Establishing clover (Trifolium spp.) into permanent pasture

Transcription

1 Sowing Method Effects on Clover Establishment into Permanent Pasture David Schlueter and Benjamin Tracy* ABSTRACT A study was conducted from 2009 to 2011 near Blacksburg, VA, to gain a better understanding of how sowing method affected establishment and persistence of clover in permanent cool-season grass pastures. Four 1.1-ha pastures were split in half and assigned a broadcasted or no-till drilled sowing treatment. Pasture treatments were sown with an equal proportion of red (Trifolium pratense L.) and white clover (Trifolium repens L.) each at 4.4 kg ha -1 in February Residual grass biomass on pastures was measured at sowing, and clover seedling density was counted 2 mo later. Grass, white clover, red clover, and weed biomass were measured four times during each growing season. Broadcast treatments had 56% more clover seedlings than drilled treatments 2 mo after sowing, but this difference was not significant (P = ). No difference (P > 0.10) for clover biomass was observed between sowing treatments in any year, yet clover establishment was considered successful (>25% of pasture composition). In the drilled treatments, clover seedling density was negatively affected by the amount of residual grass biomass present during sowing (P = ). In the broadcasted treatment, a negative quadratic relationship between clover seedling density and residual grass biomass at sowing was found (P = ). For successful establishment of clovers into permanent pastures, these data imply that removing residual grass biomass before sowing was more important than seeding method. Crop Ecology & Physiology Establishing clover (Trifolium spp.) into permanent pasture can be challenging, because many factors like competition from existing the sward, subsequent grazing management, or weather fluctuations can affect this process. Two common overseeding methods for clover include surface sowing without disturbance (broadcasting) or drilling of seed through undisturbed sod and residue (Pearson and Ison, 1997). Frost-seeding is a form of overseeding when seed is broadcasted mid-winter on top of snow and frozen ground or in early spring after snowmelt (Casler et al., 1999). Once sown, the freezing and thawing of the ground helps to incorporate seed into the soil. This allows for better seed-to-soil contact and reduces the equipment required for planting (Kankanen et al., 2001). Both no-till drilling and broadcasting have strengths and weaknesses. The advantage of no-till drilling is more control over seeding depth to ensure good seed-to-soil contact (Campbell, 1985a). A disadvantage of no-till drilling is that use can be limited by steep topography. Broadcasting requires less machinery and is virtually unrestrained by topography. However, no-till drilling is considered superior to broadcasting, because more control helps ensure a proper seeding depth (Taylor et al., 1972). Several experiments that have compared broadcasting with no-till drilling found varying results. Taylor et al. (1969) D. Schlueter and B. Tracy, Virginia Tech, Department of Crop and Soil Environmental Sciences, Smyth Hall (0404), Blacksburg, VA Received 27 Jan *Corresponding author (bftracy@vt.edu). Published in Agron. J. 104: (2012) Posted online 15 June 2012 doi: /agronj Copyright 2012 by the American Society of Agronomy, 5585 Guilford Road, Madison, WI All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. discovered that drilling clover seed was important for successful establishment of red clover (Trifolium pratense L.). They found that extreme moisture and temperature fluctuations on the surface of the soil caused low establishment of broadcasted seed. Cuomo et al. (2001) found no difference between drilling and broadcasting in establishment across several forage legumes. They learned that the main factor that determined legume establishment was the suppression of grass species. Byers and Templeton (1988) discovered that drilling produced more alfalfa (Medicago sativa L.) biomass over broadcasting, and Mueller and Chamblee (1984) found that broadcasting was more effective when sowing was done in late winter than in spring. In their experiment, white clover broadcasted in mid-february had greater establishment than when broadcasted in mid-march. When averaged between the two sowing dates, however, white clover (Trifolium repens L.) establishment was greater in the drilled treatments. Although sowing method should have a major impact on clover establishment, relatively few studies have compared broadcasting (frost-seeding) to no-till drilling. More information is needed about clover overseeding methods to provide guidelines to help improve establishment success. To add to this body of research, a pasture experiment was conducted from 2009 to 2011 near Blacksburg, VA. The primary objective of this study was to compare the effectiveness of broadcasting (frost-seeding) and no-till drilling for establishment of white and red clover. A secondary objective was to identify potential management variables that may help explain the success or failure of seeding methods. MATERIALS AND METHODS Study Site The pasture experiment was conducted at Virginia Tech College Farm near Blacksburg in Montgomery County, Virginia (37 11' N lat, 80 35' W long) during the 2009, 2010, and 2011 Agronomy Journal Volume 104, Issue

2 Table 1. Historic and monthly average temperature and total precipitation for Feb. Oct. 2009, 2010, and 2011 at the Kentland Farm study location near Blacksburg, VA. Historic avg Month Temp. Precip. Temp. Precip. Temp. Precip. Temp. Precip. C cm C cm C cm C cm Feb Mar Apr May June July Aug Sept Oct Avg. temp., C Total precip., cm growing seasons. The soils were of the Unison and Braddock series classified as fine, mixed, semiactive, mesic Typic Hapludults. Slopes at the study site ranged from 7 to 25%. Climate data were collected from the College Farm weather station and monthly historical, and actual temperature and precipitation data are presented in Table 1. Experimental Design The experiment was set up as a randomized complete block design with four replications. Four 1.1-ha permanent cool-season grass pastures (blocks) were split in half, with each randomly assigned a sowing treatment (broadcasted or no-till drilled). Ten years prior, the pastures were sown to endophyte-infected tall fescue (Schedonorus phoenix Scop.) and then used in conjunction with various research experiments. No clover was oversown into the pastures during this time. The species composition of the pastures at the beginning of this study was dominated by tall fescue and Kentucky bluegrass (Poa pratensis L.), with a smaller proportion of orchardgrass (Dactylis glomerata L.). Clover made up less than 1% of the pasture composition. Pastures were fertilized with 58 kg ha 1 of P 2 O 5 and 50 kg ha 1 of K 2 O in March 2009 based on soil test results. In January 2009, residual grass was grazed by a herd of 57 cows that were moved through the pastures during a 3-d period. Due to the location of winter water sources and topographic variation, grazing pressure was not uniform across the four pastures. Pasture treatments were overseeded during the first week of February Seeding rate for both treatments was 4.4 kg ha 1 of red clover ( Liberty ), 2.2 kg ha 1 ladino white clover, and 2.2 kg ha 1 white clover ( Kopu II ). Drilled plots were planted with a Haybuster All Purpose Seed Drill (Jamestown, ND) set to 15.2-cm rows and a target depth of 1 cm. The broadcasted plots were sown with a Herd Sure-Feed Broadcaster (Model GT-77 ATV, Herd Seeder Co./ Kasco Mfg., Shelbyville, IN) mounted to a tractor. Grazing Management Pastures were split into six equal-size paddocks using singlestrand electric poly-wire and back-fenced so that animals were allowed access to only one paddock at a time. Bred beef cows (avg. wt. = 640 kg) were used for the grazing trial. During the first three rotations (42 d) in 2009, pastures were stocked with six cows (11 animal units ha 1 ) to help suppress grass competition. After this initial grazing period, three cows grazed each 1.1-ha pasture for the remainder of the season and in subsequent years. Grazing lasted from 21 April to 28 Oct., 20 April to 4 Oct., and 28 April to 2 Oct. in 2009, 2010, and 2011, respectively. Starting in April, cows were moved between paddocks every 2 d, with the residency period progressively lengthened to 5 d per paddock by autumn. Animals were removed from pastures twice each season (10 15 d in June and d in July/August) to graze warm-season perennial grass pastures that were part of a different study. Water and trace minerals were offered free-choice at all times. Bloat blocks (Sweetlix, PM AG Products, Mankato, MN) were provided to the animals during the first three rotations in 2009, after which bloat preventives were not provided. During this experiment, no animals that were grazing the experimental pastures suffered from bloat. Measurements Herbage Mass and Seedling Density In January 2009, residual grass biomass present at the time of sowing was estimated by clipping 10, 0.25-m 2 quadrats to ground level in each pasture. Biomass was then dried, using a forced-air oven at 60 C for at least 48 h, and weighed. Clover seedling density was measured during the first week of April 2009, when clover seedlings began to emerge. Clover seedlings were counted in 10 haphazardly placed, m 2 quadrats per treatment plot. The number of seedlings in each quadrat were counted and recorded. Red and white clover seedlings were difficult to distinguish due their small size, so they were grouped together. Species composition and aboveground biomass was measured 11 times on 18 June, 12 Aug., and 10 Oct. 2009; 16 Apr., 14 June, 17 Aug., and 16 Oct. 2010; and 27 Apr., 15 June, 23 Aug., and 15 Oct Visual species composition was taken in 10 haphazardly placed, 0.25-m 2 quadrats per plot. Plant species present and groundcover percentage of the quadrat that species represented were recorded (data not shown). Quadrats were then clipped to a 7.5-cm stubble and hand-sorted to grasses, weeds, white clover, and red clover. The samples were then dried using a forced-air oven at 60 C for at least 48 h and weighed. Clover Populations and White Clover Morphology A destructive sod harvest was taken in late October 2009 and 2010 to measure the number of white clover and red clover plants in respective treatments. In addition to population counts, white clover stolon length, number of stolons per white clover plant, and average weight per stolon were also recorded. Sections of sods 1218 Agronomy Journal Volume 104, Issue

3 within 10 haphazardly placed, m 2 quadrats per plot were cut to a 2.5-cm depth using a garden spade. The samples were then washed, and red and white clover plants were separated from the sod sections. The number of red and white clover plants were counted and recorded. The number of stolons per white clover plant was counted, and stolon length was measured to the nearest millimeter. Stolons were removed from white clover plants and dried using a forced-air oven at 60 C for at least 48 h and weighed. Statistical Analysis Herbage mass for respective species groups, seedling density, clover population, white clover morphology, and nutritive value indices were analyzed using ANOVA. Analysis consisted of a treatment factor of overseeding method (broadcasted or drilled) blocked by pasture (b = 4 blocks) and with no replication of treatments within a block. Before analysis, data were checked for normality and homogeneity of variance. Data were analyzed for each measurement date and in each year. ANOVA and regression analyses were performed using SAS statistical software. A P value of 0.10 was used to determine statistical differences between treatments. RESULTS Drilled and broadcasted treatments had a mean clover seedling density of 65.9 and m 2, respectively (P = ). Table 2 provides the biomass data from species composition measurements taken during the 2009, 2010, and 2011 growing seasons. In 2009, no difference between the drilled and broadcasted treatments were found for grass, white clover, red clover, and weed biomass for either the entire growing season or any measurement date. In 2010, no difference between the drilled and broadcasted treatments was found for white clover, red clover, and weed biomass, although grass biomass in April was greater for the drilled treatment than broadcasted (P = ). In April 2010, clover accounted for 33% (24% white clover and 9% red clover) of the total biomass, but declined to 9% (6% white clover and 3% red clover) in October. The June 2010 sampling date had three times more clover biomass than June 2009 and twice the total biomass. By August 2010, clover biomass was six times lower than it was in August 2009, likely due to hot, dry weather in July 2010 (Table 1). In 2010, clover made up 24% of the total biomass compared with 13% in The proportion of red clover did not change between 2009 and 2010, averaging approximately 13% of the total biomass. In 2011, no difference between the broadcasted and drilled treatments was found for grass biomass in April, June, August, or October (Table 2). The June 2011 red clover biomass was greater in broadcasted treatments compared with drilled treatments (P = ). White clover biomass was greater in drilled treatments (P = ) and weed biomass was greater in broadcasted treatments (P = ) for the October measurement (Table 2). The 2011 total biomass average (359.8 g m 2 ) was greater than that of 2010 (161.6 g m 2 ) and 2009 (209.9 g m 2 ). Clover Population and White Clover Morphology In 2009, no difference between broadcasted and drilled treatments were found for number of red clover plants m 2, white clover plants m 2, number of stolons per white clover plant, or white clover stolon weight (Table 3). Average stolon length was greater in the drilled treatment compared with Table 2. Mean aboveground biomass (g m 2 ) for grass, clover, and weed biomass measured during 2009, 2010, and Apr. June Aug. Oct. Avg. BST Drill P value BST Drill P value BST Drill P value BST Drill P value BST Drill P value g m 2 Year Grass WC RC Weed Grass WC RC Weed Grass WC RC Weed WC, white clover; RC, red clover; BST, broadcasted; Drill, no-till drilled. Signifi cance at the 0.10 probability level. Agronomy Journal Volume 104, Issue

4 Table 3. Mean clover plant density and selected morphological characteristics from destructive sod harvests taken in October 2009 and Characteristic Broadcasted Drilled P value Broadcasted Drilled P value No. red clover plants, m No. white clover plants, m Stolon no. per white clover plant Avg. stolon length, cm Stolon wt., g broadcasted treatments in 2009 (P = ). In 2010, no difference between broadcasted and drilled were found for red clover plants m 2, white clover plants m 2, number of stolons per white clover plant, average stolon length, or stolon weight. In 2010, the number of red and white clover plants had decreased by about 30%. White clover stolon length also declined from 2009 (3.9 cm) to 2010 (2.5 cm). Grass Biomass and Clover Establishment Figure 1 shows the relationship between grass biomass at time of overseeding and seedling density. Grass biomass remaining at sowing was variable across the four pastures. Seedling density in the drilled treatments was negatively related to the amount of grass biomass present at sowing (r 2 = 0.96, P = ). A negative quadratic relationship was found between clover seedling density and residual grass biomass for broadcasted treatment (r 2 = 0.99, P = ). Clover seedling density was highest at a residual grass biomass of 175 g m 2. The relationship between clover seedling density in April 2009 and clover biomass in August 2009 is shown in Fig. 2. Clover biomass data was obtained from species composition harvest taken in August 2009, and the seedling density data was obtained from seedling density measurements taken in April The data show a positive relationship between clover biomass and seedling density (r 2 = 0.84, P = ). Although the data is not presented, a positive relationship between clover seedling density and clover biomass (r 2 = 0.53, P = ) also was found in DISCUSSION Clover Seedling Emergence The data indicate that more clover seedlings emerged in broadcasted plots than drilled plots. The treatment differences were marginally insignificant (P = ), however, due to high variation among plots. Results from other studies are inconsistent when comparing clover seedling emergence and sowing method. In contrast to our study, Taylor et al. (1972) found that no-till-drilled red clover produced more seedlings than broadcasted seeding. They determined that drilling provided protection from both moisture and temperature fluctuations on the soil surface. Mueller and Chamblee (1984) compared broadcasted and drilled ladino clover in mid-february and mid-march. They found no difference in clover seedling density in 3 of 4 yr between broadcasted and drilled at the mid-february sowing date. In contrast, Watkins and Vickery (1965) found more white clover plants 5 wk after planting for broadcasted treatments than in drilled treatments. Some studies have evaluated relationships between seeding depth and clover seedling emergence. Charles et al. (1991) found that 90% of white clover seed emerged when broadcasted, but only 43% emerged when drilled to 1.5 cm. Depth of seeding may help explain why fewer seedlings emerged in most plots with no-till drilling (Fig. 1). Soils were not thoroughly frozen when they were planted in 2009, and it is possible that the no-till drilling could have placed seeds deeper than 1.5 cm, thus reducing, or slowing, initial clover emergence (Campbell 1985b). Overall, our data indicate that broadcasting produced numerically more clover seedlings, but density was variable across plots, suggesting that other environmental variables beyond seeding method influence emergence. Fig. 1. Clover seedling density and its relationship to standing grass biomass at time of overseeding. Data points represent mean values from each pasture treatment plot. (Linear relationship: y = x, r 2 = 0.96, P = ; Quadratic relationship: y = x x 2, r 2 = 0.99, P = ). Fig. 2. Seedling density in April 2009 and its relationship to clover biomass in August Data points represent mean values from each pasture treatment plot (y = x, r 2 = 0.84, P = ) Agronomy Journal Volume 104, Issue

5 Clover Biomass Although more clover seedlings emerged in most broadcasted plots, no differences were found for subsequent clover biomass in any year. Other studies have shown that initial seedling density may not be a good predictor of subsequent clover biomass: Mueller and Chamblee (1984) found more ladino white clover seedlings in drilled plots than in broadcasted, but season-end clover biomass did not differ between planting methods; Byers and Templeton (1988) found that broadcasting produced more alfalfa biomass in establishment year, but discovered no differences the next year between planting methods. Regarding our broadcasted treatment, it was noteworthy that mean clover seedling density in April (118 m 2 ) was almost identical to the clover plant density counted at the end of the season (119 m 2 ) (Table 3). In drilled treatments, initial clover seedling density was lower (66 m 2 ) compared with end-of-season plant density (84 m 2 ). This result suggests that more clover seedlings may have established in drilled treatments after the April measurement date. Additional plants that established in the drilled treatment may have contributed to the equalization of clover biomass noted later in the growing season (Table 2). Both sowing methods produced good clover establishment in 2009 and 2010 (>25% composition), though clover biomass declined in During the 3 yr, clover biomass peaked in the second year and declined in the third. Red clover went from 5% of the total biomass in 2009 to 3% in 2010 and then 2% in Lomas et al. (1999) also found that when red clover was drilled into tall fescue pastures, it started at 5% of pasture composition and, by year three, was 2% of composition. Finding significant white clover biomass in 2011, 3 yr after planting, was somewhat surprising. White clover stands commonly last 2 yr and decline in the third (Hoveland et al., 1991; Olsen et al., 1981). By year three in our study, white clover biomass was still approximately 15% of the total biomass. Persistence of white clover may have been aided by favorable weather patterns and rotational grazing. In addition to abundant rainfall during the establishment year, white clover may have benefitted from dry weather in In contrast to the wetter summers of 2009 and 2011, grazing tended to open up sward canopies more and create bare patches during the hot, dry 2010 summer. White clover is known for its ability to colonize open spaces in the sward (Frame 2005), an above-average rainfall in autumn 2010 may have facilitated this process. Another factor that likely helped white clover persistence was controlling grass competition by rotational grazing. Brink and Pederson (1993) found that rotational grazing can benefit white clover production, especially during dry weather, while Wen and Jiang (2005) noted that frequent cuttings (once a month) of ryegrass (Lolium perenne L.) white clover swards decreased the competition between these species, favoring the legume. Oates et al. (2011) also found that rotational-grazed pastures produced greater white clover biomass than pastures used for hay. Grazing pressure in our study may have helped to reduce grass competition to the point that it was never able to eliminate white clover from the sward. Rest intervals during the grazing rotation that ranged from 12 to 30 d likely helped clover plants recover from defoliation. Additionally, the high seeding rates used in our study (4.4 kg ha 1 ) and the inclusion of Kopu II white clover, which may persist longer than ladino (Albrecht and Woodfield 1999), also may have contributed to persistence of white clover over time. Clover Populations and White Clover Morphology No differences between broadcasting and drilling for white and red clover populations were found in either 2009 or White clover had a greater number of plants in both treatments compared with red clover. This could be due to the fact that approximately 30% more white clover seeds were sown than red clover. In 2009, white clover stolons were longer in drilled treatments compared with broadcasted. Marshall and James (1988) found that white clover planted at low densities (9 and 25 white clover plants m 2 ) had longer stolons than white clover planted at greater densities (50 and 100 white clover plants m 2 ). In our study, white clover plant density was similar between treatments, so this may not explain the differences in stolon length. One theory is that interspaces between drilled rows allowed establishing clover plants more space to expand stolon growth without running into other clover plants. Dry weather in 2010 may have reduced stolon length (Karsten and Fick 1999) and eliminated any treatment differences observed in Sanderson et al. (2003) found that summer drought can reduce white clover stolon density by as much as 50%. Grass Biomass and Clover Seedling Density We found, as have other studies, negative relationships between the amount of grass biomass present at sowing and clover seedling density. Springer (1997) found a linear decrease in white clover ground cover with an increase in bermudagrass [Cynodon dactylon (L.) Pers.] height at sowing. Guretzky et al. (2004) found that clover had better emergence at lower grass sward heights compared with greater sward heights for no-till drilled clover. In contrast to no-till drilled plots, we found a negative quadratic relationship between broadcasted clover seedling density and grass biomass at the time of sowing. It appears that clover seedlings from broadcasted sowing exhibited more of a threshold response to existing grass biomass. Clover seedling density declined abruptly if seed was broadcasted into more than 200 g m 2 of standing grass biomass, while the drop-off was less steep and linear in drilled treatments (Fig. 1). Regardless of the mechanism, the data clearly point to the importance of removing or suppressing residual grass biomass before overseeding clover. Cuomo et al. (2001) evaluated the effect of sod suppression with glyphosate, planting method, and legume species on the establishment of legumes into permanent pastures. They found that sowing method had no effect on red clover, alfalfa, kura clover (Trifolium ambiguum Bieb.), and birdsfoot trefoil (Lotus corniculatus L.) establishment into grass pastures. As in our study, grass suppression, regardless of sowing method, was the main factor that affected legume establishment. We found that presence of many clover seedlings after planting was a good predictor of clover biomass later in the growing season (Fig. 2). This result agrees with Dear et al. (2007), who found that a large number of alfalfa seedlings led to a greater population of alfalfa plants at the end of the growing season. Campbell and Kunelius (1984) studied seeding rates and initial grazing management on the establishment of red clover. They found that high seeding rates, when averaged across grazing treatments, had greater red clover biomass accumulation. Gibson and Cope (1985) stated that it was important to have a large initial seedling density of white clover, as that will produce more biomass in the second and third year than stands with a small initial seedling density. Our results suggest that successful clover establishment requires the emergence of many seedlings, and that this emergence occurs Agronomy Journal Volume 104, Issue

6 during a short time frame in spring, several months after sowing. Pastures in which fewer clovers emerged supported less clover biomass. It is noteworthy that this relationship did not change through time, despite the potential for rotational grazing and high rainfall to stimulate clover biomass. CONCLUSIONS Overall, we could find no difference in clover establishment or persistence over 3 yr in the broadcasted and no-till drilled treatments. Both sowing methods produced a large amount of clover biomass (>25% of total) for 2 yr, although less in year three. Removing residual grass biomass before sowing was critical to encourage clover seedling emergence. Subsequent clover biomass also was clearly related to the number of clover seedlings that initially established, even 2 yr after planting. Pastures with large seedling establishment consistently produced the most clover biomass. Pastures with less clover seedling establishment remained devoid of clover, despite favorable weather and rotational grazing practices that should have promoted clover growth and recruitment into the sward. Our results suggest that producers can actively manage many variables to create a favorable environment for clover establishment. Of course, climate also plays an important role in clover establishment, and it is important to note that our study only evaluated one planting period at one location. Our understanding of what makes successful clover establishment possible would benefit from more factorial experiments that address how climate variables interact with pasture management. REFERENCES Albrecht, K., and D. Woodfield White clover for Wisconsin pastures. www. uwex.edu/ces/crops/uwforage/whiteclover2.pdf Brink, G.E., and G.A. Pederson White clover response to grazing method. Agron. J. 85: doi: /agronj x Byers, R.A., and W.C. Templeton Effects of sowing date, placement of seed, vegatation suppression, slugs, and insects upon establishment of no-till alfalfa in orchardgrass sod. Grass Forage Sci. 43: doi: /j tb02153.x Campbell, B.D., and H.T. Kunelius Performance of overdrilled red-clover with different sowing rates and intial grazing managements. New Zeal. J. Exp. Agr. 12: Campbell, B.D. 1985a. Winged coulter depth effects on overdrilled red-clover seedling emergence. New Zeal. J. Agr. Res. 28:7 17. doi: / Campbell, B.D. 1985b. Planting depths effects on overdrilled seedling survival in summer. New Zeal. J. Exp. Agr. 13: Casler, M.D., D.C. West, and D.J. Undersander Establishment of temperate pasture species into alfalfa by frost-seeding. Agron. J. 91: doi: /agronj x Charles, G.W., G.J. Blair, and A.C. Andrews The effect of soil-temperature, sowing depth and soil bulk-density on the seedling emergence of tall fescue (Festuca-arundinacea) and white clover (Trifolium repens). Aust. J. Agr. Res. 42: doi: /ar Cuomo, G.J., D.G. Johnson, and W.A. Head Interseeding kura clover and birdsfoot trefoil into existing cool-season grass pastures. Agron. J. 93: doi: /agronj x Dear, B.S., J.M. Virgona, G.A. Sandral, A.D. Swan, and B.A. Orchard Lucerne, phalaris, and wallaby grass in short-term pasture phases in two eastern Australian wheatbelt environments. 1. Importance of initial perennial density on their persistence and recruitment, and on the presence of weeds. Aust. J. Agr. Res. 58: Frame, J Forage legumes for temperate grasslands. Science Publishers, Inc., Enfield, NH. Gibson, P.B., and W.A. Cope White clover. In: N.L. Taylor, editor, Clover science and techonology, ASA, CSSA, and SSSA, Madison, WI. p Guretzky, J.A., K.J. Moore, A.D. Knapp, and E.C. Brummer Emergence and survival of legumes seeded into pastures varying in landscape position. Crop Sci. 44: doi: /cropsci Hoveland, C.S., D.R. Hardin, P.C. Worley, and E.E. Worley Steer performance on perennial vs. winter annual pastures in North Georgia. J. Prod. Agric. 4: Kankanen, H.J., H.J. Mikkola, and C.I. Eriksson Effect of sowing technique on growth of undersown crop and yield of spring barley. J. Agron. Crop Sci. 187: doi: /j x x Karsten, H.D., and G.W. Fick White clover growth patterns during the grazing season in a rotationally grazed dairy pasture in New York. Grass Forage Sci. 54: doi: /j x Lomas, L.W., J.L. Moyer, and G.L. Kilgore Effect of interseeding legumes into endophyte-infected tall fescue pastures on forage production and steer performance. J. Prod. Agric. 12: Marshall, A.H., and I.R. James Effect of plant-density on stolon growth and development of whtie clover (Trifolium repens) varieties and its influence on the components of seed yield. Grass Forage Sci. 43: doi: /j tb02157.x Mueller, J.P., and D.S. Chamblee Sod-seeding of ladino clover and alfalfa as influenced by seed placement, seeding date, and grass suppression. Agron. J. 76: doi: /agronj x Oates, L.G., D.J. Undersander, C. Gratton, M.M. Bell, and R.D. Jackson Management-intensive rotational grazing enhances forage production and quality of subhumid cool-season pastures. Crop Sci. 51: doi: /cropsci Olsen, F.J., J.H. Jones, and J.J. Patterson Sod-seeding forage legumes in a tall fescue sward. Agron. J. 73: doi: /agronj x Pearson, C.J., and R.L. Ison. (1997). Agronomy of grassland systems. Cambridge Univ. Press, Cambridge, England. Sanderson, M.A., R.A. Byers, R.H. Skinner, and G.F. Elwinger Growth and complexity of white clover stolons in response to biotic and abiotic stress. Crop Sci. 43: doi: /cropsci Springer, T.L Effect of bermudagrass height on clover establishment. Crop Sci. 37: doi: /cropsci x x Taylor, T.H., J.S. Foote, J.H. Snyder, and E.M. Smith, Jr Legume seedling stands resulting from winter and spring sowings in Kentucky bluegrass (Poa pratensis L.) sod. Agron. J. 64: doi: /agronj x Taylor, T.H., E.M. Smith, and W.C. Templeton Use of minimum tillage and herbicide for establishing legumes in Kentucky bluegrass swards. Agron. J. 61: doi: /agronj x Watkin, B.R., and P.J. Vickery Pasture establishment on a granite soil on the northern tablelands of New South Wales. Aust. J. Exp. Agr. Anim. Husb. 5: doi: /ea Wen, Y., and H.F.U. Jiang Cutting effects on growth characteristics, yield composition and population relationships of perennial ryegrass and white clover in mixed pasture. New Zeal. J. Agric. Res. 48: doi: / Agronomy Journal Volume 104, Issue