A.A.O. Aboud1, E. Owen2, J.D. Reed3 and A.N. Said4
1Department of Animal Science, Sokoine University of Agriculture, P.O. Box 3004, Morogoro, Tanzania2Department of Agriculture, University of Reading, Early Gate, RG 6 2AT, Reading, UK
3University of Wisconsin, Department of Animal Science, USA
4International Livestock Centre for Africa. P.O. Box 5689, Addis Ababa, Ethiopia
Abstract
Introduction
Material and methods
Results
Discussion
Conclusion
References
Comparisons were made between 24 Ethiopian rams and 24 Galla goats for voluntary intake, feed selectivity and growth performance when offered three levels of sorghum stover. The levels observed were 25,50 and 75 g/kg live weight per day. Measurements of intake, growth and feed selectivity were taken over a period of 10 weeks.
Both sheep and goats consumed more stover (P<0.001) as rate of offer was increased (22, 31, 32 gDM/kg live weight per day; 19,26,29 g/kg live weight per day, respectively, for sheep and goats). Increasing the level of offer beyond 50g/kg live weight per day did not, however, promote a substantial increase in intake. No difference (P>0.05) was observed in the pattern of feed selectivity between sheep and goats. Both species showed a similar preference for leaf and sheath matter as the level of offer was increased.
Sheep gained weight faster (P<0.001) than goats at all rates of offer, (28.0 vs 9.0; 54.0 vs 23.0; 62.0 vs 31.6 in favour of sheep). Those on a higher level of offer showed a faster rate of gain than those on a lower level of offer.
It was concluded that both species can benefit from generous offers and that the benefit, judged by improved intake and growth, is due to selective feeding.
Effet de l'offre alimentaire sur la croissance, la consommation et les choix alimentaires chez les ovins et les caprins
Résumé
L'ingestion volontaire, les choix alimentaires et les performances de croissance ont été comparés chez 24 béliers éthiopiens et 24 chèvres Galla auxquels de la paille de sorgho avait été servie selon trois niveaux de rationnement (25, 50 et 75 g/kg de poids vif/j). Les mesures de l'ingestion, de la croissance et de la sélectivité à l'égard des aliments ont été réalisées sur une période de 10 semaines.
La consommation de paille augmentait (P<0,001) avec l'augmentation du niveau de rationnement (22, 31 et 32 g de MS/kg de poids vif/j chez le mouton et 19, 26 et 29 g de MS/kg de poids vif/j chez la chèvre). Toutefois, aucune augmentation substantielle de l'ingestion n'a été obtenue lorsque le niveau de rationnement était amené à plus de 50 g/kg. Il n'y avait pas de différence (P>0,05) entre les choix alimentaires des chèvres et des moutons. Les deux espèces consommaient de préférence les feuilles et les gaines foliaires dans la mesure où les apports alimentaires augmentaient.
Les moutons réalisaient de meilleurs (P<0,001) gains de poids que les chèvres, quels qu'aient été les niveaux de rationnement (28,0 contre 9,0, 54,0 contre 23,0, et 62,0 contre 31,6, au bénéfice des ovins). Les gains de poids étaient plus rapides chez les animaux soumis à un niveau d'alimentation élevé que chez ceux soumis à un niveau d'alimentation modéré.
En conclusion, il apparaît que les deux espèces sont susceptibles de tirer parti d'un apport alimentaire généreux et, qu'à en juger d'après l'amélioration de la croissance et de l'ingestion, les avantages retirés s'expliquent par la sélection des aliments.
Previous studies (Aboud et al, 1992) clearly indicate the value of selective feeding. Sheep were shown to perform better and consume more digestible matter through selective intake of leaf and sheath fractions of sorghum stover as a result of liberal offers. This observation is of practical value for small ruminants in the tropics because the traditional feeding systems depend largely on crop residues. However, there is little evidence in the literature that the approach adopted by Aboud et al (1992) would be as effective for goats as it was for sheep. Results from comparative studies between sheep and goats of selective efficiency are generally conflicting (Huston, 1978; Devendra and Bums, 1983). Goats in the tropics are more able than sheep to consume feed fractions of higher nutritional quality when offered the opportunity to eat selectively (Devendra and Burns, 1983; Hoppe et al, 1977; Huston, 1978). This observation does not agree with studies carried out in temperate countries, where sheep have been shown to perform better than goats under most practical feeding systems. A comparative feeding and performance study on Small East African goats and the Red Maasai sheep in Tanzania (Shoo, 1986) was inconclusive, although sheep appeared to perform better than goats when offered a basal diet of Chloris gayana hay with Leucaena as a supplementary forage. However, observations in Shoo's (1986) study did not include an assessment of selective feeding ability between the two species. The inclusion of a supplementary forage may have confounded the results, as it is known that goats would preferentially consume browse species (Lu, 1988). Most of the comparative studies in the tropics were made under grazing conditions, whereas those in the temperate countries involved comparisons under stall-feeding conditions in which feed was offered at restricted levels, usually at 35 g dry matter M(D)/kg live weight per day or to achieve 10-20% rate of refusal. Comparative studies under grazing or in stalls at restricted levels of offer may give misleading results (Demment and Van Soest, 1983). Restricted levels of offer do not usually provide sufficient opportunity for selective feeding (Aboud et al, 1993). This is particularly serious where low-quality crop residues are used in comparisons.
No performance comparison has been made between sheep and goats that have been fed untreated crop residues at levels that would allow selective intake. Alimon (1989) only measured the digestible organic matter (OM) intake of goats offered increasing levels of feed, but did not measure any related growth performance. Aboud et al (1993) showed that certain levels of offer encouraged sheep to eat selectively. By offering goats and sheep the same type of sorghum stover at these levels both species may perform similarly.
The experiment was conducted at the Debre Zeit Experimental Station, an ILCA sub-station located about 1700 metres above sea level, 45 km south-east of Addis Ababa, Ethiopia.
A total of 48 animals were used in this study. Half were Ethiopian lowlands hair-type rams and half were Galla goats. The animals were between 11-13 months old and weighed between 14-16 kg. They were divided into three groups of 16 animals, eight rams and eight goats in each group. The three groups were allocated to three different levels of offer, which were 25,50 and 75 g stover DM/kg0.75 daily.
Sorghum stover (cv. Seredo: a bird resistant variety) was used as the basal diet. In addition, the animals also received 150 g/day of solvent extracted cottonseed cake (Table 1) and mineral lick (Table 2). The animals were fed twice a day. On each occasion half of the daily ration was provided after collection of refusals. The experimental period lasted 10 weeks after an initial adaptation period of 21 days.
Table 1. Composition of feedstuffs utilised during the adaptation and trial periods (g/kg).
|
Composition* |
Chloris gayana |
Wheat bran |
Seredo stover |
Cottonseed cake |
|
|
Sample 1 |
Sample 2 |
||||
|
DM |
870 |
876 |
913 |
923 |
932 |
|
CP |
37 |
169 |
46 |
427 |
286 |
|
CF |
37 |
113 |
- |
- |
- |
|
Ash |
87 |
64 |
9.4 |
67 |
70 |
|
NFE |
444 |
616 |
- |
- |
- |
|
NDF |
- |
- |
697 |
776 |
452 |
|
ADF |
- |
- |
452 |
293 |
376 |
|
Lignin |
- |
- |
444 |
88 |
92 |
* DM = dry matter; CP = crude protein; CF = crude fibre; NFE = nitrogen-free extract;
NDF = neutral detergent fibre; ADF = acid detergent fibre.
Table 2. Composition of mineral block
|
DM |
(g/kg) |
958 |
|
Phosphorus |
" |
43 |
|
Potassium |
" |
11 |
|
Calcium |
" |
71 |
|
Magnesium |
" |
4 |
|
Sulphur |
" |
7 |
|
Iron |
Mg/kg |
74.89 |
|
Manganese |
" |
79.31 |
|
Copper |
" |
197.40 |
|
Zinc |
" |
148.77 |
Sampling and measurement of feed intake
Before each meal, the refusals from previous meals were removed from troughs and weighed. The difference between amount offered and amount refused was taken as the quantity consumed.
For each animal, a handful of the stover refused during the morning and evening feedings was collected to make a single daily sample. Each daily sample was added to the previous day's sample for one week. At the end of each week, the combined sample was divided into two parts, one for chemical analysis and the other for botanical fractionation. The parts retained for chemical analysis from each individual were then mixed by treatment group and sub-sampled to form a single sample per week from each treatment group. Altogether 10 such samples were collected from each treatment group during the experiment.
Botanical fractionation was based on the weekly pooled samples for each animal. Proportions of botanical components (leaf, sheath and stem) were expressed in g/kg of sample and averaged for each treatment group per week. This provided 10 such averaged values for the experimental period, one per week based on means from eight animals per treatment. Botanical fractions were partitioned using the traditional 'winnowing' process. This involved shaking the sample in a gentle circular motion using a 'safed', a reed woven mat about 1 m across. This trapped the heavier material (stem and sheath) in grooves while the lighter material (leaf) was shaken towards the edge. The fractions were then picked by hand.
Growth performance and measurements
Once a week the animals were weighed, on each occasion at 0700 hours. Prior to weighing, water and feed were withdrawn for at least six hours. A canvas sling hung on a spring balance was used for weighing.
Chemical analyses
Chemical analyses for dry matter (DM), acid detergent fibre (ADF), neutral detergent fibre (NDF), nitrogen (N), ash, silica, ADF-ash and in vitro dry matter digestibility (DMD) were done in duplicate on all samples collected. The methods of Goering and Van Soest (1975) were used for fibre analysis and for the in vitro digestibility. Results from the in vitro digestion were expressed as the residual amount of NDF remaining after 48 hours incubation. N was determined using the macro-Kjeldahl procedure.
Intake of stover
Table 3 shows the main effect means for intake of stover by sheep and goats. Analysis of variance showed no significant interaction between the main effects (species and level of offer) on stover intake. Effects of species and level of offer were highly significant (P<0.001). Sheep consumed significantly (P<0.001) more stover than goats. Both species improved their DM intake as the level of offer was raised from 25 to 75 g/kg live weight per day. The improvement was particularly large between animals offered 25 and 50 g/kg live weight per day. Increasing the offer to 75 g/kg live weight per day resulted in a small increase in intake.
Rate of refusals and selectivity
Effects of level of offer and of differences between the species were highly significant (P<0.001). Figure 1 shows a comparison between the species and effect of level of offer on the rate of stover refusal. Rate of refusal was expressed as rate of utilisation (1-rate of refusal) to show the fall in the proportion of stover utilised as the amount offered was raised. Goats have lower utilisation values because of their higher rates of refusals. Proportionately more stover was refused as the level of offer was raised.
Summaries of analysis of variance for the comparison of the stover leaf, sheath and stem proportions (g/kg) in the refusals are given in Table 4. For leaf matter proportions, no significant interaction between main effects was observed. Differences between species and effects of level of offer were both significant (P<0.01). The stover refused by goats had more leaf (P<0.05) than that left by sheep. In both species stover refused from higher rates of offer contained more (P<0.001) leaf than that refused at the 25 g/kg live weight per day level.
Table 3. Main effects means for stover intake, rate of stover refusal and growth rate (g/d).
Figure 1. Effect of level of offer on stover intake and rate of utilisation by sheep and goats.
Figure 2. Effect of level of offer on botanical composition of DMI by sheep and goats.
Significant interactions (P<0.001) between main effects were observed in the proportion of sheath and stem in the stover refused. Effect of level of offer and differences between species were highly significant (P<0.001). In goats, the proportion of sheath and stem in refused stover increased with increasing level of offer. Sheep had no consistent pattern. Table 5 shows the main effects means of intakes of the three botanical fractions (leaf, sheath and stem). Sheep and goats consumed about the same amount of leaf and sheath fractions. Sheep had a considerably higher intake of stem fractions than goats. More leaf and sheath were consumed by both species as the level of offer was raised (Figure 2). The amount of stem consumed, however, decreased with increasing levels of offer, although differences in intake at the higher levels were about the same.
Rate of growth
Sheep gained nearly twice as much as goats (Table 3). Increasing the level of offer significantly (P<0.001) improved the rate of gain in both species, particularly when the offer was raised from 25 g to 50g/kg live weight per day.
Effect of level of offer on stover intake and stover selectivity
The results show that sheep and goats respond in almost the same manner when offered increasing quantities of sorghum stover.
Similarly to previous findings (Owen and Aboud, 1988), the response to increasing level of offer appeared to diminish in both species as the level of daily offer was raised beyond 50 g/kg live weight per day. It also appears that both species optimised their DMI and growth performance at the same level of offer. The efficiency with which this was achieved was different between the species. Sheep showed a better utilisation rate than goats (P<0.01). The DMI achieved by both species at the 50g/kg live weight per day level of offer was close to the levels reported by Devendra and Burns (1983) for Baribari goats in India where the animals were fed forages ad libitum.
Most reports in the literature suggested that goats are more able than sheep to eat selectively (Devendra and Burns, 1983; Lu, 1988). These observations are largely based on grazing animals, most of them on natural rangelands (Lu, 1988). The findings in this study with sorghum stover showed that in general both species are able to eat selectively. The results also suggest that the differences in selection between the sheep and goats used were dependent on the level of feed on offer. The botanical and chemical composition of the stover refused showed that sheep tended not to select when offered stover at 25g/kg live weight per day, whereas goats were selective at all levels of offer. At 50g/kg live weight per day and 75g/kg live weight per day there were no substantial differences between the two species in selective feeding, both in ability or efficiency. The pattern of intake of botanical components was similar in the two species. Both showed a preference for leaf and sheath against stem. Figure 2 shows that the intake of leaf and sheath was increasing even at the highest level of offer. The two components were constituting an increasing proportion of the DMI as the level of offer was increased.
Table 4. Analysis of variance for comparison of stover refusals by sheep and goats.
Table 5. Main effects means for intake of stover fractions by sheep and goats.
|
|
Species |
Level of offer (g/kg live weight per day) |
||||
|
Goats |
Sheep |
25 |
50 |
75 |
||
|
Stover offered (g/d): |
||||||
|
Total amount |
731 |
752 |
352 |
757 |
1143 |
|
|
|
leaf |
61 |
63 |
29 |
63 |
95 |
|
|
sheath |
173 |
178 |
83 |
179 |
271 |
|
|
stem |
451 |
463 |
217 |
467 |
704 |
|
Stover refused (g/d): |
||||||
|
Total amount |
281 |
224 |
36 |
282 |
627 |
|
|
|
leaf |
0.8 |
0.3 |
0.04 |
0.5 |
2.1 |
|
|
sheath |
21.0 |
22.2 |
2.1 |
18.6 |
82.8 |
|
|
stem |
237 |
197 |
31 |
255 |
519 |
|
Intake of components (g/d): |
||||||
|
|
leaf |
60 |
63 |
29 |
63 |
93 |
|
|
sheath |
152 |
156 |
81 |
161 |
188 |
|
|
stem |
214 |
266 |
187 |
212 |
185 |
|
Component proportion (g/kg stover intake): |
||||||
|
|
leaf |
133 |
112 |
9 |
132 |
180 |
|
|
sheath |
338 |
295 |
256 |
338 |
364 |
|
|
stem |
475 |
503 |
591 |
446 |
358 |
Effect of level of offer on growth performance
For both sheep and goats, growth performance improved significantly as the rate of offer was increased. The improvement in performance was remarkable when the rate of offer was raised from 25g/kg live weight daily to 50g/kg live weight daily. Feed conversion efficiency also improved, particularly in sheep.
The difference in rate of gain between the two species cannot be explained by differences in their ability to eat selectively. Because both species were compared for their absolute gain rather than their relative gain, this may have favoured sheep due to their higher initial weight. It would have been preferable to express the rate of gain as a proportion of the expected mature weight.
The findings in this study clearly support earlier observations that generous offers of feed can promote intake both qualitatively and quantitatively. This possibility should offer opportunities for modifying current practices of supplementary feeding. Where restricted feeding is practised there is always the likelihood of underestimating the potential nutritive quality of forages, especially crop residues.
Aboud A A O. Owen E, Reed J D and Said A N. 1992. Influence of stover variety and level of offer on voluntary intake, feed selectivity and growth performance of Ethiopian rams: Perspectives on practical problems. A paper presented at the 19th Scientific Conference of Tanzania Society of Animal Production.
Aboud A A O. Owen E, Reed J D, Said A N and McAllan A B.1993. Feeding sorghum stover to Ethiopian goats and sheep: effect of amount offered on intake, selection and performance. In: Gill M, Owen E, Pollott G E and Lawrence T L J (eds), Animal production in developing countries. Proceedings of a symposium organized by The British Society of Animal Production, held at Wye, September 1991. BSAP Occasional Publication 16. BSAP (British Society of Animal Production), Edinburgh, UK. pp. 202-203.
Alimon A R. 1989. An examination of strategies to maximize dietary selection for improving the intake and utilization of barley straw by goats. PhD thesis, University of Reading, Reading, UK. 348 pp.
Demment M W and Van Soest P J. 1983. Body size, digestive capacity and feeding strategies of herbivores. Winrock International Institute of Livestock Research and Training Center, Morrilton, Arkansas, USA. 66 pp.
Devendra C and Burns M. 1983. Goat production in the tropics. CAB (Commonwealth Agricultural Bureaux), Farnham House, Slough, UK. 183 pp.
Hoppe P P, Qvortrup S A and Woodford M H. 1977. Rumen fermentation and food selection in East African sheep, goats, Thompson gazelle, Grant gazelle and Impala. Journal of Agricultural Science 89:129-135.
Huston J E. 1978. Forage utilization and nutrient requirements of the goat. Journal of Dairy Science 61:988-993.
Lu C D. 1988. Grazing behaviour and diet selection of goats. Small Ruminant Research 1:205-216.
Owen E and Aboud A A O.1988. Practical problems of feeding crop residues. In: Reed J D, Capper B S and Neate P J H (eds), Plant breeding and the nutritive value of crop residues. Proceedings of a Workshop held at ILCA, Addis Ababa, Ethiopia, 7-10 December 1987. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. pp. 133-155.
Shoo R A. 1986. Effect of supplementation with leucaena to sheep and goats on intake and growth performance. MSc Dissertation, Sokoine University of Agriculture, Morogoro, Tanzania.
