Effect of breed and age on the voluntary intake and the
micromineral status of non-pregnant sheep
1. Estimation of voluntary intake
A.H. RamõÂrez-PeÂrez
a, S.E. Buntinx
a,*, C. Tapia-RodrõÂguez
b, R. Rosiles
a aDepartamento de NutricioÂn, Facultad de Medicina Veterinaria y Zootecnia,Universidad Nacional AutoÂnoma de MeÂxico, Mexico City, D.F. 04510, Mexico
bCentro de EnsenÄanza, InvestigacioÂn y ExtensioÂn en ProduccioÂn Ovina, Facultad de Medicina Veterinaria y Zootecnia,
Universidad Nacional AutoÂnoma de MeÂxico, Mexico City, D.F. 04510, Mexico
Received 9 April 1999; accepted 1 November 1999
Abstract
Twenty-four sheep were used to test breed and age differences in voluntary intake (VI). The sheep were divided into four groups of six animals each: Suffolk mature ewes (SM), Suffolk yearlings (SY), Rambouillet mature ewes (RM), and Rambouillet yearlings (RY). The animals grazed alternatively two mixed pastures containingTrifolium repens latum,Lolium perenne and Pennisetum clandestinumfrom 07:00 to 16:30 hours and remained penned overnight. Voluntary intake was estimated using chromic sesquioxide (1 g per sheep) as an external marker, administered in two gelatin capsules, one before and one after grazing, during 65 days. Faecal grab samples were collected directly from each animal for three consecutive days beginning on Day 7 of the experiment and at 14-day intervals until Day 65 (sampling times5). Hand-plucked samples of forage were used to estimate in vitro dry matter digestibilities (IVDMD), which were used with estimates of fecal output to calculate intakes (per animal and per kg of metabolic body weight (kg0.75)). The data were analyzed as a completely randomized design in a factorial arrangement with repeated measurements. Voluntary intake, expressed both per animal and per kg0.75, showed a breedage interaction (P<0.001). Suffolk mature ewes had the greatest daily VI (1.88 versus 1.26 kg DM/animal for the other three groups) because they were the heaviest animals, but their VI/kg0.75(0.071 kg DM/day) was not different (P>0.05) from that of RY (0.063 kg DM/day), which were smaller, younger animals. Suffolk yearlings and RM had similar VI/kg0.75(0.057 and 0.051 kg DM/day, respectively), not different from that of the RY. Intakes showed a cubic trend (P<0.001) with time, possibly because of changes in IVDMD and climatic conditions. The lower intake, yet better gains of RM and RY indicate a greater ability to thrive under conditions of average pasture quality. High forage availability did not result in greater forage consumption. The information on VI was later used to assess the micromineral status of the sheep in the experiment.#2000 Elsevier Science B.V. All rights reserved.
Keywords:Voluntary intake; Suffolk; Rambouillet; Ewes; Yearlings
1. Introduction
In Mexico, sheep production generally occurs under grazing conditions, and the animals depend almost *Corresponding author. Tel.:525-6225907;
fax:525-6225906.
E-mail address: sbuntinx@servidor.unam.mx (S.E. Buntinx)
exclusively on the available forage to meet their nutritional needs. Occasionally, improved pastures are grazed, but native species are the most common resource used by sheep. The performance of sheep under these conditions is in¯uenced by factors that affect forage production and quality (GonzaÂlez, 1992). Seasonal changes in forage growth may result in periodic suboptimal nutrition, which can be modi®ed by grazing management and environmental changes. Moreover, the animals' selective behavior, choosing one plant species or part over another, makes it very dif®cult to determine the overall nutritive value of the consumed forage (Allison, 1985; Van Houtert, 1996), which can be substantially better than that of the average forage on offer.
An area of special concern in sheep management is mineral nutrition because little or no supplementation is given to the animals in most of the sheep production regions in Mexico. Furthermore, there is very little information on the changing mineral needs of grazing sheep throughout their different production stages (growth, dry period, pregnancy and lactation). There-fore, recommendations for mineral feeding are dif®-cult. The mineral content of forage will depend largely on soil characteristics (Spears, 1994) and these will vary from region to region. Also, the quantity of forage consumed by the animal determines whether nutri-tional needs are met and thereby determines produc-tivity (Doyle et al., 1994). Estimating voluntary intake (VI) under grazing conditions is a cumbersome and laborious task. It can be accomplished in several ways, but the most common approach is estimating fecal output with an external marker, calculating intake using an estimate of digestibility (usually determined in vitro) (Burns et al., 1994).
The aim of this experiment was to estimate VI in dry, non-pregnant sheep of two breeds that were either mature or approaching maturity. This was part of a larger study involving changes in the micromineral status of sheep.
2. Material and methods
2.1. Location
The study was carried out at the Center for Teach-ing, Research and Extension in Ovine Production of
the School of Veterinary Medicine and Animal Hus-bandry of the University of Mexico. The center is located in the state of Morelos, at 108030N, 99
8140W
and 2810 m above sea level. The climate in the area is temperate with a cool summer; the mean annual temperature is 188C and the mean annual rainfall is 1724 mm (GarcõÂa, 1988). The experiment was con-ducted during the rainy season (July±September 1996) and the rainfall during those months was 308 mm.
2.2. Animals
Four groups with six, non-pregnant animals each were formed: Suffolk mature ewes (SM), Suffolk yearlings (SY), Rambouillet mature ewes (RM) and Rambouillet yearlings (RY). Mean body weights in kg (SE) were, respectively: 85.31.6, 68.81.7, 67.3 2.7, and 56.21.5. Mature ewes were, on average, 6 years old and yearlings, 16 months old. All animals were weighed before grazing, at 06:00 hours, at the beginning of the experiment (S0) and at Day 7 (S1), 21 (S2), 35 (S3), 49 (S4), and 63 (S5) of the experimental period. Because the animals received no supplemen-tation while indoors, they were weighed after a 13 h fast.
2.3. Grazing conditions
Before the start of the experiment, the animals were on a schedule of rotational grazing and were supplemented with oat hay and a concentrate (ground sorghum grain, citrus pulp, soybean meal, cane molasses, a mineral premix and sodium bicarbonate: CP15.9%; TDN83.1%). Seven days before the ®rst chromic oxide dosing the supplementation was stopped.
about 100% of the grazing hours, the sheep were moved.
The pastures had been sown with white clover (Trifolium repens latum), ryegrass (Lolium perenne) and kikuyo grass (Pennisetum clandestinum). Dry matter (DM) production was estimated at the begin-ning and at the end of each grazing period. The method used consisted of randomly throwing a metal square of 0.1 m2to the ground and the forage enclosed within the square was cut with scissors. Fifty samples were taken each time; they were placed in paper bags, weighed, and the proportion of grass (ryegrass and kikuyo), legume (white clover) and weeds present estimated. This was accomplished by separating the pasture components in different boxes and weighing each component. Weeds were considered to represent all plant material that was not ryegrass, kikuyo grass or white clover. There were mainly three weed spe-cies: cow's tongue (Ancusaspp.), dandelion ( Tarax-acum vulgare) and aPaspalumspecies. The samples were then oven-dried at 508C to determine DM content.
2.4. Chromic oxide dosing
Prior to the ®rst dosing of chromic oxide, fecal grab samples were collected from each animal and used as a blank (S0). All the animals received two gelatin capsules with 0.5 g of chromic oxide per day during the 65 days of the experiment, one in the morning (06:30 hours), before grazing, and one in the afternoon (17:00 hours), after grazing. Two fecal grab samples were taken from each animal on Days 7±9 (S1), 21±23 (S2), 35±37 (S3), 49±51 (S4) and 63±65 (S5) of the experiment, at 06:00 and 17:00 hours. These samples were kept separate by animal per day, frozen in sealed plastic bags and later oven-dried at 508C and ground. A composited sample (animal per sampling time) was obtained and Cr determined using the methodology of Fenton and Fenton (1979) and atomic absorption spectrophotometry.
2.5. Estimation of voluntary intake
Hand-plucked samples of forage were taken from sites a few centimeters away from where the animals were grazing (Burns et al., 1994). Samples were collected on Days 7, 21, 35, 49 and 63, three times
per day (08:00, 12:00 and 16:00 hours). The forage was quick-frozen in liquid N, stored at ÿ208C and later freeze-dried. In vitro dry matter digestibility (IVDMD) was determined with the Tilley and Terry technique (Tejada, 1992).
Once the concentration of Cr in feces and the IVDMD were obtained, VI was estimated using the following equations (Burns et al., 1994):
Fecal output g DM=day
dosed marker mg=g
concentration of marker in feces mg=g
VI g=day fecal output g DM=day 1ÿ IVDMD=100
2.6. Statistical analysis
The study was conducted as a completely rando-mized design in a factorial arrangement of treatments with repeated measurements, where the between sub-ject effects were breed, age and the interaction breedage. The within subject effects were sampling time (S1±S5), the interactions sampling timebreed and sampling timeage, and the three-way interac-tion. The experimental unit was the animal (n24). Body weight (BW) and VI data (both per animal and per metabolic weight (VI/kg0.75)) were analyzed using PROC GLM of SAS (1985) and the REPEATED statement. Comparisons among means were made using Fisher's protected lsd (Steel and Torrie, 1980). When the sampling time effect was signi®cant (P<0.05), an analysis of orthogonal polynomials (polynomial transformation) (SAS, 1989) was carried out in order to determine whether the trend was linear, quadratic, cubic, etc.
3. Results
3.1. Pasture conditions
The IVDMD of the forage for S1±S5 was, respec-tively: 54.8, 66.5, 64.8, 55.7, and 51.9%.
3.2. Body weight changes
The analysis of variance for BW indicated that the three-way interaction sampling timebreedage was signi®cant (P<0.001) and the polynomial transforma-tion revealed linear (P<0.01), quadratic (P<0.05) and cubic trends (P<0.001), which can be observed in Fig. 1. The four experimental groups lost weight from S0 to S1. Suffolk mature ewes continued to lose weight linearly until the end of the experiment, whereas RM recovered their initial weight by S3, losing some of it again from S4 to S5. The yearlings fared much better. Rambouillet yearlings gained weight from S1 to S5; SY gained weight from S2 to S4, but from S4 to S5 they lost some weight. The ®nal BW (kg) for SM, SY, RM, and RY were, respec-tively (meanSE): 70.70.4, 68.30.7, 65.51.4, and 58.31.3.
3.3. Individual voluntary intake and voluntary intake per kg of metabolic body weight
Although the analysis of variance for individual VI revealed both differences between breeds (P<0.001) and ages (P<0.01), the breedage interaction was signi®cant (P<0.001) and the means appear in Table 2. Suffolk mature ewes had the greatest DM intake (P<0.05), 1.880.04 kg/animal/day, and the DM intake of the other three groups did not differ (mean1.260.04 kg/animal/day). Voluntary intakes per sampling time were (0.03 SE) 1.29, 1.64, 1.59, 1.24, and 1.30 kg/animal/day for S1±S5, respectively, and the sampling time effect was signi®cant (P<0.001). The polynomial transformation indicated that the trend was cubic (P<0.001) (Fig. 2), intakes being highest in S2 and S3 and lowest in S1 and S4. There was no interaction between sampling time and the between subject effects.
The analysis of variance of VI/kg0.75 showed a breed effect (P<0.01), but again the breedage
inter-Table 1
Grazing periods, dry matter (DM) production, and botanical composition of pastures used in the experiment Pasture Days on pasture Before grazing After grazing
DM (t) Botanical
composition (G:L:W)a
DM (t) Botanical
composition (G:L:W) A 1±17 7.7 60.0:40.0:0.0 7.1 70.4:29.6:0.0 B 18±31 2.7 72.6:22.7:4.6 2.1 85.0:9.4:5.6 A 32±55 8.1 70.4:29.6:0.0 7.2 82.9:17.1:0.0 B 56±65 3.4 82.1:11.9:6.0 3.0 88.2:5.5:6.3
aG:L:Wpercentage of grass:legume:weeds.
Fig. 1. Body weight (BW) changes of non-pregnant Suffolk (S) and Rambouillet (R) mature ewes (M) and yearlings (Y) grazing a mixed temperate pasture (Sampling times (S0±S5)Days 1, 7, 21, 35, 49 and 63, respectively, of the experimental period.).
Table 2
Meanavoluntary intake (VI) and mean voluntary intake per kg of
metabolic body weight (VI/kg0.75) of grazing, non-pregnant Suffolk
(S) and Rambouillet (R) mature ewes (M) and yearlings (Y)
Item VI
(kg DM/day)
VI/kg0.75 (kg DM/day)
SM 1.88 b 0.071 b
SY 1.33 a 0.057 a
RM 1.17 a 0.051 a
RY 1.27 a 0.063 a,b
SE 0.04 0.002
aMeans with different letters in each column indicates statistical
action was signi®cant (P<0.001). Table 2 presents the means for each breed-age combination, and in this case, there was no difference in DM intake between SM and RY (0.071 and 0.0630.002 kg DM/kg0.75, respectively), but SM were different (P<0.05) from SY and RM (0.057 and 0.0510.002 kg DM/kg0.75, respectively). Voluntary intakes per kg of metabolic body weight for each sampling time (S1±S5) were (0.002 SE) 0.056, 0.071, 0.068, 0.053, and 0.056 kg/ day, respectively. The sampling time effect was sig-ni®cant (P<0.001), but its interactions with between subject effects were not. As in the previous analysis, the trend for the sampling time effect was cubic (P<0.01) and the shape of the curve was identical to that of Fig. 2.
4. Discussion
4.1. Pasture conditions
Dry matter production was very different for the two pastures. Pasture B had very little cover of rye-grass and white clover, which produce higher yields than kikuyo grass (Flores, 1990). The difference found in grass and legume proportions before and after each grazing period (Table 1) could have resulted from two factors. First, grasses are more resistant to trampling than legumes, and second, legumes, such as clover, are generally preferred over grasses by grazing ruminants
(Flores, 1990). The amount of weed in Pasture B (Table 1), on the other hand, did not change. This indicates that these weeds were not consumed by the animals and also resisted trampling.
During the ®rst grazing period of Pasture A, sheep only consumed 600 kg of DM out of the 7700 kg they had available (Table 1). As was explained above, sheep were moved from one pasture to the other when they appeared to spend most of their time in the pasture looking for food. With 7100 kg of DM still available forage was not limited. It was noted, how-ever, that the sheep spent the ®rst few days on Pasture A trampling it and refused to eat the trampled forage afterwards.
The IVDMD values obtained were low, considering that the experiment was carried out during the rainy season, when rapid plant growth occurs. Water, how-ever, by accelerating growth, is one of the factors that promotes plant maturity (Van Soest, 1982). This could help explain the low IVDMD values.
4.2. Body weight changes
The loss of weight shown by the four groups of animals from S0 to S1 (Fig. 1) could have resulted from the recent cessation of supplementation. New diets and diet schedules need longer adaptation times (Betteridge, 1986).
Suffolks require high diet quality (Botkin et al., 1988), which may explain the linear loss of weight by
the mature ewes throughout the experiment and the low growth rate of the yearling ewes (Fig. 1). It was expected that the yearlings of both breeds would gain weight because they were still growing. However, SY did not begin to gain weight until S2, 15 days later than the RYand when forage IVDMD was highest. Both the SY and SM showed weight loss from S4 to S5, when the animals were changed from Pasture A to Pasture B and IVDMD values were lowest. Rambouillets, on the other hand, are much hardier animals; they easily adapt to adverse conditions, including poor-quality diets (Botkin et al., 1988). The RM in this experiment managed to recuperate the weight lost from S0 to S1, but in the end, the low IVDMD of the forage from S4 to S5 prevented them from gaining more weight. The RY were the only group that accrued weight in a continuous manner despite the poor quality of the pasture.
4.3. Individual voluntary intake and voluntary intake per kg of metabolic body weight
According to the NRC (1985), the daily DM intakes of SM, SY, RM, and RY should have been 1.3, 1.5, 1.2, and 1.5 kg, respectively. However, the NRC does not make adjustments for breed or activity level. Because the Suffolk breed is larger and heavier than the Rambouillet breed and because DM intakes are related to body size (Illius and Gordon, 1999), DM intakes of Suffolks should be higher than those of Rambouillets. In this experiment, the mean daily DM intake of SM (1.88 kg), calculated with chromic oxide, was the highest one and apparently met the animals' DM requirement, but since the forage's quality was low, it is unlikely that the entirety of the animals' nutri-tional needs were satis®ed, as Fig. 1 makes clear. The estimated DM intake of RM appears to have fallen within the NRC's recommendation, but the estimated intake of the yealings for both breeds fell short of the 1.5 kg DM/day required (NRC, 1985). This appar-ently was not a problem for the RY because they continued to gain weight throughout the experiment, but the SY could not sustain their growth rate with an estimated mean intake of 1.3 kg DM/day.
A better way to express VI is VI/kg0.75, because energy requirements are proportional to metabolic BW (NRC, 1985). Thus, the energy requirements per kg0.75 of young animals are greater than those
of adult animals. Because their rumen is smaller, young animals meet their requirements by increasing their intake and passage rate (Allison, 1985). De Waal and Biel (1989) found differences among sheep breeds in VI/kg0.75. In this experiment, the VI/kg0.75of SM was different and higher than that of RM, but similar to that of RY, the smallest animals in the study. The estimated DM intakes per kg0.75of RM and SY were not different either. The DM intakes per kg0.75 esti-mated in this study were similar to those found by other authors with different breeds using alkane meth-odology and in vitro digestibility of organic matter (Champion et al., 1995; Friend et al., 1995).
The low DM intakes, either per animal or per kg0.75, observed in S1 could be explained by the change in the feeding program of the ewes, as has already been mentioned. By S2 the animals had adapted to their feeding situation and this plus the higher IVDMD values of the forage in S2 and S3 (66.5 and 64.8%, respectively) made it possible for the animals to obtain the highest DM intakes of the experiment from Days 21 to 35. From S3 to S4 intakes diminished, reaching the levels of S1. These variations in intake coincided with the changes in IVDMD observed in the forage. Voluntary intake is positively affected by increased forage digestibility because rate of passage is also increased. If, on the other hand, digestibility is low, it will take longer for forage particles to exit the rumen because of slower particle breakdown (Cheeke, 1991) and this will reduce VI. However, forage digestibility was not the only problem during S4 because IVDMD of forage during S5 was even lower than in S4 and nonetheless intakes went up a little (Fig. 2). For several days during S4, the climatic conditions were less than ideal, with heavy rains and overcast skies. It is known that daylight and humidity alter appetite and intake (Dulphy and Demarquilly, 1994).
5. Conclusions
Under the conditions of this study, breed and age differences were observed in voluntary intake. The ability of adult and yearling Rambouillet ewes to thrive under pasture conditions of average dry matter digestibility and low availability of legumes was noteworthy. Also, high forage availability does not necessarily mean greater forage consumption by the animals.
Acknowledgements
This study was made possible by a grant from the Consejo Nacional de Ciencia y TecnologõÂa (National Council for Science and Technology Ð CONACYT), project 1567P-B9507. The help from the staff at the Center is recognized and greatly appreciated.
References
Allison, C.D., 1985. Factors affecting forage intake by range ruminants: a review. J. Range Manage. 38, 305±311. Betteridge, K.A., 1986. A survey of the phosphorus content of
pastures and the serum inorganic phosphorus content of dairy cows. N. Z. Vet. J. 34, 22±26.
Botkin, M.P., Field, R.A., Johnson, C.L., 1988. Sheep and Wool Science, Production, and Management. Prentice-Hall, Engle-wood Chffs, NJ, 451 pp.
Burns, J.C., Pond, K.R., Fisher, D.S., 1994. Measurement of forage intake. In: Fahey Jr., G.C. (Ed.), Forage Quality, Evaluation, and Utilization, ASA, CSSA, SSSA. Madison, WI, pp. 494± 532.
Champion, S.C., Robards, G.E., Lindsay, A.R., Friend, M.A., 1995. Alkane technology estimates of summer and winter herbage intake of four specialty carpet wool breeds, Romneys and Merinos. Proc. N. Z. Soc. Anim. Prod. 55, 130±132. Cheeke, P.R., 1991. Applied Animal Nutrition. Feeds and Feeding.
McMillian Publishing Company, USA, 504 pp.
De Waal, H.O., Biel, L.C., 1989. Supplementation of lactating Dorper and Merino ewes on Themeda cymbopogon veld. 2. Diet quality and feed intake. S. Afr. Tydskr. Veek. 19, 148±155.
Doyle, P.T., Casson, T., Cransberg, L., Rowe, J.B., 1994. Faecal output of grazing sheep measured by total collection or using chromium sesquioxide. Small Rumin. Res. 13, 231±236. Dulphy, J.P., Demarquilly, C., 1994. The regulation and prediction
of feed intake in ruminants in relation to feed characteristics. Livest. Prod. Sci. 39, 1±12.
Fenton, T.W., Fenton, M., 1979. An improved procedure for the determination of chromic oxide in feed and feces. Can. J. Anim. Sci. 59, 631±634.
Flores, J.A., 1990. BromatologõÂa Animal. Limusa, MeÂxico, 930 pp. Friend, M.A., Robards, G.E., Lindsay, A.R., Champion, S.C., 1995. The relative intake of three Merino strains under different grazing regimes estimated using alkane technology. Proc. N. Z. Soc. Anim. Prod. 55, 127±129.
GarcõÂa, E., 1988. Modi®caciones to KoÈppen's system of climate classi®cation. 4th edn. IndianaÂpolis, MeÂxico, 217 pp. (in Spanish).
GonzaÂlez, M.S., 1992. Compensatory growth in sheep. In: Colegio de Postgraduados (Editor), Proc. International Seminar: Advances in Sheep Production. Montecillo, MeÂxico, pp. 44± 72 (in Spanish).
Illius, A.W., Gordon, I.J., 1999. The physiological ecology of mammalian herbivory. In: Jung, H-J.G., Fahey Jr., G.C. (Eds.), Nutritional Ecology of Herbivores. Proc. 5th Int. Symp. Nut. of Herbivores. San Antonio, Texas, USA, pp. 71±96.
NRC (National Research Council), 1985. Nutrient Requirements of Sheep, 6th Edition. National Academy Press, Washington, DC, 99 pp.
SAS Institute Inc., 1985. SAS/STAT, User's Guide, Version 5. SAS Institute Inc., Cary, NC.
SAS Institute Inc., 1989. SAS/STAT, User's Guide, Version 6. 4th Edition, Vol. 2. SAS Institute Inc., Cary, NC, pp. 951±957. Spears, J.W., 1994. Minerals in forages. In: Fahey Jr., G.C. (Ed.),
Forage Quality, Evaluation, and Utilization, ASA, CSSA, SSSA. Madison, WI, pp. 281±317.
Steel, R.G.D., Torrie, J.H., 1980. Principles and Procedures of Statistics, 2nd Edition. McGraw-Hill, New York, 633 pp. Tejada, I., 1992. Quality control and feed analysis for animals.
Continuing Education System in Animal Production. A.C. MeÂxico, D.F., 387 pp. (in Spanish)
Van Houtert, M.F.J., 1996. Determination of the quantity and quality of feed consumed by grazing ruminants. In: Continuing Education Division of the School of Veterinary Medicine and Animal Husbandry of the National Autonomous University of Mexico. Proc. of the course: Nutritional Aspects of Double-Purpose Cattle in the Tropics, Veracruz, MeÂxico, pp. 67±75 (in Spanish).
