Growth of White Clover Clones in Monoculture and Contrasting Bermudagrass Swards

Hybrid bermudagrass (Cynodon dactylon×C. transvaalensis) is widely used as turf in transition zone of China. Spring dead spot (SDS) is one of the most damaging diseases of hybrid bermudagrass. Symptoms of SDS appear when hybrid bermudagrass starts to break dormancy with warm temperature in early spring. The symptoms show sunken, circular or irregularly shaped, straw-colored patches, with 20 to 100 cm in diameter. The patches maintain dormant as the surrounding uninfected turfgrass resumes growth and turns green. SDS pathogens are soilborne fungi that colonize roots, stolons and rhizomes, infected roots or rhizomes become black and eventually collapse. Three species of fungi are reported to cause SDS: Ophiosphaerella herpotricha (Fr) J. Walker; O. korrae (J. Walker & A.M. Smith) Shoemaker & C.E. Babcock; or O. narmari (J. Walker & A.M. Smith) Wetzel, Hubert & Tisserat (Walker and Smith 1972; Walker 1980; Shoemaker and Babcock 1989; Wetzel et al. 1999). However, distribution of the three species may vary by geographical region (Cottrill et al. 2016). In October 2020, symptoms of SDS were observed on hybrid bermudagrass fairways of Taihu golf course in Wuxi, Jiangsu province. Root samples of SDS were collected, symptomatic roots with 3-4 cm length were cut, washed 2-3 times, surface sterilized in 0.6% NaOCl for 5 min, rinsed and blotted dry for 2 min and placed on potato dextrose agar (PDA) amended with 50 mg L-1 each of ampicillin, streptomycin sulfate and tetracycline. Plates were incubated in the dark at 25℃ for 5-7 days, Hyphae growing from the roots were transferred to new PDA plates. A total of 7 fungal isolates with morphology similar to SDS pathogens were obtained (Tredway et al. 2009). The genomic DNA was extracted from 2 of them (7-41, 8-6) and amplified with universal primers ITS5 and ITS4 (White et al. 1990). PCR products were sequenced (deposited as MW536995 and MW536994 in GenBank, not available yet) and showed 99.79% similarity to O. narmari (KP690979). Pathogenicity tests were performed on 'Tifdwarf' hybrid bermudagrass (9-week-old in 5 × 20 cm Cone-Tainers containing a sand and nutrition substrate mixture). Eight oat seeds infested with O. narmari were inserted 5 cm below the soil surface in the root zone of hybrid bermudagrass. The inoculated turfgrass grew for five weeks in the growth chamber with a 12-h day/night cycle of 25/20°C and 90% relative humidity. A control treatment was inoculated with 8 noninfested sterile oat seeds, and each treatment was replicated 3 times. The root tissues of hybrid burmudagrass inoculated with O. narmari became black and necrotic, no symptoms were observed on the roots of noninfested plants. O. narmari was consistently reisolated from symptomatic roots, and confirmed by PCR as mentioned above. To the best of our knowledge, this is the first report of O. narmari caused spring dead spot in the transition zone of China. The identification of SDS caused by O. narmari will have important implications for the management of this root-rot species on hybrid bermudagrass.

Broiler litter should be applied to coincide with crop growth to maximize nutrient uptake. We determined the influence of rate and date of litter application on bermudagrass [Cynodon dactylon (L.) Pers.] yield and nutrient uptake. Litter was applied to ‘Alicia’ hybrid bermudagrass on a Savannah fine sandy loam (fine‐loamy, siliceous, semiactive, thermic Typic Fragiudult) at 9 and 18 Mg ha−1 as a single application in early April, May, and June or as a split application in early April and June, May and July, and June and August. Increasing application rate increased yield and P uptake when litter was applied in April [12.8–15.6 Mg ha−1 dry matter (DM); 43.3–49.9 kg P ha−1] or June (14.3–16.9 Mg ha−1 DM; 45.4–51.3 kg P ha−1) but not in May (mean of 15.0 Mg ha−1 DM and 48.4 kg P ha−1). Delaying application of 9 Mg ha−1 until May or 18 Mg ha−1 until June increased annual yield and N, K, and Cu uptake compared with applications in April. Split application improved yield and nutrient uptake only when litter was applied in April and June compared with April application. Phosphorus uptake was unaffected by application date, possibly due to high soil P levels (351 mg kg−1) and to the small quantity of P required by the grass relative to that applied. The results suggest that litter should be applied to bermudagrass only after minimum temperatures exceed those necessary for optimum growth.

Broiler litter should be applied to coincide with crop growth to maximize nutrient uptake. We determined the influence of rate and date of litter application on bermudagrass [Cynodon dactylon (L.) Pers.] yield and nutrient uptake. Litter was applied to ‘Alicia’ hybrid bermudagrass on a Savannah fine sandy loam (fine-loamy, siliceous, semiactive, thermic Typic Fragiudult) at 9 and 18 Mg ha−1 as a single application in early April, May, and June or as a split application in early April and June, May and July, and June and August. Increasing application rate increased yield and P uptake when litter was applied in April [12.8–15.6 Mg ha−1 dry matter (DM); 43.3–49.9 kg P ha−1] or June (14.3–16.9 Mg ha−1 DM; 45.4–51.3 kg P ha−1) but not in May (mean of 15.0 Mg ha−1 DM and 48.4 kg P ha−1). Delaying application of 9 Mg ha−1 until May or 18 Mg ha−1 until June increased annual yield and N, K, and Cu uptake compared with applications in April. Split application improved yield and nutrient uptake only when litter was applied in April and June compared with April application. Phosphorus uptake was unaffected by application date, possibly due to high soil P levels (351 mg kg−1) and to the small quantity of P required by the grass relative to that applied. The results suggest that litter should be applied to bermudagrass only after minimum temperatures exceed those necessary for optimum growth.

Transpiration responses of warm-season turfgrass in relation to progressive soil drying

ABSTRACTPlants may survive drought stress by altering photosynthetic metabolism. This study was designed to investigate stomatal and metabolic changes associated with photosynthetic responses to drought for bermudagrass. A hybrid bermudagrass [Cynodon dactylon (L.) Pers. × C. transvaalensis Burtt Davy, cv. Tifway] and a common bermudagrass (C. dactylon, cv. C299) were subjected to drought by withholding irrigation in growth chambers. Tifway maintained higher relative water content and lower electrolyte leakage than C299 under drought. Net photosynthetic rate (Pn) and stomatal conductance (gs) decreased, with more rapid decline in C299, but Pn was significantly greater in Tifway than in C299, while gs did not differ between the two genotypes during 14 d of drought. Photochemical efficiency, ribulose‐1,5‐bisphosphate carboxylase:oxygenase (Rubisco) activity, and activation state declined during drought stress, but were higher in Tifway than in C299 within 14 d of treatment. Electrophoretic analysis revealed that at 14 and 18 d of drought, pyruvate orthophosphate dikinase (98 kDa) and Rubisco protein (49 kDa) exhibited severe degradation in C299 but to a lesser extent in Tifway. These results indicated that the superior drought resistance in Tifway could be attributed to higher photosynthetic capacity through the maintenance of higher metabolic activities, as manifested by more active Rubisco enzyme and Rubisco activase and more stable photosynthetic proteins.

The need for tall fescue (Festuca arundinacea Schreb.) pasture renovation due to a fungal endophyte (Acremonium coenophialum Morgan‐Jones and Gams) that causes fescue toxicosis in grazing cattle (Bos taurus L.) has given farmers the opportunity to add white clover (Trifolium repens L.) to their improved pastures. This study was conducted to determine the compatibility of five white clover and five tall fescue cultivars grown in association during two seeding years and to determine any interactions between cultivars. ‘Regal’, ‘Tillman’, ‘Louisiana S‐1’, ‘Osceola’, and MS Brown Loam Syn. #6 white clover were grown in all combinations with ‘Kentucky 31’, ‘Kenhy’, ‘Johnstone’, ‘Triumph’, and ‘Missouri 96’ tall fescue in two separate seeding year studies. The studies were grown on a Catalpa silty clay (fine, montmorillonitic, thermic Fluvaquentic Hapludoll) at Mississippi State, MS. Louisiana S‐1 associations produced significantly less total dry matter (DM) and white clover (WC) yield and more tall fescue (TF) yield than those of the other white clover cultivars across tall fescue cultivars. Triumph associations produced significantly less WC yield and more TF yield than those of the other tall fescue cultivars across white clover cultivars. No differences in total DM yield were observed for the tall fescue cultivars. No interactions between tall fescue and white clover cultivars were found in either seeding year. The cultivar relationships were constant over years an over harvests within years, though the botanical composition of the plots varied between harvests and years. Triumph was more competitive with white clover than the other tall fescue cultivars, and Louisiana S‐1 was less competitive with tall fescue than the other white clover cultivars during the seeding year. No advantage was found for any specific combination of white clover and tall fescue cultivars during the seeding year.

To determine yield potential, persistence, and nutrient removal, replicated blocks of two cultivars of seeded bermudagrass (‘Cheyenne’ and ‘Mohawk’) and two seeded bermudagrass blends (‘Laredo’ and ‘Ranchero Frio’) were planted in a commercial swine farm effluent spray field. Measurements were taken during a 4‐year period, including the establishment year. Total dry matter (DM) yields of 10,890 and 10,750 lb/acre for Cheyenne and Ranchero Frio were greater than for Laredo and Mohawk (10,090 and 10,080 lb/acre, respectively; P = 0.01) and decreased after a peak in the second growing season (P < 0.01). Weed contribution to DM was large following establishment, negligible the second year, and then grew progressively greater thereafter (18.7, 0.3, 12.8, and 21.3% for the first through fourth seasons, respectively; P < 0.01). Weed DM composition was greater for Cheyenne than for Mohawk or the blends (17.2, 11.7, 12.2, and 12.0% for Cheyenne, Laredo, Mohawk, and Ranchero Frio, respectively; P = 0.02). Nitrogen (N) and phosphorus (P) yields were proportional to DM yield; however, nitrate concentration was greater in forage from Cheyenne and Ranchero Frio than from Laredo and Mohawk (0.58, 0.60, 0.43, and 0.47% of DM, respectively; P < 0.01). Lower DM yields after Season 2 resulted in average DM yields that were close to the realistic yield estimates listed for common bermudagrass when applying swine effluent as source of N fertilizer, and were much lower than the realistic DM yield estimates for hybrid bermudagrass. Ranchero Frio and Cheyenne were well suited to this site because they produced the greatest average DM yield across years, but Ranchero Frio had less weed composition in the final year.

Methods are needed for utilizing nutrients contained within animal wastewater lagoons. One potential method for capturing nutrients in a useful form is to grow vegetation on the lagoon. A study was conducted from 2005 to 2008 to determine the feasibility of growing vegetation on floating platforms on a single-stage swine wastewater lagoon. Five species were selected from earlier studies as having potential for growth on a commercial swine farm wastewater lagoon: common bermuda grass (Cynodon dactylon (L.) Pers.), Tifton 85 bermuda grass (Cynodon dactylon (L.) Pers.), St. Augustine grass (Stenotaphrum secundatum (Walter Kuntze)), fall panicum (Panicum dichotomiflorum (L.) Michx.), and giant reed (Arundo donax L.). The plants were periodically harvested as needed, and the biomass was weighed and analyzed for N, P, K, Ca, Mg, S, Al, B, Cd, Cr, Cu, Fe, Mn, Mo, Na, Ni, Pb, and Zn. Giant reed and St. Augustine grass were found to be unsuitable for long-term growth on the wastewater lagoon. The greatest biomass production (sum of six cuttings) was 3.6 kg m-2 dry matter from Tifton 85 bermuda grass, followed by common bermuda grass (3.2 kg m-2 dry matter) and fall panicum (3.1 kg m-2 dry matter). All of the plant species accumulated greater than 1000 ppm Na. Nutrient (N, P, and K) uptake and removal from the wastewater with biomass harvesting was primarily a function of biomass produced. The greatest annual uptake and removal of N and P from the wastewater was by Tifton 85 bermuda grass in 2006, where three cuttings of the floating vegetation removed totals of 69 and 25 g m-2 N and P, respectively. Annual uptake and removal of K was greatest by fall panicum, where uptake and removal by three cuttings in 2007 totaled 78 g m-2. In 2008, weeds that had populated the mats were harvested for biomass and elemental uptake. Uptake and total removal of nutrients in two cuttings of the weeds in 2008 was lower than what was observed with the planted species. Total uptake and removal of N, P, and K by the weeds in 2008 was approximately 30, 10, and 30 g m-2. The study showed that plant species exist that can grow and thrive on single-stage anaerobic wastewater lagoons on floating platforms for at least two years while taking up N, P, and K from the wastewater. Harvesting of biomass (which could potentially be used as a soil amendment or as cellulosic feedstock for bioenergy) from floating mats hence could be a mechanism for animal producers to both remove and productively use nutrients contained in the wastewater.

In the southeast United States, the most popular vegetatively propagated hybrid bermudagrass (Cynodon dactylon × C. transvaalensis Pers. L.) has long been the cultivar Tifway. In recent years, however, southern turf managers choosing cultivars for lawns, golf course fairways, or athletic fields have a newer option: ‘TifSport’ hybrid bermudagrass, also a vegetative selection. Although the use of TifSport is increasing, research that studies the establishment of TifSport is largely absent from the literature, as most past work has focused exclusively on Tifway. Thus, the objective of our research was to evaluate the establishment of Tifway and TifSport hybrid bermudagrasses when established from sprigs, examining the effect of N rate and N source of the rate of groundcover, shoot density, and stolon and rhizome development. The study was conducted in 2002 and 2004; in each year, new plots were fumigated and then sprigged with Tifway and TifSport. Nitrogen source treatments were ammonium nitrate (AN) (34‐0‐0) or calcium (Ca) nitrate (CaN) (16‐0‐0; 39% Ca), both applied at 2.4, 4.9, 7.3, or 9.8 g N m−2 wk−1. Treatments were a factorial combination of cultivar, N rate, and N source. Collected data included percentage groundcover as measured via line‐transect, shoot density, and dry weight of stolons and rhizomes. In both years, TifSport had higher percentage groundcover than Tifway at early measurement dates, with >95% groundcover for both cultivars at 7 and 5 weeks after sprigging for 2002 and 2004, respectively. Shoot density often increased as N rate increased, with shoot density maximized at N rates between 6.2 and 7.6 g N m−2 wk−1. In both years, dry weight of stolons and rhizomes decreased as N rate increased.

Switchgrass (Panicum virgatum L.) is a native grass species to much of the US. It has shown great potential for use in production of fuel ethanol from cellulosic biomass (Lynd et al., 1991). Work in Alabama demonstrated very high dry matter yields can be achieved with switchgrass (Maposse et al. 1995) in the southeastern US. Therefore, this region is thought to be an excellent choice for development of a switchgrass cropping system where farmers can produce the grass for either biomass or forage. Another report has shown success with selection and breeding to develop high yielding germplasm from adapted cultivars and ecotypes of switchgrass (Moser and Vogel 1995). In the mid 1990s, however, there was little plant breeding effort for switchgrass with a potential for developing a cultivar for the southeast region. The main goal of the project was to develop adaptive, high-yielding switchgrass cultivars for use in cropping systems for bioenergy production in the southeastern US. A secondary objective was to assess the potential of alternate herbaceous species such as bermudagrass (Cynodon dactylon L.), bahiagrass (Paspalum notatum Flugge.), and napiergrass (Pennisetum purpureum Schumach.) that may compete with switchgrass for herbaceous bioenergy production in the southeast. During the conduct of the project, another goal of developing molecular markers useful for genetic mapping was added. The ''lowland'' cultivars, Alamo and Kanlow, were found to be the highest yielding switchgrass cultivars. Although most summers during the project period were hot and dry, their annual dry matter yield continue to outperform the best ''upland'' cultivars such as Cave-in-Rock, Shawnee, NE Late, and Trailblazer. The use of a breeding procedure based on the ''honeycomb design'' and multi-location progeny testing, coupled with the solid heritability and genetic gain estimates for dry matter yield in lowland type switchgrass germplasm, indicated excellent potential to isolate parental genotypes for producing higher yielding synthetic cultivars. The four experimental synthetics produced thus far, and now in performance tests, could provide this cultivar. Initial performance results of these experimentals have been very promising demonstrating a 30% yield enhancement over Alamo and Kanlow. Future testing, including testing in other states, will be critical before a determination can be made to release one or more of these into the commercial seed trade. In the genetic mapping project, 42 genotypes of switchgrass were surveyed using restriction fragment length polymorphism (RFLP) probes from different grass species. The different genotypes included 24 from Alamo, 15 from Kanlow, and 3 from ''Summer.'' A majority of the probes (87%) hybridized to the switchgrass DNA and 81% were polymorphic. Most of the polymorphism observed was between the cultivars. A mapping population consisting of 100 progeny from a cross between the most dissimilar Kanlow and Summer genotypes was produced during 2001. The parents and progeny population are now maintained at the University of Georgia and will be used to construct a map based on the polymorphic RFLP probes. When compared to ''Tifton 85'' bermudagrass, ''Tifton 9'' bahiagrass, and ''Merkron'' napier-grass, Alamo switchgrass was found to show poorer yields than Merkron and Tifton 85, but better yields than Tifton 9 in the coastal plain region. The exceptional performance of Tifton 85 bermudagrass is extremely noteworthy because this hybrid bermudagrass is also a variety of choice for many commercial hay producers in the lower south and would give any producers a very good option to produce either biomass for a biofuels initiative or sell as hay on the open market. Merkron has consistently showed the highest dry matter yields. However, there continues to be some winter damage each year on this species at the Athens location indicating its real potential lies mainly in the Gulf Coast region of the southeastern United States. The excellent characteristic of Tifton 85 and Merkron should therefore be enough to initiate more basic research for their chemical ability as a biofuels crop.

Nine white clover populations (WCP) naturalized in Chile were evaluated under controlled (greenhouse pot experiments) and field conditions for water and P stress. The objective of this work was to determine dry matter production and water use and P efficiencies to identify which naturalized white clover populations would be most useful for breeding programs. In the field experiments the clover was grown mixed with perennial ryegrass, but in pots the clover was grown alone. The treatments were the factorial combinations of WCP by water and P levels; completely randomized and complete randomized block experimental designs were used for pot and field experiments, respectively. The population WCP 9-1-X was the most efficient for using water under controlled conditions and was among the three with the highest density of stolons under non-irrigated field conditions. The naturalized WCP 8-1-X produced more clover DM at the low P level than at the high P level and reached the same DM yield as cultivar Huia, which surpassed all the naturalized populations. These genetic materials have high potential to be included in white clover breeding programs whose objective is to generate cultivars for water-restricted and/or P-deficient marginal soils.

Forage species common to the southern USA Piedmont region, Lolium arundinacea, Paspalum dilatatum, Cynodon dactylon and Trifolium repens, were established in a model pasture system to test the future climate change scenario of increasing ozone exposure in combination with varying rainfall amounts on community structure and nutritive quality. Forages were exposed to two levels of ozone [ambient (non-filtered; NF) and twice ambient (2×) concentrations] with three levels of precipitation (average or ±20% of average) in modified open-top chambers (OTCs) from June to September 2009. Dry matter (DM) yield did not differ over the growing season between forage types, except in primary growth grasses where DM yield was higher in 2× than NF treatment. Primary growth clover decreased in nutritive quality in 2× ozone because of increased concentrations of neutral detergent fibre (NDF), acid detergent fibre (ADF) and acid detergent lignin (ADL). Re-growth clover exhibited the largest decrease in nutritive quality, whereas grasses were not adversely affected in 2× ozone. Re-growth grasses responded positively to 2× ozone exposure, as indicated in increased relative food value (RFV) and percentage crude protein (CP) than NF-exposed re-growth grasses. Effects of precipitation were not significant over the growing season for primary or re-growth forage, except in primary growth grasses where DM yield was higher in chambers with above average (+20%) precipitation. Total canopy cover was significantly higher over the growing season in chambers receiving above average precipitation, but no significant effects were observed with ozone. Results indicate shifts in plant community structure and functioning related to mammalian herbivore herbivory in future climate change scenarios.

Take-all root rot caused by Gaeumannomyces graminis (Sacc.) Arx & Olivier is a destructive disease of hybrid bermudagrass (Cynodon dactylon × C. transvaalensis Burtt-Davy) (Elliott 1991; Stephens et al. 2020). In April 2018, multiple circular to irregular patches (10 to 200 cm in diameter) were evident on 'Tifway419' bermudagrass fairway of Wohu golf course (Huangshan city, Anhui province). Patches were found at five of the 18 fairways and showed a decrease in stand density. Individual plants in patches exhibited off-color to white foliage, dark and necrotic lesions were frequently observed on roots and associated stolons. Ectotrophic runner hyphae and lobed hyphopodia are ubiquitous on the surface of necrotic roots. Two diseased samples were collected from each patch by a soil corer (10 cm in diameter), and a total of ten patches were sampled across the five symptomatic fairways. Symptomatic roots or stolons were cut into 2 to 3 cm length and surface disinfested for 3 min in 0.6% sodium hypochlorite solution. The sterilized tissues were rinsed, air dried on sterile filter paper for 5 min and placed on potato dextrose agar (PDA) amended with 50 mg L-1 each of ampicillin, streptomycin sulfate and tetracycline. Plates were incubated in the dark at 20°C for 5 days, and similar colonies were consistently recovered from 85% of the symptomatic tissues. After hyphae purification, a total of 20 fungal isolates were obtained with morphology similar to G. graminis (Hernández-Restrepo et al. 2016). DNA was extracted and amplified with primers ITS5/ITS4 (White et al. 1990). PCR products were sequenced (deposited as MZ889119 and MZ889120 in GenBank), and showed 99.43 to 99.81% homology when compared with a known G. graminis isolate (accession no. KX306505). To complete Koch's postulates, healthy Tifway419 hybrid bermudagrass was grown in 5 × 20 cm Cone-Tainers containing a sand and Pindstrup substrate (Pindstrup Mosebrug, Pindstrup, Denmark) mixture (1:1, V/V) nine weeks after transplanting from the experimental field of the Institute of Botany, Jiangsu Province and Chinese Academy of Sciences. Five sterile oat seeds infested with G. graminis were inserted 5 cm below the soil surface in the root zone of Tifway419. Cone-Tainers inoculated with five noninfested sterile oat seeds served as controls, and each treatment was replicate 3 times. All Cone-Tainers were transferred into a growth chamber with a 12-h day/night cycle at 25/20°C and 95% relative humidity. They were covered with plastic bags for the first week, and watered once daily to encourage disease development. Hybrid bermudagrass inoculated with G. graminis decreased in plant density and turfgrass quality, black and necrotic roots covered with lobed hyphopodia were observed eight weeks after inoculation, no symptoms were present on the roots and above-ground of noninfested control plants. G. graminis was consistently reisolated from symptomatic roots and stolons, and confirmed by sequencing the ITS region as mentioned above. To our knowledge, this is the first report of G. graminis causing take-all root rot of hybrid bermudagrass in China, which may influence the management of this important root disease.

Replacing cool-season turf with more drought and heat tolerant warm-season turfgrass species is a viable water conservation strategy in climates where water resources and precipitation are limited. Field studies were conducted in Riverside and Irvine, CA, to investigate three methods (scalping, eradication with a nonselective herbicide, planting into existing turf) of converting an existing tall fescue ( Festuca arundinacea ) sward to warm-season turf. Cultivars established vegetatively by plugging were ‘De Anza’ hybrid zoysiagrass [ Zoysia matrella × ( Z. japonica × Z. tenuifolia )], ‘Palmetto’ st. augustinegrass ( Stenotaphrum secundatum ), ‘Tifsport’ hybrid bermudagrass ( Cynodon dactylon × C. transvaalensis ), ‘Sea Spray’ seashore paspalum ( Paspalum vaginatum ), and ‘UC Verde’ buffalograss ( Buchloe dactyloides ). Cultivars established from seeds were ‘Princess-77’ bermudagrass ( C. dactylon ) and ‘Sea Spray’ seashore paspalum. Neither scalping nor planting into existing tall fescue were effective conversion strategies, as none of the warm-season turfgrasses reached 50% groundcover within 1 year of planting. All of the species except for st. augustinegrass reached a higher percentage of groundcover at the end of the study when glyphosate herbicide was applied to tall fescue before propagation compared with the other conversion strategies. Bermudagrass and seashore paspalum established from seeds and hybrid bermudagrass from plugs provided the best overall establishment with 97%, 93%, and 85% groundcover, respectively, when glyphosate was used before establishment. Quality of seeded cultivars matched or exceeded that of cultivars established vegetatively by plugging. These results suggest that eradication of tall fescue turf followed by establishment of warm-season turf from seeds is the best and easiest turf conversion strategy.

A 6-year trial was laid out with varied PK fertilization on fen soil and analysed for the capability of undersown clover species to fix atmospheric nitrogen. No mineral nitrogen was applied. On an average of the six test years, the share of white clover in the total crop yield of the sward was 17% after sowing once and that of red clover or Swedish clover was 19% after sowing twice. The annual mean shares of these two clover variants in the total yield varied only slightly over the test period. White clover showed typical seasonal dynamics in the test years, with yield portions increasing from spring to late summer. Undersown red and Swedish clovers reached the highest yield percentages in their 2nd and 3rd growths. The variants grown without PK fertilizer and clover yielded 45 dt DM ha − 1. With PK fertilizer dry matter yields increased to 57 and 68 dt ha − 1, respectively. With supplemental white clover or red/Swedish clover sown into the sward, substantial amounts of symbiotic nitrogen were accumulated in the harvestable plant matter, giving a total dry matter yield of 76 dt ha − 1. Clover yield shares up to 30% resulted in increasing DM yields which, however, declined again with even larger shares of clover. As to the ingredients of grasses, herbs and clover, the differences were mostly significant. On an average of the six test years, nitrogen fixation was 58.4 kg ha − 1 for white clover and 46.6 kg ha − 1 for red/Swedish clover.