Animal-based techniques for grazing ecology research

ELSEVIER

Small Ruminant

Research16 (1995) 203-214

Animal-based techniques for grazing ecology research I.J. Gordon* Macaulay Land

Use

Research Institute, Craigiebuckler, Aberdeen, AB9 ZQJ, UK

Accepted15July 1994

Abstract Grazing ecologists have available to them a suite of research techniques to measure aspects of plant/animal interactions. This review focuses on animal-based techniques and describes the methods available to tackle questions concerned with intake, foraging behaviour, diet composition and location of free-ranging large mammalian herbivores. The majority of techniques have been developed and tested for agricultural livestock and have proved to be highly reliable and accurate. However, they rely on being able to handle the animal regularly, either in order to administer the measurement protocol or to gather samples or data from the animal. Current interest in the grazing ecology of domestic and wild livestock grazing extensive vegetation systems requires a less hands-on approach to data gathering. Recent developments in microelectronics (e.g. datalogging and radiotelemetry) promise to provide researchers with more accurate data on the foraging strategy of free-ranging animals. Ke,vwords: Grazing ecology;

Techniques; Intake; Diet composition; Grazing behaviour

1. Introduction Grazing ruminants form the backbone of most of the world’s ruminant livestock enterprises (US Dep. Agric., 1990), particularly in marginal areas where cereal crops are of low productive potential. The animals are pastured on a variety of vegetation systems ranging from simple sown monocultures, typical of intensively managed grassland, to complex mixtures of vegetation communities, typical of rangelands. They can have a major impact on the environment through their grazing and movement patterns (Archer and Smeins, 1991). The level of production obtained from animals grazing vegetation depends on their ability to ingest a diet adequate to meet their nutrient requirements for maintenance, growth and reprobuction. This in turn is reg* Fax: 44 1224 311556. 0921-4488/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDIO921-4488(95)00635-4

ulated by a series of short-term decisions made by the animal about which plants to select and how long to search between bites. These decisions influence rate of food intake and nutrient content of the diet. Longer term decisions concern the length of time to spend feeding and where to feed, given topographic influences on energy expenditure and distance travelled between foraging sites, water and shelter (Stuth, 1991). This suite of decision-making processes is defined as the foraging strategy of the animal (Gordon and Illius, 1992). If we are to make the most efficient use of the plant and animal resources available in marginal areas, it is essential to improve our understanding of the foraging strategies of livestock which use these ecosystems. To date, most studies of foraging strategy of domestic livestock and wild-grazing species have been descriptive catalogues of their diets at the taxonomic or plant part level and quantification of their daily

I.J. Gordon /Small Ruminant Research 16 (1995) 203-214

204

Intake

Diet composition

Shon-term Bite simulation Animal weight change Bolus size and number Oesophageal fislulates Sward presentation

Direct observation Faecal cuticle analysis Rumen content analysis Oesophageti

fistulates

Plant markers

Long-term Offtake estimation Bite simulation Markers

;ward boards

L-l Field experiments pastures sown patches across boundaries mosaics

-mixed

Foraging behaviour Event recorder

-

Bite meters Video

Location

L

I

Direct observation Time lapse photography/ video I Radiotelemetry Global positioning systems I

Fig.

1.Approachesand techniques used to measure the various components of the foraging strategy of livestock.

intake (Van Dyne et al., 1980). This type of information is of limited predictive value outside the circumstances and site from which it was obtained and does not advance our general understanding of foraging strategy and diet selection. Only through an understanding of the mechanisms involved in the decisionmaking processes of the foraging animal will it be possible to develop more widely applicable predictive models to meet agricultural, conservation or landscape objectives (Rice et al., 1983; Armstrong, 1991; Blackburn and Kothmann, 1991; Baker et al., 1992). There is a growing interest in the use of predictive models to estimate interaction between livestock and their resources (Heitschmidt and Stuth, 1991; Gordon and Hutchings, 1993). To develop the understanding required to build these models, researchers must be able to draw upon a range of techniques to measure and interpret short-term and long-term foraging decisions of free-ranging livestock. This review describes techniques for measuring the foraging strategy of ruminants focusing on four areas of measurement: intake rate (short- and long-term), foraging behaviour (e.g. bite rate, feeding time, mastication rate, movement rate), diet composition and location of the animal (Fig. 1). This will provide a background to the techniques available and the reasons for making the measurements.

2. Intake rate The principal reasons for measuring herbage intake on free-ranging animals are ( 1) to understand the relationships between resource distribution, sward structure and herbage intake, and (2) to explain between-animal variation in intake and performance in relation to animal nutritional status, grazing regime and management practice. While foraging, an animal takes a series of bites of varying size from the herbage on offer. Combination of bite size and short-term rate of biting is defined as the short-term intake rate, with units given in mg s-l or g min-‘. Over the day the animal spends a certain amount of time grazing and intake during this time period is defined as the long-term intake rate, in kg day - ‘. Short-term intake tends to be measured either as an indirect measure of daily intake or to relate intake rate to the structure of the vegetation within homogeneous patches. Long-term intake measurements are usually related to the productivity of the animal. Because of differences in approach to measuring short- and long-term intake rate the descriptions of the techniques available are split into two sections. 2.1. Short-term intake rate Two approaches have been used to estimate shortterm intake of grazing animals. First, the animal ranges

I.J. Gordon /Small Ruminant Research 16 (1995) 203-214

freely across a pasture and exhibits its normal foraging behaviour. Second, the animal is confined to a pen or arena and provided with representative plant material to consume. A variety of techniques has been developed for estimating short-term intake rate of free-ranging animals. First, an observer mimics the bite size of tame, freeranging animals by visual observation of the bites and clipping or hand-plucking simulated bites from the vegetation (Bjugstad et al., 1970; Parker et al., 1993). Concurrent visual measurements of bite rate (see Section 3) allow estimation of short-term intake rate. This technique has been applied to a wide range of domestic and wildlife species (Renecker and Hudson, 1985; Hudson and Watkins, 1986; Hudson and Frank, 1987). The advantage of the technique is that a great deal of detail on the size of individual bites of specific plant species can be assessed (Hobbs et al., 1983; Vulink and Drost, 1991). The disadvantages of this technique are that it is time consuming, measurements tend to be biased towards plants species which are easily observable at a distance, it is difficult to follow even tame animals at night and it can be unreliable due to interobserver differences in estimation of bite size. However, in certain circumstances, for example with browsing animals or where herbaceous plants occur in discrete, larger than bite size patches, the technique can be accurate (Parker et al., 1993). Secondly, short-term changes in BW, before and after free-range grazing, can be used to estimate shortterm intake rate, using very accurate balances (Penning and Hooper, 1985), or by measuring the weight changes of the animal with pressure transducers attached under each hoof (Horn and Miller, 1979). It is necessary to correct BW changes for faecal, urinary, respiratory and other obligatory weight losses. Intake (Z) can then be calculated as: z=(w*+F+U+IWL)-W,

(1)

where Wr and W, are the weights of the animal before and after grazing, F and U are weights of faeces and urine produced and IWL is the insensible weight loss (respiration and other obligatory weight losses). Insensible weight loss can be measured currently using animals penned in environmental conditions similar to that of the grazing animal but without access to food or water (Penning and Hooper, 1985), or it can be correlated to some climatic variable and then measured

205

during the period of the trial (Murray and Brown, 1993). The technique integrates biomass consumed over a short period, usually about 1 h or 1000 bites and thus gives mean bite size across plant species consumed. An estimate of the dry matter (DM) ratio of the forage needs to be gathered in order to estimate dry matter intake (DMI) . This technique appears to give accurate estimates of intake rate (g min- ‘) (Penning and Hooper, 1985) and is particularly useful when resource conditions are changing rapidly. The disadvantages are that it requires tame, trained animals; the weighing technique requires access to high precision scales (usually f 10 g) (Penning and Hooper, 1985; Murray and Brown, 1993) in the field; it is necessary to have the animal standing still on the scale for a period long enough to take about 200 weight measurements in order to measure the animal’s weight accurately; insensible weight losses have large between-animal variation; the majority of studies rely on individually calibrated correction factors (Penning and Hooper, 1985). The third technique assumes a strong correlation between number of boluses swallowed by a free-ranging animal while feeding (Forwood and Hulse, 1989) and intake of herbage. A number of techniques have been developed to determine number and rate of swallowing of food boluses by free-ranging animals by measuring the change in the geometry of the oesophagus (Stuth et al., 1981) or the change in the conductance or pressure on an oesophageal cannulae (Forwood et al., 1985) with the passing of food boluses down the oesophagus. Theoretically, this allows intake rate to be measured instantaneously. The main assumption is that the size of the swallowed bolus is constant across the grazing period. However, because of differences between individual animals in the weight of the food bolus in relation to the animal’s BW and type of forage consumed, calibration equations are required for each circumstance in order to estimate intake (Forwood et al., 1991). The method also involves surgical implants into animals to record the number of boluses swallowed. Finally, short-term intake rate and bite size of freeranging animals can be estimated from animals prepared with oesophageal fistulae through the collection and weighing of extrusa over a known time period and the simultaneous recording of bite rate (Stobbs, 1973). This technique appears to provide the most accurate

206

I.J. Gordon /Small Ruminant Research 16 (1995) 203-214

estimate of intake rate per bite, although it presents welfare problems particularly in some species (see also Section 4). Furthermore, in certain circumstances, it can provide unreliable estimates because of incomplete extrusa recovery, the extent of which varies with the individual (R.H. Armstrong, personal communication, 1994). This can be overcome by placing a foam sponge posterior to the fistula to increase recovery (Hodgson, 1982); however, this can affect the normal grazing behaviour of the animal despite pretraining (Penning and Hooper, 1985). An alternative approach to estimating short-term intake rate is to present plant material to animals confined in pens or arenas. This allows very close control over the plant material offered to the animals. The plant material may be presented in the form of turves or sods cut from representative areas of plant communities (Mursan et al., 1989), in trays managed to provide plant material of a given sward characteristic (Illius et al., 1992) or plant material anchored to a board (Black and Kenney, 1984; Spalinger et al., 1988; Laca et al., 1992). Because the latter technique requires hand construction of swards, only a very limited number of trials can be conducted each day as compared with the turf and tray methods of presentation. These techniques estimate short-term intake rate (usually less than 5 min) and its components by weighing the swards before and after the animal has grazed for a set period of time or a set number of bites. If measurements of bite rates are made concurrently then short-term intake rate can be estimated. Using an alternative approach, Burlison et al. ( 1991) used a ‘grazing cage’ to confine animals onto very small areas (2.5 m’) of pasture and bite weights were measured using the oesophageal fistula technique. Although a large amount of detailed data can be gathered from each trial by confining and fasting animals, their normal foraging behaviour is altered (see Section 2.2.). No critical test of the validity of this approach has been applied, although, in general, the results obtained using these techniques agree well with published values obtained from free-ranging animals (Mursan et al., 1989). 2.2. Long-term intake rate Long-term (daily) intake can be estimated using the results of short-term trials to estimate short-term intake rate (see Section 2.1.) multiplied by the amount of

time spent grazing per day (see Section 3). These estimates may, however, overestimate long-term intake because the techniques used to measure short-term intake necessitate withholding food from the animal for a period, increasing its short-term intake rate during the period of the trial (Greenwood and Demment, 1988; Newman et al., 1992). Short-term intake rate tends to change over the course of a feeding bout, making it difficult to measure a valid average short-term intake rate which can provide an estimate of long-term intake rate. A number of methods have been developed for measuring long-term (daily) intake rate directly (Leaver, 1982; Dove and Mayes, 1991). These can be categorised either as measurements based on the rate of depletion of offered forage or measurement of quantity of herbage ingested using internal or external markers. The majority of the pasture-based methods of intake estimation (Meijs et al., 1982) are of limited value except on simple swards because of errors associated with measurement of pasture growth and senescence (Grant, 1993), the difficulty of estimating intake from vegetation community mosaics, the inability to provide estimates of between-animal variation in intake and problems of estimating intake by individual animal species in multi-species grazing systems. A more direct approach to measuring intake using animal-based methods is frequently used (Corbett, 1981; Mayes, 1989). Animal-based methods of estimating intake have in the past relied on the measurement of faecal output (F) and diet digestibility (D) (Van Dyne, 1969; Kotb and Luckey, 1972) and estimating intake as: I=Fl(

1-D)

(2)

Total collection methods for estimating faecal output (Galyean et al., 1986) are laborious and possibly affect normal grazing behaviour (Armstrong and Eadie, 1977). Faecal output is usually estimated, therefore, from the concentration of an orally dosed indigestible marker (e.g. ytterbium, Cr203, CX6n-alkane) in the faeces (Corbett, 198 1; Mayes, 1989; Dove and Mayes, 1991). The marker can be administered continuously (intraruminal pumps or slow release capsules; Parker et al., 1990; Mayes et al., 1991) or dosed once or twice a day (capsules, impregnated paper; Le Du and Penning, 1982) or as a single dose (Hollernan and White, 1989). There appears to be little improvement in the

I.J. Gordon /Small Ruminant Research I6 (1995) 203-214

estimation of faecal output by using a single as compared with a twice daily dosing (Hatfield et al., 1990). However, there may be behavioural consequences to handling animals frequently, which result in an under or overestimate of intake (Cordova et al., 1978; Hatfield et al., 1993). Faecal output is estimated from concentrations of indigestible markers in faeces (Mf) and daily dose of marker (Md) administered: F=(M,xR’)/M,

(3)

where recovery of the marker in faeces (R) is assumed to be 1.00 for Cr,03 (Kotb and Luckey, 1972), and between 0.86 and 0.98 for C36 n-alkane (Mayes et al., 1988; Dove et al., 1989; Vulich et al., 1991). Accurate estimates of faecal output with external markers require that the marker be completely recovered in the faeces or that its recovery rate is known. However, available evidence suggests that marker recovery can be extremely variable due to marker absorption across the gut wall, faecal sampling in relation to diurnal variation in marker excretion and analytical problems (Dove et al., 1989; Dove and Mayes, 1991; Ohajuraka and Palmquist, I99 1) . An alternative method for estimating faecal output uses a single-dose marker technique (Holleman and White, 1989). In this method the marker is dosed once and faecal samples are taken at intervals over the following days. Marker concentration in the faeces vs. time after pulse dosing is adjusted using an appropriate model (Matis et al., 1989) to estimate volume of the gastrointestinal tract, passage rate, and following on faecal output per unit time assuming that input and output of the alimentary tract are in steady state:

-M,,,,)

(ri+i -ri)))/(2(T,-T,))

(4)

where Mer is the mean marker concentration (mg g - ‘) , mfci+l ), mfcij are the marker concentrations in the faeces at times ri + I and ti, Mfco, is the marker concentration in the faeces at T,,, which is calibrated for the marker background, and T2 is the time of last sampling. From this the hourly dry faecal output (F, g h-i) can be calculated as: F=Md/(Mef(T,-T,,))

(5)

The long-term intake can be estimated by substituting 24(F) into Eq. (5). This technique has the advan-

207

tage that it provides estimates of physical properties of the gastrointestinal tract which relate to the long-term voluntary food intake from grazed swards. However, animals are handled or followed (Jiang and Hudson, 1992) in order to obtain faecal samples at least every 4 h, which may affect their grazing behaviour and intake. Generally, the accuracy of intake estimates is limited to a greater extent by the accuracy of the digestibility rather than the faecal output measurements. Conventionally, the Tilley and Terry (1963) method for estimating the 48-h in vitro digestibility of extrusa samples from oesophageal fistulates or hand-plucked samples is the preferred method of determining in vitro digestibility of the diet consumed (Van Dyne, 1969). The in vitro digestibility values are then corrected to give in vivo estimates using regression equations derived from a range of feeds (Armstrong et al., 1989) This method assumes that fistulated animals consume a diet similar to that of animals for which faecal output has been measured and that the in vitro estimates of digestibility reflect in vivo digestibility. Both of these assumptions appear not to hold in all cases (Holechek et al., 1986; Coates et al., 1987). However, this error can be reduced by acclimatising oesophageal fistulated animals to the pasture conditions for at least 7 days prior to samples being collected, using fistulates from the same flock or herd and maintaining them under similar conditions as test animals. A less intrusive technique to predict digestibility is to use faecal indices, e.g. concentrations of N (Greenhalgh and Corbett, 1960) or reflectance from near infrared spectroscopy (NIRS; Stuth et al., 1989) which may correlate with forage digestibility. Although faecal N has been used extensively on simple pastures, these faecal indices require calibration for different sward types, species of animal and animals of different physiological state, and are of little value where the animal might consume a diet containing protein-precipitating plant metabolites (Hobbs, 1987; Armstrong et al., 1989; Nunez-Hernandezet al., 1992). NIRS may prove to be a useful analytical procedure for estimating digestibility from faecal samples. Prediction equations appear to be fairly robust in their estimation of digestibility of a range of forages and a range of sites (Stuth et al., 1989; Lyon and Stuth, 1992). Both in vitro and faecal indices suffer from the fact that on high quality pasture errors on estimates of digestibility can be mul-

208

I.J. Gordon /Small Ruminant Research 16 (1995) 203-214

tiplied three to five times when estimating intake, and they cannot be used where the animals are getting feed from others sources than pasture (e.g. concentrates). A preferred technique is to use internal indigestible markers (silica, indigestible NDF, acid insoluble ash, acid insoluble lignin; van Keulen and Young, 1977; Fahey and Jung, 1983; Minson, 1990) to estimate in vivo digestibility directly from concentration of the component in the diet (C,) and faeces (C,) using the equation: D=(l-C,)IC,

(6)

The use of some markers (silica and acid insoluble ash) is of limited value for free-ranging animals because of the ingestion of these elements through other means than internal to the herbage (e.g. soil, dust; Kreulen, 1985; Beyers et al., 1994). Recently, Mayes et al. ( 1986) have shown that longchain n-alkanes, which occur naturally as waxes in the cuticles of plants, can be used as internal and external markers to estimate diet digestibility and intake in sheep, goats, cattle and red deer. Alkanes with odd numbers of carbon atoms occur in relatively high concentrations in plants whereas even-chained alkanes occur only at low concentrations (Dove and Mayes, 1991). Odd- and even-chained n-alkanes are used as internal and external markers, respectively. Herbage intake can be measured using the C3#&, (natural:dosed) n-alkane ratio in faeces and herbage:

where Ax2 is the daily dose rate of dosed n-alkane C3*, Fj2 and F33 are the concentrations of the n-alkanes C32 and C& in faeces and Hj2 and H33 are the concentrations of n-alkanes Cs2 and Cj3 in herbage (Mayes et al., 1986). Digestibility can be estimated if C,, is dosed at the same time as &, allowing faecal output to be measured (Dove and Mayes, 1991). Digestibility can be also be estimated using C35 as an internal marker since it has a recovery of 93-97%; however, C35 concentrations are low in most herbages (Dove and Mayes, 1991). Techniques used to measure herbage intake must meet the criteria of causing minimal disturbance to the animal, having little impact on its welfare, having a precision suitable for the purposes of measurement and

giving unbiased results. In some circumstances, it is important to have low labour inputs in order to minimize costs. Currently, there is research being conducted on the use of slow release n-alkane markers from spring-loaded capsules (Mayes et al., 1991) similar to those widely used for Cr203 administration (Parker et al., 1990). These devices remain in the rumen and deliver a constant dose of marker over a period of up to 30 days. This would bring closer the possibility of estimating intake and digestibility of natural forages by free-ranging domestic and wild livestock, which can only be handled very infrequently.

3. Foraging behaviour Free-ranging ruminants respond to the quantity and distribution of vegetation on offer by altering their foraging behaviour. Rate of intake by sheep on homogeneous ryegrass (Lolium perenne) pastures increases in proportion to herbage mass or height until it reaches an asymptote (Hodgson, 1977; Penning, 1986; Penning et al., 1991). In an attempt to maintain intake, sheep respond to a low sward height by taking more bites per minute and grazing for longer periods. However, below sward heights of about 6 cm these behavioural mechanisms are unable to compensate for changes in bite size; as a result intake rate drops. The primary foraging behaviour variables which researchers are interested in are bite rate (prehension, harvesting), mastication rate, feeding time, feeding/ resting bout length and movement rate. These measurements can be used to estimate intake (see Intake rate) and also to estimate energy budgets (Hudson and White, 1985). Frequently measurements of foraging behaviour are made by direct observation, usually recording events with a manually operated data logger (Owen-Smith, 1979; Demment and Greenwood, 1987; Unwin and Martin, 1987). This can be extremely time consuming, it is difficult to collect data over the 24 h period without night vision equipment even with tame animals (see Section 5), and unless a video-recording is made the observer can only sample a limited number of individuals at one time. In addition it relies on animals being easily observable and the presence of a human observer can also alter the behaviour of even tame animals, However, unlike the automated techniques described below, this technique does allow the

I.J. Gordon /Small Ruminant Research 16 (1995) 203-214

Prehension bites (#.min?)

40.3 (13.2)

Mastication bites (#.min-‘)

113.6 (17.4)

Ruminating

32.5% of the day

441 minutes 1.1

Boluses (#.min-I)

Idling

29.0% of the day

406 minutes

Eating

Ruminating chews (#.min?)

98.3 (29.5)

Chews.Bolus~’

78.7 (14.4)

512 minutes

37.7% of the day

Fig 2. A portion of a typical results file from a recording of the jaw movements of a sheep grazing an 8 cm ryegrass (L&urn perenne) sward (collected on 17.7.1991; I.J. Gordon, unpublished data). Values in parentheses are standard deviations of the mean.

gathering of other contextual data, such as the distance to a dominant or subordinate animal and the slope and aspect of the feeding station. A number of mechanical and electronic devices have been developed to measure one or more of these variables automatically (Stobbs, 1970; Anderson and Kothmann, 1977; Penning, 1983; Matsui and Okubo, 1989). These systems monitor leg or jaw movements, using sound recordings of the bites and chews (Alkon et al., 1989; Laca et al., 1992)) or in digital or analogue form. For estimating bite rate and grazing time using the digital system the inter-bite interval is analysed in order to differentiate between grazing and ruminating (Janeau et al., 1987; Matsui and Okubo, 1989; Matsui, 1994). The more sophisticated devices (Penning, 1983) record the signal and allow a number of behavioural parameters to be derived from the recordings (Fig. 2). The majority of jaw movement monitoring devices are carried on the animal and require the animal to be handled regularly (normally daily) in order to retrieve data which are stored on audiocassette or datalogger. A more flexible system involves the data being transmitted to a remote receiver via radio telemetry (Nichols, 1966; P.D. Penning, personal communication, 199 1) . This reduces the frequency of handling as animals only have to be captured to replace the batteries.

209

4. Diet composition A number of techniques have been developed for assessing diet composition of free-ranging animals (McInnis et al., 1983). These techniques range from following an animal and obtaining clipped or handplucked samples which mimic each bite taken (Bryant et al., 1981; Vulink and Drost, 1991), to the use of a marker unique to each species or plant form in the diet and identified in the faeces (Mayes, 1989; Dove and Mayes, 1991). Each technique has its advantages and disadvantages. Hand-plucking requires tractable animals which can be approached closely enough to facilitate observation and is of limited use where plants occur in intimate mixtures at a scale smaller than the bite. On the other hand, the marker technique requires high cost analytical equipment to measure marker concentrations in feed and faeces, is limited at present to a small number (about five) of plant species and also integrates the diet composition over a day limiting the temporal scale over which dietary changes can be quantified (Dove and Mayes, 1991) . For many years the technique of recording the remains of plant cuticles in the faeces or alimentary tract has been used to estimate diet composition (Cracker, 1959; Ward, 1970). However, because of differential digestion of plant species within the rumen (Rice, 1970; Slater and Jones, 197 1) and the typically high proportion of unidentified cuticular remains in the faeces (McInnes et al., 1983) these techniques have little quantitative value. Where qualitative estimates are required and plant species can be grouped into functional classes (e.g. grass, forbs and browse) the faecal cuticle technique provides a valuable insight. Use of oesophageally fistulated animals has been relatively successful in determining the diets of domesticated livestock (Vavra et al., 1978). Although practical guides are now available for the preparation and maintenance of various species (Pfister et al., 1990; Goddard and Fraser, 1994)) it has not been extensively used in wild herbivore species (Pietersen et al., 1993). Overall, the results from oesophageal fistula extrusa provide, as yet, the most accurate method for determining diet composition (McInnis et al., 1983) as long as proper protocols and management are implemented. The problems with the technique are: ( 1) requirement for surgery (Rice, 1970) ; (2) the short time period for collection of samples and the small area over which the

210

I.J. Gordon /Small Ruminant Research 16 (1995) 203-214

samples are collected which may lead to biases (Lesperance et al., 1974; Coates et al., 1987), particularly where plant species are patchily distributed over a large scale. The development of remote control fistula valves (Raats and Clarke, 1992) allows the collection of diet samples throughout the day without disturbing the normal feeding behaviour of the animal and should overcome this problem; (3) that the grazing behaviour of fistulated animals may differ from that of intact animals (Engels and Malan, 1973; Forbes and Beattie, 1987; Jones and Lascano, 1992). However, the proper selection, management and experimental protocols can reduce much of the discrepancy between listulated and intact animals. Unique plant markers are increasingly being used to estimate diet composition. In tropical grassland/legume pasture, for example, the use of the carbon isotope discrimination technique would seem to be a reliable method to estimate legume proportion in the diet of grazing animals (Jones and Lascano, 1992). This method is based on differences in the ratios of 13C to “C in tropical grasses (C4 plants) and legumes (C3 plants). However, the i3C:“C ratio is of use only in a limited number of circumstances as C4 plants are rare in temperate regions. More recently, the odd-chain nalkane ( C25-C35) concentration profiles of plants have been used as plant markers (Mayes, 1989; Dove and Mayes, 1991). For example, the high ratio of C29:C33 in temperate legumes as compared with grass means that, knowing concentrations of C29:C33 in grass and legume components of herbage and faeces, simultaneous equations can be used to estimate diet composition. However, the resolving power of simultaneous equations limits the number of species which can be separated. Resolving power is greatest where there are large differences in concentrations of individual alkanes between plant species (Dove and Mayes, 199 1) . Thus far, preliminary investigations have shown the effectiveness of using this technique for quantifying diet composition for small numbers of species (Dove and Mayes, 1991) .

5. Location Even in relatively homogeneous plots, animals distribute themselves and conduct their behaviour unevenly (Arnold, 1984/1985). In more complex

ecosystems, such as rangelands, spatial dispersion of resources will affect location of animals (Gordon, 1989). The decision-making processes invoked by the foraging animal may differ in relation to the landform scale (Senft et al., 1987; Stuth, 1991). By recording the location of animals, researchers are able to answer questions about the distribution of impact of livestock on vegetation and effect of topographic features, climatic conditions and land use on relative use of resources by herbivores. By linking locational data with Geographic Information System (GIS) databases, linescale aspects of habitat can be used to develop habitat suitability modelling (Morrison et al., 1992). To date this is an understudied aspect of foraging strategy of grazing ruminants. Traditionally, information on the location of animals was gathered by direct observation ( Attwood and Hunter, 1957; Arnold, 1984/1985), with the animals located either at exact points or within grid squares on a map. However, this is a time consuming process, requires specialised night-vision equipment if information on nocturnal distribution is to be gathered (Boag et al., 1990) and movements of the observer can affect distribution of the animals. Recent advances in electronic technology have overcome many of these problems. Systems using video cameras (Sherwin, 1994; Parsons et al., 1994) or radiotelemetry (Kenward, 1987; White and Garrott, 1990) are less time consuming and are likely to provide the required information more efficiently. Video recording is limited in its application because of the field of view and the topography of the paddock or study area may result in animals being out of sight. Although animals might be individually recognised by using alphanumeric or colour markings, in order to identify markings the resolution needs to be high, again limiting the field of view. With radiotelemetry individual animals can be tracked over much larger areas, either manually (Warren and Mysterud, 1991) or automatically (Kenward, 1987). Activity monitors can also be used to assess the behaviour of animals within a location; this may be important especially if animals have geographically separate areas or habitats in which they feed or rest. Use of Global Positioning System (GPS) has long been used for wildlife tracking (White and Garrott, 1990) ; however, recently the accuracy has been improved, tracking sheep over large ranges to an accuracy of less than 5 m (Roberts et al., 1993). Video and electronic

I.J. Gordon /Small Ruminant Research 16 (1995) 203-214

location finding systems are still in their infancy but are likely to develop rapidly in the near future (Wratten, 1994).

6. Conclusions

There is an increasing awareness of the importance of understanding the foraging strategy of domestic livestock and wildlife in the development of efficient management meet agricultural systems to and environmental goals. Grazing ecologists now have a suite of techniques available to measure components of foraging strategy. Many of these techniques have been developed and tested for agricultural species grazing simple monocultures. As researchers increase the complexity of the questions they are asking and move on to understand the foraging strategy of herbivores grazing complex mixtures of plant communities, new techniques will have to be developed to meet requirements. Technique development especially using electronics and remote control is, therefore, likely to remain a fruitful area of research into the future. However, many of the newly developed systems have teething problems and are unreliable under the rigorous conditions required of field experimentation using free-ranging animals. In the end it may be as well to use tried and tested measurement techniques which provide a limited range of data rather than gather no data at all from systems which promise the earth! There is no ‘best’ technique for making measurements. The most appropriate technique will depend upon the goals of the research and the circumstances under which the measurements are made including such considerations as the time scale of the study, grain of heterogeneity, the availability of tame animals, logistics and funding.

Acknowledgements The author thanks Professor R.J. Hudson, Dr. R.W. Mayes, Dr. J.A. Milne, Professor T.J. Maxwell and an anonymous referee for helpful comments on the manuscript. W. Sherriffs kindly drew Fig. 1.

211

References Alkon, P.U., Cohen, Y. and Jordan, P.A., 1989. Towards an acoustic biotelemetry system for animal behaviour studies. J. Wild]. Manage., 53: 658662. Anderson, D.M. and Kothmann, M.M., 1977. Monitoring animal travel with digital pedometers. J. Range Manage., 30: 316-317. Archer, S. and Smeins, F.E., 1991. Ecosystem-level processes. In: R.K. Heitschmidt and J.W. Stuth (Editors), Grazing Management. An Ecological Perspective. Timber Press, Portland, Oregon, pp. 109-139. Armstrong, H., 1991. Predicting the effects of large herbivores on hill vegetation in the UK. Aspects Appl. Biol., 26: 217-220. Armstrong, R.H. and Eadie, J., 1977. The growth of hill Iambs on herbage diets. J. Agric. Sci. (Cambridge), 88: 683-692. Armstrong, R.H., Common, T.G. and Davies, G.J., 1989. The prediction of the in vivo digestibility of the diet of sheep and cattle grazing indigenous hill plant communities by in vitro digestion, faecal nitrogen concentration or ‘indigestible’ acid detergent fibre. Grass Forage Sci., 44: 303-313. Arnold, G.W., 1984/1985. Spatial relations between sheep, cattle and horse groups grazing together. Appl. Anim. Behav. Sci., 13: 7-17. Attwood, P.R. and Hunter, R.F., 1957. A method for studying the preferential grazing of hill sheep. J. Anim. Behav., 5: 149-152. Baker, B.B., Bourdon, R.M. and Hanson, J.D., 1992. FORAGE: a model of forage intake in beef cattle. Ecol. Model., 60: 257-279. Beyers, W.N., Connor, E.E. and Gerould, S., 1994. Estimates of soil ingestion by wildlife. J. Wild]. Manage., 58: 375-382. Bjugstad, A.J., Crawford, H.S. and Neal, D.L., 1970. Determining forage consumption by direct observation of domestic grazing animals. In: Range and Wildlife Habitat Evaluation. US Dep. Agric. For. Serv. Misc. Publ. No. 1147, pp. 101-104. Black, J.L. and Kenney, P.A., 1984. Factors affecting diet selection by sheep. II. Height and density of pasture. Amt. J. Agric. Res., 35: 564-578. Blackbum, H.D. and Kothmann, M.M., 1991. Modelling diet selection and intake for grazing herbivores. Ecol. Model., 57: 145163. Boag, B., McFarlane Smith, W.H. andGriffith& D.W., 1990. Observations on the grazing of double low oilseed rape and other crops by roe deer. Appl. Anim. Behav. Sci., 28: 213-220. Bryant, F.C., Taylor, C.A. and Merrill, L.B., 198 1. White-tailed deer diets from pastures in excellent and poor condition. J. Range Manage., 34: 193-200. Burlison, A.J., Hodgson, J. and Illius, A.W., 1991. Sward canopy structure and the bite dimensions and bite weight of grazing sheep. Grass Forage Sci., 46: 29-38. Coates, D.B., Schachenmann, P. and Jones, R.J., 1987. Reliability of extrusa samples collected from others fistulated at the oesophagus to estimate the diet of resident animals in grazing experiments. Aust. J. Exp. Agric., 27: 739-745. Corbett, J.L., 1981. Measurement of the forage intake of grazing animals. In: J.L. Wheeler and R.D. Mochrie (Editors), Forage Evaluation: Concepts and Techniques. CSIRO and American Forage and Grassland Council, East Melbourne and Lexington, pp. 287-312.

212

I.J. Gordon/Small Ruminant Research 16 (1995) 203-214

Cordova, F.J., Wallace,, J.D. and Pieper, R.D., 1978. Forage intake by grazing livestock: a review. J. Range Manage., 31: 430-438. Cracker, B.H., 1959. A method of estimating the botanical composition of the diet of sheep. NZ J. Agric. Res., 2: 72-85. Demment, M.W. and Greenwood, G.B., 1987. The use of a portable computer for real-time recording of observations of grazing behaviour in the field. J. Range Manage., 40: 284-285. Dove, H. and Mayes, R.W., 1991. The use of plant wax alkanes as marker substances in studies of the nutrition of herbivores: A review. Aust. J. Agric. Res., 42: 913-952. Dove, H., Mayes, R.W., Freer, M., Coombe, J.B. and Foot, J.Z., 1989. Faecal recoveries of the aIkanes of plant cuticular waxes in penned and in grazing sheep. In: R. Jarrige (Editor), Proc. XVI International Grassland Congress, Nice, France, The French Grassland Society, Versailles, pp. 1093-1094. Engels, E.A.N. and Malan, A., 1973. Sampling of pastures in nutritive evaluation studies. Agroanimalia, 5: 89-94. Fahey, G.C. Jr. and Jung, H.G., 1983. Lignin as a marker in digestion studies: a review. J. Anim. Sci., 57: 220-225. Forbes, T.D.A. and Beattie, M.M., 1987. Comparative studies of ingestive behaviour and diet composition in oesophageal-fistulated and non-fistulated cows and sheep. Grass Forage Sci., 42: 79-84. Forwood, J.R. and Hulse, M.M., 1989. Electronic measurement of grazing time and intake in free roaming livestock. In: R. Jarrige (Editor), Proc. XVI International Grassland Congress, Nice, France, pp. 799-800. Forwood, J.R., Hulse, M.M. and Ortbals, J.L., 1985. Electronic detection of bolus swallowing to measure forage intake of grazing livestock. Agron. .I., 77: 969-972. Forwood, J.R., da Silva, A.M.B. and Paterson, J. A., 1991. Sward and steers variables affecting feasibility of electronic intake measurement of grazers. J. Range Manage., 44: 592-596. Galyean, M.L., Krysl, L.J. and Estell, R.E., 1986. Marker-based approaches for estimation of fecal output and digestibility in ruminants In: F.N. Owen (Editor), Proc. Symposium on Feed Intake by Beef Cattle. Oklahoma Agricultural Experiment Station MP-121, pp. 96-133. Goddard, P.J. and Fraser, M.D., 1994. Esophageal fistulation in the guanaco. Small Rumin. Res., 13: 287-292. Gordon, I.J., 1989. Vegetation community selection by ungulates on the Isle of Rhum. II. Vegetation community selection. J. Appl. Ecol., 26: 5364. Gordon, 1.J and Hutchings, N.J., 1993. The development of sustainable ruminant livestock grazing systems: the role of modelling. In: IV World Conference on Animal Production. Vol. 1. Invited Papers, Edmonton, Canada, Canadian Society of Animal Science, University of Alberta, Edmonton, Alb., pp. 413438. Gordon, LJ and Illius, A.W., 1992. Foraging strategy: from monoculture to mosaic. In: A.W. Speedy (Editor), Progress in Sheep and Goat Research. CAB International, Wallingford, pp. 153177. Grant, S.A., 1993. Resource description: vegetation and sward components. In: A.G. Davis,R.D. Baker, S.A. Grant andA.F.Laidlaw (Editors), Sward Measurement Handbook, 2nd Edition. The British Grassland Society, Reading, pp. 69-97.

Greenhalgh, J.F.D. and Corbett, J.L., 1960. The indirect estimation of the digestibility of pasture herbage. I. Nitrogen and chromogen as faecal index substances. J. Agric. Sci. (Cambridge), 55: 371376. Greenwood, G.B. and Demment, M.W., 1988. The effect of fasting on short-term cattle grazing behaviour. Grass Forage Sci., 43: 377-386. Hatfield, P.G., Clanton, D.C., Sanson, D.W. and E&ridge, K.M., 1990. Methods of administering ytterbium for estimation of faecal output. J. Range Manage., 43: 316-320. Hatfield, P.G., Walker, J.W., Fitzgerald, J.A., Glimp, H.A. and Hemenway, K.J., 1993. The effect of different methods of estimating fecal output on plasma cortisol, fecal output, forage intake and weight change in free-ranging and confined wethers. J. Anim. Sci., 71: 618-624. Heitschmidt, R.K. and Stuth, J.W., 1991. Grazing Management. An Ecological Perspective. Timber Press, Portland, Oregon, 264 pp. Hobbs, N.T., 1987. Fecal indices to dietary quality: a critique. J. Wildl. Manage., 51: 317-320. Hobbs, N.T., Baker, D.L. and Gill, R.B., 1983. Comparative nutritional ecology of montane ungulates in winter. J. Wildl. Manage., 47: 1-16. Hodgson, J., 1977. Factors limiting herbage intake by the grazing animal. Proc. International Meeting on Animal Production from Temperate Grasslands, Dublin, An Foras Taluntais, Dublin, pp. 70-75. Hodgson, J., 1982. Ingestive behaviour. In: J.D. Leaver (Editor), Herbage Intake Handbook. The British Grassland Society, Hurley, pp. 113-138. Holechek, J.L., Wofford, H., Arthun, D., Galyean, M.L. and Wallace, J.D., 1986. Evaluation of the total fecal collection for measuring cattle forage intake. J. Range Manage., 39: 2-4. Hollernan, D.F. and White, R.G., 1989. Determination of digesta fill and passage rate from nonabsorbable particulate phase markers using a single dose method. Can. J. Zool., 67: 488-497. Horn, F.P. and Miller, G.E., 1979. Bovine-boots-a new research tool. Oklahoma Agric. Exp. Stn. Misc. PubI. MP-104, pp. 4466. Hudson, R.J. and Frank, S., 1987. Foraging ecology of bison in aspen boreal habitats. J. Range Manage., 40: 71-75. Hudson, R.J. and Watkins, G.W., 1986. Foraging rates in wapiti on green and cured pastures. Can. J. Zool., 64: 1705-1708. Hudson, R.J. and White, R.G., 1985.Computersimulationsofenergy budgets. In: R.J. Hudson and R.G. White (Editors), Bioenergetits of Wild Herbivores. CRC Press, Boca Raton, FL, pp. 261290. Illius, A.W., Clark, D.A. and Hodgson, J., 1992. Discrimination and patch choice by sheep grazing grass-clover swards. J. Anim. Ecol., 61: 183-194. Jiang, Z. and Hudson, R.J., 1992. Estimating forage intake and energy requirements of free-ranging wapiti (Ceruus eluphus). Can. J. Zool., 70: 675-679. Janeau, G., Spitz, F., L&vain, E., Dardaillon, M. and Kowalski, C., 1987. Systeme automatique d’acquisition en temps reel et a distance de param&es biological sur animal liberte (An automatic, remote, real time recording system for biological parameters on free ranging animals). Acta Oecol. Gecol. Appl., 8: 333-341.

I.J. Gordon /Small Ruminant Research I6 (1995) 203-214

Jones, R.J. and Lascano, C.E., 1992. Oesophageal fistulated cattle can give unreliable estimates of the proportion of legume in the diet of resident animals grazing tropical pastures. Grass Forage Sci., 47: 128-132. Kenward, R., 1987. Wildlife Radio Tagging. Equipment, Field Techniques and Data Analysis. Academic Press, London, New York, 222 pp. Kotb, A.R. and Luckey, T.D., 1972. Markers in nutrition. Nutr. Abs. Rev., 42: 813-845. Kreulen, D.A., 1985. Lick use by large herbivores: a review of benefits and banes of soil consumption. Mammal Rev., 15: 107123. Laca, E.A., Ungar, E.D., Seligman, N.G., Ramey, M.R. and Demment, M.R., 1992. An integrated methodology for studying shortterm grazing behaviour of cattle. Grass Forage Sci., 47: 81-90. Leaver, J.D., 1982. Herbage Intake Handbook. The British Grassland Society, Hurley, 143 pp. Le Du, Y.L.P. and Penning, P.D., 1982. Animal based techniques for estimating herbage intake. In: J.D. Leaver (Editor), Herbage Intake Handbook, The British Grassland Society, Hurley, pp. 37-75. Lesperance, A.L., Clanton, D.C., Nelson, A.B. and Theurer, C.B., 1974. Factors Affecting the Apparent Chemical Composition of Fistula Samples. West Region Coordinating Committee Publication No. 8. Nevada Agricultural Research Station, Reno, 30 PP. Lyons, R.K. and Stuth, J.W., 1992. Fecal NIRS equations for predicting diet quality of free-ranging cattle. J. Range Manage., 45: 238-244. Matis, J.H., Wehrly, T.E. and Ellis, W.C., 1989. Some generalised stochastic compartment models for digesta flow. Biometrics, 45: 703-720. Matsui, K., 1994. A new ambulatory data logger for a long-term determination of grazing and rumination behaviour on free-ranging cattle, sheep and goats. Appl. Anim. Behav. Sci., 39: 123130 Matsui, K. and Okubo, T., 1989. A method for quantification of jaw movements suitable for use on free-ranging cattle. Appl. Anim. Behav. Sci., 32: 107-l 16. Mayes, R.W., 1989. The quantification of dietary intake, digestion and metabolism in farm livestock and its relevance to the study of radionuclide uptake. Sci. Tot. Environ., 85: 21-51. Mayes, R.W., Lamb, C.S. and Colgrove, P.M., 1986. The use of dosed and herbage n-alkanes as markers for the determination of herbage intake. J. Agric. Sci. (Cambridge), 107: 161-170. Mayes, R.W., Lamb, C.S. and Colgrove, P.M., 1988. Digestion and metabolism of dosed even-chained and herbage odd-chained nalkanes in sheep. Proc. 12th General Meeting of the European Grasslands Federation, Dublin, Wicklow Press, Dublin, pp. 159163. Mayes, R.W., Dove, H., Lamb, C.S. and Ellis, K.J., 199 1. Evaluation of an intraruminal device for the controlled release of alkane markers for intake estimation in sheep. Proc. 42nd Annual Meeting of the European Association of Animal Production, Berlin, European Association for Animal Production, Rome, Vol. 1, pp. 457.

213

Mclnnis, M.L., Vavra, M. and Krueger, W.C., 1983. A comparison of four methods used to determine the diets of large herbivores, J. Range Manage., 36: 302-306. Meijs, J.A.C., Walter, R.J.K. and Keen, A., 1982. Sward methods. In: J.D. Leaver (Editor), Herbage Intake Handbook. The British Grassland Society, Hurley, pp. 1 l-36. Minson, D.J., 1990. Forage in Ruminant Nutrition. Academic Press, San Diego, 461 pp. Morrison, M.L., Marcot, B.C. and Mannan, R.W., 1992. WildlifeHabitat Relationships: Concepts and Applications. University of Wisconsin Press, Madison, 448 pp. Murray, M.G. and Brown, D., 1993. Niche separation ungulates in the Serengeti: an experimental 62: 380-389.

of grazing

test. J. Anim. Ecol.,

Mursan, A., Hughes, T.P., Nicol, A.M. and Sugiura, T., 1989. The influence of sward height on the mechanics of grazing in steers and bulls. Proc. NZ Sot. Anim. Prod., 49: 233-236. Newman, J.A., Parsons, A.J. and Harvey, A., 1992. Not all sheep prefer clover: diet selection revisited. J. Agric. Sci. (Cambridge), 119: 275-283. Nichols, G. DeLaM., 1966. Radio transmission ments. NZ J. Agric. Res., 9: 468473.

of sheep’s jaw move-

Nunez-Hemandez, G., Holechek, J.L., Arthun, D., Tembo, A., Wallace, J.D., Galyean, M.L., Cardenas, M. and Valdez, R., 1992. Evaluating fecal indicators for assessing energy and nitrogen status of cattle and goats. J. Range Manage., 45: 143-147. Ohajuruka, O.A. and Palmquist, D.L., 1991. Evaluation of n-alkanes as digesta markers in dairy cows. J. Anim. Sci., 69: 1726-1732. Owen-Smith, N., 1979. Assessing the foraging efficiency of a browsing ungulate, the kudu. S. Afr. J. Wildl. Res., 9: 102-l 10. Parker, K.L., Gillingham, M.P. and Hartley, T.A., 1993. An accurate technique for estimating forage intake of tractable animals. Can. J. Zool., 71: 1462-1465. Parker, W.J., Morris, S.T., Garrick, D.J., Vincent, G.L. and McCutcheon, S.N., 1990. Intraruminal chromium controlled release capsules for measuring herbage intake in ruminants-a review. Proc. NZ Sot. Anim. Prod., 50: 437442. Parsons, A.J., Newman, J.A., Penning, P.D., Harvey, A. and Orr, R.J., 1994. Diet preference in sheep: effects of recent diet, physiological state and species abundance. J. Anim. Ecol., 63: 465478. Penning, P.D., 1983. A technique to record automatically some aspects of grazing and ruminating behaviour of sheep. Grass Forage Sci., 38: 89-96. Penning, P.D., 1986. Some effects of sward conditions on grazing behaviour and intake by sheep. In: 0. Gudmundsson (Editor), Grazing Research at Northern Latitudes. Plenum Press, London in association with NATO Scientific Affairs Division, pp. 219226. Penning, P.D. and Hooper, G.E., 1985. An evaluation of the use of short term weight changes in grazing sheep for estimating herbage intake. Grass Forage Sci., 40: 79-84. Penning, P.D., Parsons, A.J., Orr, R.J. and Treacher, T.T., 1991. Intake and behaviour responses by sheep to changes in sward characteristics under continuous stocking. Grass Forage Sci., 46: 15-28.

214

I.J. Gordon /Small Ruminant Research 16 (I995) 203-214

Pfister, J.A., Hansen, D. and Malechek, J.C., 1990. Surgical establishment and maintenance of esophageal fistulae in small ruminants. Small Rumin. Res., 3: 47-56. Pietersen, L.M., Meissner, H.H. and Pietersen, E.W., 1993. Food selection and intake by male impalas in the Timbavati area. S. Afr. J. Wild]. Res., 23: 6-l 1. Raats, J.G. and Clarke, B.K., 1992. Remote control collection of oesophageal fistula samples in goats. Small Rumin. Res., 7: 245251. Renecker, L.A. and Hudson, R.J., 1985. Estimation of dry matter intake of free-ranging moose. J. Wildl. Manage., 49: 785-792. Rice, R.W., 1970. Stomach content analysis: a comparison of the mmen vs. esophageal techniques. In: Range and Wildlife Habitat Evaluation-A Research Symposium. US Dep. Agric. For. Serv. Misc. Publ. 1147, pp. 127-132. Rice, R.W., MacNeil, M.D., Jenkins, T.G. and Koong, L.J., 1983. Simulation of the herbage/animal interface of grazing lands. In: W.K. Lauenroth, G.V. Skogerboe and M. Flug (Editors), Analysis of Ecological Systems: State-of-the-art in Ecological Systems. Vol. 5. Elsevier, Amsterdam, pp. 475488. Roberts, G., Williams, A. and Last, D., 1993. Do Welch hill farmers dream of radioactive sheep? GIS Europe, Nov., 34-36. Senft, R.L., Coughenour, M.B., Bailey, D.W., Rittenhouse, L.R., Sala, O.E. and Swift, D.M., 1987. Large herbivore foraging and ecological hierarchies. Bioscience, 37: 789-799. Sherwin, C., 1994. Farm animals. In: S.D. Wmtten (Editor), Video Techniques in Animal Ecology and Behaviour. Chapman and Hall, London, pp. 125-144. Slater, J. and Jones, R.J., 1971. Estimation of diets selected by grazing animals from microscopic analysis of the faeces-a warning. J. Aust. Inst. Agric. Sci., 37: 238-240. Spalinger, D.E., Hanley, T.A. and Robbins, CT., 1988. Analysis of the functional response in foraging in the Sitka black-tailed deer. Ecology, 69: 1166-I 175 Stobbs, T.H., 1970. Automatic measurement ofgrazing time by dairy cows on tropical grass and legume pastures. Trop. Grassl., 4: 237-244 Stobbs, T.H., 1973. The effect of plant structure on the intake of tropical pastures. I. Variations in the bite size of grazing cattle. Aust. J. Agric. Res., 24: 809-819. Stuth, J.W., 1991. Foraging behaviour. In: R.K. Heitschmidt and J.W. Stuth (Editors), Grazing Management. An Ecological Perspective, Timber Press, Portland, Oregon, pp. 65-83. Stuth, J.W., Kanouse, K.J., Hunter, J.F. and Pearson, H.A., 1981. Multiple electrode impedance pleythsmography system for mon-

itoring grazing dynamics. Biomed. Sci. Instr., 17: 121-124. Stuth, J.W., Kapes, E.D. and Lyons, R.K., 1989. Use of near infrared spectroscopy to assess nutritional status of cattle diets on rangelandsIn: R. Jarrige (Editor), Proc. XVI International Grassland Congress. Nice, France, pp. 889-890. Tilley, J.M.A. and Terry, R.A., 1963. A two-stage technique for the in vitro digestion of forage crops. J. Brit. Grass]. Sot., 18: 104111. Unwin, D.M. and Martin, P., 1987. Recording behaviour using a portable microcomputer. Behaviour, 101: 87-100. US Dep. Agric., 1990. World Agriculture, 1970-1989. Statistical Bulletin No. 815, US Dep. Agric., Washington, DC. Van Dyne, G.M., 1969. Measuring quantity and quality of the diet of large herbivores. In: F.G. Follet and H.K. Beuchner (Editors), A Practical Guide to the Study of the Productivity of Large Herbivores. Blackwell Scientific Publications, Oxford, pp. 5994. Van Dyne, G.M., Brockington, N.R., Szocs, Z., Duek, J. and Ribic, C.A., 1980. Large herbivore subsystem. In: AI. Breymeyer and G.M. Van Dyne (Editors), Grasslands, Systems Analysis and Man. International Biome Programme 19. Cambridge University Press, Cambridge, pp. 269-528. Van Keulen, J. and Young, B.A., 1977. Evaluation of acid-insoluble ash as a marker in ruminant digestibility studies. J. Anim. Sci., 44: 282-287. Vavra, M., Rice, R.W. and Hansen, R.M., 1978. A comparison of esophageal fistula and fecal material to determine steer diets. J. Range Manage., 31: 11-13. Vulich, %A., O’Riordan, E.G. and Hanrahan, J.P., 1991. Use of nalkanes for the estimation of herbage intake in sheep: accuracy and precision ofestimates. J. Agric. Sci. (Cambridge), 116: 319323. Vulink, J.T. and Drost, H.J., 1991. A causal analysis of diet composition in free ranging cattle in reed-dominated vegetation. Oecologia, 88: 167-172. Ward, A.L., 1970. Stomach content and fecal analysis. In: Range and Wildlife Habitat Evaluation. US Dep. Agric. For. Serv. Misc. Publ. No. 1147, pp. 146-158. Warren, J.T. and Mysterud, I., 199 1. Summer habitat use and activity patterns of domestic sheep on coniferous forest range in southern Norway. J. Range Manage., 44: 2-6. White, G.C. and Garrett, R.A., 1990. Analysis of Wildlife Tracking Data. Academic Press, San Diego, 372 pp. Wratten, S., 1994. Video Techniques in Animal Ecology and Behaviour. Chapman and Hall, London, 211 pp.

RQumCn Gordon, I.J., 1995. Ttcnicas

basadas en el animal en la investigaci6n

de la ecologia de1 pastoreo. Small Rumin. Rex., 16: 203-214.

Los ec6logos de pastoreo disponen de varies tecnicas de investigaci6n para medir aspectos de la interacci6n planta/herbiovor. Esta revisi6n se centra en thcnicas basadas en el animal y describe 10s m&odos disponibles para enfrentaese a cuestiones sobre consume, comportamiento alimentario, composici6n de dieta y locaIizaci6n de grandes herfvoros en pastoreo libre. La mayorfa de las tdcnicas han sido desarrolladas y veriticadas para animales domesticados y se ha comprobado que son altamente fiables y precisas. Sin embargo, todas ellas requieren manejar al animal regularmente, bien para adminstrarles las medidas de1 protocol0 6 bien para obtener muestras 6 datos de1 animal. Sin embargo, debido al inter& actual en la ecologfa de pastoreo de animales dom&ticos y salvajes en sistemas extensivos, se necesita una menor aproximaci;d al animal para la obtenci6n de datos. Los recientes desarrollos de la microelectr6nica (recogida de datos aytomatizada y radiotelemetria) pueden proveer a 10s investigadores de datos m&s precisos sobre las estrategias de forrageo de 10s animales en pastoreo libre.