Estimation of Intake in High Producing Holstein Cows Grazing Grass Pasture

Estimation of Intake in High Producing Holstein Cows Grazing Grass Pasture L. A. HOLDEN,' L. D. MULLER,' S. L. FALES2 The Pennsylvania State University University Park 16802 ABSTRACT

spectroscopy, RDP = ruminally degradable protein, PD-MI = pasture DMI, S P =-soluble protein, TDMI = total DMI.

The objectives were to estimate intake of pasture and total DMI by high producing cows grazing grass pastures-and to measure changes in nutrient composition of grass pasture during the grazing season. Sixteen multiparous Holstein cows averaging 31 kg of 4% FCWd at the start of the trial grazed grass pastures at a stocking rate of 2.5 or 3.9 cowsha from April until October 1990. Intake was estimated using Cr2O3 as an indigestible fecal marker. Pasture samples were analyzed for nutrient composition at six times during the grazing season corresponding to the times of intake estimation. Total daily DMI increased from 21.3 kg in early spring to 22.4 kg in late spring and then decreased as lactation progressed; however, DMI exceeded NRC recommendations during most of the grazing season. Daily pasture DMI varied with season, ranging from 11.6 to 15.6 kg and was lowest (11.6 kg) in the summer. Estimated NEL intakes were lower than NRC recommendations in early spring. During the grazing season, pasture ranged from 39 to 48% NDF and from 22 to 30% CP with 15 to 20% ruminally degradable protein on a DM basis. Grazing cows consumed adequate DM from pasture except in early spring. Although nutrient composition of pasture varied with season, quality remained high. (Key words: intake, pasture, orchardgrass, lactating cows)

INTRODUCTION

Because of the potential economic advantages of intensive grazing over conventional stored feeding of forages (10, 24), many dairy producers in the northeastern United States have replaced stored forages with pasture during all or most of the grazing season. One of the challenges in using pasture as the primary or sole source of forage for lactating dairy cows is estimating pasture DMI (PDMI). Researchers outside of the United States (8, 18, 27) have estimated PDMI of lactating cows, but the forage species, environmental conditions, and animal production potential differ from those in the northeastern United States. Estimates of PDMI of high producing cows grazing naturalized grass pastures in the United States have not been published. Likewise, few published data exist that quantify nutrient changes in naturalized grass pasture throughout a grazing season. The objectives of this research were to estimate PDMI and total DMI (TDMI) of high producing Holstein cows grazing naturalized grass pastures in the northeastern United States and to measure changes in nutrient composition of that pasture during an entire grazing season. MATERIALS AND METHODS

Pastures and Cows

Abbreviation key: IVDMD = in vitro DM digestibility, NIRS = near infrared reflectance

Received November 19, 1993. Accepted March 15. 1994. 'Department of Dairy and Animal Science 2Department of Agronomy. 1994 J Dairy Sci 77:2332-2340

Pastures were located at The Pennsylvania State University dairy production research center (University Park). A botanical survey for species presence, conducted using randomly sited transects, indicated that pastures contained approximately 38% orchardgrass (Duct y l i ~glomeratu L.), 34% Kentucky bluegrass (Poa prateneis), 18% smooth bromegrass (Bro-

2332

2333

PASTURE INTAKE OF HIGH PRODUCING COWS TABLE 1. Schedule for application of NH~NOJ. Date April 6 May 30 July 6 August 6 August 30

(kg NHqNOjh) 247.6 194.1 164.7

115.3 98.8

mus inemis L.), and small amounts of assorted

herbaceous weeds. Because of the greater individual mass of orchardgrass, its relative contribution to the total sward mass was higher than reflected in the 38% figure. Very little clover (Trifolium spp.) was observed. Pastures were divided into a replicated paddock system with 14 paddocks per stocking rate, and half of the paddocks were set aside for silage in the spring. Once silage was harvested and regrowth was sufficient, cows were allowed to graze these paddocks. Cows grazed each paddock for 2 d so that the intervals between grazing paddocks were 14 d in the spring and 28 d in the summer and fall. Intervals between grazing were the same for each stocking rate, and paddock size differed such that stocking rate was either 2.5 of 3.9 cowska. A temporary electric fence divided each paddock in half so that cows had access to only the front half of the paddock on the 1st d and to the entire paddock on the 2nd d. Pastures were fertilized with NH4N03 five times during the grazing season and were clipped once in June (Table 1). Cows were allowed to graze during the day (0800 until 1500 h) for 1 wk before the start of the trial to adapt to the new diet and the grazing system. During this time, cows were fed a T M R indoors at night. Sixteen multiparous Holstein cows, averaging 31 kg of 4%FCWd (SEM = 2.76), were blocked by parity, day of lactation, and milk production. Cows averaged 133 d of lactation at the start of the trial and were randomly assigned to a stocking rate of either 2.5 or 3.9 cowska with 8 cows per stocking rate. Cows grazed pastures from April 23 until October 15, 1990. Cows were milked twice daily at 0530 and 1530 h and were fed a concentrate mix at the approximate rate of 1 kg of concentrate DW5 kg of milk with a maximum of 10 kg and a minimum of 4 kg of concentrate DM

TABLE 2. Ingredient and chemical composition of the concentrate mix. (% of

Item

DM)

Ingredients Shelled corn Oats Renprol Beet pulp Molasses, liquid Corn gluten meal Animal-vegetable fat Magnesium oxide Limestone Monosodium phosphate Dynamatem2 Salt Trace mineral and vitamin premix' Selenium premix Sodium bicarbonate Nutrient CP Soluble CP Ruminally degradable CP ADF NDF Ca P Mg K

62.6 7.3 6.1 60 5.2 3.7 2.6 .5 1.2 .7 .2 .9 .4

,1

1 .o

13.9 11.8 9,1 10.8 22.5 I .o .2 .4

1 .o

'Renpro (Baker Commodiues, Rochester, NY) consists pnmanly of meat and bone meal with some blood meal and IS 56% CP with 34% mmnally undegradable protein, 4% soluble protein, and NEL of I 7 8 McaVkg of DM 2Pitman-Moore, Inc (Mundelein. JL) 3Vitmns were added at a rate of 15,000 IU of vitamin f i g , 4000 IU of vitamm Dkg,and 40 IU of vitamin E/ kg

given twice daily in the barn after milking. Concentrate intakes were recorded daily and were adjusted according to the previous week's milk production every 28 d. Ingredient and chemical composition of the concentrate mix are in Table 2. Cows grazed pasture at all times and were fed small amounts (about 2 kg of DWd per cow) of grass silage (harvested from paddocks set aside in spring) in the barn during July. Cows used in this study were part of a larger experiment that evaluated the effects of stocking rate (1 1). An additional year of experimentation was planned for 1991, but data were only collected for one grazing cycle in the spring. Drought conditions in 1991 necessitated feeding large Journal of Dairy Science Vol. 77, No. 8, 1994

2334

HOLDEN ET AL.

amounts of silage during much of the grazing season, so intake estimates were discontinued after the spring cycle. Data from spring 1991 are included only for informational purposes. No statistical analyses were possible. Experimental Measures and Sample Analyses

(FO) was calculated by the equation FO = (grams of Cr dosed per day)/(grams of Cr per gram of fecal DM). The first run of the calculations utilized the IVDMD values from pasture alone. Pasture DMI was determined by difference between the estimated TDMI and the known grain and silage (if fed) DMI. Once DMI of all diet components were estimated, the diet IVDMD was calculated. The second run of the calculations used the diet IVDMD rather than the pasture IVDMD for a more accurate estimate. Only values from this second set of calculations are presented.

Milk production was recorded daily at each milking. Milk samples were taken once weekly at consecutive a.m. and p.m. milkings for analysis of fat and protein. Milk samples were analyzed for fat and protein by Pennsylvania DHIA (Foss 605B Milko-Scan; Foss Elecuic, Statlstical Analyses Hillerad, Denmark). Cows were weighed at the beginning of the trial and at 4-wk intervals The experimental design was a randomized thereafter. Three independent observers scored block design. Within-block replication was accows for body condition based on a five-point complished using field replicates, and there scale (1 = thin to 5 = fat) (33) as cows were was no replication within a block x replicate being weighed. unit. The model used for all cow data was Y = Intake was estimated at six times during the block + replicate + (block x replicate) + stockgrazing season using Cr2O3 as an indigestible ing rate + (block x stoclung rate) + (replicate x fecal marker. Cows were dosed twice daily at stocking rate) + (block x replicate x stocking 0600 and 1600 h for 10 d with 5 g of Cr2O3 rate) + time + (time x stocking rate) + (time x via gelatin capsule. Fecal grab samples were block) + error. The block x replicate term was taken twice daily at dosing on d 7 to 11. used as an error term to test the effect of block Concentrate, pasture, and silage (if fed) sam- and replicate. Time refers to the period when ples were also taken during d 7 to 11. Pasture intake measurements were made and samples samples were plucked by hand from 10 differ- were taken, the month when BW and body ent areas of the paddock to the approximate condition score were analyzed, or the week height to which cows grazed. The fvst dosing when milk data were analyzed. The model used to test forage data was Y = stocking rate of marker began on the 1st d of the mal. Fecal, pasture, and silage samples were + replicate + error. Data were analyzed using freeze-dried, ground through a 1-mm screen, the general linear models procedure of SAS and composited on an equal weight basis (26), and all means presented are least squares across sampling week prior to analyses. Fecal means. composites were analyzed for Cr by atomic absorption spectroscopy according to the RESULTS AND DISCUSSION procedure of Parker et al. (25). Pasture, silage, and concentrate samples were analyzed for Pasture Composltion DM, OM, and CP (2). Ruminally degradable No differences in nutrient composition of protein (RDP)and soluble protein (SP)were determined according to methods of Krish- pasture were found for either stocking rate or namoorthy et al. (21). The ADF and NDF replicate, so the least squares means are the contents were measured according to methods pooled means (n = 4) (Table 3). In the larger of Goering and Van Soest (12) and in vitro DM experiment (data not shown), Fales et al. (11) digestibility O M D ) according to methods of found significant differences in IVDMD and Tilly and Terry (28). Hemicellulose was deter- NDF content of pasture that were due to stockmined by calculating the difference between ing rate. Although the method of sample collection was similar between the two trials, NDF and ADF. Intake was estimated using the equation Fales et al. (11) collected more samples on DMI = fecal output/(l - IVDMD). Fecal output different days, and sometimes in different padJournal of Dairy Science Vol. 77, No. 8. 1994

2335

PASTURE INTAKE OF HIGH PRODUCING COWS TABLE 3. Nutrient composition of grass pasture for six sampling times throughout the grazing season Date' Item

Apr 30

May 28

Jul 2

Jul 30

Aug 27

Sep 24

SEM2

OM CPa Soluble protein& Degradable protein ADP NDP Hemicellulose' In vitro DM digestibility

90.2 24.3 3.3 20.2 24.6 41.3 16.7 68.5

91.3 22.2 5.1 18.6 25.6 44.9 19.3 70.0

89.9 26.6 7.2 16.3 26.6 45.1 18.5 68.8

90.3 25.0 5.4 16.1 26.3 48.2 21.9 64.4

90.5 26.7 7.1 14.9 26.7 44.7 18.0 67.7

91.1 29.7 14.6 19.3 23.6 39.3 15.7 70.0

.56 .EO .75 1.09 .48 1.18 .70 1.21

aMeans differ because of date (f < .05). 'Dates shown are for the 1st d of the 5-d sampling period 2n = 4

docks, compared with those of the current trial. Also, Fales et al. (11) analyzed samples using near infrared reflectance spectroscopy (NIRS). When samples from the present study were analyzed using NIRS, NDF content tended (P = .13) to be higher at the low stocking rate. Using the NIRS technique, the NDF content of pasture averaged 55.1 and 53.8% for low and high stocked paddocks, respectively. The nutrient composition of the pasture (Table 3) showed high but somewhat variable CP throughout the grazing season. Pasture CP was highest in the fall (September 24) and lowest in late spring (May 28). The high CP in the fall reflected the cool wet conditions, and the low CP in late spring might have been the result of N fertilizer application after sampling in May. Even at the lowest CP content of 22% of DM, the pasture was as high or higher in CP than typical stored grass silage and grass hay (23). A h g h proportion of the CP (60 to 80%) in pasture was RDP. Other researchers (1, 30) have shown by both in vitro and in situ techniques that fresh forages have hlgh amounts of RDP. Holden et al. (16) found that in vivo measurements of ruminal NH3 N were higher for dry cows grazing grass pasture than for those fed grass hay or silage. In the present study, the RDP in pasture was highest in the spring and the fall and lowest in the summer, but the SP increased throughout the grazing season. The high RDP content of pasture corresponded to the cooler portion of the growing season and also to the

time when fiber content is lower. Because RDP is lower when fiber is higher, more of the N may be bound in NDFN and may not be as ruminally degradable. The increase in SP throughout the grazing season was not expected and may be related to the accumulated effects of N fertilization. Also, the increase in SP may possibly be associated with sample handling procedures. Kohn and Allen (20) found that freezing decreased the amount of phosphate buffer SP by more than 40% for bromegrass; however, the length of time the samples were frozen before analysis did not affect the amount of SP. In the current study, all samples were lyophilized simultaneously, so samples from the spring were frozen longer than samples from the fall. The increase in SP as the grazing season progressed may be an artifact of the freezing time, but, more importantly, the SP content was probably affected by freezing and may not be truly representative of fresh forages. Further research is warranted on the handling of samples in relation to nutrient content. The composition of protein and fractions for the 1991 early spring pasture was 27.5% CP, 20.9% RDP, and 6.4% SP. As expected, ADF and NDF increased (P < .05) during the warmer summer months and were lowest in the cooler spring and fall months (Table 3). The highest NDF value was 48.2% in late July, and the highest ADF value was 26.7% in late August. The low fiber content of the pasture seems to be uncharacteristic Journal of Dairy Science Vol. 77, No. 8. 1994

2336

HOLDEN ET AL.

TABLE 4. Average nutrient intake from pasture, total DMI, and milk production and composition for 16 cows at six times during the grazing season. Date' Item

Apr 30

May 28

Jul 2

Jul 30

Aug 27

Sep 24

SEM2

Pasture intake DM, kg/da CP, kg/da Degradable protein, kg/da Soluble protein, kg/da Grain DMI, kg/da Total DMI, kdda.3 DMI, % of BW Milk production, kg/da 4% FCM, kg/da Milk fat, % Milk protein, %

13.9 3.3 2.8 .5 7.3 21.3 3.7 36.0 29.7 2.8 2.6

15.1 3.3 2.8

12.9 3.4 2.1 .9 7.6 22.4 3.8 26.9 22.2 2.8 2.7

11.6 2.9 1.9 .6 7.5 20.6 3.4 22.7 18.9 2.9 2.8

14.3 3.8 2.1 1 .o 5.5 19.9 3.3 19.6 16.8 3. I 2.8

15.6 4.6 3.0 2.3 4.8 20.3 3.3 17.4 14.9 3.0 3.0

.60 .15 .ll .07 .12 .57 .19 2.49 2.76 33 .32

.8

7.2 22.4 3.8 30.1 26.2 3.1 2.7

aMeans differ because of date (P < .05). 'Date is the 1st d of the 5-d sampling period. *n = 16. 3Silage intake during July 2 averaged 1.2 kg of DM/d per cow and during July 30 averaged 2.0 kg of DWd per cow. The remainder of the intake, in addition to pasture, was from grain. In all other months only grain was fed.

of orchardgrass but may be partly the result of curate prediction of nutritional characteristics the contribution of other forage species, in based on calculation, particularly for NEL. some cases small amounts of legumes, mixed in with the grass. Intake Estimates Despite differences (P < .OS) in both NDF There were no differences from stocking and ADF because of sampling cycle, there was no difference in IVDMD because of sampling rate in any of the cow measures; thus, least cycle. The hemicellulose content of pasture squares means are the pooled means (n = 16) in varied (P e .OS), ranging between 15.7 and Table 4. Both TDMI and PDMI varied over 21.9%. Hemicellulose is recognized to be the time (P c .05), and TDMI declined with milk most digestible fiber fraction (5, 19). Although production during the grazing season. Average the fiber content did increase during the sum- PDMI was highest in late spring and autumn mer months, the increase was in the hemicellu- and was lowest in summer (Table 4). Because lose fraction, and the IVDMD of the forage did pasture intake is related to availability (14), not change significantly, ranging between 64 some of the variation in PDMI in the current and 70% on a DM basis. Pasture samples from trial may have been the result of variation in spring 1991 averaged 41.6% NDF, 24.3% available herbage. During the spring, estimates ADF, and 17.3% hemicellulose, on a DM ba- of available pasture DM averaged 3600 kg of DMha for the low stocking rate and 2815 kg sis, and had an IVDMD of 71.8%. Using prediction equations (23) that related of DMha for the high stocking rate. During TDN to NEL and using IVDMD in place of the summer, estimates of available pasture DM TDN, the NEL content of pasture ranged be- were 3164 and 2662 kg of DMha for the low tween 1.46 McaYkg of DM in late July to 1.60 and high stocking rates, respectively. During Mcalkg of DM in both late spring and fall. autumn, estimates of available pasture DM Data from these calculations are interpreted averaged 3818 and 3098 kg of DMha for the with caution because the equations used were low and high stocking rates, respectively. Estivalidated using stored forages and may not be mates of available pasture DM were made applicable to fresh forages. More data for com- prior to grazing according to the methods of position of fresh forages are needed for ac- Fales et al. (1 1). Cows were fed small amounts, Journal of Dairy Science Vol. 77, No. 8, 1994

PASTURE INTAKE OF HIGH PRODUCING COWS

approximately 2 kg of DWd per cow, of grass silage in the barn during the summer when pasture availability was low. Increases in PDMI in autumn may be related to reduced grain intake (Table 4). It is not clear how the reduced concentrate intake may have influenced the PDMI. In 1991, TDMI averaged 22.2 kg/d per cow, and PDMI averaged 14.2 kg/d per cow for the first spring cycle when milk production averaged 32 kg/d per cow and concentrate was fed at a rate of 1 kg of concentrate D W 4 kg of milk. Comparison of TDMI values from this study with the NRC (23) recommendations for DMI based on BW and milk production indicates that cows met or exceeded recommendations for DMI at all times except in early spring. Given a similar energy density, a decrease in DMI decreases energy intake. Because NRC recommendations are based on stored forages, comparison of energy intake in megacalories of NEL, as well as DMI, is useful. Figure 1 shows that, similar to DMI, NEL intake met or exceeded NRC requirements except in early spring when cows were deficient in NEL by nearly 3 McaUd. The average TDMI increased from 21.3 kg/d in late April to 22.4 kg/d in late May, increasing energy intake by 2.2 Mcal/d. It is not clear why DMI in early spring was less than optimal when cows were grazing high quality pasture and were at the highest milk production. Mertens (22) showed that intake is related to the NDF content of the diet and that DMI was depressed when forage NDF in the diet fell to .7% of BW or below. Although the NDF content in early spring pasture was somewhat low (41.3% of DM), it was not below the critical level at which DMI may be depressed. Additionally, the NDF content of the pasture remained relatively low throughout the grazing season; however, DMI was only below NRC recommendations in the early spring. Conrad et al. (7) concluded that DMI was governed by the capacity of the digestive tract (gut fill) and by the cow’s energy requirements. In general, diets of low digestibility are subject to physical control. and diets of high digestibility are subject to physiological control. Conrad et al. (7) found the transition point between physical and physiological control to be 66.7% digestible DM for a 454-kg cow producing 16.8 kg of milk/d. Waldo (32) suggested that

2337

this transition point was not fixed but could occur at a higher level of digestibility for higher producing cows. Using this line of reasoning and considering the low NDF content of the pasture, the energy demand of the cow could be the major factor influencing DMI. However, it is unclear why both TDMI and PDMI are low in early spring, when milk production and energy demands are greatest and when pasture quality is high. Bywater (4) suggested that intake models should be dynamic rather than static and that multiple factors influence PDMI and TDMI simultaneously. In the current study, multiple regression analysis of individual cow data using BW, milk production, milk composition, and pasture ADF and NDF contents as predictors of TDMI produced equations with R2 greater than .6 for only 8 of the 16 cows. Inclusion of factors such as herbage availability, cow selection and preference, and differences in grazing management into an intake prediction model for grazing dairy cattle may enable more accurate predictions than traditional approaches. The high water content of the fresh pasture may also limit DMI, and, consequently, DM values for pasture may help to improve prediction equations. Chase (6) reported that DMI declined about .02 kg/100 kg of BW for each percentage unit increase in moisture content of silage-based diets. However, changes in silage quality may have influenced these results because the moisture content of the silage is directly related to its fermentation properties. Dulphy and Demarquilly (9) found that fresh forage in cow diets resulted in higher DMI than either silage or hay. In contrast, Holden et al. (16) found similar DMI for grass pasture, hay, and silage with dry cows. Comparison between studies is difficult because DMI of lactating cows in the current study differs from that of dry cows. The moisture content of forages seems to have variable effects on intake, and it is unclear whether or not the moisture content of early spring grass pasture significantly influenced DMI in the current trial. The low TDMI occurred only in early spring, and the TDMI increased 5% from early to late spring, remaining equal to or above NRC recommendations for the rest of the grazing season. The increase in TDMI on May 28 Journal of Dairy Science Vol. 77, No. 8, 1994

2338

HOLDEN ET AL.

compared with that on April 30 was the result of a 9% increase in PDMI. This increase in TDMI from early to late spring was not caused by increases in DMI resulting from peak lactation because these cows averaged 133 DIM at the start of the trial. Only 3 of the 16 cows were less than 100 DIM. As TDMI increased 5% from early to late spring, milk production declined 16%, partially because of the lower than optimal DMI and NEL intake in early spring. The rapid decline in milk production in early spring in the present study was similar to the decline in milk production for the other cows that were part of the larger study (1 1) as well as those on a previous trial with similar pastures (15). The adaptation period at the beginning of this study may not have been long enough to allow for changes in digestive tract function to be complete before the first samples were collected. A reasonable explanation for the lower DMI in early spring may be related to the retention time of digesta in the rumen. Holden et al. (16) showed a tendency for increased turnover of rumen fluid in cows grazing grass pasture compared with the fluid turnover in cows fed grass silage or hay. Additionally, Bergazhi and Polan (3) observed high fractional passage rates of both rumen fluid and DM from grazing cows. With a faster fractional passage rate and lower retention time, DMI could be increased without a change in gut fill. However, the change in fractional passage rate would not be expected to occur immediately, but rather over a period of time as the cow adapted to the fresh forage diet. Verite et al. (31) measured changes in digestion over a grazing season with fresh ryegrass (Lolium perenne L.), and the time at which cows had the highest intake (1 1.5 kg of OWd) was also the time when fractional passage rate of rumen fluid was the highest (14.6% h). Because TDMI was above NRC recommendations (23) for a majority of the grazing season, high TDMI could partially be the result of rapid digesta passage. Early spring was the exception to this high TDMI, possibly because increasing digesta passage may be an adaptive process. Less than optimal TDMI may partially explain the rapid drop in milk production at the beginning of the grazing season. The NEL values of the pasture, silage, and grain can be calculated by an equation from Tyrrell and Journal of Dairy Science Vol. 77, No. 8, 1994

Moe (29) where IVDMD (Table 3) replaces TDN. For cows in the current study, maintenance energy requirements adjusted for grazing activity (23) reflect a 17.5% increase in NEL for maintenance, 10% for good pasture and 7.5% for walking .7 km/d. Comparison of the adjusted NEL requirements and the calculated NEL intakes showed that cows were deficient in energy by almost 3 Mcal/d per cow in early spring, but met or exceeded the adjusted requirements for the rest of the grazing season (Figure 1). Milk production per unit of TDMI (kilogram per kilogram) of 1.69 in April and 1.34 in May partly reflects a change in forage to concentrate ratio with increasing PDMI in May. Milk production per unit of PDMI was 2.59 in April and 1.99 in May. Clearly, the energetic efficiency of feed utilization or the mobilization of body reserves, or both, are altered from April to May. Despite a lack of gut fill limitation and a demand for additional energy, high producing cows did not consume sufficient DM to meet energy needs during early spring grazing. The CP intake from pasture was highest in the fall, which was reflective of higher pasture intakes and higher CP in the pasture (Table 4). High RDP from pasture in the spring and fall is the result of higher pasture intakes rather than a change in forage composition because RDP did not change with time. However, a

Estimated N E

PZ N R C N E

45

40

.-

35

gm 8 25 I i20 w 15

10

5 n "

Apr

May

Jun

Jul

Aug

Sep

Month Figure I . A comparison of the calculated NEL intake and the NRC requirements adjusted for grazing activity and averaged over 16 cows.

PASTURE INTAKE OF HIGH PRODUCING COWS +Weight

+

2339

recommendations and higher estimated NEL intakes than NRC requirements (23) for most of the grazing season.

Score

CONCLUSIONS

2 540

-2

520 500' Apr

May

Jun

Jut

Aug

Sep

'1.5 Oct

Month

Figure 2. Average BW and body condition score (Bcs) over the grazing season.

larger portion of the CP intake was from RDP in the spring compared with that in the summer and fall. If high RDP from pasture resulted in excess N in the rumen and the excess N was converted to urea, then this energy expenditure could also increase maintenance energy requirements for grazing cows. Research (13, 17) indicates that availability of carbohydrate and protein in the rumen must be synchronized and in the correct proportions for optimal utilization. In the current trial, feeding of concentrate separate from pasture and the high RDP of pasture may have increased the likelihood of uncoupled fermentation in the rumen. BW and Body Conditlon Score

The average BW and body condition score for each month of the grazing season are in Figure 2. Although average BW increased as the grazing season and lactation progressed, appreciable increases in BW did not occur until July. Cows gained an average of 47 kg of BW from the beginning to the end of the grazing season. Body condition score tended to decrease from April to May when TDMI was low and cows were not consuming adequate energy, and average body condition score did not exceed 2.5 until September. Cows finished the grazing season in less than optimal condition despite higher actual DMI than NRC

When properly managed, grass pasture can be high in quality throughout the grazing season. Higher fiber and lower protein can be expected in the summer months when environmental temperatures are warmer, and lower fiber and higher protein contents can be expected during the cooler months. Much of the protein in pasture is degradable, especially in the spring. A complete change from a diet with stored forage to one with pasture may require extensive adaptation to prevent lower than desirable DMI in early spring. High producing cows may not consume enough pasture DM in early spring to prevent rapid decreases in milk production when 65% of the diet is pasture DM and 35% is concentrate. With pasture as the sole forage source, DMI was higher than NRC recommendations for most of the grazing season, but low body condition at the end of the grazing season suggested that increases in maintenance energy for activity that were due to grazing may have been underestimated. ACKNOWLEDGMENTS

The authors gratefully acknowledge the assistance of The Pennsylvania State University farm staff, including W. McNeill, M. Rudy, M. Amsler, and R. Hoffman. We also thank Ron Hoover for assistance with field work and Teni Cassidy for assistance with laboratory work. The authors appreciate the critical review of the manuscript by Gabriella Varga and Harold Harpster. REFERENCES 1 Abdalla, H. 0.. D. G. Fox, and R. R. Seaney. 1988.

Variation in protein and fiber fractions in pasture during the grazing season. J . Anim. Sci. 66:2663. 2 Association of Official Analytical Chemists. 1984. Official Methods of Analysis. 14th ed. AOAC. Washington, DC. 3 Berzaghi, P..and C. E. Polan. 1992. Digesta passage in grazing lactating cows fed with or without corn. J . Dairy Sci. 75(Suppl. 1):233.(Abstr.) 4Bywater. A. C. 1984. A generalized model of feed intake and digestion in lactating cows. Agric. Syst. 13: 167. 5 Collins. M. 1988. Composition and fibre digestion in Journal of Dairy Science Vol. 77. No. 8, 1994

2340

HOLDEN ET AL

morphological components of an alfalfa-timothy sward. Anim. Feed Sci. Technol. 19:135. 6Chase. L. E. 1979. Effect of high moisture feeds on feed intake and milk production in dairy cattle. Proc. Cornell Nutr. Conf., Ithaca, NY. 7Conrad, W. R.. A. D. Pran, and 1. W. Hibbs. 1964. Regulations of feed intake in dairy cows. 1. Changes in the imponance of physical and physiological factors with increasing digestibility. J. Dairy Sci. 46: 1431. 8Dillon. P. 1989. The use of dosed and herbage nalkanes to estimate herbage intake with dairy cows. M.Sc. Thesis, University College, Cork, Ireland. 9 Dulphy, J. P., and C . Demarquilly. 1983. Voluntary feed consumption as an attribute of feeds. Feed Information and Animal Production. G. E. Robards and R. G. Packham, ed. Commonw. Agric. Bur., Slough, England. 10Emmick. D. L. and L. F. Toomer. 1991. The economic impact of intensive grazing management on fifteen dairy farms in New York State. Page 19 in Proc. Am. Forage Grassl. Conf. Am. Forage Grassl. Counc., Georgetown, TX. 11 Fales, S . L , L. D. Muller. M. O'Sullivan. L. E. Lanyon, R. J. Hoover, and L. A. Holden. 1993. Intensive grazing of high-producing Holstein cows: mdk production, forage utilization, and profit potential at three stocking rates. Page 7 in XVII Int. Grassl. Congr. Program Proc. N.Z. Grassl. Assoc., Palmerston North, N.Z. 12Goering. H. K.. and P. J . Van Soest. 1970. Forage Fiber Analyses (Apparatus. Reagents, Procedures, and Some Applications). Agric. Handbook No. 379. ARSUSDA. Washington, DC. 13 Herrera-Saldana. R , R. Gomez-Alarcon. M. Torabi, and J . T. Huber. 1990. Influence of synchronizing protein and starch degradation in the rumen on nutrient utilization and microbial protein synthesis. 1. Dairy Sci. 73:142. 14 Hodgson. J . 1990. Grazing Management: Science into Practice. John Wiley & Sons, Inc.. New York. NY. 15 Hoffman, K.. L. D. Muller, S . L. Fales. and L. A. Holden. 1993. Grazing systems: pasture quality evaluation and concentrate supplementation of lactating cows. J. Dairy Sci. 76:2651. 16 Holden, L. A,. L.D. Muller, G. A. Varga, and P. J. Hillard. 1994. Ruminal digestion and duodenal nutrient flows in dairy cows consuming grass as pasture, hay or silage. J. Dairy Sci. 77:(in press). 17 Hoover, W. H.. and S. R. Stokes. 1991. Balancing carbohydrates and proteins for optimal microbial yield. J. Dairy Sci 74:3630. 18 Journet, M. 1983. Capacite d'ingestion. Bull. Tech. Ctr. Rech. Zootech. Vet. Theix 38:57.

Journal of Dairy Science Vol. 77, No. 8, 1994

19 Keys. J. W., Jr.. P. J. Van Soest, and E. P. Young. 1969. Comparative study of the digestibility of forage cellulose and hemicellulose in ruminants and nonruminants. J . Anim. Sci. 29:361. 20 Kohn. R. A,, and M. S . Allen. 1992. Storage o f fresh and ensiled forages by freezing affects fibre and crude protein fractions. J. Sci. Food Agric. 58:215. 21 Knshnamoorthy, U.. T. V. Muscato, C. J. Sniffen, and P. J. Van Soest. 1982. Nitrogen fractions in selected feedstuffs. J. Dairy Sci. 65:217. 22 Mertens. D. R. 1985. Effect of fiber on feed quality for dairy cows. Page 7 in Proc. 46th Minnesota Nutr. Conf.. Madison, WI. 23 National Research Council. 1989. Nutrient Requirements of Dairy Canle. 6th rev. ed. Natl Acad. Sci.. Washington, DC. 24Parker. W. J., S. N. McCutcheon, and D. H. C a n 1989. Effect o f herbage type and level of intake on the release of chromic oxide from intrmminal controlled release capsules in sheep. N.Z. J. Agric. Res. 32537. 25 Parker, W. J.. L. D. Muller, S . L. Fales, and W. T. McSweeny. 1993. A farm survey of pasture and confined feeding systems for dairy cows in Pennsylvania. Prof. Anim. Sci. 9:13. 26 SAS@ User's Guide: Statistics, Version 5 Edition. 1985. SAS Insi., Inc., Cary, NC. 27 Stakelum. G., and P. Dillon. 1990. Dosed and herbage alkanes as feed intake predictors with dairy cows: the effect of feeding level and frequency of perennial ryegrass. Proc. VI1 Eur. Grazing Workshop, Wageningen, Netherlands. 28 Tilly, J.M.A., and R. A. Terry. 1963. A two-stage technique for the in vitro digestion of forage crops. J. Br. Grassl. Soc. 18:104. 29 Tyrrell, H. R., and P. W. Moe. 1975. Production efficiency in the h g h producing cow. Effect of intake on digestive efficiency. J. Dairy Sci. 58:1151. 30Van Vuuren, A. M., S. Tamminga, and R. S. Ketelaars. 1991. In sacco degradation of organic matter and crude protein of fresh grass (Lolium perenne) in the rumen of grazing dairy cows. J . Agric. Sci. (Camb.) 116:429. 31 Vente, R., B. Remond, and M. J o m e t . 1984. Sites of organic matter and protein digestion in lactating cows fed fresh grass from spring to autumn. Can. J. Anim. Sci. 64(Suppl.):328.(Abstr.) 32 Waldo, D. R. 1966. Forage utilization by the lactating cow. Effect of forage quality on intake and forageconcentrate interactions. J. Dairy Sci. 69:617. 33 Wildman, E. E.,G. M. Jones, P. E. Wagner, R. L. Boman, H. F. T r o w Jr., and T. N. Lesch. 1982. A dairy cow body condition scoring system and its relationship to selection production characteristics. J. Dairy Sci. 65:495.