| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Texas Agric. Exp. Stn., 1619 Garner Field Road, Uvalde, TX 78801-6205
* Corresponding author (h-lippke{at}tamu.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONESTIMATING INTAKECONCLUSIONSREFERENCES |
|---|
Abbreviations: CRD, controlled release device • DDM, digestible dry matter • IVDMD, in vitro dry matter disappearance
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONESTIMATING INTAKECONCLUSIONSREFERENCES |
|---|
|
ESTIMATING INTAKE |
|---|
|
TOPABSTRACTINTRODUCTIONESTIMATING INTAKECONCLUSIONSREFERENCES |
|---|
 | [1] |
 | [2] |
Since estimations of fecal output and, to a lesser extent, DDM have predominantly been based on the use of dietary markers, this paper will deal primarily with developments in dietary marker techniques since the thorough review of that topic by Kotb and Luckey (1972).
Estimation of Fecal Output
The possibility or necessity of total collection of feces from grazing animals, that is, measurement instead of estimation, should be considered at the outset, but the procedure has disadvantages. The labor requirement is high, not only for the actual collection, but also for the required training and maintenance of a group of suitably docile animals. Acquiring the animals once or twice daily to empty fecal collection bags has the potential to alter grazing behavior and, therefore, intake. Feces may escape the collection bag. Except by chance observation, the magnitude of these errors is unknown and unknowable. Hatfield et al. (1993) concluded that for wethers the collection bags per se posed no added stress, but their data suggest that the twice daily acquisition and restraint of the animals did impose an added level of stress. Weight and composition of the collected feces make the procedure more suited to sheep and goats than to cattle. However, refinements in harnesses and techniques have been numerous and have encouraged the continued use of this procedure (Ingleton, 1971; Johnson et al., 1998). Furthermore, as pointed out by Adams et al. (1991) and Momont et al. (1994), total fecal collection may be necessary with at least a few animals in order to monitor the recovery rate of the marker being administered to all animals in experiments where estimates of intake are needed in absolute terms.
According to Kotb and Luckey (1972), fecal markers should ideally "be inert with no toxic, physiological or psychological effects, be neither absorbed nor metabolised within the alimentary tract and therefore be completely recovered from either raw or processed food; have no appreciable bulk; mix intimately with the usual food and remain uniformly distributed in the digesta; have no influence on alimentary secretion, digestion, absorption, normal motility of the digestive tract or excretion; have no influence on the microflora of the alimentary tract which is of significance to the host; have qualities that allow ready, precise quantitative measurements; (and) have physical-chemical properties which make it discernible throughout the digestive process." Much of the research with fecal marker techniques has focused on ways to remedy failures to meet these requirements, particularly those for complete recovery and ease and precision of measurement.
External Markers
Functionally, markers are classified as being internal (endogenous) to the feedstuff or external, that is, added to the feedstuff or dosed separately to the animal (i) in a single large pulse at the beginning of a trial, (ii) in uniform daily pulses, or (iii) from a controlled release device (CRD) that is continually active throughout the trial. If the external marker has a constant and known recovery rate and if it is uniformly distributed in the feces, then the concentration of the marker in a fecal sample taken at any time during a dose step can be used to estimate fecal output according to the equation:
 | [3] |
However, one can rarely assume uniform marker distribution with any type of pulse dosing, and fecal sampling at least twice daily to assess the variance in marker concentration must be accommodated in most experiments. As noted earlier, marker recovery rate may also need to be measured rather than being assumed if a relative ranking of treatment effects on intake is not sufficient for the objectives of the experiment.
Chromium sesquioxide (Cr2O3), also called chromic oxide, has been the most widely used external marker for estimation of intake during the last 50 yr (Kotb and Luckey, 1972; Adams et al., 1991; Soder et al., 1995; Huston et al., 1999). The primary problem with this marker is that it moves through the digestive tract independently of undigested particles of the diet, and fecal concentrations of Cr2O3 may, consequently, exhibit strong diurnal variation. Dosing twice daily has generally been considered a minimum. Some research has shown that dosing six times daily greatly reduces diurnal variation, but such a regime is considered impractical in any situation. The accepted solution to this problem has been to find, during a pretrial period, a couple of sampling times in the diurnal pattern when the fecal concentration of Cr2O3 was equal to the concentration of Cr2O3 for the daily total fecal collection. This problem with diurnal variation has led to the development and commercial marketing of a CRD containing Cr2O3, which has also been widely used. Although diurnal variation is greatly reduced with the use of a CRD, the release rate of these devices appears to be diet dependent and must be verified with at least a few animals in each experiment where intake estimates in absolute terms are needed. Earlier problems with Cr2O3 analysis have also been resolved with the development of relatively benign sample preparation procedures prior to determination by atomic absorption spectrophotometry.
Whereas Cr2O3 has been the most prominent external marker used for daily dosing and in CRDs, the rare earth elements, primarily Yb, have been most commonly used in the single pulse dose technique. These elements bind with varying affinity to organic matter, particularly to forage fiber. This characteristic is used to obtain information about the flow dynamics of feedstuffs through the digestive tract of ruminants, while also obtaining an estimate of fecal output (Ellis et al., 1994). Carrier material is prepared by extracting and discarding the soluble carbohydrates. The rare earth element is applied, and the excess is then rinsed from the carrier material so that the final product will resist displacement of the rare earth element from its binding sites on the carrier when placed in the rumen environment. The quantity of marker that can be bound in this way is relatively small (<30 mg g-1), necessitating sensitive analysis techniques such as neutron activation analysis (Pond et al., 1985), plasma emission spectroscopy (Combs and Satter, 1992), or flameless atomic absorption spectrophotometry to avoid dosing large amounts of marked material.
Fecal sampling is generally intensified around the time of anticipated peak marker concentration following administration of a single pulse dose. Depending on rate of passage of undigested residues, first appearance of marker in feces is commonly 6 to 15 h after dosing. Additional samples (total samples per animal is 10–15) are taken for another 3 d to provide an adequate description of the shape and magnitude of the marker excretion curve. With a sufficient number of observers, each defecation can be sampled from the ground, avoiding undue disturbance to grazing animals during the period of intensive sampling (Lippke et al., 2000).
Ellis et al. (1994) presented a series of models that may be used as the basis for nonlinear least squares analysis of the data from fecal concentrations of pulse dosed markers. The model apparently most appropriate for grazing animals describes two functional compartments, one with time dependent rates of passage and a second with exponential rates of passage, and a time delay usually associated with linear flow (Ellis et al., 1994; Lippke et al., 2000). Fitting the data to this model provides estimates for marker concentration at the moment of dosing, passage rates from each digestive compartment, and time delay. Calculation then yields the estimate:
 | [4] |
Estimates of mean residence times for undigested particles in each compartment and for the entire digestive system may also be calculated. With the addition of dry matter disappearance rates from in vitro fermentation studies, these data may be used in a deterministic model of digestive dynamics to validate our understanding of the interaction of system components (Lippke et al., 2000).
Estimation of Digestible Dry Matter
Estimates of DDM are frequently taken from procedures that measure the disappearance of dry matter during in vitro fermentation (IVDMD) of a sample of the sward. These procedures are commonly based on the method of Tilley and Terry (1963) and on that of Goering and Van Soest (1970). Samples from forages of known DDM must be run simultaneously with the samples in question to have an appropriate correction factor for the raw IVDMD values. However, this correction cannot properly account for variations in DDM caused by the level of intake and for variations in rate of passage through the digestive tract, and consequently in DDM, caused by physical form of the forage, that is, fresh vs. dried or ensiled. These systematic errors from estimates based on IVDMD justify extensive efforts to develop precise and accurate procedures based on internal markers.
Internal Markers
Although IVDMD has perhaps been most widely used to generate estimates of DDM, a variety of feedstuff entities has been used as internal markers to estimate DDM. These markers have included chromogen (Reid et al., 1950), lignin (Drennan et al., 1970; Momont et al., 1994), indigestible neutral detergent fiber (Lippke et al., 1986; Waller et al., 1980), and indigestible acid detergent fiber (Berger et al., 1979; Waller et al., 1980). With any of these methods, including the IVDMD method, the primary challenge is the collection of representative samples of plant parts or, in a mixed sward, the plant species or parts being consumed. In monocultures, samples hand plucked by an experienced observer are likely to be adequate (Lippke et al., 2000). In mixed swards with only a few species constituting >90% of the herbage, esophageally fistulated animals having experience with the sward(s) being tested are probably needed to obtain representative samples of the diet (Lippke and Evers, 1986). For rangelands with numerous native species, a fecal index method (Lancaster, 1949; Reid et al., 1952) may have to be used to estimate DDM, since collecting a sample representative of the wide variation likely in the daily diet is improbable. Electronic monitoring of the species and plant parts being consumed by domestic range animals has been proposed for development (H. Lippke and T.D.A. Forbes, 1994, unpublished data).
With diet and fecal samples in hand, the marker of choice is determined, and DDM is estimated according to the equation:
 | [5] |
While excellent recovery of the marker entity has been reported in some experiments, most of the internal markers mentioned have imperfect and variable recovery rates (Lippke et al., 1986) that must be ascertained in each experiment where they are used. Predominantly, these markers are not unitary entities and are defined by the analysis procedure. They are unlikely to be identical chemically or structurally in feed and feces, making their continued use tenuous at best.
Alkanes as Markers
Following the suggestion by Grace and Body (1981) that naturally occurring long-chain fatty acids in herbage might be used as indigestible fecal markers, the work of Mayes et al. (1986) on the use of n-alkanes for the determination of forage intake began a series of investigations on the use of this class of compounds as both internal and external markers for estimating digestibility and intake of forages (Dove and Mayes, 1996). Using perennial ryegrass fed to lambs, Mayes et al. (1986) examined the suitability of odd-chain alkanes in the range, C27 to C35, and the dosed alkanes, C28 and C32, under a variety of dietary conditions and found that (i) fecal recovery increased from 71 to 93% as chain length increased (Fig. 1)
, (ii) recovery of dosed alkanes appeared slightly higher than recovery of naturally occurring alkanes, and (iii) recovery was generally unaffected by level of intake or diet composition. The recovery rates of the dosed C32 and the naturally occurring C33 were practically identical (89%). Because the analysis errors for all alkanes in a sample are also likely to be identical, C32 and C33 bias the estimates of fecal output and digestibility proportionally, making them a particularly appropriate pair for accurate estimation of intake, Eq. [2]. Duncan et al. (1999) found generally lower rates of recovery in one experiment, but recovery rates similar to those of Mayes et al. (1986) in a second experiment, while Dove et al. (1999) reported recoveries ranging from 0.89 to 0.98 for C28 and 0.93 to 1.00 for C32 in a series of trials with sheep. Lippke et al. (1999), who dosed C32 twice daily to steers grazing ryegrass and measured C32 and C33, found that the coefficient of variation for estimated intake was only 7%.
|
Alkane-based procedures for partitioning the total intake of grazing sheep into component plant parts were demonstrated by Dove et al. (1999). In their review, Dove and Mayes (1996) pointed out that most plant parts and plant species have a characteristic pattern of alkanes. Using least squares statistical procedures, Dove et al. (1999) exploited the differences in these patterns, as measured in plant parts, in esophageal extrusa of fistulated sheep, and in feces of fistulated and intact sheep, to show that spraying annual grass pastures with glyphosate dramatically changed dietary preferences of sheep grazing the pastures from leaf to stem in the first month after spraying. Dove and Mayes (1996) also discussed the potential of other plant wax components such as alkenes to further differentiate plant components and species. Dove and Moore (1995) proposed that, as research and experience increase the number of usable marker compounds in plant wax, canonical variates analysis should be used to select the best markers in any given situation. This procedure, too, was demonstrated by Dove et al. (1999).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONESTIMATING INTAKECONCLUSIONSREFERENCES |
|---|
Received for publication May 29, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONESTIMATING INTAKECONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. C. Burns Grazing Research in the Humid East: A Historical Perspective Crop Sci., December 2, 2005; 46(1): 118 - 130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Smit, H. Z Taweel, B. M. Tas, S. Tamminga, and A. Elgersma Comparison of Techniques for Estimating Herbage Intake of Grazing Dairy Cows J Dairy Sci, May 1, 2005; 88(5): 1827 - 1836. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
