| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Agron., Univ. of Wisconsin–Madison, 1575 Linden Dr., Madison, WI 53706
b Dep. of Animal Sci., Univ. of Wisconsin–Madison, 1675 Observatory Dr., Madison, WI 53706
c Dep. of Animal Sci., Univ. of Padova, Padova, Italy
* Corresponding author (kaalbrec{at}facstaff.wisc.edu)
Received for publication March 12, 2002.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Abbreviations: ADF, acid detergent fiber • ADG, average daily gain • CC, carrying capacity • CP, crude protein • DM, dry matter • IVTD, in vitro true digestibility • KC–G, kura clover–grass • LW, live weight • NDF, neutral detergent fiber • RC–G, red clover–grass
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Burns and Standaert (1985) reported, based on an extensive review of grazing experiments in the USA, that average daily gain (ADG) and gain per hectare are usually greater on legume–grass systems until N application rates on N–grass systems exceed 200 kg ha-1. Unfortunately, legumes are more difficult to grow, manage, and maintain than grasses. Persistence of legumes is recognized as a major limitation worldwide (Barnes et al., 1985; Marten et al., 1989). The most important forage legumes in the North-Central USA are alfalfa (Medicago sativa L.), red clover, birdsfoot trefoil (Lotus corniculatus L.), and white clover (Trifolium repens L.) (Knight, 1985), but their ability to remain in the pasture under grazing is often limited (Forde et al., 1989; Van Keuren and Matches, 1988).
Kura clover is a rhizomatous perennial legume with a wide range of adaptation that makes it potentially useful as a pasture crop and for soil conservation purposes (Bryant, 1974; Speer and Allison, 1985). This potential is primarily based on its greater ability to persist compared with other forage legumes (Brummer and Moore, 2000; Sheaffer et al., 1992; Zemenchik, 1998). Kura clover yield is usually lower than that of alfalfa, red clover, or birdsfoot trefoil during the earlier years of production, but this trend reverses later because kura clover maintains its plant stand better than the above-mentioned legumes (Sheaffer and Marten, 1991; K.A. Albrecht, unpublished data, 2001). The initial lower yield and subsequent longevity have been attributed to the development of an extensive rhizome system (Bryant, 1974; Kim, 1996; Speer and Allison, 1985). Nutritional characteristics of kura clover are comparable to or better than those of the forage legumes currently seeded in the North-Central USA (Allison et al., 1985). On the other hand, ruminal bloat potential and the need for careful management during establishment are constraints to the use of kura clover (Sheaffer et al., 1992).
Grazing experiments in which animal production is measured are vital in forage evaluation programs. They provide the final assessment of pastures in terms of marketable products and allow economic evaluation of the results. Kura clover has been evaluated in monoculture and binary mixtures with birdsfoot trefoil in terms of animal performance by lambs. Sheaffer et al. (1992) reported lamb gains of 878 kg ha-1 on kura clover monoculture compared with 705 kg ha-1 on birdsfoot trefoil in the same study. But assessments of cattle performance on grazed kura clover have not been reported in the USA.
We hypothesized that a mixture of KC–G would outperform a mixture of RC–G in terms of gain per hectare by increasing herbage mass and maintaining a constantly high quality diet because kura clover would be more persistent than red clover in mixed pastures under grazing. Accordingly, the specific objectives of this experiment were to (i) quantify and compare beef gain per hectare, (ii) estimate productivity and quality of the two pasture types to relate these variables to animal performance, and (iii) assess the botanical composition of the swards.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
The experimental layout, a randomized complete block design, consisted of two blocks with one replication of each treatment: (i) KC–G and (ii) RC–G. Each experimental unit of 2.43 ha was subdivided into six paddocks equal in area and considered a grazing management unit.
Pasture Establishment and Management
The research site was a predominantly grass pasture that had been grazed for most of the past 30 yr. Of the four grazing units, three were dominated by smooth bromegrass (Bromus inermis Leyss.), but a mixture of other grasses with relatively little smooth bromegrass dominated the fourth. Vegetation in this last pasture was killed with glyphosate [N-(phosphonomethyl)glycine] and 2,4-D (2,4-dichlorophenoxy acetic acid) in April 1995. ‘Badger’ smooth bromegrass and ‘Endura’ kura clover were no-till–drilled (Krause model 5200, Krause Corp., Hutchinson, KS) into this pasture at rates of 4.2 and 13.8 kg ha-1, respectively, in May 1995. The other three experimental units were suppressed with paraquat (1,1'-Dimethyl-4, 4'-bipyridinium) at the time of seeding and interseeded with Endura kura clover at 11.4 kg ha-1 or ‘Marathon’ red clover at 7.3 kg ha-1 in mid-April of 1996. Clover seeds were inoculated with appropriate strains of Rhizobium.
After seeding, vegetative competition was controlled by a combination of clipping and grazing to allow adequate establishment of the sown species. Pastures were subjected to rotational stocking after establishment. In addition to smooth bromegrass, Kentucky bluegrass (Poa pratensis L.), orchardgrass (Dactylis glomerata L.), reed canarygrass (Phalaris arundinacea L.), perennial ryegrass (Lolium perenne L.), and tall fescue (Festuca arundinaceae Schreb) grew spontaneously in all of the trial pastures. The legume component of the RC–G treatment was supplemented by frost-seeding 4 kg ha-1 Marathon red clover seed using a spinner-type seeder (Herd Seeder Model I-92, Herd seeder Co., Logansport, IN) every March from 1998 to 2000.
Phosphorus and K were applied every year during the experiment based on soil test recommendations for red clover (Kelling et al., 1991) while no lime application was needed to maintain soil pH at 
6.6. Nitrogen fertilizer was not applied on a routine basis in either treatment. However, N fertilizer was applied on the RC–G treatment in 2000 when the frost seeding of red clover was not successful and the proportion of legume in the sward was estimated to be below 15%. Nitrogen was topdressed in the form of urea [(NH2)2CO] at a rate of 150 kg N ha-1. The total amount of fertilizer was split in three equal amounts and applied in early May, early July, and mid-August.
All pastures were clipped after grazing to a height of about 20 cm to remove reproductive stems once in early summer each year. Isolated patches of Canada thistle [Cirsium arvensis (L.) Scop.] were controlled by hand wiping or localized spraying of glyphosate or clopyralid (3,6-dichloro-2-pyridinecaboxylic acid).
Cattle Management and Grazing Criteria
A variable stocking method was used to maintain both treatments at a similar herbage allowance within a range of 1.5 to 2.0 kg dry matter (DM) kg-1 live weight (LW) at entry into a new paddock in each treatment. All adjustments required removal but not addition of steers. Fifteen or 16 Holstein steers (211 ± 17 kg, 218 ± 13 kg, and 204 ± 23 kg of LW in 1998, 1999, and 2000, respectively) were randomly placed in each grazing unit. Steers were fed on grass–legume pastures for at least 10 d before the trial started. Grazing of the trial pastures began on 28 Apr. 1998, 29 Apr. 1999, and 19 Apr. 2000 (when the rate of growth of both experimental pastures was estimated to be enough to support the assigned steers for at least one complete grazing cycle) and lasted for 184, 140, and 183 d, respectively (when forage growth was estimated to be insufficient to maintain at least four steers at about the average performance for the trial). Paddocks were grazed once every 13 to 35 d with grazing periods of 2 to 7 (typically 3 or 4) d per paddock. Despite this broad range, most of the defoliation intervals were close to the average across seasons of 19 d. Yearly mean for herbage mass remaining after grazing ranged from 1.36 to 1.92 Mg ha-1 DM across treatments.
Cattle received identical veterinary treatment through each season. All steers were implanted with a hormonal growth promoter [Synovex-S (20 mg of estradiol benzoate + 200 mg of progesterone; Fort Dodge Animal Health, Overland Park, KS) in 1998 and 1999 and Ralgro (36 mg of zeranol; Pitman-Moore, Mudelein, IL) in 2000] and reimplanted after 84 d to keep the growth-promoting implant active. Continual access to clean, fresh water; trace mineralized salt; and dicalcium phosphate was provided. Poloxalene was added to the trace mineralized salt as a preventative against bloating for discrete periods of time for the KC–G treatment.
Data Collection and Forage Analysis
Estimates of animal response, such as gain per hectare and ADG, were derived from LW records of all the steers in the trial. Steers were individually weighed on two consecutive days without previous fasting whenever they were placed in or taken away from the trial pastures. In addition, all steers were weighed without fasting 1 out of every 28 d. Gain per hectare was calculated by dividing LW gained in a specified time period by the total surface area of the grazing unit. Carrying capacity (CC) was calculated by dividing the total amount of steer LW by the area of the grazing unit. Herbage allowance was estimated by dividing herbage mass (Mg ha-1) at the beginning of each grazing event by CC. Because herbage mass was estimated with data collected only on the first day of each grazing event, this calculation is not an estimate of the actual forage available for the cattle on a daily basis.
Forage samples were taken before and after each grazing event. Four 0.37-m2 quadrats for pregrazing or eight for postgrazing were placed randomly in each paddock, and vegetation was harvested to ground level. Samples were oven-dried at 60°C until they reached a constant weight. Dry weight was used to estimate herbage mass. The pregrazing samples were ground to pass through a 1.0-mm screen for subsequent laboratory analysis. Estimates of pregrazing nutritive value were performed on one composite sample for each paddock. Samples were analyzed for neutral detergent fiber (NDF) and acid detergent fiber (ADF) concentrations by the Robertson and Van Soest (1981) method as modified by Hintz et al. (1996), in vitro true digestibility (IVTD) by Goering and Van Soest (1970), and N concentration by rapid combustion (850°C), conversion of all N combustion products to N2 and subsequent measurement with a thermoconductivity cell (LECO Model FP-528, LECO Corp., St. Joseph, MI). Crude protein (CP) was calculated as N x 6.25.
Mixture grass, legume, and weed proportions were estimated by near-infrared reflectance spectroscopy. At pregrazing, additional herbage (n 30 samples treatment-1 yr-1) was randomly collected from each of the two pasture types and sorted into two fractions, grass and legume, that were dried as described above. Pure fractions were scanned in duplicate using a single reflectance monochromator near-infrared reflectance instrument (NIRSystem 6500, Foss, Silver Spring, MD). Samples were placed in a rotating cup, and near-infrared spectra (log 1/R) were collected by averaging 32 scans taken between wavelength 400 and 2498 nm at 2-nm intervals. In addition, electronic grass–legume mixes (range from 90 to 10% legume at 10% intervals) using the pure fraction spectra of each season were produced for each corresponding year. The best samples of each year were chosen by the algorithm SELECT (Shenk and Westernhaus, 1991), and samples with a Mahalanobis distance lower than 0.6 were discarded. Mixes of the chosen subset were recreated combining pure fractions based on dry weight. The recreated mixtures were scanned as described above. One spectral data file was created for each of the 3 yr that data were collected. Data acquisition, calibration, and analysis were performed using WinISI 1.50 (Infrasoft Int. Limited, Port Matilda, PA).
Equations for predicting species composition were developed combining the spectral data files of all seasons using wavelength between 1108 and 2482 nm at 8-nm intervals. The calibration equations were computed using modified partial least-squares regression, detrending and standard normal variate as scatter correction (Barnes et al., 1989), and transformation of spectral data into first-derivative terms with curve smoothing calculated over four data points. The coefficients of determination (r2) for estimating the components of the mixtures were 0.99 for grass and 0.98 for legume. The standard error of cross validation was 36 and 57, and the component means ranged from 1000 to 0 g kg-1, for grass and legume respectively. Sward composition of the pasture was predicted from the pregrazing samples that were scanned in the same manner as the calibration samples. Legume and grass content was predicted directly by the near-infrared spectroscopy equation, and weed proportions were calculated by difference.
Statistical Analysis
Analysis of variance (ANOVA) using the general linear model (GLM) procedure of SAS (SAS Inst., 1990) was used to test the effects of block, year, treatment, year x block, and treatment x year interaction of all quantities measured. Year effect was tested against the year x block error term. Estimates of forage nutritive value (NDF, ADF, CP, and IVTD) and botanical composition were weighted by herbage mass available at the moment that the samples were taken. All of the estimates recorded were averaged over periods concurrent with measurements of animal performance. For all of the responses, yearly means or total amount, as appropriate, were used in the ANOVA analysis. Within years, ANOVA was used to check for treatment, block, and time effect. Treatment and block were tested against the block x treatment error term. Regression analysis of relationships between quality estimates and sward composition was evaluated using the GLM procedure.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
|
|
Based on visual estimates at the beginning of the experiment, the components of the grass fraction were mainly smooth bromegrass, with less and variable quantities of Kentucky bluegrass, orchardgrass, reed canarygrass, perennial ryegrass, and a small quantity of tall fescue distributed in patches and were similar for both treatments. By the end of the experiment, the contribution of the grass species segregated according to treatment. The primary grass components of the KC–G pastures were orchardgrass and reed canarygrass, with a lower contribution of Kentucky bluegrass and almost no smooth bromegrass. On the other hand, the species contribution of the grass fraction for RC–G remained similar to the original. In both treatments, the contribution of tall fescue increased but still remained low. This observation supports data reported by Peterson et al. (2001), in which they observed the compatibility and high yield of a kura clover–reed canarygrass mix and the observation by Zemenchik et al. (2001), who found that kura clover can be overly competitive in mixtures with short grasses compared with mixtures with tall grasses.
Greater content of legumes in a pasture usually implies a higher risk of bloat. During this study, three animals died from bloat (3.2% of the herd fed on KC–G), one each year, on the KC–G pasture. All three steers died in the paddocks with the highest proportion of legume and concurrent to the time of the year when kura clover was more productive than the grasses. The use of poloxalene in the trace mineralized salt after the first observation of bloat prevented additional instances of bloating.
Herbage mass at entry into each paddock averaged 2.79 vs. 2.56 Mg ha-1 DM in 1998, 2.81 vs. 2.33 in 1999, and 2.53 vs. 2.06 in 2000 for KC–G and RC–G pasture, respectively. Herbage mass was significantly affected (P < 0.01) by treatment and year, but there was not a treatment x year interaction (P > 0.05). Herbage mass followed a similar pattern for both treatments each year (Fig. 1) and was the typical pattern of forage production in the upper Midwest (Undersander et al., 1991).
|
Both treatment (P < 0.05) and year (P < 0.01) had significant effect on residual herbage mass with no treatment x year interaction (P > 0.10). Differences between treatments in the amount of herbage remaining after grazing may have been due to dissimilar amounts of herbage ingested, different pasture growth rate, or a combination of both. Yearly mean residual herbage was 9% greater in 1998, 15% in 1999, and 16% in 2000 for KC–G than for RC–G. Different residual herbage mass could have accounted, to a certain extent, for greater herbage mass for KC–G pasture by direct contribution to DM and due to the different rates of growth that usually occur at different levels of defoliation (Bryan et al., 2000; Parsons et al., 1988).
Laboratory analyses are intended to be a rapid and economic means to predict animal response. The KC–G pasture was lower (P < 0.01) in NDF and ADF and higher (P < 0.01) in CP and IVTD than the RC–G pasture (Table 3). Year did not affect (P > 0.10) quality assessments, except for CP (P < 0.01), and no treatment x year interaction was detected (P > 0.05). The estimates of nutritive value were evidence of higher potential for animal production on KC–G compared with RC–G pasture. Yearly trends for estimates of nutritive value were the same for both treatments within years (data not shown). All quantities displayed a narrow range of variation as a result of grazing all paddocks at similar stages of maturity.
|
Mean forage IVTD was greater every year in KC–G than in RC–G pastures. Thus, steers on KC–G were exposed to forage of better potential digestibility than those that grazed RC–G pasture. Acid detergent fiber has been widely used to predict digestibility and energy concentrations in feedstuffs (Bath and Marble, 1989; Rohweder et al., 1978) despite the fact that ADF was originally intended merely as a transitional step in the estimation of lignin and other compounds (Van Soest, 1994). The lower concentration of ADF displayed by KC–G is further evidence of the superior nutritive value of this pasture.
According to the National Research Council (1984), the highest requirement for protein by the steers used in this study is 16.6% of their DM intake, and both pastures exceeded this requirement (Table 3). Nevertheless, KC–G had greater CP concentration every year, and this would allow for grazing categories of livestock with higher requirements for this nutrient, such as lactating dairy cows.
The relatively high nutritive value of both pastures in this study is partly a result of the grazing management imposed. The rapid grazing frequency allowed grazing at a vegetative stage of most of the species in the pasture, and the residual herbage, 1.64 Mg ha-1 on average, allowed for rapid regrowth. Although herbage mass was of considerably better nutritive value in KC–G than in RC–G pasture, it cannot be assumed that those differences existed in the diet consumed. Bertelsen et al. (1993) have shown that steers grazing grass–legume pastures were able to select a diet that had lower NDF and ADF and higher CP than the average herbage mass that was available. At the grazing pressure imposed in this trial, it is very likely that steers were able to select a better diet than the average offered.
It is well documented that legumes usually display lower concentration of fiber and greater digestibility than grasses (Buxton and Brasche, 1991; Waldo and Jorgensen, 1981). Difference in forage nutritive value between treatments was primarily due to differences in species composition of the sward. Across years and treatments, NDF and IVTD were the estimates of quality with greatest correlation (r2 = 0.78, P < 0.01 and r2 = 0.61, P < 0.01, respectively) to the proportion of legume in the sward.
As a result of greater forage production and higher nutritive value, KC–G pasture supported better CC, ADG, and gain per hectare than RC–G pasture (Table 4). Carrying capacity is a function of forage yield, its quality, and the animal's efficiency in using it (Van Soest, 1994). The greater CC of KC–G compared with RC–G is to some extent attributed to its higher herbage mass. Neither year (P > 0.05) nor treatment x year interaction (P > 0.10) affected CC. Carrying capacity varied throughout each grazing season following the variation of herbage mass (Fig. 1), being greater early in the season and decreasing as forage availability decreased (Fig. 2, 3, and 4) .
|
|
|
|
Due to higher CC and ADG, the KC–G pasture outperformed RC–G pasture in terms of gain per hectare. Gain per hectare averaged 1021 kg for KC–G pasture, 28% more (P < 0.01) than the 800 kg ha-1 gained on RC–G pasture over the 3 yr of this study (Table 4). This gain per hectare was almost two times greater than the previously reported (Scholl et al., 1974) 430 kg ha-1 on a bromegrass–orchardgrass–alfalfa mix grazed for approximately 116 d at the same location. Likewise, both pasture types produced greater gain per hectare than the 505 kg ha-1 on a birdsfoot trefoil–grass or 556 kg ha-1 on a smooth bromegrass–orchardgrass mixture fertilized with 224 kg N ha-1 and grazed for 142 d reported by Paulson et al. (1977). The superior ADG and gain per hectare of steers in this grazing trial, compared with past research, are attributed primarily to the flexibility afforded by the two legume systems employed. The excellent persistence of kura clover and the recruitment of new red clover plants each spring through frost seeding allowed frequent grazing and grazing through autumn, management that previously was not practiced with grass–legume pastures because of concern for legume persistence. Advances in animal genetics, grazing management, and farm technology probably also played a role.
Year also had an effect (P < 0.05) on gain per hectare, but no treatment x year interaction (P > 0.10) was detected. The unusually short grazing season of 1999 compared with 1998 and 2000 partially explained the differences in gain per hectare due to year. However, during 1998 and 2000, grazing seasons were of similar length (184 vs. 183 d), and these 2 yr still had different (P < 0.01) gain per hectare (1052 vs. 925 kg, respectively). This difference is mainly attributed to the lower herbage mass and subsequent lower CC during 2000 compared with 1998. Gain per hectare was greater early in each season (Fig. 2, 3, and 4), concurrent with the greater herbage mass (Fig. 1).
|
SUMMARY AND CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
The KC–G had greater herbage mass availability and greater CC than RC–G pastures. In addition, KC–G pasture had lower levels of fiber, higher protein concentration, and better digestibility than the RC–G pasture. All these factors are associated with greater proportion of legume in the sward of KC–G pastures. As a result, KC–G pastures displayed greater ADG and gain per hectare than RC–G pastures. Additionally, kura clover demonstrated excellent persistence under rotational grazing in mixture with grasses. This study documents unprecedented steer performance on pasture containing a mixture of grass and legume and a new alternative for beef production and soil conservation in the Upper Mississippi Valley.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. M. Sulc and B. F. Tracy Integrated Crop-Livestock Systems in the U.S. Corn Belt Agron. J., February 6, 2007; 99(2): 335 - 345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. E. Contreras-Govea and K. A. Albrecht Mixtures of Kura Clover with Small Grains or Italian Ryegrass to Extend the Forage Production Season in the Northern USA Agron. J., January 1, 2005; 97(1): 131 - 136. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. Moore, T. A. White, R. L. Hintz, P. K. Patrick, and E. C. Brummer Sequential Grazing of Cool- and Warm-Season Pastures Agron. J., July 1, 2004; 96(4): 1103 - 1111. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
, *, or ** were different at P < 0.10, 0.05, or 0.01, respectively. Pooled standard error of each mean was 12, 26, and 0.12 for GAIN, CC, and ADG, respectively.
