1976; Valdivié et al 1977a; Valdivié et al 1977b). However, results concerning torula yeast utilization by broilers fed diets based on this single cell protein revealed a lower nitrogen retention and poor dry matter utilization by birds (Alvarez and Valdivié 1980) together with an unusual humidity in droppings, observed since the first trials (Valdivié 1976; Lon Wo and Valdivié 1981). Further studies related to this unfavorable effect of torula yeast were carried out by Tillán et al (1986) in colectomized birds. In this respect, it was possible to find a straight dependence of humidity in droppings and enhanced water drinking, thus increasing daily urine volume output as a consequence (table 7), with no influenced on faecal moisture. The cause(s) provoking this phenomenum remain still obscure, although it has been considered that the high mineral content of torula yeast stimulates water drinking (Alvarez and Valdivié 1980). For this reason, torula yeast could be included in poultry diets only at moderate levels.

Contrary to the experience with poultry, it has been demonstrated that pig diets can be formulated with torula yeast as the only source of protein (see Maylin 1988). Moreover, a liquid feeding system has been designed by preparing a high-test high-protein molasses in the sugar mill (Cervantes et al 1984).

Recent studies concerning experiments carried out with this industrial mixture of high-test molasses and a cream of torula yeast were recently reviewed by Maylin (1987ab, 1988), Lan (1967) and Figueroa (1987, 1988). Results related to nitrogen digestibility and retention in pigs fed diets containing high proportions of torula yeast suggest that intensive pig production can be sustained with this protein source.

Table 7: Water status and digestibility indices of colectomized broilers fed graded levels of torula yeast in the diet.

	------------------ Torula yeast, % dry basis ------------------
	0	10	20	27

Water status
- Water consumption, ml	705	1 000	888	1 153
- Urine output, ml	112	234	226	425
- DM in faeces, %	20,9	24,5	22,6	24,7
Digestibility indices
- DM, %	83,7	75,6	73,5	69,3
- Nitrogen, %	91,0	85,6	87,4	87,1

The pattern of feed intake on non-conventional liquid feeding system in pigs

The change from a dry system to a system based on a liquid feed, where sugar cane products as the major fractions of the diet has been a decisive step taken in poultry nutrition hence permitting the use of some products like sugar cane molasses and juice in many tropical countries. In spite of some anatomical peculiarities of birds like hens and turkeys, somewhat different to ducks and geese, the etiology of the animals included in this type of feeding system has not been described as yet. On the other hand, contradictory results concerning feed intake have arisen from different studies (Pérez and Preston 1970; Valarezo and Pérez 1972; Valdivié and Pérez 1974), although there is a trend towards a decreased feed intake by chickens fed increased levels of final molasses in the diet. If compared to farm birds, a considerable amount of observation has been accumulated concerning the pattern of feed intake by the pig fed either swill or sugar cane feeds (Ly and López 1979; Ly and Muñóz 1979; Ly and Castro 1984; Ly et al 1987).

At first sight, the main physiological restriction for a strategy for pig feeding in diets based on swill could be derived from the liquid nature of the diet, together with the relatively low dry matter content of the diet. In fact, the pattern of swill intake by pigs is quite different to that of conventional, cereal and grain based diets. In this feeding system, pigs spend a long time eating (Ly et al 1987). Nonetheless, it has been claimed that the increase in water content in the diet does not have a deleterious effect either on diet digestibility or on animal performance in the case of the pigs (Kornegay and Vander Noot 1968).

The pattern of feed intake in pigs fed sugar cane molasses based diets is also quite different if compared to swill and cereal and grain feeding systems or raw sugar. On the other hand, an inhibition (Marrero and Ly 1976) or a stimulation of voluntary feed intake can occur (Figueroa et al 1988) in pigs fed a high-test cane molasses. In contrast, feed intake in pigs fed final molasses tends to be lower than expected as the low energy density of the molasses cannot be compensated for by an increase in the ingestion of the feed by the animals (see Ly 1987).

Further information concerning long and short term control of feed intake in the pig is needed in order to put into practice the invaluable data which are being produced in different laboratories (Forbes 1985; Baile and McLaughlin 1986; Houpt 1986) if it is taken into account that some indices of the pattern of feed intake may be associated with performance traits of economic interest, at least in the pig (Ly and Castro 1984).

Since it has been observed that, as with other species, pigs drink water mainly at meal time (Auffrey et al 1974; Houpt et al 1983), it is logical to find an increase in the amount and frequency of water drunk by the pig when sugar cane molasses are offered in the ration as the major source of energy (Marrero and Ly 1977; Ly and Castro 1984). An important point here is the hypertonicity of the diet due to an extremely high content of low molecular weight soluble carbohydrates, like sucrose, glucose and fructose (see table 6) and minerals. The same phenomenum seems to be present in the chick (Aragón et al, quoted by Valdivié and Fraga 1988).

It is noteworthy, that even in pigs exhibiting the typical laxative state induced by sugar cane final molasses, there is no change in the ratio of water excretion between faeces and urine (Marrero and Ly 1976; Savón 1984). The same seems to be true of chickens (Rodríguez et al 1980). On the other hand, an increase has been found in kidney weight of chickens and pigs fed increasing levels of sugar cane final molasses (Aragón et al 1974; quoted by Valdivié and Fraga 1988; Marrero and Ly 1976). Therefore, the same physiological response would appear to be present in pigs and gallinaceous birds fed high levels of sugar cane final molasses. This enhanced drinking behavior has not been observed with farm animals fed diets containing other sugar cane derived feeds, such as raw sugar, high-test molasses or sugar cane juice.

These results suggest at a first approach, that water supply and hence liquid manure are very high on final molasses diets if compared to cereals or even other sugar cane feeds. This is likely to be an important issue if substantial amounts of sugar cane final molasses are going to be included in the diet for pigs and poultry.

Anatomical modifications of the digestive tract in pigs and poultry fed sugar cane feeds

Certain slight changes in weight or length of different organs of the digestive system of pigs take place when the animals are fed sugar cane molasses, either as high-test or as final molasses, if compared to maize (Ly and Mollineda 1984) or final molasses plus moderate amounts of a fibrous source (Rodríguez et al 1986).

Final molasses tend to decrease the stomach and pancreas weight, as well as to diminish haustra development in the large intestine.

In contrast, sugar cane molasses tend to increase liver weight (table 7). This same trend appears in diets with variable proportion of swill and sugar cane final molasses (Pérez and Ly 1978).

Anatomical modifications have been found to be dramatic in broilers fed sugar cane final molasses (Alvarez 1976). In fact a remarkable increase has been observed in the crop, proventricle, small intestine and caecal empty weight of the bird as the final molasses proportion in the diet increases; the reverse occurred with gizzard weight. On the other hand, a substantial increase in the apparent area of the mucosa in the crop and caeca of birds fed final molasses was also evident (table 8).

Table 8: Sugar cane molasses and organ weights in pigs.

Fresh weight, g/kg BW	Maize meal	High-test molasses	Molasses type B	Final molasses	Source of data

Entire GIT	27.9	29.2	-	28.1	Ly and Mollineda 1983

- Stomach	6.6	4.6	-	4.8

- Small intestine	11.9	12.9	-	13.3

- Caecum and colon	9.4	11.7	-	10.0

Liver	14.6	-	22.8	18.1	Ly et al 1989

Pancreas	1.1	-	0.8	0.8

The changes in the digestive system of turkeys fed sugar cane molasses are concentrated in the crop, according to Valarezo and Preston (1973). As it has been argued, pendulous crop incidence in turkeys fed sugar cane molasses decreased bird performance and provoked a high mortality rate.

In line with recent Swedish and French investigations concerning nutritional physiology and anatomy interrelationships, summarized by Thomke (1986) and Février et al (1988), the strategy of a new feeding system for pigs and poultry based on sugar cane feeds indicates an anatomical effect on the gastrointestinal tract, due to these non-conventional dietary ingredients. These results may help in turn to a better understanding of the physiology of the digestion of the pig and farm birds.

As expected, pigs and poultry tend to show a fast rate of passage of digesta through the gastrointestinal tract when sugar cane final molasses are included in large proportions in the diet (Alvarez 1976; Marrero and Ly 1979; Ly 1984). Furthermore, the rate of either gastric or crop ingesta emptying as well as digesta retention time in the caecum and colon appear to be accelerated and diminished respectively when this same sugar cane feed is given to pigs or poultry (Torres et al 1974; Pérez and Ly 1978; Ly 1985), when the comparison is made, not only with swill and maize (Pérez et al 1978; Ly 1979; Ly and López 1979; Pérez et al 1981), but also with high-test cane molasses (Marrero and Ly 1979, Ly 1984). On the other hand, mouth to rectum digestibility of several nutrient constituents tend to decrease if sugar cane final molasses is compared to maize, swill or other molasses.

As it has been claimed (Willet et al 1946; Rossemberg 1955; Rossemberg and Palafox 1956ab; Obando et al 1969; Maner et al 1969) the laxative effect of sugar cane final molasses, defined as a rapid rate of passage of digesta through the entire gastrointestinal tract, brings about a sharp decrease in the digestibility of the diet, thus causing a deterioration in the daily gain and feed efficiency of the diet. In this connection, working hypotheses developed in order to neutralize these adverse effects of sugar cane final molasses have long been put forward. The cation content of molasses and particularly potassium has been argued to be the cause of the diarrhoeic state in pigs and poultry (Kondo and Koss 1962; Rossemberg 1956; Rossemberg and Palafox 1956ab; Maner et al 1969; Obando et al 1969), but this hypothesis has failed to be viable in practice (Cuervo et al 1972a; Cuervo, Restrepo, Bushman and Rondón 1972b; Obando et al 1968; Maner et al 1969).

Another factor not previously considered could be the influence of the exacerbated water intake which might produce the laxative condition in pigs and poultry fed high levels of sugar cane final molasses. However, water dilution of diets formulated with grains and cereals suspended in water, or even swill, do not support this hypothesis (Kornegay and Vander Noot 1968). On the other hand, in spite of its accepted heterogeneity due to its diverse origin, swill digestibility is not low, if it is taken into account that digestion in the large intestine does not seems to play a substantial role, at least in the pig, as indicated in table 9.

Table 9: Weight and area of different organs of the gastro-intestinal tract of chickens fed graded levels of sugar cane final molasses.

	--------------- Sugar cane final molasses in the diet, % -----------------
	0	16	32	46	66

Weight, g
- Crop	7.56	7.74	8.20	7.28	8.34
- Proventricle	7.84	12.24	13.24	14.20	12.62
- Gizzard	39.24	36.72	35.36	30.50	23.12
- Small intestine	49.52	50.42	58.24	61.08	58.76
- Caeca	10.38	10.20	9.84	11.20	13.30

Area, cm²
- Crop	62.45	65.64	73.00	89.47	92.24
- Caeca	51.40	53.54	66.49	70.51	77.36