daily herbage allowance, herbage intake by grazing dairy · original article the effect of daily...

17
Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing dairy cows JL Peyraud EA Comeron MH Wade G Lemaire 1 Station de recherches sur la vache laitière, Centre de recherche Inra de Rennes, 35590 Saint-Gilles, France; 2 INTA, Estacion Exp Agrop RafaelaZ CC 22, 2300 Rafaela, Santa Fe; 3 Facultad Cs Veterinaria, UNCPBA, Pinto 399, 7000 Tandil, Buenos Aires, Argentina; 4 Station d’écophysiologie des plantes fourragères, Centre de recherche Inra de Poitou-Charentes, 86600 Lusignan, France (Received 24 April 1995; accepted 5 September 1995) Summary &mdash; The effects of daily herbage allowance (DHA = herbage mass [HM] x daily offered area [DOA]) and cow characteristics upon herbage intake at grazing were assessed in two experiments. In n experiment 1, two DHA (low and medium, 19 and 26 kg organic matter [OM]/cow/day) were com- pared in a continuous design using two groups of five cows. In experiment 2, three DHA (low, medium and high, 19, 29 and 46 kg OM/cow/day) were compared in a 3 x 3 latin square design using three groups of five cows. Mid-lactating cows (six first lactation per trial) were used. Fat-corrected milk at turnout (FCMt) ranged between 17 to 35 kg and live weight (LW) from 510 to 680 kg. Cows strip-grazed plots of veg- etative Lolium perenne and did not receive concentrates. Herbage mass cut to ground level (HM) ranged from 3.5 to 7.1 t OM/ha. In experiment 1, herbage organic matter intake (HOMI) (13.5 vs 14.9 kg/day) and FCM yield (20.6 vs 22.0 kg/day) tended to increase from low to medium DHA but dif- ferences were not significant. In experiment 2, HOMI increased in a quadratic manner (13.8, 16.2 and 16.7 kg/day) and FCM increased linearly (P < 0.01 ) with DHA (20.4, 21.7 and 23.0 kg/day for low, medium and high DHA, respectively). In both experiments, HOMI was consistently lower in first lac- tation compared to adult cows and large between-cow variations within lactation were noted. From the pooled data, HOMI was related to DHA and cow characteristics: HOMI = 7.9 - 98 DHA- 1 + 0.264 FCMt + 0.0073 LW (n = 95, R 2 = 0.60, rsd = 1.77 kg). However, splitting DHA into its two compo- nents accounted for more of the variance in HOMI: HOMI =-20.4 - 115 DOA- 1 + 9.63 HM - 0.873 HM 2 + 0.266 FCMt + 0.0095 LW (R 2 = 0.70, rsd = 1.56 kg). These relationships showed that HOMI was affected by DHA but the original sward herbage mass/structure does have an independant effect in reg- ulating intake. Moreover, voluntary intake increases with the potential of milk yield and this increase could account for the two-thirds of the supplementary energy requirements. dairy cows / grazing / intake / herbage allowance / animal requirements

Upload: others

Post on 03-Jul-2020

10 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

Original article

The effect of daily herbage allowance, herbage massand animal factors upon herbage intake

by grazing dairy cows

JL Peyraud EA Comeron MH Wade G Lemaire

1 Station de recherches sur la vache laitière, Centre de recherche Inra de Rennes,35590 Saint-Gilles, France;

2 INTA, Estacion Exp Agrop RafaelaZ CC 22, 2300 Rafaela, Santa Fe;3 Facultad Cs Veterinaria, UNCPBA, Pinto 399, 7000 Tandil, Buenos Aires, Argentina;

4 Station d’écophysiologie des plantes fourragères, Centre de recherche Inra de Poitou-Charentes,86600 Lusignan, France

(Received 24 April 1995; accepted 5 September 1995)

Summary &mdash; The effects of daily herbage allowance (DHA = herbage mass [HM] x daily offered area[DOA]) and cow characteristics upon herbage intake at grazing were assessed in two experiments. In nexperiment 1, two DHA (low and medium, 19 and 26 kg organic matter [OM]/cow/day) were com-pared in a continuous design using two groups of five cows. In experiment 2, three DHA (low, mediumand high, 19, 29 and 46 kg OM/cow/day) were compared in a 3 x 3 latin square design using three groupsof five cows. Mid-lactating cows (six first lactation per trial) were used. Fat-corrected milk at turnout (FCMt)ranged between 17 to 35 kg and live weight (LW) from 510 to 680 kg. Cows strip-grazed plots of veg-etative Lolium perenne and did not receive concentrates. Herbage mass cut to ground level (HM)ranged from 3.5 to 7.1 t OM/ha. In experiment 1, herbage organic matter intake (HOMI) (13.5 vs14.9 kg/day) and FCM yield (20.6 vs 22.0 kg/day) tended to increase from low to medium DHA but dif-ferences were not significant. In experiment 2, HOMI increased in a quadratic manner (13.8, 16.2and 16.7 kg/day) and FCM increased linearly (P < 0.01 ) with DHA (20.4, 21.7 and 23.0 kg/day forlow, medium and high DHA, respectively). In both experiments, HOMI was consistently lower in first lac-tation compared to adult cows and large between-cow variations within lactation were noted. From thepooled data, HOMI was related to DHA and cow characteristics: HOMI = 7.9 - 98 DHA-1 + 0.264FCMt + 0.0073 LW (n = 95, R2 = 0.60, rsd = 1.77 kg). However, splitting DHA into its two compo-nents accounted for more of the variance in HOMI: HOMI =-20.4 - 115 DOA-1 + 9.63 HM - 0.873 HM2+ 0.266 FCMt + 0.0095 LW (R2 = 0.70, rsd = 1.56 kg). These relationships showed that HOMI wasaffected by DHA but the original sward herbage mass/structure does have an independant effect in reg-ulating intake. Moreover, voluntary intake increases with the potential of milk yield and this increase couldaccount for the two-thirds of the supplementary energy requirements.

dairy cows / grazing / intake / herbage allowance / animal requirements

Page 2: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

Résumé &mdash; Effet des quantités d’herbe offertes, de la biomasse et des caractéristiques ani-males sur l’ingestion des vaches laitières au pâturage. L’effet de la quantité d’herbe allouée(DHA = biomasse (HM) x surface offerte journellement (DOA)) et des caractéristiques des animaux surles quantités de MO d’herbe ingérées au pâturage (HOMI) ont été analysées au cours de deux essais.Dans l’essai 1, deux niveaux de DHA (Bas et Moyen, 19 et 26 kg MO/vache/jour) ont été comparés dansun schéma en continu en utilisant deux groupes de cinq vaches appariées. Dans le second essai,trois niveaux d’herbe allouée (Bas, Moyen et Haut ; 19, 29 et 46 kgljour) ont été comparés dans unschéma en carré latin 3 x 3 en utilisant trois lots de cinq vaches. Tous les animaux étaient en milieu delactation et six primipares ont été utilisées par essai. À la mise à l’herbe, la production de lait à 4 % (FCMt)et le poids vif (LtN) variaient de 17 à 35 kgljour et de 510 à 680 kg. Les vaches n’ont pas reçu deconcentré et ont été conduites en pâturage rationné sur prairies de ray-grass anglais pendant toute ladurée des essais. La biomasse coupée au niveau du sol a varié de 3,5 à 7,1 t MOlha entre les

périodes. Dans l’essai 1, HOMI (13,5 versus 14,9 kglj) et la production de laü 4 % (20,6 versus22,0 kgljour) ont eu tendance à augmenter entre le traitement Bas et Moyen mais les différencesn’ont pas été significatives. Dans l’essai 2, HOMI a augmenté de manière curvilinéaire (13,8 ; 16,2 ;16,7 kgljour) et la production de lait 4 % a augmenté linéairement avec le niveau d’herbe allouée(20,4; 21,7 et 23,0 kgljour). HOMI a toujours été beaucoup plus faible pour les vaches primipares queles vaches adultes et des différences interindividuelles importantes sont apparues pour un mêmerang de lactation. À partir de l’ensemble des données individuelles nous avons montré que HOMIpeut être prédite à partir de DHA et des caractéristiques des animaux : HOMI = 7,9 - 98 DHA-1 +

0,264 FCMt+ 0,0073 LVV (n = 95, R2 = 0,60, etr= 1,77kg). Cependant, la précision de ce modèle estsensiblement améliorée lorsque l’on considère séparément les deux composantes de DHA : HOMI =- 20,4 - 115 DOA-I + 9,63 HM - 0,873 HM2 + 0,266 FCMt + 0, 0095 LW(R2 = 0, 70, etr = 1,56 kg). Cesrelations montrent que HOMI est affectée par DHA mais aussi que la biomasse et/ou la structure ini-

tiale de la prairie affecte également les quantités ingérées. De plus les quantités ingérées augmententavec le potentiel de production des animaux et cet accroissement couvre environ les deux tiers desbesoins supplémentaires de production pour des pâturages de bonne qualité.

vache laitière / pâturage / quantités ingérées / herbe offerte / besoin nutritif

INTRODUCTION

Using pasture as a primary source of energyin the diet of dairy cows has potential eco-nomic advantages. However, one of the chal-lenges in optimizing the nutrition of dairycows at grazing is knowing to what extentfresh grass can meet the high energy require-ments of the dairy cows. The regulation ofherbage organic matter intake (HOMI) is acrucial point but the effects of source of vari-ation in HOMI are not well quantified. Theseinclude mainly the milk yield of the cows,grazing management and provision of con-centrate supplements to the cows.

Among factors of grazing managementaffecting HOMI, daily herbage allowance(DHA), defined as the quantity of herbagecut above a sampling height (generallyground level) offered daily per cow, is of major

importance (Greenhalgh et al, 1966; Combel-las and Hodgson, 1979; Le Du et al, 1979).The relation between HOMI and DHA is basi-

cally curvilinear and HOMI was near the max-imum when DHA was about 25 to 30 kgorganic matter (OM)/day (Le Du et al, 1979).However, these aforementioned trials deal

only with cows producing less than 15 kg ofmilk and the response curve of HOMI to DHA

must be described in high producing cowswhich may be more sensitive due to their

high energy requirements. Moreover, theinteraction between DHA and potential intakeof the cows has not yet been studied.

Many studies have focused on the detailsof the effects of sward factors upon the rate

of herbage intake (Hodgson 1982, 1986).These studies were essentially developedon continuously stocked swards with graz-ing managements which minimize changes

Page 3: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

in sward conditions. They have generallynot gone further than analysis of swardheight. Under rotational grazing systems,where the changes in sward conditions arerapid, the rate of intake appears to be moreclosely related to the green leaf mass thanto the sward height (Hendricksen and Min-son, 1980). Moreover, there are still fewindications about the effects of sward param-eters at the level of daily intake, althoughsuch quantitative information is required toimprove the predictability of HOMI.

The two experiments reported in thispaper aimed at studying the effect ofherbage allowance and cow characteristicsupon HOMI in the case of rotationally grazedperennial ryegrass swards differing by thegrowth number of age of regrowth. Somesward characteristics were recorded to

examine more precisely their effects priorto grazing upon HOMI. These experimentswere part of a larger work designed to quan-tify the effect of controllable factors affectingintake and to analyse the precise swardcharacteristics which determine the avail-

ability of herbage for grazing dairy cows.

MATERIALS AND METHODS

Treatments and experimental design

Two experiments were carried out in spring 1988(experiment 1) and spring 1989 (experiment 2)with grazing dairy cows. In experiment 1, two lev-els of daily herbage allowances (DHA) of 20 (low)and 30 kg (medium) OM/cow measured to groundlevel were compared in a continuous design usingtwo groups of five cows during five successiveperiods. Cows were allocated to one herbageallowance throughout the experiment. In experi-ment 2, three levels of herbage allowances of 20(low), 30 (medium) and 40 kg (high) were com-pared in a 3 x 3 latin square design using threegroups of five cows. Cows were rotationallygrazed on a 1 day paddock on swards of peren-nial ryegrass (Lolium perenne). In the two exper-iments, the contrasting herbage allowances were

achieved by varying the size of the grazing area.Each experimental period lasted 7 (experiment1) and 8 days (experiment 2).

Animals

In both years, autumn calving Holstein cows ofdifferent levels of energy requirements wereselected from a larger group of cows which wererotationally grazed. In experiment 1, ten cows werepaired into two groups of five animals each on thebasis of lactation number, fat-corrected milk yieldat turnout (FCMt), live weight (LW) and calvingdate. There were six first lactation cows. FCMt

ranged from 18 to 32 kg and LW from 550 to 650kg. In experiment 2, 15 Holstein cows were dividedinto three groups of five according to the samecriteria. There were six first lactation animals. FCMt

ranged from 23 to 35 kg and LW from 530 to 690kg at turnout. State of lactation was 140+30 dayspostpartum at the beginning of the experiments.

Between the date of calving and the begin-ning of the grazing season the cows were fed ona maize silage based diet to appetite and receivedconcentrates according to their level of milk pro-duction. The quantities of maize silage consumedwere recorded individually in February and Marchfor all cows in experiment 1. The cows wereturned out to pasture during the first week in Aprileach year. Maize silage and production concen-trate were reduced during 2 weeks and finallycows received grass alone the week before the

beginning of the experiments. No production con-centrates were given during the experiments. Thecows were milked between 0630 and 0730 hoursand between 1600 and 1700 hours each day.

Grazing and pasture managements

The experiments were carried out on swards ofperennial ryegrass (Lolium perenne) cv Vigor(1.5 ha) sown in 1985 in experiment 1 and cvBelfort (2.5 ha) sown in 1986 in experiment 2. Theswards had received liberal amounts of fertilizer N.Each year, at the end of March, the swards weregrazed by heifers, the refusals were cut with aforage harvester and residues removed. At thisstage the swards received 50 (experiment 1) or 75kg N/ha (experiment 2). The fields were then lon-gitudinally divided into two unequal areas withpermanent fences to provide for the two herbage

Page 4: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

allowances in experiment 1 and into three equalareas in experiment 2. The groups of cows wereassigned to one area throughout the experiments.

The cows went to a new paddock every dayafter the morning milking. The areas necessary toprovide the prescribed allowances were basedon estimates of the herbage mass present prior tograzing made on 2 days, once the day before theexperimental period and once on day 3. Becauseof the delay between cutting grass and dry mat-ter determination, a previous estimate of areawas calculated from the herbage height and itsrelation to herbage mass obtained on similarswards. Front and back electric fences weremoved daily at 0700 hours. Back fences weremoved 1 day after front fences to increase thearea allowed to the cows and thus avoidingpoaching. Water and minerals were freely sup-plied in every 1 day paddock.

In experiment 1, three periods were on firstgrowth during the 1 st, 2nd and 3rd weeks of May(24, 31 and 38 days since the sward was toppedfor periods 1, 2 and 3, respectively) and two onregrowths during the last week of May and thefirst week of June (21 and 35 days regrowths forperiods 4 and 5, respectively). During these lasttwo periods, cows grazed area previously used inperiod 2 and period 1, respectively. In experi-ment 2, measurements were realized on firstgrowth in May (respectively, 32, 40 and 48 daysafter topping).

Animal measurements

The HOMI was calculated from estimates of fae-cal output and the digestibility of herbage con-sumed. Faecal output was calculated by refer-ence to chromic oxide (Cr2o3) with the

assumption that Cr203 was totally recovered infaeces (M61ix et al, 1987). Small amounts of con-centrate pellets containing chromic oxide werefed in two meals of 400 g each immediately afterthe morning and afternoon milking in order toadminister 16 g of Cr2o3 per cow/day. The con-centrate included (g/kg DM) Cr2o3 20, barley700, soya-bean meal 240 and molasses 40. Apreliminary dosing period of 10 days was carriedout before the measurements. Any uneaten frac-tion of the concentrate was oven-dried and

weighed daily to determine the actual intake ofchromic oxide. Refusals were scarce except fromone cow during the first two periods of experi-ment 2 (about 200 g/day). Faecal sampling took

place during the last 5 (experiment 1 ) or the last4 (experiment 2) days of each period. All freshdung was sampled (about 50 g of fresh material)each morning on the plot grazed the day beforeand a bulk sample for each cow was realizeddaily. Compared to grab sampling, this samplingtechnique limits the risk of bias in the estimationof daily mean concentration of Cr203 (M61ix andPeyraud, 1987). The faeces were marked by indi-vidual feeding of 60 g/day of different colouredplastic particles (Schulman SA) with the chromicoxide concentrate. These daily bulked sampleswere oven-dried at 80 °C for 72 h and mixed in a

composite sample for the period and the cow inquestion. The cow spent less that 2 h daily outof the plot and it could be assumed that morethan 80% of the defaecations were sampled. Theplastic particles were then discarded and the sam-ples were ground through a 0.8 mm screen forthe analysis.

The digestibility of organic matter of grazedherbage was estimated from the concentrationof nitrogen (N) and acid detergent fibre (ADF) infaecal organic matter and from the pepsin-cellu-lase digestibility (PCD) of herbage (cut at 8 cm ofextended tiller height) according to the relation:DOM = 0.624 + 0.284 PCD + 0.0165 N - 0.00354

ADF (n = 23, r= 0.93, rsd = 0.011 This equationwas derived from 23 indoor trials in which three tofour cows were fed on an ad libitum basis withfresh ryegrass (10), cocksfoot (4), pure whiteclover (7) and mixture of cocksfoot and whiteclover (2 trials).

Estimates of the time spent grazing were car-ried out in experiment 2 on 1 day in each period.Observations were made at intervals of 10 minfrom morning milking time to nightfall. Grazingwas recorded when a cow was seen to prehendgrass with its head close to the sward. Milk yieldswere recorded at each milking. Milk compositionwas determined three times per period fromaliquot sample of two consecutive milkings.These samples were analysed for fat and pro-tein using an infrared milk analyser (Milkoscan,Foss Electric DK 3400 Hillerod, Denmark). Thecows were weighed once a week at the end ofeach experimental period, immediately after themorning milking.

Sward measurements

Herbage mass present prior to grazing was mea-sured by cutting two strips of 5 m long and 0.5 m

Page 5: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

wide at random on the area ahead of the grazingcow. Herbage was cut with a motorscythe (Agria)with a reciprocating blade leaving a stubble ofabout 5 cm. In addition, a single quadrat of 0.30x 0.30 m was cut to ground level with kitchenscissors to estimate the stubble mass in each

strip. Herbage samples were weighed fresh andsubsampled for drying at 80 °C for 48 h. Becauseof the risk of soil contamination, all subsampleswere ashed to express results as organic mat-ter. The stubble herbage mass was added to themotorscythe estimate to give the total quantityoffered. Chemical analyses were only performedon the herbage cut with the motorscythe. Theextended tiller height of the longest leaftip (ETH)and leaf sheath (LSH) from ground level weremeasured before and after grazing. Twenty read-ings were taken at random per treatment duringthe days of intake measurements.

Sample analyses

Organic matter (OM) was determined by ashingfor 5 h at 550 °C. Nitrogen content was obtainedby the Kjeldahl method. The NDF and ADF con-tents were measured according to the method ofVan Soest and Wine (1967) on a Fibertec Mextraction unit (Tecator, Denmark) as describedby Giger and Pochet (1987). Cellulase digestibil-ity (PCB) was determined according to Aufrbreand Demarquilly (1989). Chromic oxide was anal-ysed using the method outlined by Poncet andRayssiguier (1980). One g of faeces or 0.1 g ofconcentrate were digested in hot nitric and per-chloric acids prior to an automated colorimetricdetermination of the hexavalent chromium with

diphenyl-carbazide. The method was adapted toan auto-analyser (Technicon).

Statistical analysis

In experiment 1, animal data were analysedaccording to a split-plot design using the followingmodel: Yiki = Mean + T + Lj + Ck(Ti*Lj) + P, + T¡P1+ LJPI + (Q fl + eijkl, where T = treatment effect (i= 1 to 2); Li = effect of the lactation U 1 to 2);Ck(Ti*Lj) = cow effect (within lactation and treat-ment); P, = period effect (I= 1 to 5); T¡P1 = inter-action treatment x period; LiPi = interaction lac-tation x period; TiLjP, = interaction treatment xlactation x period; e,iki = error term with 24 degrees

of freedom. Treatment, lactation and treatmelactation were tested with Ck(Ti*Lj) as an Eterm.

In experiment 2, animal data were analysed asa 3 x 3 latin square design according to the fol-lowing model: Yiikl = Mean + T + Lj + Ck(Li) + P, +T¡P1 + Ljpl + 8¡¡kl, where T = treatment effect (i= 1 1to 3); Li = effect of the lactation (j= 1 to 2); Ck(Lj)= cow effect (within lactation); P, = period effect (I= 1 to 3); LiT = interaction lactation x treatment;

LiPi = interaction lactation x period; 8¡¡kl = residualterm. The residual term had 22 degrees of free-dom. L¡ was tested with Ck(L¡) as an error term.For herbage data, similar procedures were usedbut only treatment and period were taken intoaccount. All data were analysed using the GLMprocedure of SAS (1987).

To examine the relationships between HOMI,animal factors which can influence voluntaryintake, grazing management and sward charac-teristics before grazing, data from the two exper-iments were combined by multiple regressionanalysis. Predictions of HOMI were performedusing individual estimates of HOMI (n = 95) andtaking into account cow characteristics at turnout(25 cows), herbage allowance and sward char-acteristics during the experiments (19 herbageallowances x periods).

RESULTS

In both years, all the cows completed theexperiment satisfactorily and results fromall animals were included in the statistical

analysis. In experiment 1, temperatureswere normal (mean value of 13.3 °C withmaximum value reaching 21 °C). Experi-ment 2 was characterized by warmer anddryer weather conditions (mean 16.3 °C withmaximum value reaching 28 °C, total rainfallless than 14 mm), which dramaticallyreduced herbage accumulation after grazingand prevented a repetition of the completelatin square. Estimates of herbage mass cutby motorscythe (HMm, t OM/ha) and toground level (HM, t OM/ha) were highly cor-related (HMm = 0.78 HM - 0.90, n = 22,R2 = 0.94) and only HM was presentedthereafter.

Page 6: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

Experiment 1

The actual levels of DHA were near to those

intended although there was some betweenperiod variation. Herbage mass and swardheight were slightly higher (P < 0.10; table I)in medium than in low DHA. This mainlyarose from a much lower herbage accumu-lation after grazing the first growth in lowDHA, herbage mass being 5.16 and 4.38 tOM/ha, respectively, for medium and lowDHA in periods 4 and 5. The proportion ofgreen leaf lamina in the pregrazing biomassaveraged 43% and did not differ accordingto the DHA. Chemical composition of theoffered herbage and PCD were similarbetween the two treatments. There were

significant differences in sward structureand chemical composition of offered

herbage between periods (table I). The high-est herbage mass, ETH and NDF content

of the grass and the lowest crude proteincontent of the grass were recorded duringperiods 3 and 5 (ie, with the oldest growths).The leaf sheath height expressed in pro-portion of ETH was also slightly higher inperiods 3 and 5 than during the three otherperiods (0.37 vs 0.33, P < 0.05). Dry mattercontent of the grass averaged 171 g/kg freshmaterial and was not greatly modifiedaccording to period. PCD was high and wasonly slightly reduced during period 5.

Milk yield was higher and fat content ofmilk was lower in medium than in low

allowance (table II). Fat-corrected milk, milkprotein content and body live weight werenot affected by the level of DHA. Nitrogenand ADF contents of faeces did not varyand thus the digestibility of the selectedherbage was similar for the two levels ofDHA. Cows consumed nearly 1.5 kg moreon medium than on low allowance but the

Page 7: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

difference failed to be significant. Extendedtiller height after grazing was higher inmedium than in low DHA (81 vs 104 mm,P < 0.01 ).

Milk yield and herbage intake were sub-stantially higher in adult than in first lactationcows (+6.3 and +3.2 kg/day, respectively,for milk yield and HOMI). There were nointeractions between lactation and allowance

for any aspects of milk production andherbage intake. Within a lactation number,there was large between-cow variation forHOMI (standard error for the cow effect was4.6 kg/day). These variations were corre-lated to the quantity of maize silage andconcentrate eaten during the winter feed-ing period (r= 0.74). Only one cow grazingin the medium DHA showed a much lower

HOMI than expected according to this rela-tionship (-2.0 kg OM/day). When this cowwas taken out, the correlation was 0.89.

Herbage intake was low in period 1,intermediate in period 5 and highest valueswere recorded during the three other periods

(P < 0.05; table 111). There were noallowance x period interactions for HOMIand milk production.

Experiment 2

The actual levels of DHA were near to those

intended for the low and medium

allowances and similar to those in experi-ment 1. The actual level of DHA was slightlyhigher than intended for the high allowancetreatment. Herbage mass, sward heightand chemical composition of the offeredherbage were similar for the three levels ofDHA. Only crude protein content of thegrass tended to be lower for the low DHA

(P< 0.12). Herbage mass and sward heightincreased during the experiment (P< 0.01;table IV) with highest values observed inperiod 3 when the sward was also charac-terized by a higher LSH/ETH ratio (0.37 vs0.32, P < 0.05), a lower crude protein con-tent and a lower PCD than in periods 1

and 2.

Page 8: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

Milk yield, fat-corrected milk yield, pro-tein content of milk and body live weightincreased linearly with increasing DHA (tableV). Faecal output increased from low to

medium DHA, but did not differ betweenmedium and high DHA. Nitrogen content offaeces increased, ADF content of faecesdecreased and thus digestibility of selected

Page 9: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing
Page 10: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

herbage increased linearly with increasingDHA. Herbage intake increased in a

quadratic manner with herbage allowance(13.8, 16.2 and 16.7 kg/day, respectively,for low, medium and high DHA). Grazingtime was depressed in low DHA and did notdiffer significantly between medium and highDHA. The difference occurred during thesecond half of the day (ie, between after-noon milking and nighfall); the cows grazedonly 130 min in the low DHA compared to180 min for the medium and high DHA.Extended tiller height after grazing increasedlinearly (P < 0.01 ) with the level of DHA (93,146,173 mm, respectively, for low, mediumand high allowance).

Fat-corrected milk yield, body live weight,faecal output and HOMI were consistentlyhigher in adult than in first lactation cows(table V). Although the interaction betweenlactation and allowance was not significant,HOMI was fairly constant between mediumand high allowance for adults cows buttended to increase (P < 0.11 ) for first lacta-tion animals (+0.1 and +0.9 kg/day, respec-tively, for adult and first lactation cowsbetween high and medium DHA). Grazingtime was not affected by lactation (P > 0.10)and consequently the mean rate of intake,calculated by the ratio HOMI/grazing time

tended to be higher in adult than in first lac-tation cows (45 vs 38 g OM/min, P < 0.10).

Digestibility of the selected herbage andHOMI were similar during the first two peri-ods but were significantly reduced (P < 0.01 )during the third period (table VI) despite ahigher grazing time (P < 0.01). Thus, themean rate of intake was consistently lowerin period 3 (33 g OM/min) than in the twoother periods (45 g OM/min, P < 0.01 ).

DISCUSSION

Estimates of energy allowancesand requirements

Without a reference method, validation ofHOMI is difficult. One approach could be tocompare net energy (NE) intake and NErequirements for actual cow’s performances.Calculations were based on the mean datafor the whole time of the experimentsbecause reliable estimates of live weightchange are difficult to obtain in short periods.Thus, the two experimental DHA were con-sidered in experiment 1 and a mean theo-retical DHA was considered in experiment 2.Body live weight changes were calculated by

Page 11: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

the difference between LW in the last and

first period in each experiment, data beingpreviously adjusted for treatment effect inexperiment 2. NE requirements for mainte-nance, milk yield and body reserve deposi-tion were calculated according to Inra (tableVII). NE content of grazed herbage wasderived from a mean digestibility of 0.815.NE intake exceeded NE requirements byabout 1.0 UFL/day in all cases. Althoughsome slight biases in the method used toestimate HOMI should not be excluded, this

may be because maintenance requirementat grazing was underestimated in our cal-culation. According to Langlands et al(1963), energy requirement is increased byabout 20% in good grazing conditions withample quality grass. This is equivalent to1.0 UFL/day for a dairy cow weighing 600 kgand is in good agreement with the previ-ously mentioned difference. Such anincrease in maintenance requirement is nottaken into account in the Inra net energysystem.On the other hand, the method was pre-

cise. The residual standard deviation was1.0 and 1.1 kg OM/cow/day, respectively,in experiment 1 and 2 (ie, 7% of HOMI).

Similar values (5-9%) were obtained byother workers using indirect methods to esti-mate HOMI in grazing dairy cows (Combel-las and Hodgson, 1979; Le Du et al, 1981;Arriaga Jordan and Holmes, 1986).

Effect of herbage allowanceon herbage intake, grazing behaviourand milk yield

The response of HOMI between low and

medium DHA was of a similar magnitude inexperiment 1 and 2 (+0.19 and 0.23 kg/kgextra allowance, respectively) although it

failed to be significant in experiment 1 prob-ably because the effect of allowance wastested with cow variation as an error term

and also because one cow showed an unex-

pectedly low herbage intake in medium DHAtreatment. In experiment 2, the relationshipbetween intake and allowance was curvi-linear. Intake increased by 2.4 kg OMbetween low and medium allowance and

by only 0.5 kg (not significant) with a fur-ther increase of allowance. Similar curvilin-ear relationships have been previouslyreported in several trials under strip-graz-

Page 12: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

ing management and involving similar shortexperimental periods (Greenhalgh et al,1966, 1967; Combellas and Hodgson, 1979;Le Du et al, 1979). However, these previ-ous results referred to dairy cows weighingless than 500 kg, producing less than 16 kgFCM/day and eating less than 13 kgOM/day. Moreover, our results showed thatthe response to allowance might be differentaccording to the animal characteristics.Herbage intake of first lactating cowsincreased between medium and highallowance and this was not the case foradult cows. This might be explained bysome social competition when young cowsgrazed together with older and more expe-rienced cows.

Taken together the two experimentsshow that individual herbage intake can bepredicted when taking into account somecow characteristics and the daily herbage

allowance measured to ground level (tableVIII, eq 1 The quadratic function of DHAhas often been proposed to predict intake(Meijs and Hoekstra, 1984; Stockdale, 1985;Caird and Holmes, 1986). However, intakeshould not decrease at very high herbageallowances and a hyperbolic model mustbe preferred, even though the precision isquite similar (table VIII, eq 2) for the twoequations. Gibb and Treacher (1976) pro-posed an asymptotic function of DHA butthey did not consider the potential intake ofthe animal. Extended tiller height after graz-ing (ETHa) and DHA were correlated(r= 0.81, n = 19) and, therefore, ETHa couldalso be used to predict intake. However, themodel was less accurate when grazing man-agement was characterized with ETHa thanwith DHA (table VIII, eq 3).

Measurement of milk yield was not themain interest in our experiments because

Page 13: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

of the sudden suppression of the concen-trate at turnout and the short duration of the

experimental periods. However, there weresubstantial increases of milk yield withherbage allowance in both experiments:milk protein content increased with allow-cance in experiment 2 and milk fat contentdecreased with DHA in experiment 1. Theseresults would be expected where the level ofenergy intake was increased. They werealso in general agreement with those ofstudies on the effect of stocking rate andinvolving much longer periods of experiment(Hoden et al, 1991 ).

The grazing efficiency, defined as intakeper unit of allowance, was 0.56 in mediumDHA where intake was hardly depressedbut decreased up to 0.36 for high allowanceand increased to 0.71 for the low allowance.A value of about 0.55 was also obtained in

experiment 1 for the medium DHA. Thesedata support the suggestions of Combellasand Hodgson (1979) and Le Du et al (1979)that under strip-grazing management,herbage intake depression occurs whenabout one-half of the herbage mass hasbeen consumed. However, this practicalrule does not in itself assist the under-

standing of sward factors which may affectherbage availability and intake.

On low DHA treatment, grazing time wasslightly reduced and the difference occurredin the second part of the grazing day afterthe afternoon milking. This agrees with pre-vious observations reported by Le Du et al(1979) and Combellas and Hodgson (1979)for strip-grazing dairy cows and Jamiesonand Hodgson (1979) for grazing calves. Nomeasurements of grazing time wererecorded after nightfall. However, furtherexperiments (Astigarraga, 1994; Peyraudet al, unpublished) showed that grazing timeduring night hours is always short (about30 min) for similar grazing conditions. Thus,it is reasonable to assume that the differ-

ence in daylight grazing time may reflectthe between treatment difference in total

daily grazing time. It has been suggestedthat this decrease reflects the ability of cat-tle to anticipate the change to the next stripof fresh grass (Tayler, 1953; Jamieson andHogdson, 1979). Alternatively, cows mayalso abandon grazing because sward struc-ture may represent a physical limit to pre-hend the grass. Indeed, this decrease in

grazing time with low DHA and hence withlow residual sward height contrasted withthe increase in grazing time to compensatefor low herbage height generally reportedin continuous stocked animals (Zoby andHolmes, 1983; O’Sullivan, 1984; Hodgson,1986). Indeed, short continuously grazedtemperate swards generally contain a densemat of green leaf whereas in our study, thesward grazed at low DHA ended with verylittle green leaf mass compared to high DHA(86 vs 840 kg/ha, ie, 3 and 20% of residualherbage mass, respectively, for low andhigh DHA; Wade, 1991 The data presentedby Penning et al (1994) also strongly sug-gested that content of leaf lamina was oneof the most important factors determiningherbage availability in a sward under rota-tional grazing.

Effect of pregrazing herbagemass/structure upon intake

Variations of intake between periods werenoted in the two experiments. They cannotbe related to the stage of lactation. In exper-iment 1 variation of intake appeared to beerratic on a time basis and in experiment 2less than 1 week elapsed between the firsttwo periods and the last period where intakewas sharply decreased. Indeed, splittingherbage allowance into its two components(offered area and herbage mass) accountedfor more of the variance in individual

herbage intake (table VIII, eq 4), thequadratic term of HM being highly signifi-cant. Herbage mass and extended tillerheight before grazing (ETHb) were closelycorrelated (r= 0.94, n = 19); therefore, ETHb

Page 14: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

had also a highly significant effect (tableVIII, eq 5) when added to DHA. These rela-tionships clearly show that the pregrazingsward structure does have an independenteffect upon herbage intake of grazing dairycows and that DHA or ETHa alone is notsufficient to predict herbage intake accu-rately. Using multiple regression analysis,Stockdale (1985) detected a small positiveeffect of herbage mass upon herbage intakeand Stakelum (1986b) also demonstratedexperimentally an increase of herbageintake with herbage mass. Meijs (1981)failed to identify such an effect in an exhaus-tive study but in his study herbage massand maturity were confounded.

The quadratic component for herbagemass or height supports an optimum range ofmass or height for intake. From these rela-tions, it appears that HOMI might not begreatly affected between 4.0 and 5.5 t OM/habut might be reduced by 2 kg/day when HMfalls from 4 to 3 t/ha (ie, 2.2 and 1.4 t OM/hafor HMm) or increases from 5.0 to 6.5 tOM/ha (ie, 3.0 to 4.2 t OM/ha for HMm). Adecline of HOMI for high herbage mass hasbeen previously reported (Hodgson andWilkinson, 1968; Jamieson, 1975) but inthese studies high herbage mass and lowquality of herbage were confounded. This isnot the case in our study where taking intoaccount the pepsin-cellulase digestibilityimproved slightly the precision of the modelbut had little effect upon the coefficients forHM and HM2 (table Vill, eq 6). These datacould explain the results of Combellas andHodgson (1979), who found that intake ofcows was lower at high than at low herbagemass at comparable levels of digestibility andallowance. Because of the narrow range ofthe proportion of green leaf lamina in HM (41to 47%) the precision of eq [4] in table VIIIwas not improved when considering thegreen leaf mass instead of HM (table VIII,eq 7). Moreover, despite its theoritical advan-tage when contrasting proportion of greenleaf material occurs, eq [7] (table VIII) requirestedious work to estimate green leaf mass.

It is reasonable to assume that when

herbage mass is not related to herbagequality, the increased mass of the swardincreases intake primarily because cowshave access to a greater proportion of eas-ily harvestable material before they had tograze the deeper horizons. The higher pre-hensibility of taller swards is clearly demon-strated in continuous grazing where rate ofintake is linearly related to sward height(Allden and Whittaker, 1970; Hodgson,1982). The reason why intake decreasesfor the highest herbage mass is not clear.One possibility is that the amount taken perbite was too large, thus resulting in manymanipulations of the bites in the mouthbefore swallowing. This may dramaticallyreduce biting rate as the rate of jaw move-ment is remarkably constant across a largerange of sward conditions (Black and Kenny,1984). Biting rate was not recorded in ourstudies but it was shown to decline as sward

height or herbage mass increased (Hodg-son, 1985; Penning et al, 1994).

No interaction between allowance and

period was noted in experiment 1 and theterm DHA x HM was never significant in theequations proposed in table VIII. The resultsreported by Greenhalgh et al (1966) withdairy cows showed a remarkably constantrelationship between intake and herbageallowance in a series of short-term studieson different swards. Combellas and Hodg-son (1979) also failed to show a clear DHAby herbage mass interaction in an experi-ment designed to examine the interactionbetween these two factors. However, thedata of Stakelum (1986a and b) suggest aherbage allowance by herbage mass inter-action. This aspect merits further study.

Between-cow variation of intake

Because of the sudden suppression of theconcentrate at the beginning of the experi-ments, mean individual FCM yield was con-

Page 15: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

sistently lower, but well correlated with FCMat turnout (FCMt: FCM = 5.1 + 0.60 FCMt,n = 25, R2 = 0.81, rsd = 1.6 kg). This meansthat cows did not achieve their potential pro-duction during the trials. Moreover, becausecorrelations between intake and milk yieldmay be due to intake limiting milk secretion,we have preferred to use FCM at turnoutas a covariate in the equations proposed intable VIII.

Voluntary intake changes in proportionto potential milk yield and, according to therelationships found in table VIII, increased byabout 270 g OM/kg FCMt. This numericalvalue agrees with previously published par-tial coefficients of multiple regression anal-ysis. Greenhalgh et al (1966) and Curranand Holmes (1970) reported values of250 g/kg milk. Values ranged between 170to 330 g/kg milk in the work of Caird andHolmes (1986) according to the modelsused. In our situation, this additionalincrease in HOMI accounts for the two-thirdsof the supplementary net energy require-ments for 1 kg of FCM. This suggests thatwhen plentiful amounts of high quality grassare available, high yielding cows could sat-isfy a large proportion of their energyrequirements from grazing alone. Accord-ing to eq [4], [5], [6] and [7], HOMI increasesby 0.9 to 1.0 kg/100 kg LW. Increases ofabout 1.5 kg/100 kg LW were obtained indirect comparisons between cattle differingin size (Hodgson and Wilkinson, 1967; Zobyand Holmes, 1983). Similarly partial coeffi-cients obtained by Greenhalgh et al (1966)and Caird and Holmes (1986) were between0.8 and 1.1 kg/100 kg LW. No intrinsic effectof the lactation number was detected andthus milk yield and live weight accountedfor the observed difference of intakebetween first lactating and adult cows. Stageof lactation was not significant in modelsprobably because it varied in a narrow range(from 100 to 200 days postpartum). Cairdand Holmes (1986) also showed that HOMIis hardly affected by the stage of lactationafter the second month of lactation.

CONCLUSION

The results indicate that herbage intake ofindividual grazing cows is greatly affectedby herbage allowance but the original swardstructure plays a role in regulating intake.Intake of grass is likely to decline at a pro-gressively faster rate when herbage mass isbelow 4.0 t OM/ha (ie, 2.2 t OM/ha with amotorscythe cut). In addition, the resultsshow that voluntary intake increases withthe potential of milk yield and this may havesome practical implications for feeding con-centrates.

ACKNOWLEDGMENTS

The authors wish to thank B Marquis, R Papinand the dairy staff at Mejussaume for care of theanimals and work in the fields and A Stephantfor chemical analysis.

REFERENCES

Allden WG, Whittaker IA (1970) The determinants ofherbage intake by grazing sheep: the interrelationshipof factors influencing herbage intake and availability.Aust J Agric Res 21, 755-766

Arriaga Jordan CM, Holmes W (1986) The effect of con-centrate supplementation on high yielding dairy cowsunder two systems of grazing. J Agric Sci (Camb)107, 453-461

Astigarraga L (1994) Effet de la fertilisation azotée du ray-grass anglais sur la digestion et les quantit6sing6r6es chez la vache laiti6re au p5turage. Theseuniversité Rennes I, France, 62 p

Aufr6re J, Demarquilly C (1989) Predicting organic mat-ter digestibility of forage by two pepsin-cellulasemethods. XVI International Grassland Congress,Nice, France, 877-878

Black JL, Kenny PA (1984) Factors affecting diet selec-tion by sheep. II. Height and density of pasture. AustJ Agric Res 35, 565-578

Caird L, Holmes W (1986) The prediction of voluntaryintake of grazing dairy cows. J Agric Sci (Camb)107, 43-54

Combellas J, Hodgson J (1979) Herbage intake andmilk production by grazing dairy cows. 1. The effectsof variations in herbage mass and daily herbage

Page 16: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

allowance in a short-term trial. Grass For Sci 34,209-214 4

Curran MK, Holmes W (1970) Prediction of the voluntaryintake of food by dairy cows. 2. Lactating grazingcows. Anim Prod 12, 213-224

Gibb MJ, Treacher TT (1976) The effect of herbageallowance on herbage intake and performance oflambs grazing perennial ryegrass and red cloverswards. J Agric Sci (Camb) 86, 355-365

Giger S, Pochet S (1987) M6thodes d’estimation desconstituants pari6taux dans les aliments destinesaux ruminants. Bull Tech CRZV Theix, INRA70, 49-60

Greenhalgh JFD, Reid GW, Aitken JN, Florence E (1966)The effects of grazing intensity on herbage con-sumption and animal production. I. Short term effects

in strip-grazed dairy cows. J Agric Sci (Camb) 67,13-23

Greenhalgh JFD, Reid GW, Aitken JN (1967) The effectof grazing intensity on herbage consumption andanimal production. II. Longer-term effects in strip-grazed dairy cows. J Agric Sci (Camb) 69, 217-223

Hendricksen R, Minson DJ (1980) The feed intake andgrazing behavior of cattle grazing a crop of Lablabpurpureus. J Agric Sci (Camb) 95, 547-554

Hoden A, Peyraud JL, Muller A, Delaby L, Faverdin P(1991) Simplified rotational grazing management ofdairy cows: effects of rates of stocking and concen-trate. J Agric Sci (Camb) 116, 417-428

Hodgson J (1982) Ingestive behavior. In: Herbage IntakeHandbook (JD Leaver, ed), Br Grass Soc, Oxford,UK, 113-138

Hodgson J (1985) The control of herbage intake in thegrazing ruminant. Proc Nut Soc 44, 339-346

Hodgson J (1986) Grazing behavior and herbage intake.In: Grazing (J Frame, ed), Br Grass Soc, Lavenham,UK, Occasional Symposium no 19, 51-63

Hodgson J, Wilkinson JM (1967) The relationshipbetween live-weight and herbage intake in grazingcattle. Anim Prod 9, 365-376

Hodgson J, Wilkinson JM (1968) The influence of thequantity of herbage offered and its digestibility onthe amount eaten by grazing cattle. J Br Grass Soc23, 75-80

Institut National de la Recherche Agronomique (Inra)(1988) Ruminant Nutrition: RecommandedAllowances and Feed Tables (R Jarrige, ed), INRAJohn Libbey, London, Paris

Jamieson WS (1975) Studies of the herbage intake andgrazing behaviour of cattle and sheep. PhD Thesis,University of Reading, Reading, PA, USA

Jamieson WS, Hodgson J (1979) The effect of dailyherbage allowance and sward characteristics uponthe ingestive behavior and herbage intake of calvesunder strip-grazing management. Grass Forage Sci34, 261-271

Langlands WS, Corbett JL, McDonald I, Reid GW (1963)Estimates of the energy required for maintenanceby adult sheep. 2. Grazing sheep. Anim Prod 5,11-16 6

Le Du YLP, Combellas J, Hodgson J, Baker RD (1979)Herbage intake and milk production by grazing dairycows. 2. The effects of level of winter feeding anddaily herbage allowance. Grass Forage Sci 34, 249-260

Le Du YLP, Baker RD, Newberry RD (1981) Herbageintake and milk production by grazing dairy cows.3. The effects of grazing severity under continuousstocking. Grass Forage Sci 36, 307-318 8

Meijs JAC (1981) Herbage intake by grazing dairy cows.Pudoc, Wageningen, the Netherlands. Agric ResReport no 909, p 276

Meijs JAC, Hoekstra JA (1984) Concentrate supple-mentation of grazing dairy cows. 1. Effect of con-centrate intake and herbage allowance on herbageintake. Grass Forage Sci 39, 59-66

M61ix C, Peyraud JL (1987) Utilisation de l’oxyde dechrome chez les vaches laitibres pour la previsiondes quantités de feces 6mises. 2. Comparaison desm6thodes de pr6l6vement de feces par voie rectaleet par collecte globale. Reprod Nutr Dev 27(IB), 217-218

M61ix C, Peyraud JL, Vérité R (1987) Utilisation del’oxyde de chrome chez les vaches laitibres pour laprévision des quantit6s de feces 6mises. 1. Etudedes variatons du taux de recuperation et ses con-sequences sur I’estimation de la digestibilité et desquantit6s ing6r6es de rations d’herbe et d’ensilage demafs. Reprod Nutr Dev 27(I B), 215-216 6

O’Sullivan M (1984) Measurement of grazing behaviorand herbage intake on two different grazing man-agement systems for beef production. In: GrasslandBeef Production (W Holmes, ed), Martinus NijhoffPublishers, The Hague, 141-150

Penning PD, Parsons AJ, Orr RJ, Hooper GE (1994)Intake and behaviour response by sheep to changesin sward characteristics under rotational grazing.Grass Forage Sci 49, 476-486

Poncet C, Rayssiguier Y (1980) Effect of lactose sup-plement on digestion of lucerne hay by sheep. 1.Sites of organic matter and nitrogen digestion. J AnimSci 51, 180-185

Statistical Analysis Systems (1987) SASlSTAT Guidefor Personal Computers, Version 6. SAS Institute,Cary, NC, USA

Stakelum G (1986a) Herbage intake of grazing dairycows. 2. Effect of herbage allowance, herbage massand concentrate feeding on the intake of cows graz-ing primary spring grass. IrJAgric Res 25, 41-51

Stakelum G (1986b) Herbage intake of grazing dairycows. 3. Effect of herbage mass, herbage allowanceand concentrate feeding on the herbage intake ofdairy cows grazing on mid-summer pasture. IrJAgricRes 25, 179-189

Page 17: daily herbage allowance, herbage intake by grazing dairy · Original article The effect of daily herbage allowance, herbage mass and animal factors upon herbage intake by grazing

Stockdale CR (1985) Influence of some sward character-istics on the consumption of irrigated pastures grazedby lactating dairy cattle. Grass Forage Sci 40, 31-39

Tayler JC (1953) The grazing behavior of bullocks undertwo methods of management. Br J Anim Behav 1,72-77

Van Soest PJ, Wine RH (1967) Use of detergents in theanalysis of fibrous feeds. VI. Determination of plantcell wall constituents. J Ass OffAgric Chem50, 50-55

Wade MH (1991) Factors affecting the availability ofvegetative Lolium perenne to grazing dairy cowswith special reference to sward characteristics, stock-

ing rate and grazing method. PhD dissertation, Uni-versit6 de Rennes, France, 77 p

Zoby JLF, Holmes W (1983) The influence of size ofanimal and stocking rate on the herbage intake andgrazing behavior of cattle. J Agric Sci (Camb) 100,139-148