Steam treated bagasse for fattening cattle. Effect of supplementation with Gliricidia sepium and urea/molasses

Steam treated bagasse for fattening cattle. Effect of supplementation with Gliricidia sepium and urea/molasses

Convenio Interinstitucional para la Producción Agropecuaria en el Valle del rio cauca (CIPAV)

The potential use of sugar-cane bagasse in animal feeding was eva-luated in commercial Zebu steers. A basal diet of steam-treated bagasse was supplemented with two levels (2% or 3% of animal live weight) of foliage of the legume tree Gliricidia sepium and four levels (ad libitum, 1, 1.5 or 2 kg/day) of a urea/molasses mixture containing 10% urea. Liveweight gains were in the range 0.55 to 0.75 kg/d; there were no statistical differences in animal live weight change between the levels of Gliricidia (P=0.40) nor between the quantities of urea/molasses mixture used (P=0.76). In a second experiment, comparisons were made between animals fed on bagasse treated in two different ways (steam or steam-ammonia), supplemen-ted with two levels of Gliricidia (1% or 1.5% of live weight) and three levels of urea (12, 14 or 16%) in a urea/molasses mixture fed ad libitum; a higher rate of live weight change (P=0.00) was obtained for the animals fed on steam-treated bagasse (0.635, 0.624 and 0.564 kg/day for 12, 14 and 16% urea in urea/molasses, respectively) than those fed on steam-ammonia-treated bagasse (0.297 kg/day); there were no differences due to level of Glirici-dia (P=0.87) and the percentage of urea in the molasses (P=0.40). In a third experiment, rumen degradability tests (nylon bag tech-nique) were carried out on raw bagasse, steam-ammonia, and steam-treated bagasse. There were significant differences (P=0.003) between the treatments. Mean values for dry matter degradability at 24 hours in the rumen of cattle were 14, 20 and 31% for untreated, steam-ammonia and steamed bagasse respectively; the corresponding value for cotton wool was 35%. It was concluded that steam-treated bagasse can be used successfully in the fattening of cattle.

Key words: ammonia, bagasse, cattle, Gliricidia sepium, molasses, steam-hydrolysis, sugar-cane, urea.

Cellulose is the world's most widely available renewable resource, amounting to about fifty per cent of the cell-wall material of woody and herbaceous plants. Due to this abundance and renewabili- ty, there has been a great deal of interest in utilizing cellulose as an energy resource and as a feedstock (Fan et al 1982). Many fibrous by-products have a substantial potential value as animal feedstuffs. Ruminants, especially, have the unique capacity to utilize cellulose, because of their microbes (Boucqué and Fiems 1988). The shortage and high costs of conventional raw materials necessitates the use of these agricultural residues in animal feeds. However, the high lignin content, the low level of soluble carbohydrate and the relative absence of both fermentable nitrogen and bypass protein are responsible for the low nutritional value of untreated residues (Hamad and El-Saied 1982; Preston and Leng 1974; Sundstol 1988).

Bagasse is the highly fibrous residue remaining after sugar-cane is pressed to remove sucrose. Sugar-cane mills produce more bagasse than can be utilized as a fuel source for sugar processing; few commercial uses for the excess bagasse have been developed and its accumulation presents a waste problem for the sugar industry. One potential use of bagasse is as a feedstuff for cattle. Because the components of bagasse are in their natural, resistant conformation, susceptibility to enzymic hydrolysis is extremely limited (Rivers 1988).

Agricultural by-products like cereal straws and sugar-cane bagasse are high in ligno-cellulose. Roughly three-quarters of straw is cellulose plus hemicellulose. Bagasse also contains more than 60% of its dry matter in the form of cellulose and hemicellulose but its degradability is very poor. One of the main reasons for this depression in degradability is the presence of lignin which protects carbohydrates from attack by the rumen microbes. Sugar- cane bagasse contains around 50% cellulose, 27.9% hemicellulose, 9.8% lignin and 11.3% cell contents (Kewalramani et al 1988).

It has been recognized that in order to improve the nutritive value of ligno-cellulosic materials for livestock, some form of pretreatment of processing of the plant material is required (Helmling et al 1989).

Polysaccharide degradation in the rumen takes place as the result of the action of a consortium of anaerobic bacteria, protozoa and phycomycete fungi and, even so, its bioconversion is far from complete. Hence a pretreatment of the substrate is required to increase the rate and extent of holocellulose hydrolysis, in order to alter significantly the structural characteristics of the ligno- cellulosic matrix. Such a pre-treatment must enhance the close contact between microbe and fibres to provide an efficient enzyme action (Rolz et al 1987).

A technique which has shown considerable potential for the cost- effective pre-treatment of ligno-cellulosic materials is that of steam explosion (Wong et al 1974). With this method, the substrate is loaded into a pressure vessel and heated by steam injection for a defined time-temperature period. At the end of this period, the contents are explosively discharged into a pressure vessel; a substantial portion of the hemicellulose fraction is made water soluble and the lignin fraction is modified (Wang et al 1974, Morjanoff and Gray 1987).

The most noticeable effect of treating raw bagasse with steam under pressure is a large reduction in certain fibre components. Crude fibre was reduced from 43 to 34% and NDF was reduced from 83 to 51%; however, there was little change in ADF, cellulose and lignin content. These data indicate that steam-pressure treatment com- pletely modified the hemicellulose fraction of raw bagasse, chan- ging it into more soluble components, but did not affect the ligno- cellulose components (Wong et al 1974; Pate 1982; Kling et al 1987)

Chemical or combined chemical/physical treatments are generally more effective than physical methods, the most successful to date being treatment with alkali (usually sodium hydroxide) to raise the degradability of straws, bagasse, husks, hulls, cobs, etc., but the success of this approach requires application of relatively large amounts of alkali (Crosthwaite et al 1984). The production of antimicrobial substances has been demonstrated during pre-treatment of ligno-cellulose with certain chemicals at high temperatures (Al-Ani and Smith 1988).

The possibility of utilizing suitably treated ligno-cellulose as an efficient ruminant feed could have major impact on livestock production practices in areas where indigenous supplies of ligno- cellulose are abundant and where cereal grain supplies are limited or are utilized primarily for human food (Gould et al 1989).

When sugar-cane bagasse (and fibrous residues in general) is used as the basal diet, it is important to give the correct supplementa- tion in order to obtain satisfactory physical and economic respon- ses. The supplementation must take account of the productive stage of the animals (eg: growing, fattening, lactating, etc) and the locally available sources of supplements in the region (Preston and Leng 1987). The nutrients arising from rumen fermentation must be balanced using supplements containing dietary by-pass protein to provide essential amino-acids, dietary by-pass starch which provide extra glucose, and dietary fat to provide long chain fatty acids for the synthesis of body tissues and milk fat. The low N content of sugar-cane and its by-products, impairs rumen fermetation due to the low availability of ammonia and sulphur, in animals fed on these diets (Leng 1989).

Gliricidia sepium (Jacq.) Steud, is a legume tree which has been used for a long time in Colombia and other countries, mainly as a live fence. It is a tropical species which grows at altitudes from 0 to 1500 metres above sea level (Baggio 1982). The foliage of Gliricidia sepium is a combined source of micro-nutrients for the rumen bacteria, vitamin A and some fermentable and by-pass protein to balance the energy from the rest of the diet. Moreover, it also provides long fibre, which is needed in diets in which bagasse is the basal diet. Diets based on bagasse are deficient in long chain fatty acids; supplementation with rice bran provides a suitable supply and also gives additional by-pass nutrients in the form of starch, lipids and protein (Elliot et al 1978). Poultry litter is a local and natural resource, rich in macro-(Ca, P, S) and micro- minerals (Cu, Co, Zn, etc), because of the routine supplementation with balanced minerals in intensive poultry enterprises; it also provides fermentable nitrogen, mostly as uric acid which is hydro- lysed to ammonia by rumen micro-organisms (Preston and Leng 1984). Urea supplementation increases the utilisation of low-nitrogen fibrous diets. A liquid mixture of urea/molasses is attractive and palatable to ruminants because of the smell and taste of molasses; the animals lick it almost continuously, and as a result its ingre- dients are continuously available to the rumen micro-organisms.

Although chemical analysis and IVOMD are valid first steps in assessing the relative values of chemical pre-treatments, responses may be modified in the in vivo situation. For example, although ammonia treatments increase the nitrogen contents of by-products, the availability of the retained nitrogen to the animal may not be high (Ibrahim and Pearce 1983a).

The purpose of this experiment was to find out if the physical treatment (steam) and the combination of a physical and a chemical treatment (steam-ammonia) of sugar cane bagasse were suitable treatments for this sugar industry by-product, in order to use it later as an animal feed. The optimum levels of supplementation with foliage of a legume tree (Gliricidia sepium), and urea/molasses liquid mixture were also tested. In vivo degradability tests were made to obtain additional information about the effect of the treatments on the sugar cane bagasse, and the effect of the supplementation on the degradability of the sugar cane bagasse.

Experiments were carried out at sites adjacent to the Ingenio del Cauca and Ingenio Providencia, sugar factories, both situated in the Department of Valle del Cauca, Colombia (4°N, 76°W, 957 metres above sea level, mean temperature 24°C).

A first experiment was carried out to study the effect on live weight change in Zebu cattle (Bos indicus) of steam-treated sugar- cane bagasse supplemented with two different levels of a legume tree foliage (Gliricidia sepium) and four quantities of an urea/molasses liquid mixture which contained 10% of urea.

On the basis of the results obtained in that first experiment, a second one was designed in which lower levels of Gliricidia sepium and three different percentages of urea in the molasses/urea liquid mixture were used, and measurements were made of the live weight change of Zebu cattle. In this second experiment a comparison was also made between bagasse treated with steam (as in the first one) vs. steam and ammonia treatment and their effects on the live weight change of Zebu cattle.

In vivo degradability tests by the nylon bag method (Orskov et al 1980) were carried out in Experiment 3, using animals of the same type as those in the previous two experiments, mainly to assess the effect of the treatments on the degradability of bagasse, and the effect, if any, of supplementation with Gliricidia sepium and poultry litter.

The animals used in all the experiments were of the so called "Zebu commercial" type, which is mainly derived from Zebu (Bos indicus) with some genes of European breeds (mainly Brown Swiss and some Holstein). It is the most widely used kind of cattle for fattening in Colombia. These animals are well suited to the tropical conditions, mainly through adaptations in their skin, which is highly pigmented, loose and light in colour.

For Experiment 1, 24 animals were used, allocated to eight treat- ments with 3 replicates each. For the second experiment, 5 animals per treatment were used (replicates) and the number of treatments was 8, giving a total of 40 animals. Two animals fitted with rumen fistulae were used in Experiment 3 for the degradability tests.

Fresh sugar-cane bagasse was treated with steam in a non-continous digester which worked under the following parameters: pressure 10 to 17 atmospheres, temperature 180 to 200°C, treatment time 5 to 7 minutes. After this treatment a product was obtained with a softer texture and a darker colour than the raw bagasse. Pate (1982) and Kling et al (1987) report similar texture to that found in this experiment.

For this experiment two different treatments of the sugar-cane bagasse were used. The first treatment was exactly the same as that already described in Experiment 1. The second treatment was a combination of physical and chemical treatments: sugar-cane bagasse treated with steam (as described for Experiment 1) was then treated with ammonia. This last treatment consisted of the injection of anhydrous ammonia into a plastic (polythene) bag, which contained the steam-treated bagasse. After treatment the bag was kept sealed for a period of 15 days before use. The quantity of NH₃ injected into the bag corresponded to 3% of the weight of the bagasse on a dry basis. The texture of the steam-treated bagasse was not noticeably improved by the treatment with anhydrous ammonia and, in general, there were no apparent changes in the steam-treated bagasse after this later chemical treatment.

The levels of supplements fed were in proportion to the live weight of the animals, adjusted every 28 days. Thus, where the level is described as 1% of live weight, 1 kg fresh material was offered per 100 kg live weight.

For Experiment 1, this supplement was used at two different levels: 2 and 3% of animal live weight (fresh matter basis). The percentages of Gliricidia sepium used in the second experiment were: 1 and 1.5% of animal live weight on a fresh matter basis.

For both Experiments 1 and 2, this supplement was included in the diet at a quantity equivalent to 0.2% of animal live weight (fresh matter basis).

This was litter with a rice bran and/or wood shavings base from broiler houses. The level of inclusion of this supplement in the diet was 0.2% of the animals' live weight (fresh matter basis).

A restricted system (in several of the treatments in Experiment 1), and a free-choice system (in Experiment 2) were used, in which the molasses was the vehicle for urea and also provided trace elements. The urea/molasses liquid mixture used in the first experiment contained (% fresh matter basis): 85 molasses, 10 urea, 3.5 mineral/vitamin pre-mixture, 1.5 salt; for the second experiment the urea/molasses liquid mixture was varied in its urea content (12, 14 and 16% fresh matter basis), and the level of molasses was reduced (83, 81 and 79% fresh matter basis respectively), while the percentages of mineral/vitamin pre-mixture and salt were the same as in the first experiment.

For this experiment, twenty-four Zebu steers were allocated to eight treatments (Table 1), with three animals (replicates) in each. The animals were kept in eight pens (one pen per treatment) of approximately 3 m wide x 10 m long, in which a third of the area was covered by a roof and the floor of this covered area was con- creted; the rest of the area was soil surface. The three animals (replicates) from each treatment were kept together in the same pen.

All the animals were fed with steam-treated sugar-cane bagasse ad libitum; the differences between treatments were the level of Gliricidia sepium (1 or 1.5% of animal live weight, fresh basis) and the quantity of urea/molasses liquid mixture (ad libitum, 1, 1.5 or 2 kg/animal per day) fed to each animal. Additionally the animals were supplemented with rice bran and poultry litter (each one of them at the level of 0.2% (fresh basis) of animal live weight). The quantities of Gliricidia sepium, rice bran, poultry litter and the restricted levels of urea/molasses liquid mixture were calculated on the basis of the total weight of the three animals on the same treatment, because they were not separated in the pen in which they were kept.

The supply of the dry ingredients of the ration was made in a wooden trough in the following way: the treated bagasse was put in first, then the mixture of the respective quantity of poultry litter and rice bran was sprinkled on top and finally the foliage of Gliricidia sepium was added to the former ingredients. Once all the ingredients were in the container, they were lightly mixed by hand. The urea/molasses liquid mixture was provided in a separate plastic trough, beside the other containing the dry ingredients; both of them were protected from the rain by the roof. The animals had free access to fresh drinking water at all times of the day. Previous to the experimental period, the animals were kept for an adaptation period of four weeks on the diet which they were going to receive and, immediately after this adaptation period, every animal was weighed and the experimental period began; the experimental period lasted for a total of 141 days.

Table 1: Description of the treatments in Experiment 1

Experimental	Molasses/urea mixture	Gliricidia sepium
treatment	(kg/animal/day)	(% of live weight)

(1)2%	1.0	2.0
(1)3%	1.0	3.0
(1.5)2%	1.5	2.0
(1.5)3%	1.5	3.0
(2)2%	2.0	2.0
(2)3%	2.0	3.0
(a)2%	ad libitum	2.0
(a)3%	ad libitum	3.0

Every twenty-eight days each of the animals was weighed in order to calculate the live weight change (kg/day) between these periods of time, and this calculated value was the main response variable used to compare treatments.

Forty Zebu steers were used in this experiment, allocated to eight treatments (Table 2) with five animals (replicates) in each. The animals were kept in the same pens already described for Experiment 1; in each pen five animals were kept, all of them belonging to the same treatment.

All the animals were supplemented with rice bran and poultry litter, both at the level of 0.2% (fresh basis) of the animal live weight. Additionally, every treatment was supplemented with either 1 or 1.5% (fresh basis) of the live weight with Gliricidia sepium. The quantities of these three supplements were calculated for the five animals on the same treatment, because they were kept toge- ther, and this total amount fed to the group; re-calculation of these quantities was made after every weighing (four weeks). In six of the treatments the animals were fed with steam-treated sugar- cane bagasse ad libitum, and they were supplemented with urea/molasses liquid mixture ad libitum, containing one of three levels of urea (12, 14 or 16% fresh basis). In the other two treatments the animals were fed with steam-ammonia-treated sugar- cane bagasse ad libitum and they were not supplemented with urea/molasses liquid mixture since the bagasse treated in this way contains ammonia-N, which can be used by the rumen micro-organisms. The diet was supplied in the same type of troughs and in the same way already described for Experiment 1. Fresh drinking water was available to the animals at all times.

Table 2: Description of the treatments in Experiment 2

Experimental	Treatmet of	%urea in	Gliricidia sepium
treatment	the bagasse	molasses	(% of live weight)

SPT+Amon(1%Gs)	SPT+ammonia	....	1.0
SPT+Amon(1.5%Gs)	SPT+ammonia	....	1.5
SPT(12%u)(1%Gs)	SPT	12	1.0
SPT(12%u)(1.5%Gs)	SPT	12	1.5
SPT(14%u)(1%Gs)	SPT	14	1.0
SPT(14%u)(1.5%Gs)	SPT	14	1.5
SPT(16%u)(1%Gs)	SPT	16	1.0
SPT(16%u)(1.5%Gs)	SPT	16	1.5

In this experiment, the animals were given an adaptation period in the same way as in Experiment 1, and for the same period of time (four weeks). The experimental period lasted for 169 days; the animals with the diet of steam-ammonia-treated bagasse were on experiment for just 113 days, and then they were taken out of the experiment.

Weighing of every animal in the experiment was done every twenty- eight days and the live weight change calculated to elucidate differences between treatments.

The animals used for this experiment were kept in individual pens in order to control their diet, which consisted of steam-treated bagasse and urea(10%)/molasses liquid mixture, both of them fed ad libitum, and rice bran (0.2% of the live weight); one of the animals was supplemented with Gliricidia sepium (1.5% of the live weight) for a period of twelve days, after which the other animal was supplemented with Gliricidia sepium for another twelve days. During the last three days of every period of twelve days, degradability tests were conducted by the nylon bag technique (Orskov et al 1980). The bags had a pore size of 44 µm and the dimensions were 9 cm wide x 12 cm long. The samples analysed were raw bagasse, steam-treated bagasse and steam-ammonia-treated bagasse; cotton wool (almost pure cellulose) was used as a fourth sample to verify if the microbial ecosystem was optimal for the digestion of a fibrous feed resource. For the tests, five grammes of air-dried sample, previously ground using a mill fitted with a 2.5 mm sieve, were placed in the bag, then the bags were carefully closed and fitted firmly to a pipe and introduced into the rumen of the animals. After 6 hours the first set of bags was taken out, washed thoroughly and squeezed under running water from a tap for 3 minutes and then were placed in an oven at 70 ^oC for 24 hours in order to obtain the dry matter content and weighed again. The same procedure was made for the sets of samples used to determine degradability after 24, 48 and 72 hours. The degradability was calculated by the difference of weight of the sample before and after incubation in the rumen.

For the statistical analysis of the data, in Experiments 1 and Experiment 2, the linear regression of live weight change on time was calculated for each animal, and these values were used in the analysis of variance. For Experiment 3, the analysis of variance were done with the data measured, since the degradabilities were analysed for each period of incubation separately.

The steam-treated sugar-cane bagasse was well consumed by all the animals on every treatment (Table 3). There were no problems of urea toxicity on any of the treatments.

There were no significant differences (P=0.92) in live weight change between the eight treatments evaluated; despite this, the treatment which displayed the lower live weight change (0.567 kg/d) was that with 3% of the live weight of Gliricidia sepium and 1.5 kg/day of urea/molasses liquid mixture, while the treatment supple- mented with 2% live weight of Gliricidia sepium and urea/molasses liquid mixture ad libitum was the one which produced the highest live weight change (0.753 kg/d). The analysis of variance for the same variable between the percentages of Gliricidia sepium did not show significant differences (P=0.40) for the two levels supplied (0.743 and 0.687 kg/d for 2% and 3% of the live weight levels respectively), nor between the four quantities of urea/molasses liquid mixture (P=0.76) supplemented to the animals (0.739, 0.651, 0.738 and 0.731 for 1, 1.5, 2 kg and ad libitum respectively).

The intake of steam-ammonia-treated sugar-cane bagasse recorded for the animals supplemented with 1.5% of Gliricidia sepium was similar to that obtained for the animals fed on steam-treated bagasse, while those fed on steam-ammonia-treated bagasse but supplemented with 1% of Gliricidia sepium showed a lower intake (Table 4). The steam-treated bagasse was well consumed by all the animals and values higher than 11 kg fresh weight/day were obtained on two of the treatments; the lowest intake of the bagasse treated in this way was never less than 9 kg/day. These values are higher to those reported by Joshie et al 1984 (4.2 kg/day) with a similar treatment of the sugar-cane bagasse.

Table 3: Summary of the results obtained for all the treatments in Experiment 1 (three animals per treatment for 141 days)

	-------------------------------------------------- Treatments --------------------------------------------------
Urea/molasses	1	1	1.5	1.5	2	2	a	a
Gliricidia	2%	3%	2%	3%	2%	3%	2%	3%

Liveweight (kg)
- Initial	204	177	219	273	225	234	255	239
- Final	310	281	321	357	329	337	363	333

*Daily gain	0.742	0.736	0.736	0.567	0.742	0.735	0.753	0.708
SD	±.002	±.179	±.115	±.081	±.074	±.079	±.092	±.113

Mean intake of fresh bagasse (kg/d)
	6.3	5.9	6.9	7.9	7.8	7.9	6.4	6.3

The analysis of variance for the live weight change between the treatments with steam-ammonia-treated bagasse and those with steam- treated bagasse showed significant differences (P=0.00) for the treatment of bagasse (0.297 and 0.608 kg/day for steam-ammonia and steam, respectively).

The analysis of variance of the effect on live weight change of the levels of Gliricidia (P=0.87) and the percentages of urea in the urea/molasses mixture (P=0.40) in the animals fed on steam-treated bagasse showed no significant differences between the two levels of Gliricidia (0.604 and 0.611 kg/day for 1% and 1.5% of animal live weight), the three percentages of urea in the urea/molasses mixture (0.635, 0.624 and 0.564 kg/day for 12%, 14% and 16% urea in urea/ molasses mixture), nor for the interaction level of Gliricidia x urea in urea/molasses mixture. There was a tendency for reduced gain when the level of urea in the molasses liquid mixture rose. The intake of urea/molasses liquid mixture was slightly different for each level of urea (12, 14 or 16%) used; the animals with the mixture at 12% of urea had an intake of 1.59 kg/day (0.190 kg of urea/day), while in those supplemented with the mixture at 14% the intake was 1.55 kg/day (0.217 kg of urea/day) and the animals with the highest percentage of urea in the molasses (16%) had an intake of the mixture equal to 1.31 kg/day (0.208 kg of urea/day); there was a tendency for the animals to reduce the intake of the mixture (but not of the urea) when the percentage of urea was increased (Table 4).

Table 4: Summary of the results of Experiment 2 (5 animals per treatment)

	---------------------------------------------------- Treatements ----------------------------------------------------
	Ammon.	Ammon.	12%u	12%u	14%u	14%u	16%u	16%u
Gliricidia	1%	1.5%	1%	1.5%	1%	1.5%	1%	1.5%

Liveweight (kg)
- Initial	203	292	255	238	274	218	272	268
- Final	233	324	356	356	389	315	368	369

*Daily gain	0.268	0.325	0.587	0.683	0.671	0.576	0.553	0.575
SD	±.029	±.061	±.078	±.034	±.029	±.055	±.064	±.055

No. days	113	113	169	169	169	169	169	169

Feed intake (fresh basis) (kg/d)
- Bagasse	7.0	9.8	9.8	9.7	11.6	9.1	11.4	10.6
- Urea/molasses	...	...	1.6	1.6	1.8	1.3	1.5	1.1

Table 5: Effect of supplementation with Gliricidia sepium and treatments of the sugar-cane bagasse on its degradability in vivo (Experiment 3)

Material in	---------- With Gliricidia ----------			---------- Without Gliricidia ----------
the bag	24h	48h	72h(a)	24h	48h	72h(a)

Raw bagasse	13.9	22.9	28	14.4	20.5	13
SD	±5.4	±2.5		±3.4	±3.3

Steam-treated bagasse	31.3	44.1	60	31.2	38.2	55
SD	±8.7	±6.7		±7.4	±13.5

Steam-ammonia-treated bagasse	19.8	32.8	51	16.1	26.0	39
SD	±4.8	±7.1		±2.5	±10.1

Cotton wool	34.7	62.7	86	33.3	63.8	74
SD	±3.9	±4.5		±7.3	±0.2

SE of mean	±6.0	±5.5		±5.6	±8.6

The rumen degradability of sugar-cane bagasse was improved after treatment with steam under pressure (Table 5). Subsequent treatment with ammonia surprisingly resulted in lower values of degradability than with steam treatment alone.

There were significant differences (P=0.04) in the degradability at 24 hours by the nylon bag technique between the bagasse treated with steam-ammonia (17.9%) and that treated with steam (31.2%); there were no statistically differences for the effect of supple- mentation (P=0.84), nor for the interaction sample x supplementa- tion. The degradability of the steam-ammonia-treated bagasse at this time of incubation was only slightly higher than that of the raw bagasse (14.2%).

The analysis of variance for the degradability at 48 hours reported significant differences (P=0.00) between the samples analysed, but no statistically differences (P=0.21) were found between the degra- dabilities of steam-ammonia (29.4%) and steam-treated bagasse (41.1%). Once again, there was no significant difference (P=0.71) between the treatments without (37.1%) and with Gliricidia (40.6%). The interaction sample x supplementation was not significant.

The results obtained in Experiment 1 suggested that supplementation with Gliricidia sepium could be made at a lower level than 2% of the live weight, since increasing it to a level of 3% did not show a better response in live weight change. This strategy would permit substantial savings in the cost of the final diet. In Experiment 2 lower levels of supplementation with Gliricidia sepium were evaluated and, although there were no significant differences between the two levels used, the live weight change (analysed for the animals fed on steam-treated bagasse) tended to be lower than that obtained with the levels of Gliricidia sepium used in experiment 1.

As stated in the Results section, there were no significant differences between the three percentages of urea in the urea/mo- lasses liquid mixture. However, the animals supplemented with the lower one (12%) reached the highest live weight change (0.635 kg/day), while those supplemented with the higher level (16%) obtained the lowest value for the same variable (0.564). The decision as to which is the best level to use must be made on the basis of the cost of the additional quantity of molasses which is consumed when the urea/molasses liquid mixture contains just 12% of urea. Despite the fact that the intake of urea/molasses liquid mixtures used in Experiment 2 was inversely related to the level of urea, the quantity of urea consumed was maintained more or less at the same value. This suggests that the intake of the mixture was regulated in accordance with the percentage of urea in the urea/molasses liquid mixture.

The low live weight change obtained for the animals fed with steam- ammonia-treated bagasse could be due to an increase in pH of the rumen and maybe a change in the proportion and/or number of the protozoal population, but since measurements were not made, these are just speculations that require further research. It has been stated by some researchers (e.g. Ibrahim and Pearce 1983b) that when physical and chemical treatments are combined, the time of addition of the chemical (before, during or after physical treatment) would play an important role in the composition of the final product; in some studies the same authors concluded that the best time for the treatment of bagasse with ammonium hydroxide was during steaming.

There are a lot of available feed resources which could be more effectively used as animal feeds. Some of the fibrous residues such as the sugar-cane bagasse require physical or chemical treatments to improve their utilization. More investigations must be carried out in order to determine the potential local resources which could be incorporated in the diets of animals. Effort must be focussed on the best way to use them, from both a technical and economic point of view, appropiate dietary levels and the necessary physical or chemical treatments. The live weight change recorded for the animals fed with the steam-treated sugar-cane bagasse was greater than that obtained in Colombia with cattle on pastures.

The sugar-cane factories have a surplus of bagasse and in some cases the accumulation of this by-product has become a problem. The sugar-cane bagasse when treated, in order to improve its degradability, is a potentially good resource to use as the basal diet in the feeding of cattle and ruminants in general. The treatment of bagasse with steam alone results in a big increment in the degradability of this by-product. This type of treatment is relatively easy to apply in the sugar-cane factories since they use steam in the process of extraction of sugar from the sugar-cane. The subsequent treatment with ammonia of the steam-treated bagasse did not improve its degradability but resulted in lower values. Investigation must be made to try to identify the reason for this detrimental effect.

The supplementation of diets based on treated sugar-cane bagasse with urea/molasses liquid mixture can be made with levels of urea in that mixture as high as 16% (or possibly higher) without problems of toxicity in the animals. This method resulted in savings in time and money, since the intake of molasses, which is used mainly as a vehicle for the urea, is reduced.

In Colombia there are a lot of small farmers who use sugar-cane not for the production of sugar but for the production of panela (a brown solid block of sugar). Their income and standard of living could be improved if alternative methods for using sugar-cane bagasse are developed; this system must be investigated in order to find better small-scale treatments, which may be chemical methods, since the small farmer does not have steam available. Supplementa- tion with Gliricidia sepium will not be a problem for them because this legume is widely distributed throughout rural areas of the country.

Al-Ani F and Smith J E 1988 Effect of chemical pretreatment on the fermentation and ultimate degradability of bagasse by Phanerochaete chrysosporium. Journal of the Science of Food and Agriculture (42)1:19-28

Baggio A J 1984 Posibilidades de Gliricidia sepium (Jacq.) Steud, para uso en sistemas agroforestales en Brasil. Pesquisa Agropecuaria Brasileira. Brasilia pp 241-243

Boucqué Ch V and Fiems L O 1988 Vegetable by-products of agro- industrial origin. Livestock Production Science (19)1/2:97-135

Crosthwaite C, Ishihara M and Richards G N 1984 Acid ageing of lignocellulosics to improve ruminant digestibility - Applica- tion to bagasse, wheat and rice straw and oat hulls. Journal of the Science of Food and Agriculture (35)10:1041-1050

Elliot R, Ferreiro H M, Priego A and Preston T R 1978 Rice polishings as a supplement in sugar-cane diets: The quantities of starch (glucose polymers) entering the proximal duodenum. Tropical Animal Production 3:30-35

Fan L T, Lee Y and Gharpuray M M 1982 The nature of lignocellulosics and their pretreatments for enzymatic hydrolysis. Advances in Biochemical Engineering (23):157-187

Gould J M, Jasberg B K, Fahey Jr. G C and Berger L L 1989 Treatment of wheat straw with alkaline hydrogen peroxide in a modified extruder. Biotechnology and Bioengineering (33)2:233-236

Hamad M A and El-Saied H 1982 The ammoniation of agricultural residues. Journal of the Science of Food and Agriculture (33)3:253-254

Helmling O, Arnold G, Rzehak H, Fahey Jr. G C, Berger L L and Merchen N R 1989 Improving the nutritive value of lignocellulosics: the synergistic effects between alkaline hydrogen peroxide and extrusion treatments. Biotechnology and Bioengineering (33)2:237-241

Ibrahim M N M and Pearce G R 1983a Effects of chemical pretreatments on the composition and in vitro digestibility of crop by-products. Agricultural Wastes (5)3:135-156

Ibrahim M N M and Pearce G R 1983b Effects of chemical treatments combined with high-pressure steaming on the chemical composition and in vitro digestibility of crop by-products. Agricultural Wastes (7)4:235-250

Kewalramani N, Kamra D N, Lall D and Pathak N N 1988 Bioconversion of sugar-cane bagasse with white rot fungi. Biotechnology letters (10)5:369-372

Kling S H, Neto C C, Ferrara M A, Torres J C R, Magalhaes D B and Ryu D D Y 1987 Enhancement of enzymatic hydrolysis of sugar- cane bagasse by steam explosion pretreatment. Biotechnology and Bioengineering (29)8:1035-1039

Leng R A 1989 Livestock Feed Resources and Constraints to their utilization in Tropical Developing Countries. Integration of Livestock with Crops in Response to Increasing Population Pressure on Available Resources. Ed: T.R. Preston. Centre for Technical Cooperation in Agriculture (CTA), Wageningen. [In Press].

Morjanoff P J and Gray P P 1987 Optimization of steam explosion as a method for increasing susceptibility of sugar-cane bagasse to enzymatic saccharification. Biotechnology and Bioengineering (29)6:733-741

Orskov E R, Hovell F D DeB and Mould F 1980 The use of the nylon bag technique for the evaluation of feedstuffs. Tropical Animal Production (5):195-213

Pate F M 1982 Value of treating bagasse with steam under pressure for cattle feed. Tropical Agriculture (59)4:293-297

Preston T R 1987 Informe técnico de las actividades de CIPAV 1986- 1987 Convenio Inter-Institucional para la Producción Agropecuaria en el Valle del rio Cauca, Cali, Colombia

Preston T R and Leng R A 1984 Supplementation of diets based on fibrous residues and by-products. In: Straw and Other Fibrous By-products as Feed [F Sundstl and E C Owen, Eds] Elsevier Press, Amsterdam pp. 373-413

Preston T R and Leng R A 1987 Matching Ruminant Production Systems with Available Resources in the Tropics and Subtropics. PENAMBUL Books Ltd: Armidale NSW, Australia

Rivers D B 1988 Factors affecting the enzymatic hydrolysis of bagasse and rice straw. Biological Wastes (26)2:85-95

Rolz C, de Arriola M C, Valladares J and de Cabrera S 1987 Effects of some physical and chemical pretreatments on the composition, enzymatic hydrolysis and digestibility of lignocellulosic sugar-cane residue. Process Biochemistry (22)1:17-23

Sundstol F 1988 Straw and other fibrous by-products. Livestock Production Science (19)1/2:137-158

Wong You Cheong Y, d'Espaignet J T, Deville P J, Sansoucy R and Preston T R 1974 The effect of steam treatment on cane bagasse in relation to its digestibility and furfural production. Proceedings of the 15th Congress of ISSCT (South Africa).