.. ~J...=='. . ' ‘ - #:‘g. ‘ W 'W 2- ..:_J.. :‘5‘ “—— ..._ fi: .35.... .....::=;; :5}? i4— —_ _,._..,— 3.? :‘m‘ ~2- .....:_. L_—:>C-_- ,. Ll : '- '-'5::::::- #4 __.~_~ m...“ _‘_ ...- .—. “:_ W “'._ ____..5_,...:..- :3 a: :__.2..:,‘.:.5‘:.t 5"“...- L *3. g? _— --.:_ g...“ m: — :33“ .325": .3... z“ ”213—“ L _‘ b .‘ ~ * ____ ‘35-‘55” m h. .r- ’— 4a..” .~ “5.5.1: W ”—- w .. 0—‘55 “M “W: *4-.- H -- W?’ Ln”: -M- *— M 6:3:- — _A . a :2, .l- ~ 2' -' _:._-‘ «of: 'w " 3:: - _-_ - “’P \O. ‘ ‘5 t- g, _L -_ ,- 2;: . - U. - .4. ‘ -~— _ - -— : 3;- x. - 5.- ' w ‘ ' -‘ ' - :7i‘: _- - ._ -. _ _ s .3.» -4 . -5::‘_.:‘ __..-__ < ‘_‘_ - ”4;. '-'—— “55-55315... .-—’-—-. :.- - ’ ._._.,-_ ‘ - . W— ........ V“ ._ '“ztz': ._ - -- flw- wan—fig.— ‘v ‘ 5‘5 .25; . _. t - __ “A ...,.. ‘ 1;: - .L. m . . w - .....______.......... C“ w .— ~. _- . u -- - A- 4. {EL} 1’1 “1“} 5111 1,, § 1 11111.11 11 11 1TJM‘““L! “[3111 [‘3‘ W11 11 I“ W Lu H g") 21 11"“ 5‘3 “I “1‘51 ‘I‘M “Him 5“ 151 “111111 “R“ 1““ H1 [011 ‘21! ‘W‘n 1‘ mm -~; w..- -—..p -m- “2—- 7.4 2 I :I 3““? 31 1'3}: 2 {if 1"; ”.11.- 1 111 1'1 333 1 1 3 1111111111 111.111 111 3111111?“ B "2 ii \. THESIS " This is to certify that the dissertation entitled EVALUATION OF PROBLEMS IN THE IMPROVEMENT OF GRASS PASTURES BY SOD SEEDING presented by Clive Holland has been accepted towards fulfillment of the requirements for PhD degree in Crop & Soil Sciences W/réz / (“is Major professor 43m :24 @851 \ MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 )V1531_} RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from -;_. your record. FINES will be charged if book is returned after the date stamped below. ‘ z "'r .'»_ ‘ c. _ . '. 3 Q 5‘. {L i n‘ '- !' ‘3 . I. s. l H I C? #1}? E g 0' I“ 3 _'. R “A 955'" £15” r "J a? F ”'3 A It.» \.s4fi:'lf“rii§- 5 /4// ~53 6 75" EVALUATION OF PROBLEMS IN THE IMPROVEMENT OF GRASS PASTURES BY SOD SEEDING BY CLIVE WILLIAM HOLLAND A DISSERTATION Submitted to Mflchigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop and Soil Sciences 1983 ABSTRACT EVALUATION OF PROBLEMS IN THE IMPROVEMENT OF GRASS PASTURES BY SOD SEEDING BY CLIVE WILLIAM.HOLLAND Grass pastures without nitrogen fertilization are more pro- ductive when the award contains a nitrogen fixing legume. The intro- duction of forage legumes into grass swards is generally most success- ful when the sod is plowed and tilled. In areas that are erosive or too steep to till, improving pastures by sod seeding is a satisfactory alternative. Difficulties in establishing a good stand of legumes are often encountered when sod seeding. Four specific problems investigated were: First, lowpr soils and the effects of surface-applied lime when sod seeding alfalfa;, second, the killing of legume seedlings in the field by early spring freezing temperatures and cold hardiness of three legume seedlings under controlled conditions; third, grass suppression by cutting to simulate grazing compared to a herbicide in establishing two legumes in a sod and; fourth, effects of fertilizer on seedling survival when placed in contact with legume seeds under controlled condi— tions. Clive William Holland Equally as much alfalfa was produced when lime was broadcast on the surface at the rate of 11.2 t ha.1 or incorporated into the plow layer as recommended. No differences in stand density were obtained in three of four trials when lime at this high rate was surface applied or incorporated. Nitrogen fixation occurred only in areas of the lowth soil where surface-applied lime had penetrated. Trefoil stands were simdlar but alfalfa stands were consis- tently poorer when broadcast on 15 Mar. compared to 15 Apr. Freezing temperatures after March seedings killed many alfalfa seedlings. Greenhouse data indicated the LT for 2.3 hours of freezing as 50 follows: alfalfa, -4.44 C > red clover, -5.39 C > and trefoil, -6.67 C. Suppression of sown grasses by defoliation was as effective as a herbicide in legume establishment. Four cuttings after seeding sup- pressed grass competition more consistently than a herbicide. Quack- grass sods required a herbicide for adequate suppression. Phosphorus in contact with legume seeds reduced seedling sur- vival more severely than similar rates of K. Extremely low fertilizer- band pH caused low seedling survival with P. Potassium in contact with seeds did not reduce stands at rates less than 34.5 kg ha-l. ACKNOWLEDGMENTS As my major professor, Dr. Milo B. Tesar provided many challenges and unlimited stimuli to my professional growth and under- standing. Above all, he provided a friendship that is treasured, sincerely appreciated, and valued for the closer-than-normal professor- student relationship that has developed. His strong professional guidance and criticism have been equalled by support and encouragement in every aspect of research and academic advancement. It is with humbleness and sincere gratitude that I say thank you to "Dr. T" for the opportunity to have been his student. Drs. Boyd Ellis, Bernie Knezek, and Stephen Stephenson have been truly professional and empathetic committee members. Advice and guidance in their areas of speciality have been given freely and this has greatly enriched my study program. It is with strong apprecia- tion for their concern and understanding that I am encouraged to emulate the example they have displayed. A study program such as this is not borne by the student alone. Family members offer support and encouragement and tolerate uncomplainingly many inconveniences and disruptions to a 'normal' family life. Carmen and Andrea are such a family and have contrib- uted greatly to the successful completion of this study program. Carmen also provided invaluable assistance in the preparation and typing of this manuscript. ii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O 0 LIST OF FIGURES . . . . . . . . . . INTRODUCTION 0 O O O O O O O O O O 0 CHAPTER 1. 2. EFFECTIVENESS OF SURFACE-APPLIED LIME ON ACID SOILS WHEN SOD SEEDING ALFALFA (Mbdicago Abstract . . . . . . . . . Introduction . . . . . . . Materials and Methods . . . Results and Discussion . . Summary and Conclusions . . Literature Cited . . . . . ESTABLISHMENT OF FORAGE LEGUMES AS AND METHOD OF SEEDING AND FREEZING Abstract . . . . . . . . . Introduction . . . . . . . Materials and Methods . . . Section I-—Field Studies . . . . Section II-—Greenhouse studies . Results and Discussion . . Section I-Field Studies . . . . Section II-Greenhouse studies . Summary and Conclusions . . Literature Cited . . . . . sativa L.). . . . . INFLUENCED BY DATE TEMPERATURES . . . SIMULATED GRAZING COMPARED TO HERBICIDE SUPPRESSION OF GRASS COMPETITION WHEN SOD SEEDING Abstract . . . . . . . . . Introduction . . . . . . . Materials and Methods . . . Results and Discussion . . Summary and Conclusions . . Literature Cited . . . . . SURVIVAL OF THREE FORAGE LEGUMES SEEDED IN FORAGE LEGUMES . . CONTACT WITH PHOSPHORUS AND POTASSIUM . . . . . . . . . . . . iii Page 11 15 23 25 42 42 44 47 47 50 52 52 56 6O 62 78 78 80 84 87 93 94 106 Page Abstract . . . . . . . . . . . . . . . . . . . . 106 Introduction . . . . . . . . . . . . . . . . . . 108 Materials and Methods . . . . . . . . . . . . . . 111 Results and Discussion . . . . . . . . . . . . . 113 Summary and Conclusions . . . . . . . . . . . . . 118 Literature Cited . . . . . . . . . . . . . . . . 119 GENERAL BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . 128 iv LIST OF TABLES TABLE CHAPTER 1 1. Stand density and dry matter yield of alfalfa seeded into a tilled Kalamazoo sandy loam, pH 4.8, with various rates of lime surface applied after seeding or incorporated into the plow layer before seeding (Exp. 1) . . . . . . . . . . . . . . . . . . . . . . . Stand density and dry matter yield of alfalfa seeded into a tilled Miami sandy loam, pH 5.3, with various rates of lime surface applied after seeding or incorporated into the plow layer before seeding (Exp. 2) . . . . . . . . . . . . . . . . . . . . . . . Stand density and dry matter yield of alfalfa sod seeded into a Hillsdale sandy loam, pH 5.9, with various rates of lime surface applied after seeding (Exp. 3) . . . . . . . . . . . . . . . . . . . . . . . Stand density and dry matter yield of alfalfa sod seeded into a Kalamazoo sandy loam, pH 4.9, with various rates of lime surface applied or incorpo- rated into the plow layer before seeding (Exp. 4) . . CHAPTER 2 1. Climatological data from the National Oceanic and Atmospheric Administration for the three experimental locations in Michigan . . . . . . . . . . . . . . . . Stand density three months after seeding forage legumes at three locations on three different dates during 1979’ 1980, or 1981 O O O O O O O O O O I O 0 Yield of forage legumes from three harvests in the year after seeding . . . . . . . . . . . . . . . . . Percent survival of three forage legumes after two days of hardening and 1, 2, or 4 hours of freezing . Percent survival of three forage legumes after four days of hardening and 1, 2, or 4 hours of freezing . Page 28 29 3O 31 65 66 67 68 69 TABLE Page CHAPTER 2 6. Percent survival of three forage legumes after six days of hardening and l, 2, or 4 hours of freezing . . 70 CHAPTER 3 1. Stand density of red clover 3 and 15 months after sod seeding into 3 grass awards that were suppressed with a herbicide (check) or different cutting fre- quencies (Exp. 1) . . . . . . . . . . . . . . . . . . 97 2. Yield of red clover (red clover plus grass in parenthesis) sod seeded on two dates into three awards that were suppressed with a herbicide (check) or by different cutting frequencies (Exp. 1) . . . . . 98 3. Stand density of birdsfoot trefoil 3 and 15 months after sod seeding into two grass awards that were suppressed with a herbicide (check) or different cutting frequencies (Exp. 2) . . . . . . . . . . . . . 100 4. Yield of birdsfoot trefoil (trefoil plus grass in parenthesis) sod seeded on two dates into two grass awards that were suppressed with a herbicide (check) or different cutting frequencies (Exp. 2) . . . . . . 101 5. Stand density of red clover and birdsfoot trefoil 3 and 15 months after sod seeding into an 80% quack? grass sward that was suppressed with a herbicide (check) or different cutting frequencies (Exp. 3) . . 102 6. Yield of red clover and birdsfoot trefoil sod seeded on two dates into an 80% quackgrass award that was suppressed with a herbicide (check) or different cutting frequencies (Exp. 3) . . . . . . . . . . . . . 103 7. Stand density of red clover and birdsfoot trefoil 3 and 15 months after sod seeding into a 90% quack- grass sward that was suppressed with a herbicide (check) or different cutting frequencies (Exp. 4) . . 104 8. Yield of red clover and birdsfoot trefoil sod seeded on two dates into a 90% quackgrass sward that was suppressed with a herbicide (check) or by different cutting frequencies (Exp. 4) . . . . . . . . . . . . . 105 CHAPTER 4 1. Percent survival of three forage legumes grown in the greenhouse after seeding in contact with various kg ha-1 Of P O O I O O O O O O O O O O O O O I O O O O 1 21 vi TABLE Page CHAPTER 4 2. 3. Percent survival of three forage legumes grown in the greenhouse after seeding in contact with various kg ha-1 Of K . O O O O O O O O O O O O O O O O O I O O 122 Yield of three forage legumes obtained in the green- house 90 days after seeding in contact with various kg ha-l Of P O O O O O O O O O O O O O O O O O O O O O 12 3 Yield of three forage legumes obtained in the green- house 90 days after seeding in contact with various kg 118-1 Of K O O O O O I O O O O O O O O O O O O O O O 124 vii LIST OF FIGURES FIGURE Page CHAPTER 1 1. Soil pH obtained in the fall of each year from five depths of a Kalamazoo sandy loam after lime was broadcast on the surface or incorporated into the plow layer (Exp. 1) . . . . . . . . . . . . . . . . . 33 2. Soil pH obtained in the fall of each year from five depths of a Miami sandy loam after lime was broad- cast on the surface or incorporated into the plow layer (Exp. 2) . . . . . . . . . . . . . . . . . . . . 35 3. Soil pH obtained in the fall of each year from five depths of a Hillsdale sandy loam after lime was broadcast on the surface or incorporated into the plow layer (Exp. 3) . . . . . . . . . . . . . . . 37 4. Soil pH from five depths of a Kalamazoo sandy loam one (top) and two years (bottom) after lime was broadcast on the surface or incorporated into the plow layer (Exp. 4) . . . . . . . . . . . . . . . . . 38 5. Soil pH from four sub-surface levels of a Kalamazoo sandy loam (Exp. 1 and 4), Miami sandy loam (Exp. 2) and a Hillsdale sandy loam (Exp. 3) . . . . . . . . . 39 6. Total (top) and exchangeable (bottom) soil aluminum from four sub-surface levels of a Kalamazoo sandy loam (Exp. 1 and 4), Miami sandy loam (Exp. 2) and a Hillsdale sandy loam (Exp. 3) . . . . . . . . . . . 4O 7. Ethylene production from alfalfa grown on a Kalamazoo sandy loam soil, pH 4.9, and sampled two years after lime was broadcast on the surface or incorporated into the plow layer (Exp. 4) . . . . . . . . . . . . . 41 CHAPTER 2 1. Selected early spring daily maximum and minimum temperatures for 1979 at Location 2 (East Lansing) . . 71 2. Selected early spring daily maximum and minimum temperatures for 1980 at Location 2 (East Lansing) . . 72 viii FIGURE CHAPTER 2 3. 4. Selected early spring daily maximum and minimum temperatures for 1981 at Location 1 (Lake City) . . . Selected early spring daily maximum and minimum temperatures for 1981 at Location 3 (Kellogg) . . . . Percent survival of three forage legumes after being frozen for one (opposite top), two (opposite lower), and four hours (above) and averaged over the three hardening periods of two, four and six days . . . . . Percent survival of three forage legumes averaged over three freezing times (1, 2, and 4 hours) and three hardening periods (2, 4, and 6 days) . . . . . . CHAPTER 4 1. Fertilizer-band pH sampled in situ 3 and 10 days after placing various rates of K (top) and P (bottom) in a Brookston loam soil . . . . . . . . . . . . . . . . . Soil crease (0.5 cm deep) for seed and fertilizer place- ment and vacuum-operated seeder . . . . . . . . . . . Placement of alfalfa seed and K fertilizer by commer- cial equipment when sod seeding into a herbicide- suppress ed 80d 0 O O O O O O O O O O O O O O O O O O 0 ix Page 73 74 76 77 125 126 127 INTRODUCTION Production of many permanent pastures in the United States and in the entire world is limited because of a lack of adequate legumes in the award. Kentucky bluegrass and other low producing grasses are an integral part of many forage programs, particularly in the North Central and Northeastern United States. For many years efforts have been made to improve production of these pastures through pasture renovation ranging from improved and increased fertilizer applications and better grazing management to a complete killing of the existing sod by tillage to enable legumes and more productive grasses to be established. Over the last 50 years, an improvement in productivity of a pasture by any cultural practice has been known as a regeneration or renovation procedure. Graber (1936) worked extensively in improving unproductive pastures which he first called pasture renovation and described it as the establishment of dry weather legumes in grass sods without plowing. Pasture renovation is defined by the Crop Science Society of America as "the improvement of a pasture by the partial or complete destruction of the sod, plus liming, fertilizing, seeding, and weed control as may be required to establish desirable forage plants" (Decker et al., 1973). With the development in recent years of suitable chemical herbicides, pasture renovation has become known more generally as the establishment or re-establishment of high 2 yielding and well-adapted legumes or grasses in an existing sward without completely destroying the sod. The introduction of high-yielding forage legumes into less- productive grass awards by drilling the seed into the sod with a grain drill or other specialized equipment is now commonly referred to as sod seeding (White, 1966; Decker et al., 1969; Olsen et al., 1981). The broadcast application of legume seeds on the surface of grass sods has also become included in this generalized description of sod seeding (Tesar, 1980). While pasture renovation is understood to be the improvement of pastures by any method, including plowing and tilling, it is used synonymously throughout this study with sod seeding to refer to the introduction of forage legumes into a grass award by drilling the seed in the sod or broadcasting it on the surface. Plowing and field cultivation has been the most traditional, and considered the best method for re-establishing unproductive pastures, particularly when there has been no consideration for the cost of establishment or interruption to production. This method has been a more conventional and assured way of gaining a good stand (Roberts, 1960; Smith, 1975; Tesar and Hildebrand, 1975). There are, however, many areas that are too steep for plowing or too rough for conventional cultivating and seeding machinery. Renovation is par- ticularly suited to the steeper slopes and to soils more susceptible to erosion where exposure of the bare soil surface is not desirable. Smith (1975) states that the problem of weed control in sod seedings makes this a less desirable method of pasture renovation than plowing and complete seedbed preparation. Better stands, higher seeding year 3 yields (Scholl et al., 1970), and less work in establishing good stands are a result of plowing and seeding into a prepared seedbed, compared to renovation by surface cultivation (Smith, 1975). Other problems have been observed in making successful sod seedings into pastures in Midhigan. Many areas in need of pasture improvement are too steep and erosive for row crops and often have soils with pH values too low for alfalfa, the only permanent legume adapted to droughty soils. Accord- ing to present recommendations, lime must be incorporated into the plow layer if soil pH is to be increased satisfactorily (Christenson et al., 1981). This is not possible without tillage and it is not known how successful broadcast applications of lime on the surface would be in reducing soil acidity sufficiently for successful sod seedings of alfalfa. Adequate moisture is essential for good germination and early growth of legumes sod seeded into grass awards. This is especially crucial for seeds broadcast on the surface. The most successful sod seedings have been made in early spring where moisture from spring rains has been the most abundant (Taylor, et al., 1969; Tesar, 1980). "Frost seeding" is the broadcasting of forage legume seeds on the surface early in the spring while the ground is still frozen. The movement of the soil from alternate freezing and thawing has helped to cover seeds and benefit germination. It is not known, however, how much the early spring freezing temperatures reduce germination and seedling survival. Even though temperatures may have been high enough for germination of introduced legume seeds, stands may be depleted by subsequent freezing periods. The extent or duration of 4 freezing temperatures low enough to kill forage legume seedlings is also not known. Adequate moisture to germinate sod-seeded legumes is the most important factor for successful stands. Almost as important is the reduction of competition from the grass sward to the inter-seeded species. This has been achieved in many different ways such as close grazing, burning, disking, field cultivating, and the use of herbi- cides, all with varying degrees of success. Unfortunately,the most successful suppression methodsturherbicides and complete tillage to kill existing grasses have also been the most costly. Grasslands most often involve herbivore animals that could be utilized in the reduction of inter-species competition by grazing the sod-seeded areas. Several researchers (Roberts, 1910; Love, 1944; Cullen, 1970) have described sod seedings where the grass competition was reduced by grazing. Comparisons of grazing with other methods of grass suppression have not been reported in the literature. It is not known if grazing of grasses in an inter-seeded pasture is as effective as herbicides in reducing competition for good legume estab- lishment. Considerable research reported shows the value of added fer- tilizer for early vigorous growth of seedlings (Brown, 1959; Tesar et al., 1954; Sheard et al., 1971). The greatest benefit has been from the placement of fertilizer in bands below the seeds where developing tap roots of legume seedlings utilize the phosphorous within two weeks (Tesar et al., 1954). Fertilization of seedings drilled in the sod has frequently been with the seed and fertilizer being placed together in contact in the same slit in the sod. The 5 firm sod of the grasses in sod-seeded areas and the impracticality of machinery placing fertilizer under seeds in a sod have precluded separate placement. Injury to germinating seeds placed in contact with fertilizer has been studied in many crops, but not in forage legumes. Injury would have to be considerable to be noticeable in stands in the field. Prior to this study it was not known which rates and concen- trations of phosphorus and potassium in contact with forage legume seeds would be beneficial without being toxic to the seedlings. Sod seedings require better management than seedings in a pre- pared seedbed because of additional difficulties encountered in pre- cise depth placement of seed. In some cases, control of insects and snails in the sod are essential for good stands (Kalmbacher et al., 1979; Holland and Tesar, 1980). The primary objective of this study was to investigate some of the problems encountered when making sod seedings of forage legumes into various grass sods. Specific objectives were to (l) evaluate the effects of surface-applied lime on an acid soil when sod seeding alfalfa; (2) compare forage legume seedling survival at various freezing temperatures; (3) compare lowbcost simulated grazing of grass competition with a recommended herbicide when sod seeding; and (4) determine at what level phosphorus and potassium become toxic to seedlings when placed in contact with the legume seeds. CHAPTER 1 EFFECTIVENESS OF SURFACE-APPLIED LIME ON ACID SOILS WHEN SOD SEEDING ALFALFA (MEDICAGO SATIVA.L.) ABSTRACT Many areas of unproductive pastureland could be improved with the introduction of forage legumes into the existing grass sod. Pas- tures have often been grown in areas that have erodible and acidic soils, frequently droughty and usually not considered suitable for row crops. Seeding alfalfa (Medficago sativa L.) the only drought- resistant, long-lived legume adapted to these soils has generally not been successful in low pH soils. Recommendations state that lime must be incorporated into the plow layer to reduce soil acidity for good alfalfa growth. In this study alfalfa was grown on acid soils to determine the effectiveness of lime broadcast on the surface either before or after seeding compared to lime incorporated into the plow layer. Equally as much forage was produced when lime was broadcast at 11.2 t had1 on the surface or incorporated into the plow layer as presently recommended. No differences in stand density were obtained in three of the four trials when lime at this high rate was either surface applied or incorporated into the plow layer. Alfalfa 7 grown with lime broadcast at 2.8 t ha”1 produced better yields in comparison to no lime only on soils below pH 5.0. On soils of higher pH yields were equally as good when lime was broadcast at O or 2.8 t ha-l. Lime at 2.8 t ha-1 on the surface after seeding alfalfa into a soil of pH 5.94 increased stand density but not yields. At this low rate of surface-applied lime, stand establishment was better when alfalfa was seeded into an untilled sod but not in a plowed and prepared seedbed. Nitrogen fixation, determined by acetylene reduction, occurred only in areas of the soil where lime had penetrated adequately to reduce the acidity. The quantity of nitrogen fixed was not well correlated with soil pH and decreased with high exchangeable soil Al but increased proportionately with added lime. Additional index words: pasture renovation, glyphosate, lime incorporation, acidic soils, aluminum toxicity, acetylene reduction. INTRODUCTION Grasslands occupy about one-half the total land area of the 48 contiguous states (Sprague, 1974). This is more than that uti- lized by all other crops combined, yet this area, according to the American Forage and Grassland Council (1974), is producing less than 25% of its potential. Yields of pastures and grassy hayfields in Michigan were shown to be easily doubled or tripled by the establish— ment of productive legumes in them (Tesar, 1975). Wedin et a1. (1965) reported that yields of mixtures of grass with 30 to 40% legumes were equivalent to pure grass stands heavily fertilized with nitrogen. In the eastern humid region more than 75% of the pastures are located on land too steep for conventional tillage which is essential for com- plete pasture renovation (USDA, 1971). Sod seeding is the introduc- tion of a legume (or grass) into a suppressed sod without tillage and is an attractive alternative to conventional methods of pasture estab- lishment, particularly where there are hills and slopes where it is impractical or conservationally unwise to plow and prepare a seedbed. This method, however, has not been without problems. The suppressed sod has on many occasions harbored insects and slugs that damaged or destroyed introduced seedlings (Braithwate et al., 1958; Kalmbacher et al., 1979; Holland and Tesar, 1980). Difficulties have also been encountered in not being able to place the introduced legume seed into the soil through the sod of heavily rooted grasses 9 (Holland and Tesar, 1982). Further problems have also been encoun- tered when sod seedings have been made on soils of a low pH or during months when rainfall was not optimum (Holland, 1980). Considerable work has shown the essentiality of incorporating agricultural limestone into the plow layer for maximum benefit to alfalfa, an acid-sensitive forage legume (Weidemann, 1936; Longenecker and Merkle, 1952; Hourigan et al., 1961). When the liming materials were not incorporated by tillage, penetration was relatively slow. Brown et a1. (1956) applied 4.6 and 13.8 metric tons of lime per hectare to the surface of a grass sod and measured the pH with depth over a ten-year period. These researchers found that the rate of lime had less effect on pH adjustment than time. Their work indi- cated that it may take as long as 10 years to neutralize the plow layer. A similar study by Longenecker and Sprague (1940) showed that lime penetration was dependent on soil type and time. Acid soils provide an unfavorable environment for most legumes, and low pH soils, without a calcareous sublayer, often contain high levels of toxic aluminum (A1). Buss et a1. (1975) demonstrated that alfalfa cultivars have a narrower range of acid tolerance than other crops while Al has been shown to be the principal cause of poor growth in low pH soils (Fay and Brown, 1964; Kamprath, 1970). Munns and Fox (1976) showed that alfalfa growth increased relatively more by the adjustment of pH from 5.5 to 6.0 than within the ranges of 5.0 to 5.5 and 6.0 to 6.5. Coleman et a1. (1959) determined that Al saturation was reduced to less than 10% of saturation in several soils by increasing the pH to 5.6. Alfalfa produces the best yields on near-neutral soils. 10 Nodulation and nitrogen fixation are greatly enhanced by favorable soil conditions. Sod seeding of forage legumes is the most practical and conservationally sound method of pasture improvement on erodible slopes and hillsides. Frequently the soils in these areas are too acid to sustain a good stand of alfalfa. This study was instigated to: (1) compare alfalfa stand establishment, relative forage yield, and soil pH changes when agricultural limestone was broadcast on the surface or incorporated, as presently recommended, into the plow layer of acid soils; and (2) determine the relationship, if any, between the soil depth at which nitrogen fixation occurs on alfalfa grown in an acid soil and lime surface applied or incorporated into the plow layer. MATERIALS AND METHODS Four field experiments were established at two locations on three different acid soils. Experiment 1. This study was conducted on a Kalamazoo sandy loam (fine-loamy over sandy mixed, mesic Typic Hapludalfs) soil of pH 4.8. Treatments were replicated four times in a split—plot, ran- domized, complete block design. Alfalfa was clear seeded at 9 kg ha.-1 in the spring of 1978 after the area was plowed and fertilized with 45 kg P ha-l. Prior to plowing, 5.6 t ha-1 of agricultural limestone was broadcast on the check treatments with an additional 5.6 t ha-l added after plowing. This was incorporated by disking to a depth of 10 cm into the tilled surface. The rate of 11.2 t ha-1 is the maxi- mum rate of lime recommended for application in one year on Michigan soils and is similar to recommended amounts in other North Central states (Warncke and Christenson, 1980). Other treatment blocks received 0, 2.8 and 11.2 tons of lime per hectare broadcast on the surface after seeding the alfalfa. Recommended levels of P and K fertilizer were applied annually. Stand density was determined six weeks after seeding by counting alfalfa seedlings in four directed, 35-cm quadrat samples from each plot. Soil samples were obtained to a depth of 30 cm in five increments (O-2.5, 2.5-5, 5-10, 10-20, 20-30 cm), four months 11 12 after seeding, and in each subsequent year. Soil pH was determined from these samples by using a 1:1 soil/water ratio. Four additional soil samples were also obtained at 15-cm increments below the 30-cm level and analyzed for pH, total acidic and exchangeable aluminum. To determine total acidic soil Al, samples were extracted with 1N NH4OAc (pH 4.8) and analyzed by Directly Coupled Plasma Emission (DCPE). Exchangeable Al was determined by 1N KCl extraction and analyzed by the same DCPE. One forage harvest was made in the year of seeding and three in each succeeding year. Prior to each harvest percent alfalfa growing in each plot was estimated visually and these figures were used to calculate legume yield. Harvests were made from a 0.9 x 9.1 m area with a self-propelled, direct-chop harvester. A 1 kg forage sample from selected plots was dried with forced air at 65 C for 48 hours and used to determine dry matter. All yield data are reported in dry matter t ha.1 of the legume portion of the total forage yield. Experiment 2. This trial was established on a Miami sandy loam (fine-loamy, mixed, mesic Typic Hapludalfs) soil of pH 5.3 and differed from Exp. 1 only in location, soil type, and initial soil pH. All data were obtained in an identical manner on similar dates and reported as in Exp. 1. Experiment 3. This trial was seeded on a Hillsdale sandy loam (coarse-loamy, mixed, mesic, Typic Hapludalfs) soil of pH 5.9 at the same location as Exp. 2. Treatments were replicated three times in a split-plot, complete block design. Fertilizer was not 13 added at seeding but recommended levels were applied annually. A com- mercial grain-fertilizer drill wtih a small-seeded-legume box adapted to provide precision seed setting was used to sod seed alfalfa at 13.5 kg ha-1 in the spring of 1980 into a NB(phosphonomethyl)glycine (g1yphosate)-suppressed quackgrass (Agropyron repens L.) sod. Lime at O, 2.8, 5.6 and 11.2 t ha—1 was surface applied to designated blocks after seeding. Alfalfa stand density, annual yields, soil aluminum, and pH were evaluated in the same way as in Exp. 1. Experiment 4. In the spring of 1981 alfalfa was sod seeded into a quackgrass-infested sward on a Kalamazoo sandy loam soil. Treatments were replicated four times in a split-plot complete block design. Fertilizer was not added at seeding but recommended levels were applied annually. The same commercial drill used in Exp. 3 was used to make all seedings, including those of the check plots in a prepared seedbed. Glyphosate was used in the fall of 1980 to suppress the grasses. During late fall prior to seeding the alfalfa, lime was broadcast on the surface of designated blocks at 0, 2.8, 5.6, and 11.2 t ha-l. Check blocks received 5.6 t he"1 before plowing and 5.6 t ha-l incorporated into the tilled surface after plowing. Data were obtained similarly and reported as in the previous trials. Acetylene reduction analysis as described by Hardy et a1. (1968) was conducted on randomly selected plants prior to the final harvest in the fall of 1982. Replicated core samples 10 x 30 cm deep were obtained from all plots with each core sampled directly over one alfalfa crown. Samples were discarded where the tap root was not completely contained within the core. Each soil core containing the 14 alfalfa root was divided at 2.5- and 5.0-cm depths and thereafter at 5.0-cm increments to a total depth of 30 cm. The lower five depths of the soil cores were each placed in a 1-L container and sealed with a metal lid equipped with a serum stopper. The first and second depths were sealed in 0.5-L containers so as to adjust for the smaller bulk quantity of these samples. Ten percent of the atmosphere in each container was evacuated and replaced with calcium-carbide-generated acetylene. The samples were then incubated under a cover at ambient field conditions for one hour. All sampling was conducted between 0900 and 1200 hours to min- imize variation due to diurnal fluctuations in the rate of nitrogen fixation. One ml of gas was withdrawn from each container at the end of the incubation period and analyzed on a gas chromatograph (Varian aerograph series 1400) with a flame ionization detector. Ethylene production rates were quantified by peak height and are expressed in micromoles sample.1 hour-1. Four controls were utilized to determine ethylene source: (1) complete samples (soil and roots) were assayed but did not have acetylene added; (2) plant samples (crowns and roots) without soil were treated similarly; (3) cores of soil without legume roots were segmented and assayed with quantities of acetylene added to each con- tainer; and (4) sealed empty assay vessels were treated as complete samples and a 10% atmospheric concentration of acetylene was added. RESULTS AND DISCUSSION Experiment 1, Kalamazoo sandy loam, pH 4.8: Alfalfa stand establishment was equally as good with lime surface applied at 11.2 t ham1 or incorporated into the plow layer as presently recommended (Table 1). When lime was not added to this acid soil, fewer seed- lings survived than when 11.2 t haul was either broadcast on the surface or incorporated into the plow layer. Stand establishment was similar when lime was added at 0 or 2.8 t ha-l. Seedling counts made six weeks after seeding indicated that lime at 2.8 t ha.l was not adequate to neutralize the soil acidity for good germination. Soil samples taken four months after seeding indicated an insignificant change in pH at the 2.8 t ha—l lime application rate (Fig. 1). In comparison, 11.2 t ha.1 of lime on the surface or incorporated decreased the surface soil acidity over the same period by 1.26 and 1.77 pH units, respectively. There was no significant difference in total four-year yields of alfalfa on soil treated with 11.2 t ha-1 of lime broadcast or incorporated into the soil as recommended. Alfalfa grown where O, 2.8, and 11.2 t ha-1 of lime was applied to the surface produced 42, 83, and 92%, respectively, of the alfalfa in soil with lime incorpo- rated into the plow layer at 11.2 t ha-l. When the 11.2 t ha.1 lime application was split with one-half plowed under and the remainder surface applied and incorporated, 15 l6 noticeable pH changes occurred to a depth of 30 cm (Fig. 1). This would be more indicative as an effect of plowing depth rather than lime movement through the soil profile. When the same amount of lime was surface applied the pH in the surface 2.5 cm was nearly as high as when the lime was incorporated. The surface-applied lime at 11.2 t ha”1 did not appreciably increase the pH of the 2.5 to 5.0 cm depth in the year of application but in the third, fourth, and fifth years after application pH increased significantly from 4.7 to 5.5, 5.6, and 6.0, respectively. No significant changes occurred below 5 cm. At the lime application rate of 2.8 t ha-l the pH of the surface 2.5 cm changed from 4.8 to 5.6 in two years in comparison to a pH change to 6.3 for 11.2 t ham1 on the surface over the same period. A decline over time in soil surface pH when the lime was incorporated was most likely due to lower lime concentration in the surface 2.5 cm and leaching of this lime through the tilled soil. When no lime was applied, the soil surface became slightly less acidic in the first two years. This most likely occurred as a result of additional plant and root growth from the introduced alfalfa crop. Subsurface soil pH was extremely low (Fig. 5) and total soil acidic aluminum was very high (Fig. 6). Exchangeable aluminum was present in quantities sufficient to have a toxic effect on alfalfa and very likely was the limiting factor for good plant growth and production on the unlimed soil (Fig. 6). Experiment 2, Miami sandy loam, pH 5.3: Alfalfa, with lime surface applied at 11.2 t ha-l, produced as good yields over four years as when the same amount was incorporated into the plow layer l7 and alfalfa stand establishment was equally good at all lime treat- ments (Table 2). The initial soil surface pH of 5.3 was 0.5 pH unit more alkaline than in Exp. 1 and would have contributed to the higher stand density and lack of treatment differences. Four-year yields of alfalfa when lime was surface applied at 11.2 t ha-1 were 14% higher than yields from treatments of 2.8 t ha"1 and 24% greater when no lime was added (Table 2). No yield increases were obtained with the addi- tion of lime at 2.8 t ha-l. In the first year after seeding (1979), however, yields were highest from the incorporated lime treatment. This may indicate an early, short-term advantage for this soil by incorporating the lime into the plow layer. The pH of the surface 2.5 cm was higher in each of the four years with 11.2 t ha"1 of lime surface applied than when the lime was incorporated as recommended (Fig. 2). In the third, fourth, and fifth years after application of 11.2 t ha.1 on the surface pH of the 2.5 to 5.0 cm depth was increased from 5.1 to 6.3, 6.6, and 6.7, respectively. These pH values were equal in year three and higher in the last two years than when the lime was incorporated. There was a noticeable, but lesser improvement of pH in the 5 to 10 cm depth in the last three years with pH values in the fourth and fifth years (6.1 and 6.3) equalling those when the lime was incorporated. There was a slight increase in pH in the fourth and fifth years at the 10-20 cm.depth indicating that lime applied on the surface at the recommended rate increased pH below 10 cm.after three years. When lime was surface applied at 2.8 t ha-l, only four months were required to decrease the surface soil acidity 0.94 pH unit (Fig. 2). This was a five-fold greater change than that obtained 18 over the same period from the addition of lime to the soil in Exp. 1 (Fig. 1). Higher initial pH (Fig. 2), less acidic subsoil (Fig. 5) and relatively very little soil aluminum (Fig. 6) all may have con- tributed to the more rapid decrease in acidity of this soil. Sub-surface acidity decreased with depth indicating a calcar- eous sublayer (Fig. 5). Only traces of exchangeable soil A1 were present and this was not great enough to cause plant toxicities (Fig. 6). The yearly decrease of the surface pH when lime was incorporated into the plow layer resulted from lower lime concentra- tion per unit of soil and possible leaching in the disturbed profile. Experiment 3, Hillsdale sandy loam, pH 5.9: Stand establish- ment was better at all levels of liming than with no lime (Table 3). Stands were 29, 52, and 69% better when lime was applied at rates of 2.8, 5.6, and 11.2 t ha-l, respectively, than when no lime was applied. The stand was 31% better when lime was applied at 11.2 com— pared to 2.8 t ha-l. These differences are particularly significant since the lime was not applied until after seeding and stand determi- nations were made ten weeks later. The soil had a relatively high initial pH (Fig. 3) but a definite benefit in seedling establishment was derived from the lime broadcast after seeding. The surface-applied lime increased the pH appreciably in the surface 2.5 cm at all three levels of application and increases were similar but somewhat less at the 2.5 to 5.0 cm depth (Fig. 3). At the 5 to 10 cm depth, pH increases were not significant until the second and third years after application. Below 10 cm there was only a slight change in pH. 19 Alfalfa yield was equally as good in all years from limed or unlimed areas (Table 3). The alfalfa on this soil of pH 5.9 produced 23.2 t ha_1, however, this was not significantly less than the 27.4 tons from alfalfa on the soil with lime surface applied at 11.2 t ha-l. A high initial soil pH of 5.9 (Fig. 3), an extremely alkaline sub- surface (Fig. 5), and negligible quantities of exchangeable soil alu- minum (Fig. 6) likely contributed to the good yields from the unlimed plots. This agrees well with data from Rice et a1. (1977) that showed relative yields of alfalfa increased with increasing soil pH to 6.0, and then became constant at higher pH levels. Experiment 4, Kalamazoo sandy loam, pH 4.9: Stands were 66,89, and 77% better when lime was broadcast on the surface at 2.8, 5.6, and 11.2 t ha-l, respectively, than when no lime was applied (Table 4). Alfalfa, with lime incorporated into the plow layer at 11.2 t ha-l, had a 16% greater stand density than when no lime was added. Yields in the seeding year were low but equal following surface applications of lime at 2.8, 5.6, or 11.2 t ha-l, but all were better than when no lime was applied. When the lime was incorporated into the plow layer, yields in the year of seeding were more than double the yields from any surface-applied treatment. In the year after seeding (1982), however, yields were high and equally good when lime at 11.2 t ha.1 was broadcast on the surface or incorporated into the plow layer. When lime was broadcast at 2.8 t ha-l, alfalfa yield in the second year (1982) was 30% lower than when 11.2 t ha.-l was broad- cast but almost five times greater than when no lime was applied. Surface application of lime at 5.6 and 11.2 t haml increased yields 20 31 and 43% more, respectively, than yields obtained at the 2.8-ton level. Within 12 months after broadcasting lime on the surface at 2.8, 5.6, and 11.2 t ha-l, soil pH of the surface 2.5 cm was increased at all three levels from 5.0 to 6.4, 6.7, and 6.8, respectively (Fig. 4). When the recommended rate of 11.2 t ha-1 was incorporated the surface pH was only slightly higher (6.9) than from the same amount surface applied. The pH in the 2.5 to 5.0 and 5 to 10 cm depths, however, was greater in both years when the high rate of lime was incorporated but was much lower (5.8) in the second year when sur- face applied. This collective neutralization of the pH for the three surface-applied treatments would have helped produce the similar seeding year yields and lack of stand differences. The lime was applied six months prior to seeding and most likely weathered suf- ficiently to provide a more suitable seedling environment than the unlimed acidic soil. The soil pH did not change significantly between the first and second year after liming (Fig. 4). In the same period, however, forage yields increased markedly as lime applications were increased to 11.2 t ha-l. At this level, yields of alfalfa.were similar regardless of whether the line was surface applied or incorporated as recommended. Greenhouse studies with incremental lime rates incorpo- rated into a similar Kalamazoo soil (Ross et al., 1964) produced pro- portionally comparable, but larger, yield increases. Low subsurface pH (Fig. 5) and toxic quantities of exchangeable soil A1 (Fig. 6) most likely resulted in the lower yields associated with these low levels of lime. With increased calcium available from the higher lime 21 rates, A1 toxicity to the alfalfa would likely have been reduced and yields increased accordingly. Acetylene reduction tests demonstrated distinct differences in atmospheric nitrogen fixation ability between alfalfa grown in areas of high and low lime rates (Fig. 7). When no lime was added to this acid soil, the alfalfa plants were small and produced relatively no ethylene. Alfalfa grown where lime was applied at 2.8 t ha.l pro- duced only small amounts of ethylene. When the lime rate was doubled to 5.6 t ha-l, however, the ethylene produced by nodules in the sur- face 5 cm was more than four times greater. Alfalfa grown with lime surface applied at 11.2 t ha.1 produced almost eight times as much ethylene as alfalfa on soils limed with 2.8 t ha-l. Only nodules growing at soil depths affected by the surface—applied lime produced a significant amount of ethylene. Difficulties in sampling the sur- face 2.5 cm may have caused the inconsistent ethylene production where lime was broadcast on the surface at 2.8 t ha-l. From the data presented, it appears that soil pH (Fig. 4) was not the only factor that affected alfalfa nodulation and, there- fore, subsequent nitrogen fixation. Ethylene production (Fig. 7) did not correlate well with soil pH data presented in Fig. 4. Soil surface acidity was neutralized sufficiently for nodulation and growth even at the lowest lime rate. Alfalfa was shown by Munns (1970) not to nodulate below pH 4.8 even at high levels of calcium. At the same time he demonstrated that low levels of calcium (0.2 mM in solution) inhibited nodulation regardless of pH. Calcium was available in this acid soil in quantities ranging from 1000 kg ha.1 in the no lime treatments to 5000 kg ha.1 in the surface 2.5 cm when lime was broad- 22 cast on the surface at 11.2 t ha-l. A combination of relatively lower quantities of available calcium from the various lime treat- ments and high levels of soil A1 most likely prevented greater acetylene reduction at the lower liming rates. It is not known why ethylene production was comparatively low when lime was incorporated into the plow layer when the 2.5-cm surface layer had a pH of 6.7, equal to that where the lime was surface applied. Roots and nodules may have grown over a wider area of the less acidic soil profile and a lower percentage of nodules possibly was sampled within each soil core used for analysis. Webel et a1. (1976) found that incorporation of lime provided a uniform distribution of alfalfa nodules throughout the root system, whereas broadcasting the lime on the surface resulted in a large cluster of nodules at the crown.with very few nodules on the rest of the root system. Excellent stand establishment and yields from alfalfa when lime was incorporated indicated that nodulation and total nitrogen fixation was more than adequate. In areas where plowing and tilling of the soil is not recom— mended because of the danger of erosion, greater productivity can be gained by introducing high yielding alfalfa. Surface application of lime applied at recommended rates in these experiments, even on extremely acid soils, reduced surface acidity sufficiently for good alfalfa stand establishment and subsequent yields, similar to those in a soil with lime incorporated into the tilled surface as recom- mended. SUMMARY AND CONCLUSIONS Four field experiments were conducted at two locations on three different low pH soils. Alfalfa was seeded into a tilled seed- bed or drilled into a herbicide-suppressed sod. Lime was incorporated into the plow layer before seeding or broadcast on the surface without incorporation either before or after seeding. Equally good stands and yields of alfalfa were produced in two experiments when lime was applied in the same quantity on the surface or incorporated into the plow layer. Surface-broadcast lime resulted in increased stand density on a relatively high pH soil even when applied after seeding, but yields were not increased. In the fourth trial, a better stand was obtained by incorporating the lime but yields were not different from those produced by alfalfa when the lime was broadcast on the surface. Subsoil pH levels were very low in the two trials that produced the poorest alfalfa yields when lime was not added. Potential toxic levels of exchangeable soil aluminum corresponded to the low pH of these areas. Greater quantities of lime were required to successfully grow alfalfa in the trials with high levels of exchangeable subsoil aluminum. Significant nitrogen fixation occurred only at soil depths to which surface applied lime had penetrated or where lime was incorporated into the plow layer. The data in these four experiments on acid soils indicate 23 24 that alfalfa can be grown successfully on low pH soils with adequate amounts of lime surface applied to reduce acidity. Many existing unproductive grass pastures of low pH could be improved simply and economically by sod seeding alfalfa following an application of adequate amounts of lime broadcast on the surface of sods suppressed to permit alfalfa establishment. 3. 10. LITERATURE CITED American Forage and Grassland Council. 1974. In H.B. Sprague (ed.) Grasslands of the United States. Iowa State Univ. Press. Ames, Iowa. Braithwaite, B.M., A. Jane, and F.G. Swain. 1958. Effects of insecticides on sod-sown subterranean clover. J. Aust. Inst. Brown, B.A., R.I. Munsell, R.F. Holt, and A.V. King. 1956. Soil reactions at various depths as influenced by time since application and amounts of limestone. Soil Sci. Soc. Amer. Proc. 20:518-522. Buss, G.R., J.A. Lutz, Jr., and G.W. Hawkins. 1975. Yield response of alfalfa cultivars and clones to several pH levels in Tatum subsoil. Agron. J. 67:331-334. Coleman, N.T., J.B. Weed, and K.J. McCracken. 1959. Cation exchange capacity and exchange cations in Piedmont soils of North Carolina. Soil Sci. Soc. Amer. Proc. 23:146-149. Cowan, J.R. 1974. American Forage and Grassland Council. In H.B. Sprague (ed.) Grasslands of the United States. Iowa State Univ. Press. Ames, Iowa. Foy, C.D., and J.C. Brown. 1964. Toxic factors in acid soils. II. Differential tolerance to aluminum in plant species. Soil Sci. Soc. Amer. Proc. 28:27-32. Hardy, R.W.F., R.D. Holsten, E.K. Jackson, and R.C. Burns. 1968. The acetylene-ethylene assay for nitrogen fixation; laboratory and field evaluation. Plant Physiol. 43:1185-1207. Holland, Clive. 1980. The establishment of alfalfa (Medicago sativa L.) and birdsfoot trefoil (Lotus corniculatus L.) in various grass sods as affected by date and method of seeding, and herbicide application. M.S. Thesis, Mich. State Univ. East Lansing, Mich. , and M.B. Tesar. 1980. Unpublished observations of insect destruction of alfalfa seedlings sod seeded into a herbicide suppressed sward. Mich. State Univ. East Lansing, Mich . 25 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 26 , and . 1982. Unpublished field trials of sod seedings made into grass sods suppressed with a herbicide three days before seeding. Mich. State Univ. East Lansing, Mich. Hourigan, W.R., R.E. Franklin, Jr., E.O. McLean, and D.R.Bhumbla. 1961. Growth and Ca uptake by plants as affected by rate and depth of liming. Soil Sci. Soc. Amer. Proc. 25:91-494. Kalmbacher, R.S., D.R. Minnick, and F.G. Marten. 1979. Destruction of sod seeded legumes by the snail Pbegyia cereolus Kamprath, E.J. 1970. Exchangeable aluminum as a criterion for liming leached mineral soils. Soil Sci. Soc. Amer. Proc. 34:252-254. Longenecker, D., and F.G. Merkle. 1952. Influence of placement of lime compounds on root development and soil characteristics. Longnecker, J.F., and H.B. Sprague. 1940. Rate of penetration of lime in soils under permanent grass sod. Soil Sci. 50:277-288. Munns, D.N. 1970. Nodulation of Medicago sativa in solution culture. V. Calcium and pH requirements during infection. Plant Soil 32:90-102. , and RJL. Fox. 1976. Comparative lime requirements of tropical and temperate legumes. Plant Soil 45:610-705. Rice, W.A., D.C. Penny, and M. Nyborg. 1977. Effects of soil acidity on rhizobia numbers, nodulation and nitrogen fixation by alfalfa and red clover. Can. J. Soil Sci. 57:197-203. Ross, G.J., K. Lawton, and B.G. Ellis. 1964. Lime requirements related to physical and chemical properties of nine Michigan soils. Soil Sci. Soc. Amer. Proc. 28:209-212. Sprague, H.B. 1974. Significance of our grasslands. p. 3-5. In H.B. Sprague (ed.) Grasslands of the United States. Iowa State Univ. Press. Ames, Iowa. Tesar, M.B. 1975. Establishment of alfalfa on a quackgrass sod in northern Michigan. Michigan State Univ. Res. Rep. 288:148-156. USDA Conservation Needs Inventory Committee. 1971. Basic sta- tistics. National Inventory of soil and water conservation needs. 1967 Stat. Bull. 461. Warncke, D.D., and D.R. Christenson. 1980. Fertilizer recom- mendations, vegetable and field crops in Michigan. Mich. State 25. 26. 27 Webel, D., P.R. Henderlong, and F.L. Himes. 1976. Lime and phosphorus response of Madicago sativa seedlings on low pH hill land soils. p. 177-180. Proc. Int. Sym. Hill Lands. W. Virginia Univ. Mbrgantown, W. Virginia. Wedin, W.R., J.D. Donker, and G.C. Marten. 1965. An evaluation of nitrogen fertilization in legume-grass and all-grass pastures. Agron. J. 57:185-188. 28 no.0 m.m ©.H m.H O.N N.H m.o mN de O.mm N.MH O.MH N.MH m.NH m.N RON AQHOUflfiv N.HH m.m¢ m.NH O.mH m.NH 5.0H m.m me NoHH O.€c 0.0H m.NH w.HH N.m m.N med w.N N.NN o.h N.w o.¢ o.m N.H NOH 0 Him; a Hence «was Haas owes «has new» Nye sue; a umlc wafimmom vaofiw mwswavoom mafia .AH .mxmv wcwpoom ouowon “whoa scan was ousH canon nonsense no woaooom Houmm voHHmam moonwam mafia mo moumu m50fium> sues .w.¢ mm .amOH henna ooumamamx voHHeu m ease powwow memmHm mo vaowh wounds hum can hufimcoe venom .H manna 29 no.0 0.5 m2 ¢.H m.H H.¢ m2 m2 emu «.mo m.¢H m.oa n.5H m.mH m.~ «mm Announav N.Ha m.mo «.mH «.ma m.wH n.0H N.N Hon N.HH m.nn m.ma “.ma m.cH n.HH N.~ com w.~ «.mm m.ma e.ca H.0H n.w o.~ can 0 Hum: e Hmwmw mama meH owma anma mLMWMWm Nua Him: u mama» mmcfiapoom mafia .AN .mxmv mswwomm ouowon momma 30am ago some eoumuomuousa no mcuvomm woumm pmeaaqm oomwunm mafia mo mmumn unawum> nufi3_.m.m an .amoa meson Hana: cmHHHu m coca common mwammam mo mama» wouuma map can Anemone esmum .N manna 30 Table 3. Stand density and dry matter yield of alfalfa sod seeded into a Hillsdale sandy loam, pH 5.9, with various rates of lime surface applied after seeding (Exp. 3). Lime Seedlings Yield _1 _2 Seeding 2-yr t ha m Year 1981 1982 Total t ha-1 0 83 1.1 11.3 11.9 23.2 2.8 107 1.2 12.8 13.7 26.5 5.6 126 1.5 12.3 13.7 26.1 11.2 140 1.9 13.3 14.1 27.4 LSD 20 NS NS NS NS 0.05 31 Table 4. Stand density and dry matter yield of alfalfa sod seeded into a Kalamazoo sandy loam, pH 4.9, with various rates of lime surface applied or incorporated into the plow layer before seeding (Exp. 4). Lime Seedlings Yield t he’1 m’2 Seeding Yr 1982 t ha-1 0 61 0.2 1.5 2.8 101 1.3 7.2 5.6 115 1.5 9.4 11.2 108 1.5 10.3 11.2 (incorp) 254 3.2 11.8 LSD 37 0.5 1.6 0.05 32 259:. ~._. 889.: ~._~ can... .3 UZH... m.~ a N.“— I. a . ‘I a). .. to GNIQ~ Ihmun :-er e. quu N.—— a 'D In a .N O a)— as :0 mtm.~ Ihmun :82: 9.2. 3 or: ~._— ~.__ o.~ a 0"). a..." .1- a. tU Quin Ihaun 58;: 2.21 3 “123 ~.__ N.__ m.~ 0 Han)» 09" alaa arena to m.mla Ihlun Hd 110$ Hd 1105 33 .AH .nxmv poked 30am osu ouou moumwonwousa no momwwnm ago so unmovmoun mos.oaHH Houmo amoa ensue ooumamHmM m we anyone o>am sown you» soon no Hana one aw codename mm qum .H .wfim .32.: 2.21 3 us: ~._— N._~ o.~ D I. ..1. "Cl- 3' I... 3". 'Is . .. a tu OMION Ihaun 34 .E89:. m... .ecouzn. N."— ATnz 0v UIHJ ~.~_ o.~ a .0). '0'). In" at..' to DNIO— thmun n70: 0g UZHJ ~.__ m.~ o .a" 21.0. al- a. to mlm.m Ihmun £82: 9-2 3 we: N.__ N.~— o.~ 6 .'.I__ a..." .1- a. to O—lm Ihmwn :82: 9-21 3 ”123 N.—_ ~.—_ m.~ a '3'). 9‘9)— aolan at... to m.~la Ihmun 35 .AN .mxmv momma 30am o:u can“ vmumuomuouofi Ho oommunm ago so unmoomoun mus mafia umumm amoa meson Hamfiz m mo mnunop o>wm aoum home soon we Hana on» me codename we HHom «gnu Adi“: 9v UZHI— N.u— ~.—— D.N D ...‘ '. . at.. al- I: .l- ' .. a .1- 3. t0 ONION Ihlun .N .me 57m; av UZHJ u.m m.~ Q a " II all to QNID_ Ihmun “To: 9v UZHJ m.m o.~ O a. .a. In. eel. tu mlm.~ Ibmua Hd 1105 Hd 1105 nToL vv UZHJ ~.__ m.m m.~ D n a.‘ at». to O_lm Ihlua 57o: av UtHJ ~.__ m.m m.~ 0 :0 w.~lu Ihlun Hd 1Ios Hd 1Ios 37 .Am .933 momma Beam o5 35" woumwonfioog uo oommunm on”. so unmovmoun mm: mafia won—mm Emoa bosom mamvmadm a. mo mnunmv ~53 Scum you.» some mo 3mm 93 a.“ powwouno ma :om NJ— :fi: .3 Uta.— . II Hd 1105 to ONION Ihlun .n .wE 7.5 SOIL pH '3 in 7.5 SOIL pH In in 5.0 4.5 4.0 Fig. 4. 38 ME I 2 3 4 5 DEPTH (CH) 0-2.5 2.5-5 5-10 10-20 20-30 p r _1 '1 F . .. i" _ r _ 7,. 12345 12345 12345 12345 12345 a 4' 2.9 5.6 I 11.2 11.2 (INCORP) LIME (r hi“) SME I 2 3 4 5 DEPTH (CH) 0-2.5 2.5-5 5-10 10-20 20-30 F. r- " ha I a i. - F i“ I _ F ’ F" 12345 12345 12345 12345 12345 a 4 2.9 5.5 11.2 11.2 (INCORP) LIME (t he?” Soil pH from five depths of a Kalamazoo sandy loam one (top) and two years (bottom) after lime was broadcast on the surface or incorporated into the plow layer (Exp. 4). 39 .Am .nxmv amoa momma mamemaafim m use Am .ame amoa meson Heme: .Aq mom H .ame EmoH henna ooumEmHmM m we mHm>oH mommunmIASm poem Baum an Hwom hzuz Hmumxw e m m Golan nnlam Dmlmo avian “to. Ihlun "NF!" qutcm j .m .wE O.v n.n Hd 1105 aoaianssns 40 '03 . smart: DEPTH (cm x F I 33-43 F— ; 3'33) 2 45-63 g - 3 63-73 V 6.87 _ 4 75-33 _ E . :1 5.33- — o (D . ‘3’ 3.33 » h H I H '— F'— F'— _F- U —i m '—"1 (if 1.8?r '5 r1234 1234 1234 1234 " 0.00 l 2 3 4 EXPERIMENT 5.67 mu: DEPTH (cm .-. ’ I 33-43 To, 2 43-63 .x 5,39. 3 33-75 ° L s ‘5 T- E 3.33" g . T m T7 :1 1.67'P u x U r- 1234 1234 1234 1234 0.00 ~ “‘ _—‘ _— l 2 3 4 EXPERIMENT Fig. 6. Total (top) and exchangeable (bottom) soil aluminum from four sub-surface levels of a Kalamazoo sandy loam (Exp. 1 and 4), Miami sandy loam (Exp. 2) and a Hillsdale sandy loam (Exp. 3). 41 .Aq .nxmv Momma 30am osu Oucw moumwonuoosfi no mommusm onu no ammovmoun mm? mafia Houmm mums» oBu emanamm pom .m.¢ an .HHom EMOH apnea ooumamHmM m :o naouw mwammam Eowm :oeuonvoum ocloSum 9.2. 3 ”1:3 almOUZHu ~._~ m.~_ m.m m.~ a «unvn~_ kumvnwu numvnm. newcnm. nomvnm. onlmm nmlflm Owin— mutu— O—lm mln.~ n.~ln “toy :hmun “Jutcm K.®E ,_BnOH 3N31AH13 SBWON-N CHAPTER 2 ESTABLISHMENT OF FORAGE LEGUMES AS INFLUENCED BY DATE AND METHOD OF SEEDING AND FREEZING TEMPERATURES ABSTRACT Sod seedings of forage legumes are most successful when made in early spring with adequate moisture and suitable warm tempera- tures. Producers have been encouraged to broadcast legume seeds on frozen ground so freezing and thawing in early spring would aid in seed coverage. With these early seedings it is not known how many seeds germinate and are killed by subsequent freezing periods. Five field and one greenhouse/growth-chamber trial was con- ducted to determine the resistance of newly germinated alfalfa (Medficago sativa L.), red clover (Trifblium pretense L.), and birds- foot trefoil (Lotus corniculatus L.) seedlings to sub-zero tempera- tures. In field experiments, alfalfa and trefoil were seeded by three methods-broadcast and drilled in a sod compared to band seeding in a prepared seedbed (check) on three spring dates-—15 Mar., 15 Apr., and 15 May. Seedling counts and forage yields were used to evaluate stand establishment and freezing injury. Alfalfa stands were consistently poorer when broadcast-seeded 42 43 on 15 Mar. compared to 15 Apr. but trefoil seedings were equally good when made on the same dates. Freezing temperatures in late March and early April likely injured alfalfa but not trefoil. Alfalfa and trefoil seedings broadcast in mid-April were satisfactory and seedling density approached drilled seedings made on the same date. Yields were generally highest from mid-April band and drilled seedings, intermediate when broadcast on 15 Mar. and 15 Apr., and lowest when broadcast on 15 May. Trefoil yields from lS-May-drilled seedings were similar to March and April broadcast seedings. Under controlled conditions, alfalfa, red clover, and trefoil were hardened for 2, 4, or 6 days, and frozen for 1, 2, or 4 hours at -2.22, -3.33, -4.44, -5.56, or -6.67 C. Resistance to freezing injury was in the order of trefoil > red clover > alfalfa. Averaged freezing periods of 2.3 hours were lethal to 50% (LTSO) of alfalfa seedlings at -4.44 C, red clover at -5.39 C and trefoil at -6.67 C, corroborating freezing injury of seedlings in the field of alfalfa >> trefoil. Additional index words: Medficago sativa L., Trifblium pretense L., Lotus corniculatus L., sod seeding, frost seeding, cold resistance, stand depletion. INTRODUCTION Grassland production in humid climates is often limited by the supply of available nitrogen, but this may be overcome through the use of nitrogenous fertilizers or by growing legumes in associa- tion with the grasses. The practice of introducing legumes into an established grass sod has been commonly referred to as pasture renovation and is now also known as sod seeding. The advantages and problems associated with sod seeding have been studied by many researchers. As early as 1878 Roberts (1910) worked on pasture improvement at Cornell and later Graber (1928) introduced legumes into grass awards in Wisconsin. More recent studies have involved the use of chemicals to reduce grass competition (Blackmore, 1965; Sprague, 1960; Taylor et al., 1964; Tesar, 1980; Mueller-Warrant and Koch, 1980) and the evaluation of seeding methods for the best stand establishment (Dowling et al., 1971; Sund et al., 1966; Taylor et al., 1969). For sod seeding to be successful, high moisture is desirable at or soon after seeding (Decker et al., 1976; Holland, 1980; Tesar, 1980), consequently seedings are most likely to be successful if made at times of the year when rainfall is plentiful. Taylor and co-workers (1969) found that sod seedings made in early spring were the most successful, later summer ones were intermediate, and those made in mid-summer were poorest. The amount and distribution of 44 45 precipitation following seeding was judged by these researchers to affect germination and stand establishment more than most other factors. Early spring seedings have been shown to be the most suc- cessful and are recommended for maximum stand establishment (Tesar, 1980; Holland and Tesar, 1981). It has been a long-standing practice of farmers to broadcast clover seeds on snow or frozen ground in late winter or early spring into fall established winter wheat. Roberts (1910) maintained clovers in a cool-season grass pasture from 1878 to 1903 by sowing early in the spring every second year, up to 2 kg ha-1 of mixed clover seed. Evaluation of dates and methods of seeding alfalfa and red clover in wheat were begun in Ohio in 1928 by Willard (1934) and co-workers. These researchers stated that, "Alfalfa appeared not to be as sure as red clover to make a stand if broadcast in late February or early March because the seedlings were sometimes killed by later hard freezes." Similar observations have been made in Michigan, on early- spring sod seedings of alfalfa that were considerably poorer than later seedings (Holland and Tesar, 1980). Many studies have been conducted on freezing susceptibility and winter hardiness of field crops. Only a few researchers, how- ever, have studied the ability of forage legume seedlings to with- stand freezing temperatures. Results obtained by various investi- gators differ somewhat as to when seedlings are the most sensitive to freezing temperatures. White and Horner (1943) obtained 100% sus- ceptibility to freezing injury from unemerged winter-sown alfalfa, while Peltier and Tysdal (1932) concluded that five-day-old alfalfa seedlings are more resistant to freezing than ten-day-old seedlings. 46 Arakeri and Schmid (1949) found no injury to unemerged seedlings but they noted a sharp decline in resistance to freezing, from emergence to the three-to-four leaf stage, after which resistance gradually increased. There seems to be a greater consensus that cold hardiness of various legumes is minimal when they are forming the first trifoli- ate leaf (Steinbauer, 1926) and subsequently increases with age up to 60 days (Peltier and Tysdal, 1926). Unquestionably, because of favorable conditions of adequate moisture and increasingly warmer temperatures, spring is the ideal time to make seedings. Encouragement and advice have traditionally been given to farmers to broadcast legume seeds before the ground thaws, so early freezing and thawing will promote seed coverage. With these early seedings, it is not known how many seeds germinate and are then killed by subsequent freezing periods. The objectives of this study were: (1) to evaluate, under field conditions, if early-spring freezing temperatures are injurious to forage legume seedlings during establishment; and (2) to determine under controlled growth-chamber conditions which freezing temperatures kill forage legume seedlings during establishment. MATERIALS AND METHODS Section I—-Fie1d Studies Five forage sod-seeding evaluations were conducted at three locations of diverse early-spring temperatures. The three field locations provided a range of early spring temperatures ideally suited to testing freezing injury of forage legume seedlings (Table 1). Location 1 was in an area of considerably shorter growing season than either locations 2 or 3. Annual average temperatures of these locations are indicative of the range of earliness of growth in the spring. Growth at location 3 generally preceded that at loca- tion 2 by one to two weeks and that at location 1 by three to four weeks. Experiment 1. This study was conducted at the Lake City (LC) Experiment Station (location 1) at Lake City, Michigan, the most northerly site (44°18’N 85°12’W). The soil was an Iosco sandy loam (sandy over loamy, mixed, frigid Alfic Haplaquods) of pH 6.2. Plots were 1.8 x 7.6 m and all treatments were replicated three times in a randomized, complete block design. Alfalfa (Medicago sativa L.) was broadcast into the sod at 13.5 kg ha.1 on three different dates in l981-15 Mar., 15 Apr., and 15 May-and drilled into the sod on 15 Apr., and 15 May. Control plots were band seeded into a plowed 47 48 and tilled surface on 15 Apr. Band seeding on a prepared seedbed is considered the best method of establishing a legume stand (Tesar et al., 1954; Tesar and Jackobs, 1972) and was utilized as a check for comparing other methods of efficacy in stand establishment and production. In this and the other field experiments, actual seeding dates may have varied by two to four days but are reported as indi- cated for clarity of comparisons. All seedings were made by the same commercial grain-fertilizer drill with a small-seeded legume box adapted to provide precision seed setting to completely standardize seeding rates. The disk openers and coulters of the drill were retracted manually during the broadcast seeding. In the fall prior to establishing the study, NL(Phosponomethyl)glycine (glyphosate) was used to suppress the grasses. No fertilizer was added at seeding but recommended levels were applied annually. Stand density was deter- mined on 19 July in four, directed 35-cm-quadrat counts from each plot. Before each harvest percent forage species in each plot was estimated visually and used to calculate the legume portion of the total forage yield. Yields were obtained from an area of 0.9 x 6.7 m with a self-propelled, direct-chop harvester. A.l-kg forage sample from representative plots was dried with forced air at 65 C for 48 hours and used for dry matter determinations. Yields are expressed in t ha.1 of the legume portion of the total yield. Experiments 2 and 3. Both studies were conducted on the Crop Science Research Farm at East Lansing (EL), Michigan, (location 2) which was the most central site (42°42’N 84°28’W). Alfalfa at 13.5 and birdsfoot trefoil (Lotus corniculatus L.) at 7.0 kg ha"1 were 49 sod seeded into a Hillsdale sandy loam (coarse-loamy, mixed, mesic, Typic Hapludalf) soil of pH 5.8. The experimental design was a randomized complete block in a split-plot arrangement with four rep- licates for Exp. 2 and three for Exp. 3. Agricultural limestone was applied at 11.2 t ha.1 prior to initiating the studies and no fer- tilizer was added at seeding but recommended levels were applied annually. Experiment 2 was seeded in 1979 and Exp. 3 in 1980. Seed- ling counts were obtained in the year of seeding for Exp. 2 on 17 July and Exp. 3 on 10 July by the same procedure as described for Exp. 1. Seeding dates, methods of seeding, grass suppression, and methods of obtaining data were the same as for Exp. 1. Experiment 4. This study was conducted at East Lansing (loca- tion 2) on a Hillsdale sandy loam of pH 5.9 adjacent to and at the same time (1980) as Exp. 3. Alfalfa was broadcast on 15 Mar. and 15 Apr. and drilled into the sod on 15 Apr. Seedling counts were obtained on 3 July by four directed samplings from each plot, with all other treatments, statistical design, and methods of obtaining data the same as for Exp. 1. Experiment 5. This study was conducted at the Kellogg Bio- logical Station (KBS) near Battle Creek, Michigan (location 3), the most southerly location (42°24’N 85°24’W). Alfalfa was sod seeded into a glyphosate-treated sward on a Kalamazoo sandy loam (fine- loamy over sandy, mixed, mesic, Typic Hapludalfs) soil of pH 4.9. The experimental design was a randomized complete block in a split- split plot arrangement with four replicates. Seedings were made as 50 follows: broadcast on 15 Mar., drilled on 15 Apr., and band seeded on a prepared seedbed on 15 Apr. (check). Seedlings were counted on 10 July by the same directed sampling method used in Exp. 1. Seeding rates, fertilization, and methods of obtaining data, were the same as in the previous trials. Section II-—Greenhouse Study A greenhouse/growth-chamber trial was used to evaluate the specific effects of precisely controlled freezing temperatures on the killing of legume seedlings. Three forage legumes-—a1falfa, red clover (Trifblium pratense L.), and birdsfoot trefoil-dwere seeded in six replicates of 100 seeds each with 600 seeds per tray, and germi- nated in the greenhouse at diurnal temperatures of 18/24 C. The experimental design was a randomized complete block with a split- split plot arrangement. Seedings were made with a specially designed vacuum-operated seed head that placed 100 seeds 1 cm apart in 10 parallel rows 1 cm apart. A fungicide, Pantachloronitrobenzine (Terra-coat L025), was applied as a soil drench in a 1:400 ratio with water immediately after seeding to control seedling damping off diseases. All treatments were based on imposing freezing treatments on the legumes at the same morphological dicotyledonary stage which occurred at varying times after seeding as follows: alfalfa-5 days; red clover-7 days; and birdsfoot trefoil-9 days. The trays of seedlings in the dicotyledonary stage were placed in growth chambers for hardening periods of 2, 4, or 6 days 51 at diurnal temperatures of 2/4 C. Seedlings were counted at the end of each hardening period and transferred to a freezing chamber for five hours at -0.5 C. At the end of this period, the trays of seed- lings were sprayed with a super-fine mist of water to simulate field moisture conditions on the plant surface. The temperature was then lowered to one of five predetermined levels: -2.22, -3.33, -4.44, -5.56 or -6.67 C. The six replicates of legume seedlings were sub- jected to these freezing temperatures for periods of l, 2, or 4 hours. Freezing was conducted in darkness during the "night" period. After freezing, seedlings were maintained at 3 C for 12 hours and then returned to the greenhouse. Counts were made of the surviving seed- lings 48 hours later. This period was found necessary because ungerminated seeds, probably "hard" seeds, germinated after freezing and confounded counts made at longer periods after freezing. All data was analyzed using the arcsin transformation and are reported as percentages. RESULTS AND DISCUSSION Section-—I Field Studies Stand density was averaged separately for alfalfa for Exp. 1 to 5 and birdsfoot trefoil for Exp. 2 and 3 (Table 2) since a general pattern of performance of treatments was noted. The averaged stand densities for alfalfa and birdsfoot trefoil establishment were both ranked best, as expected (Tesar et al., 954), when band seeded into a prepared seedbed (Table 2). Band seeding, however, was better or equal to drilling the seed into the sod on the same date of 14 Apr. in all the experiments but was significantly better in only one-half of the experiments. Band seeding has been shown in many trials to produce superior stands (Tesar et al., 1954; Decker et al., 1976; Tesar and Jackobs, 1972) but data in these experiments show that in some cases with favorable rainfall and soil conditions, nearly as good stands can be obtained when drilling the seed into the sod. Broad- casting the seed on 15 Apr. was ranked third in stand establishment. It was never as good in any of the five experiments as band seeding on the same date. Drilling in the sod on 15 Apr. produced better stands than broadcasting the seed on the surface on the same date in only two of four alfalfa experiments and one of the two trefoil trials. Drilling alfalfa seed on 15 May produced, over all experiments, the fourth-best, and trefoil, the fifth-best stands. Seedings made on 52 53 this later date were all much poorer than similar seedings made one month earlier on 15 Apr. Reduced soil moisture was likely the pri- mary reason for the poorer stands obtained in these mid-May seedings. Broadcasting alfalfa and trefoil on 15 May produced the poorest stands which were significantly lower than drilled seedings made on the same date. These poorer seedings made in May indicate the impor- tance of seed placement when moisture is limiting, and that broad- casting seed at this late date should generally not be considered because of the likelihood of poor stands. Early broadcast seedings of alfalfa in mid-March produced the fifth-best stands from the various seeding methods on different dates. Trefoil stands seeded on 15 Mar. were fourth best. This species varia- tion was extremely significant as freezing temperatures were suspected to have killed seedlings of alfalfa but not of trefoil. Broadcast seedings of alfalfa made on 15 Apr. produced better stands at all loca- tions than similar seedings made on 15 Mar. There was no difference, however, in stand establishment of trefoil when broadcast on 15 Mar. or 15 Apr. Trefoil germinates and develops much more slowly than alfalfa and, even though germination had occurred at the time of the freezing temperatures, trefoil was shown to be more resistant to freezing than alfalfa. This is sub- stantiated by additional work conducted in the greenhouse and reported later in this study that shows trefoil is more resistant than alfalfa to freezing temperatures in the seedling stage. An examination of temperatures recorded at the experiment sites will help explain why stands of alfalfa were established more readily on one date compared to another. Broadcast seedings of alfalfa at East Lansing, location 2 54 (Exp. 2) in the spring of 1979 produced a better stand (64%) when broadcast on 15 Apr. than when seeded on 15 Mar. Figure 1 indicates two freezing periods (25 to 29 Mar., and 3 to 11 Apr.) when seedings made on 15 Mar. could have been killed. Temperatures dropped to ~10 C in the first and to -9 C in the second period. Time and temp- eratures were not sufficient for germination of the alfalfa seeds before the first freezing period in late March but germination had occurred prior to the freezing temperatures on 3 to 11 Apr. Birds- foot trefoil seedings in the same experiment (2) were not affected by these freezing temperatures because of a greater resistance to freezing injury. Seedings made a year later (1980) at the same location (EL), (Exp. 3 and 4) produced very similar results (Table 2) to those of the 1979 seeding (Exp. 2). Alfalfa stands were 72 and 86% better in Exp. 3 and 4, respectively, from the 15 Apr. broadcast method than from those broadcast on 15 Mar. During 13 to 17 Apr. temperatures dropped to almost -6 C at location 2 (EL) (Fig. 2). This was after temperatures had been high enough to permit germination of the alfalfa seeds and killing of the young seedlings may have resulted from this period of freezing. When seeded on 15 Mar., trefoil stands in Exp. 3 were not reduced as were the alfalfa stands. In Exp. 1 seeded at Lake City (location 1) in 1981, the alfalfa stand was 1372 better when broadcast on 15 Apr. than when broadcast on 15 Mar. (Table 2). Four periods of freezing temperatures (Fig. 3) ranging from -1 to -9 C during 5 to 6, 15 to l6, 18 to 22, and 24 to 27 Apr. were likely low enough to kill alfalfa seeded on 15 Mar. At the sOuthern-most location (3) at the Kellogg Biological 55 Station (KBS), alfalfa seeded in Exp. 5 showed the same trends of the effect of freezing temperatures after various seedings dates as in each of the other trials. Broadcasting alfalfa in mid-April pro- duced a 178% better stand than the mid-March broadcast seeding (Table 2). Trefoil stands, however, were not adversely affected when broadcast in mid-March. Temperatures (Fig. 4) indicated four periods of freezing (-l to -4 C) after the mid-March seeding date that could have killed the alfalfa seedlings. The magnitude of freezing required to significantly reduce a stand of alfalfa was not known at the time of these trials but subsequent work reported in this chapter shows that the freezing temperatures of -l to -4 C after seeding likely killed some of the alfalfa seedlings. The temperatures may not have killed the seedlings outright but may have had a weakening effect by the repeated freezing at the two-to three-day intervals from mid-to late April. Similarly to averaging stand densities for the various loca- tions because of similarities noted, yields from the year after seed- ing are also averaged for each legume. These yields (Table 3) showed no difference in dry matter production of alfalfa or trefoil between the broadcast mid-March and band, broadcast or drilled mid-April seedings, except in Exp. 5. This lack of yield differences was not unexpected even though stands were better from the mid-April seedings as work reported by Tesar (1978) showed that yields in the year after seeding from 32 alfalfa plants ha"1 (13. 9 t hafl) was almost equal to yields from 160 plants ha.1 (14.6 t ha-l). Bolger and Meyer (1983) reported on work in North Dakota that showed no difference in yield in the year after seeding from alfalfa stands ranging from 54 56 to 484 plants mfz. Other work has also been done demonstrating a yield plateau above certain stand density levels and a lack of sig- nificant yield increases proportional to higher stand density (Palmer and wynnAWilliams, 1976; Tysdal and Kiesselbach, 1939). Reduced competition to the quackgrass regrowth by the poorer stand of alfalfa in the March broadcast seeding (Exp. 5) may have contrib- uted to the lower yield. Broadcast seedings made in mid-May (Table 3) produced unsat- isfactorily low yields except in Exp. 3 at East Lansing where yields were equally as good as those from all the other seeding methods. Favorable rainfall (31 mm) two days after seeding on 15 May in Exp. 3, followed by a well-distributed, above-normal precipitation over the next six weeks, helped establish a stand of 55 plants mfz. This stand density has been shown by other researchers (Tesar, 1978; Bolger and Meyer, 1983) to be adequate for maximum yields. Precipi- tation was lower and not well distributed after May seedings in other years at each location showing that broadcast seeding in Mid-May is not a satisfactory method of stand establishment. When the seed was drilled into the sod in mid-May stands were more satisfactory, even though not as good as mid-April seedings, and produced yields almost as high as these earlier seedings. Section II-—Greenhouse Study The greenhouse trials produced a definite ranking of the three legume species indicating differences in cold resistance dependent on the length of hardening period and, more importantly, 57 on the freezing temperature. Differences in survival were only minor when the legumes were hardened for 2, 4, or 6 days (Tables 4, 5, 6) but some important differences were noted. One hour of freezing at the three lowest temperatures showed red clover to be as cold resis- tant as trefoil (Table 4). After two hours at these same tempera- tures, trefoil was the most cold resistant and red clover did not survive any better than alfalfa (Table 5). The distinct ranking of cold resistance between the legumes was clearly shown after four hours of freezing at -6.67 C (Table 6): trefoil > red clover > alfalfa. Two (Table 4) and four days (Table 5) of hardening pro- duced similar survival rates with slightly higher legume survival after six days (Table 6) of hardening. The data in Tables 4, 5, and 6 are averaged (Fig. 5), therefore, to show the effects of various lengths of freezing-1, 2, or 4 hours-at temperatures ranging from -2.22 to -6.67 C. This comparison is considered justifiable since it is likely, under actual field conditions, that the greatest vari— ance in seedling mortality would likely be related more to this com- bination of duration and degree of freezing, rather than to the length of hardening. Differences in legume resistance to freezing injury, averaged over the hardening periods (Fig. 5) show that at 1, 2, and 4 hours of sub-zero temperatures, significant differences among the legumes were first produced at -5.56, -4.44, and -3.33 C, respectively. At all temperatures lower than this, survival was consistent: trefoil > red clover > alfalfa. It is not known why red clover did not with- stand -5.56 C for two hours any better than alfalfa. Since all other data with higher and lower freezing temperatures indicated red clover 58 seedlings were more resistant to freezing than alfalfa, it is sug- gested that these data at -5.56 C, for some unknown reason, do not accurately represent differences at this level. When frozen for four hours at -2.22 C (Fig. 5), red clover and trefoil were more cold resistant than alfalfa. This indicated that alfalfa was also more adversely affected by the length of the freezing period than either red clover or trefoil. Only 32% of the alfalfa seedlings survived -3.33 C for four hours which was less than one-half the survival rate of red clover (732) or trefoil (85%). The averaged freezing period (1, 2, and 4 hours) data in Fig. 6 show that the three legumes had the following "cardinal" freezing temperatures at which 50% of the seedlings in the dicotyle- donary stage were killed (LT alfalfa -4.44 C (24 F); red clover so) ‘ -5.39 C (22.3 F); and trefoil -6.67 C (20 F). These data (Fig. 6) showing the LT5 sented in Section I of this chapter where mid-March seedings of 0 for each legume corroborate the field data pre- alfalfa were reduced by freezing temperatures but trefoil stands were not. Alfalfa seedings made in Michigan in mid-March will be injured and the stand depleted by likely freezing temperatures of -4.44 C, (24 F) that are common during this time, but seedings of red clover, and especially trefoil, are less likely to be injured. These two legumes which are more resistant to freezing than alfalfa in the seedling stage, should be established just as successfully if seeded in mid-March or mid-April. Broadcast seeding of alfalfa, then, would likely be more successful if made in early to mid-April to benefit from spring rains with a lower probability of freezing temperatures. 59 These data also substantiate why early-spring (Feb. to Mar.) broad- cast seedings of alfalfa in winter wheat are generally less sucessful than similar seedings of red clover because of the freezing tempera- tures encountered at this time. SUMMARY AND CONCLUSIONS Five field trials were conducted at three locations of diverse early-spring temperatures to evaluate the resistance to freezing of alfalfa and birdsfoot trefoil in the dicotyledonary stage. A greenhouse/growth-chamber study was used to evaluate, under controlled conditions, the precise temperatures at which seedlings of these two forage legumes and red clover were killed. Alfalfa stands were consistently poorer but trefoil seedings were equal when broadcast seeded in mid-march than in mid-April. Band seeding into a prepared seedbed produced the best stands over- all, of alfalfa and trefoil. Seedings drilled in the sod in mid- April, however, produced stands equal to band seedings in one-half the field experiments, showing this to be a satisfactory seeding option. Drilled seedings in mid-May were satisfactory but broadcast seedings made on 15 May were unsatisfactory in all but one experiment. Yields, obtained in the year after seeding, generally, did not reflect the differences in stand density unless the stands were very poor. Yields of alfalfa were best when seeded in mid-April, poorer when broadcast in mid-March or drilled in mid-May and poorest when broadcast in mid-May. Trefoil produced yields in a similar pattern to alfalfa, but mid-May drilled seedings also produced as good yields as mid-April seedings. It was concluded from these field and controlled temperature studies that: 60 61 l. Early-spring freezing temperatures in the field killed many alfalfa, but not trefoil, seedlings. 2. Alfalfa was least resistant to freezing temperatures, red clover was intermediate, and trefoil the most resistant. 3. After freezing for an average of 2.3 hours, the LTSO for alfalfa was -4.44, red clover -5.39, and trefoil -6.67 C. 4. Sod seedings of alfalfa are best made in Michigan in early to mid-April to reduce the likelihood of freezing injury to the seedlings. Red clover is more cold resistant and can be seeded two to three weeks earlier than alfalfa. Trefoil can be seeded equally as well in mid-March or mid—April. 5. Satisfactory stands and yields of birdsfoot trefoil can be obtained on herbicide-suppressed sods by the economical broad- casting of seeds earlier in the Spring than is possible to drill the seed into the sods. lO. LITERATURE CITED Arakeri, H.R., and A.R. Schmid. 1949. Cold resistance of various legumes and grasses in early stages of growth. Agron. J. 41:182-185. Blackmore, L.W. 1965. Chemical establishment and renovation of pastures in southern Hawkes Bay and northern wairarapa in New Zealand. 1:310-312. Proc. 9th Int. Grassl. Congr. Sao Paulo, Brazil. Bolger, T.P., and D.W. Meyer. 1983. Influence of plant density on alfalfa yield and quality. p. 37-44. Amer. Forage Grassl. Council. Eau Claire, Wis. Decker, M.A., T.H. Taylor, and C.J. Willard. 1976. Establish- ment of new seedings. In.M;E. Heath, D.S. Metcalfe and R.F. Barnes (eds.). Forages, the science of grassland agriculture. Iowa State Univ. Press. Ames, Iowa. Dowling, P.M., R.J. Clements, and J.R. McWilliams. 1971. Establishment and survival of pasture species from seed sown on the soil surface. Aust. J. Agric. Res. 22:61-64. Graber, L.F. 1928. Evidence and observations on establishing sweet clover in permanent pastures. J. Amer. Soc. Agron. 21:1197-1202. Holland, Clive. 1980. Establishment of alfalfa (Medficago sativa L.) and birdsfoot trefoil (Lotus corniculatus L.) in various grass sods as affected by date and method of seeding, and herbicide application. M.S. Thesis. Mich. State Univ. East Lansing, Mich. , and M.B. Tesar. 1980. Unpublished observations on freezing injury to early spring seeded alfalfa. Mich. State Univ. East Lansing, Mich. , and . 1981. Establishment of alfalfa (Medicago sativa L.) in quackgrass sods with herbicides using conventional and sod seeding methods. Mich. State Univ. Res. Rep. 420:9-18. Mueller-warrant, G.W., and D.W. Koch. 1980. Establishment of alfalfa by conventional and minimum-tillage seeding techniques in a quackgrass-dominant sward. Agron. J. 72:884-889. 62 11. 12. 13. 14. 15. 16. l7. 18. 19. 20. 21. 22. 23. 63 Palmer, T.P., and R.B. Wynn-Williams. 1976. Relationship between density and yield of lucerne. New Zealand J. Exp. Peltier, G.L., and E.M. Tysdal. 1932. A method for the determination of comparative hardiness in seedling alfalfas by controlled hardening and artificial freezing. J. Agric. Res. 44:429-444. Roberts, I.P. 1910. The Roberts pasture. p. 505-511. In Pastures in New York. Cornell Agric. Exp. Sta. Bull. 280. Sprague, MgA. 1960. Seedbed preparation and improvement for unplowable pastures using herbicides. p. 264-266. Proc. 8th Int. Grassl. Cong. Reading, England. Steinbauer, G. 1926. Differences in resistance to low temperatures shown by clover varieties. Plant Physiol. 1:281-286. Steponkus, F.L. 1978. Cold hardiness and freezing injury of agronomic crops. Advan. Agron. 30:51-98. Sund, J.M., G.P. Bairington, and J.M. Scholl. 1966. Methods and depths of sowing forage grasses and legumes. p. 319-323. Proc. 11th Int. Grassl. Cong. Queensland, Aust. Taylor, T.H., J.M. England, R.E. Powell, J.F. Freeman, C.K. Cline, and W.C. Templeton, Jr. 1964. Establishment of legumes in old Pba pratensis L. sod by use of paraquat and strip tillage for seedbed preparation. Proc. 7th Brit. Weed Control Conf. 2:792-803. , J.S. Foote, J.H. Snyder, E.M. Smith, and W.C. Temple- ton, Jr. 1972. Legume seedling stands resulting from.winter and spring sowings in Kentucky bluegrass. Agron. J. 64:535-538. , E.M. Smith, and W.C. Templeton, Jr. 1969. Use of minimum tillage and herbicide for establishing legumes in Kentucky bluegrass (Pba pratensis L.). Agron. J. 61:761-765. Tesar, M.B. 1978. Yield and persistance of alfalfa (Msdflcago sativa L.) as affected by rate and date of seeding and annual fertilization. Mich. State Univ. Agric. Exp. Stn Res. Rep. 353:5-12. . 1980. Sod seeding birdsfoot trefoil and alfalfa. Mich. State Univ. Ext. Bull. E-956. , K. Lawton, and B. Kawin. 1954. Comparison of band seeding and other methods of seeding legumes. Agron. J. 46:189-194. 24. 25. 26. 27. 28. 64 , and J.A. Jackobs. 1972. Establishing the stand. p. 415-435. In C.H. Hanson (ed.) Alfalfa science and technology. Amer. Soc. Agron. Madison, Wis. Tysdal, J.M.. and T.A. Kiesselbach. 1939. Alfalfa nursery technic. J. Amer. Soc. Agron. 31:83-98. , and A.J. Pieters. 1934. Cold resistance of three species of lespedeza compared to that of alfalfa, red clover, and crown vetch. J. Amer. Soc. Agron. 26:293-298. White, W.J., and W.H. Horner. 1943. The winter survival of grass and legume plants in fall sown plots. Sci. Agric. 23:399-408. Willard, C.J., L.E. Thatcher, and J.S. Cutler. 1934. Alfalfa in Ohio. Ohio Agric. Exp. Sta. Bull. 540. 65 Table l. Climatological data from the National Oceanic and Atmo- spheric Administration for the three experimental locations in Michigan. Annual temperatures C Frost-free Location days Max. Min. Average 2-East Lansing 151 33.3 -24.4 7.6 3-Kellogg 159 33.3 -25.5 9.3 66 .Gvumouooe ou Qvumon wo $50“ch .Hwoa ow m was H momma ea c was m “mnma aw powwow N usoaaummxm + S on Na 8 S on mm 8.093 Amv sq mm mm Aev cm I I we we we vow voaawuv mm: ma so m u n 3v 2 .. u mm 3 H e8 “3385 .3: 2 AHV NNH «ma me Aav mod «mm I NwH noH oma onHAu noon .ue4 ma ANV nm em om Auv «ma mma qu mHH flea omH pom poaawuv .ue< ma any we Hm no Amv ONH mNH mud nHH mNH woa vow unmovmouo .u@< ma Aev Ho mm no HAmv am me me we as mq pom unmovwouo .umz ma Hwomoua.llllll mwamwa< «La mmafiHvoom owmuo>< m N owmuo>< m c m N .n oomwusm ponuoz sumo uomaaumnxu wowvoom AMIN mmMIm AMIN UAIH coaumooa +.meH no .omma .anma madman mouse udoumwmap oops» co msoaumooa moans um moaswoa owmuom wcfivomm umumm mzuooa manna huamcov vcmum .N wanna .onumouoomnou Aavumon mo wafixommuu .Nme cu m was H .mea a“ q can m momma aw woumo>ums N uooaauonxm 67 + m.~ n.a m.o m2 m2 m.a m.~ no cams Amy m.m m.e m.m Amv m.m . u H.~H a.m m.m eon emaflaue an: ma Ass «.0 u ~.o Ase o.s u . m.HH w.H H.o com ummuemoun an: we adv o.m o.m o.m ANV a.oH e.HH s w.HH m.HH N.“ amass“ vamp .ua< ma ANV m.w m.m o.m fiflv H.HH a.oa o.mH ~.~H w.HH N.h com emafiaue .ua< ma Aev m.m m.s «.0 Amy o.oH o.oH m.~a o.NH o.w m.o eon ammoemoun .uaa ma Amy H.m a.m ~.e Hfisv m.m m.m m.mH o.NH H.a s.m eon ammuemoun .umz ma Haomuua moammaa m u H: ; mmmuo>¢ m N owmuo>< n q n N .H momuusm eonumz puma uaoafluoaxm , wawvoom amum mmeum gala cola ooaumoog +.w:fipoom Houmm pooh onu ca mumo>ums mounu Scum moasmma mwmuom mo vamaw .m manna 68 Table 4. Percent survival of three forage legumes after two days of hardening and 1, 2, or 4 hours of freezing. Temperature C -2022 -3033 -4044 -5056 -6o67 One hour freezing Alfalfa 100 79 a 73 a 60 a 61 a Red clover 100 94 b 97 b 74 a 72 a Trefoil 100 96 b 95 b 87 b 84 b Two hours freezing Alfalfa 98 a 48 a 42 a 18 a 16 a Red clover 99 a 90 b 56 b 27 ab 28 b Trefoil 100<2 91 b 67 b 35 b 46 b Four hours freezing Alfalfa 91 a 26 a 18 a 3 a 4 a Red clover 98 b 72 b 28 ab 14 b 12 b Trefoil 10019 80 b 37 b 18 b 18 b Means followed by the same letter within columns and freezing periods do not differ at the 5% level of probability according to Duncan's multiple range test. 69 Table 5. Percent survival of three forage legumes after four days of hardening and l, 2, or 4 hours of freezing. Temperature C -2022 -3033 -4044 -5056 -06067 One hour freezing Alfalfa 100 87 a 77 a 61 a 60 a Red clover 100 99 b 95 b 82 b 79 b Trefoil 100 100 b 98 b 90 b 86 b Two hours freezing Alfalfa 100 a 69 a 43 a 23 a 17 a Red clover 98 a 93 b 68 b 26 a 30 b Trefoil 100 a 96 b 70 b 44 b 39 b Four hours freezing Alfalfa 93 a 33 a 22 a 8 a 9 a Red clover 100 b 71 b 29 ab 12 ab 13 ab Trefoil 100 b 81 b 35 b 21 b 17 b Means followed by the same letter within columns and freezing periods do not differ at the 5% level of probability according to Duncan's multiple range test. 70 Table 6. Percent survival of three forage legumes after six days of hardening and l, 2, or 4 hours of freezing. Temperature C -2.22 -3.33 —4.44 -5.56 -6.67 One hour freezing Alfalfa 100 100 80 a 77 a 74 a Red clover 100 100 97 b 90 b 87 b Trefoil 100 100 98 b 92 b 91 b Two hours freezing Alfalfa 100 96 a 69 a 34 a 27 a Red clover 100 93 a 74 a 36 a 39 ab Trefoil 100 98 a 91 b 63 b 43 b Four hours freezing Alfalfa 99 a 38 a 25 a 10 a 9 a Red clover 100 a 77 b 35 a 20 a 18 b Trefoil 100 a 95 c 85 b 42 b 29 c Means followed by the same letter within columns and freezing periods do not differ at the 5% level of probability according to Duncan's multiple range test. .Awofimamg ummmv N oofiumoog um mmma Mom mouaumuomawu asaficfia was asaaxma haamw magnum manna pmuooaom .H .wfim JHmmm 10mm: 0m mm m— : v mm _N @— 71 I i I - .. ZHZ >1.ch l l"||l’ 'lll II"I| ‘ I xmz >4H¢n . ~_ (3) dlnlll. 72 .Awafimcmq ummmv N coaumooq up owma mom mousumuonaou auafiafia was asaaxma mHHmo wcuuem haumo wouooaom JHmnE Iummz an mm m— 2 V mm —N mu # 1 q 1 — a q 2H: >1.ch II" II ' II III I .III . 9 'II'II A J n * XI! >JH¢D J .N .5 NT mI a m m N d m. ) 3 a. vm an 73 .A%uHU omev H aoHumooH on meH How monoumuomaou EsaHoHa was asamee mHHmv wsHuam %Humo wouooHom JHmmm 10mm: an mm mu __ v mm _N mg q a 1 d - q q 2H: >4Hmn . T XE: >JHED . .m.3& NT ml. 9 m u. N d n: ) 3 2 cm an 74 .AwwOHHon m coHumoOH um HmmH you monououooamu asEHcHa was asamea hHHmv wcHuom aHumm wouomHom .e .me JHmnE 10mm: an mm mg 2 v mm _N mu 1| J 1 q d 4 u Nul- H ZHZ >JH¢D 4 ml i II III I- ..II ||||| I I'll' III' IIIII a . m m .. N ( a 4 N— J t 3 4 OH xmt >JH¢D .. em 4 an PERCENT SURVIVHL PERCENT SURVIVFIL 1 10 108 so ea 79 so so 49 39 as no I 18 108 90 BB 70 BB 50 48 30 20 19 75 TEMP (C) a a n a a l-FILFFILF’H A ._ Z-RED CLO 17- F"! ii 3-TREFOIL a C c H 1 _ s a '1 ,1 a a "'1 r— 1 a 3 1 a 3 a 2 3 1 a 3 l 2 3 -2.22 -3.33 -4.44 -5.56 -6.67 TEMP (C) a a a 7__ a a l-FILFFILFFI _ a-RED CLOVER c 3-TREF'OIL a 7" s r .1 a —- s c a P— a — n 1 z 3 l a 3 1 2 3 1 2 3 n a 3 -2.22 -3.33 -4.44 -5.55 -s.s7 1 1o 100 so so 7o so so 40 3o PERCENT SURVIVHL 29 18 76 s a RF)— l-FILFFILFFI r c 2-Rsu CLOVER '1 a 3-TREFOIL r. E 9 s '1 -' C a c 3 o H H 123 123 #123 123 123 -2.22 -3.33 -4.44 -5.56 -6.6? TEHP (C) Fig. 5. Percent survival of three forage legumes after being frozen for one (opposite top), two (opposite lower), and four hours (above) and averaged over the three hardening periods of two, four and six days. Dis-similar letters at the top of columns within each temperature indicate differences between species at the 5% level of probability according to Duncan's multiple range test. 77 . 35. e was .c .Nv muoHuon wchovuos mou£u can Amuaos c was .N .Hv moaHu wsHuomum mouSu Ho>o vowmuo>m moason owmuom manna mo Hm>H>usm unmouom “UV aim... mm.ml mm.ml ¢¢.¢I nm.ml NN.NI . Each: on! . .040 cum T .. . datum» IIIII II IIIIIIIIIIIIIIII ones 1 1 .o .wE o a" an d an 3 m“ at 3 N am 1 5 am 0 mm on I mm _Um_ HI am an— a: CHAPTER 3 SIMULATED GRAZING COMPARED TO HERBICIDE SUPPRESSION OF GRASS COMPETITION WHEN SOD SEEDING FORAGE LEGUMES ABSTRACT Grass pastures are more productive when the sward contains a nitrogen fixing legume. Complete tillage and reseeding, or sod seed- ing and suppression of existing foliage with herbicides, have been two very successful methods for improving pastures by introducing legumes. Both methods require specialized equipment and costly inputs such as fossil energy and chemical herbicides. Many of the grassland areas in need of improvement are operated by marginal producers and livestock farmers who do not own the necessary equipment or cannot afford the higher costs involved in chemical renovation. This study was designed to (1) compare mechanical defoliation, simulating grazing, of grass sods with an accepted herbicide to sup- press competition to sod seeded forage legumes and; (2) determine stand establishment and yields from seed broadcast on the surface in early spring or drilled into the sod four weeks later. Red clover (Trifblium pratense L.) and birdsfoot trefoil (Lotus corniculatus L.) seed was broadcast on 26 Mar. and drilled on 78 79 21 Apr. into bromegrass (Bromus inenmls L.), reed canarygrass (Phalaris arundinacea L.), orchardgrass (Lactylis gZomerata L.), and quackgrass (Agropyron rspens L.) sods. Competition from these grasses defoliated 0, 2, or 4 times during the seeding year was come pared to suppression by the herbicide 3,5-dichloro-NL(1,l-dfimethyZ-Z- propynyl)benzamide (pronamide) applied in the fall prior to seeding. Defoliation dates were 15 May and 15 July for the two- and four-cut series, with additional cuts made on 5 June and 26 Aug. for the four- cut treatment. Seedling counts obtained 3 and 15 months after seeding and legume yields from the first and second years after seeding were used to evaluate treatments. Suppression of the grasses by cutting or with a herbicide was essential for legume establishment and good yields. Four cuts of the sown grasses improved red clover and trefoil establishment more con- sistently than herbicide suppression. Herbicide control of vigorous quackgrass was more successful than the defoliation treatments. Less vigorous quackgrass was suppressed equally well by 4 cuts or the herbicide. Early (26 Mar.) broadcast and later (21 Apr.) drilled seedings of trefoil were equally good. Red clover stands were best when drilled into quackgrass sods but equally good from both seeding methods in sown grasses. Year-after-seeding stand density indicated no long-term.advantage of drilling seed in the sod, compared to broadcasting it on the surface. Additional index words: Pasture improvement, grass suppression, defo- liation, frost seeding, Trifoliwn pmtense L., Lotus comiculatus L. INTRODUCTION Introducing and maintaining legumes in grass swards has been tried in a variety of ways, from broadcasting seeds on the surface in the winter (Roberts, 1910; Dowling et al., 1971), herbicide appli- cation to the sod to reduce competition (Sprague, 1952; Triplett et al., 1975; Van Keuren and Triplett, 1970; Tesar, 1976); strip tillage (Taylor et al., 1969; Decker et al., 1964); improved fertil- ization (Baker, 1980; Decker et al., 1969); and the use of special- ized seeding equipment (Ackley, 1975; Harris, 1974; Olsen et al., 1981). Partial tillage with disk harrows and field cultivators has been utilized on sods to obtain a suitable seedbed (Smith et al., 1973) and lime coating of the introduced seeds has been used success- fully in drier environments (Dowling et al., 1971; White, 1970). The availability of moisture to the introduced seedlings is perhaps the most crucial factor in the success of all sod seedings (Suckling, 1976; Dowling et al., 1971; Taylor et al., 1969). Almost as impor- tant is the reduction of competition from the existing sward to the inter-seeded species. This has been carried out in many different ways such as close grazing, burning, disking, field cultivating, and the use of herbicides (Graber, 1927; Sprague, 1960; Cullen, 1966). Moisture is not a controlled input but seedings can be made at a time when it is most abundant. Competition to introduced seedlings is con- trollable but is also costly in time, equipment and/or chemicals. 80 81 Grasslands most often involve herbivore animals that could be uti- lized in the reduction of plant inter-species competition by grazing the sod-seeded areas. As early as 1878, Roberts (1910) utilized dairy cattle to graze a grass pasture inter-seeded with a mixture of red and alsike clover seeds. He observed that one mechanical mowing was necessary to remove rank growth of the grasses in late June after which a good clover stand was produced. After the clovers were well established the pastured area carried three times as many cattle as the average pastures of New York state. Graber (1927) suggested that poor pasture production was often the result of overgrazing. He maintained that deferred grazing was necessary to allow regrowth of plants to replen- ish food reserves in storage organs if the pasture vigor and produc- tivity were to be maintained. Severe overgrazing without added fertilizer resulted in heavy weed infestation in bluegrass pastures (Fuelleman and Graber, 1938). They listed 47 weed species in the pasture and with pasture renovation and improved grazing management they markedly reduced the weed population. Burcalow et a1. (1940) showed that without judicious management, the duration and value of pasture improvement was greatly reduced. Moderate grazing of reno- vated pastures resulted in better legume persistence and fewer weeds than excessively grazed renovated pastures. While acknowledging grazing as a common practice in aiding renovation of pastures, Ahlgren et a1. (1940) successfully sod-seeded pastures with sweet and red clover without any summer defoliation in the year of seeding. Love (1944) compared grazing by sheep and mechanical mowing of species to reduce competition in stand establishment and concluded that the 82 timing and management of the grazing produced a vigorous stand of perennials. Mowing was not managed on a similar schedule to the grazing and resulted in poorer stands. In areas of steep hillsides, mowing is not possible and grazing of garss competition with sheep or cattle is the only prac- tical method of reducing competition to introduced legumes. On New Zealand hill lands, close, frequent grazing generally proved superior to infrequent grazing for grass germination and survival, but it seldom aided clover germination although clover survival was improved (Cullen, 1970). Robinson and Cross (1960) found grazing of permanent-pastured hill lands beneficial during early establish- ment with sod seeding. Suckling (1976) found no difference in the establishment of red clover, subterranean clover, and lotus major, subjected after seeding to three grazing regimes of continuous grazing, and variable periods of rotational grazings. He did find, however, that white and red clover were established better on sods grazed before seeding to l to 2 cm than pasture 5 to 10 cm long. Close grazing or clipping is essential for successful legume stand establishment if other suppression methods are not used. Barn- hart and Wedin (1981) found that it was necessary to clip a brome- grass sward every two to three weeks when sod seeding trefoil as light penetration of the canopy was reduced by over 50% within 20 days following clipping. Decker and Dudly (1976) concluded that complete suppression of a grass award with herbicides was neither necessary or desirable as severe weed invasions often occurred. Inter-seeding forage legumes into a sod and reducing grass competi- tion with grazing or clipping would prevent weed infestations and 83 maintain greater productivity than that obtained with a complete herbi- cide kill. Taylor and Allinson (1983) established alfalfa and trefoil in various grass sods by reducing competition with several clippings in the year of seeding. Livestock producers could, with judicious management, continue to graze sod-seeded pastures without interruption to productivity while at the same time reducing competition to intro- duced legumes. Costly inputs and production interruptions are con- sidered major drawbacks in producer acceptance of pasture improvement by sod seeding. This study was designed to (1) determine how successful reno- vation of low-producing sown grass and quackgrass pastures would be by broadcasting seed compared to using specialized, expensive machinery in sod seeding; and (2) compare several levels of mechani- cally simulated grazing with a recommended herbicide in reducing grass competition to introduced seedlings when sod seeding forage legumes in sown grasses or quackgrass sods. MATERIALS AND METHODS Four field experiments were conducted on two different soil types in the same general area at the Lake City Experiment Station, Michigan, in three different sown grass swards and two predominately quackgrass (Agropyron repens L.) sods. Two legumes, 'Arlington' red clover (Trifblium pratense L.) and 'Viking' birdsfoot trefoil (Lotus corniculatus L.) were sod seeded on two dates into grass swards sup- pressed before seeding by a herbicide, or by mowing at various fre— quencies after seeding to reduce grass competition. The use of herbicides is a well established and approved method of reducing grass competition when sod seeding (Sprague, 1960; Triplett et al., 1975; Decker and Dudly, 1976; Tesar, 1980). The two methods of sod seeding were: 1. Drilling seed into the sod at the earliest practicable date on 21 Apr. with a commercial grain-fertilizer drill with a small-seeded legume box adapted to provide precision seed setting. 2. Broadcasting seed on the sod surface on 26 Mar. to maxi- mize the benefit of early spring thaws and rains in the germination of the uncovered seeds. The same commercial drill was used for both seeding methods so as to completely standardize the seeding rates. The disk openers and coulters of the drill were retracted manually during the broadcast operation. The four methods of grass suppression were: 84 85 1. No cutting in the year of seeding. 2. Two cuttings, to simulate rotational grazing, made on 15 May and 15 July after seeding. 3. Four cuttings, to simulate more frequent rotational grazing, made on 15 May, 5 June, 15 July, and 26 Aug. 4. Herbicide suppression of the grasses with 3,5-dichloro- NL(1,l-dimethyZ-Z-propynyl)benzamidb (pronamide) which was fall applied at 1.7 kg ha"1 in 1979 to selected plots within each grass species. This herbicide was utilized as a check treatment and sup- presses grasses, except orchardgrass, adequately for legume estab- lishment when sod seeding (Holland, 1980; Triplett et al., 1977). Experimental design for the four experiments was a split- split block replicated four times for Exp. 1 and 2 and three times for Exp. 3 and 4. Recommended rates of fertilizer were applied on the surface at seeding and during each subsequent year of the trials. Other differences among the experiments were as follows: Experiment 1. This study was located on a Kent silt loam (Fine, illitic Typic Eutroboralfs) of pH 5.3 with well established bromegrass (BC) (Bromus inermis L.), reed canarygrass (RCG) (Phalaris arundinacea L.), and orchardgrass (0G) (Dootylis gZomerata L.) in adjacent blocks. Red clover was seeded into these grasses at 13.5 kg ha-l. Experiment 2. Birdsfoot trefoil was seeded at 6.7 kg ha-1 into reed canarygrass and orchardgrass sods adjacent to Exp. 1. Experiment 3. Red clover and birdsfoot trefoil were seeded into an 80% quackgrass sod in an area adjacent to Exp. 1 and 2 on a soil of pH 6.3. 86 Experiment 4. This trial was established on an Iosco loamy sand (sandy over loamy, mixed, frigid Alfic Haplaquods) of pH 5.7. Red clover and birdsfoot trefoil were seeded into the 90% quackgrass sod as in Exp. 3, except that flooding prevented the broadcast seed- ings of trefoil from being made until the later date of the seedings made by drilling in the sod. Data were obtained from all the experiments in an identical manner on the same dates. Stand density was determined on 15 July by counting legume seedlings in four directed, 35-cm quadrat samplings from each plot. A permanent lS-cm stake placed in the soil at ground level in the northeast corner of each quadrat sampled enabled precise counts to be made on plant survival by counting the same area 12 months later. Two harvests were made in each of the first and second years after seeding. Prior to each harvest the percent legume in each plot was estimated visually and used to calculate pure legume yield. Har- vests were made from a 0.9 x 8.2 m area with a self—propelled, direct- chop harvester. A l-kg forage sample from selected plots was dried with forced air at 65 C for 48 hours and used to determine dry matter. Yield data are reported in dry matter of pure legume and total (grass plus legume) yield as an average of seeding dates and methods. RESULTS AND DISCUSSION Stands Grass suppression was vital for seedling establishment and yields when sod seeding red clover or birdsfoot trefoil into sown grasses (BG, RCG, 0G) or a weedy quackgrass sod. Four cuts were as effective as the herbicide pronamide in suppressing the sown grasses for good seeding-year stands of both legumes (Tables 1 and 3). Two cuts were less effective but better than 0 cuts. Pronamide was more effective than the cutting treatments, however, in suppressing quack- grass on the droughty soil in Exp. 4 (Tables 7 and 8). In comparison to the sown grasses, quackgrass is a noxious perennial weed with aggressive rhizomatous growth making it difficult to eradicate. Quackgrass competition to the introduced legumes was not reduced adequately by 2 cuts for better stand establishment and yields. Legume stands in the year after seeding of 26-Mar.-broadcast seedings were equally good as 21-Apr.-drilled seedings into the sown grasses (Tables 1 and 3). These year—after-seeding counts were obtained at a time that reflected more accurately the complete grass suppression treatments than the year-of-seeding counts. All seeding- year counts were made three months after seeding, after only 1 and 2 cuts, for the 2- and 4-cut treatments, respectively. After only two of the scheduled four cuttings during the 87 88 seeding year, better red clover stands were established in all sown grass sods, compared to no grass suppression. Better seeding-year red clover stands were also obtained when competition to the grasses was reduced by cutting twice, compared to only once, except with drilled seedings in bromegrass. Seeding—year drilled seedings of red clover were consistently better than broadcast seedings in sown grasses in reed canarygrass sods only. various other treatments were benefited initially by drilling in sown grasses but stand decline measured in the year after seeding indicated no long-term advantage of drilling the seed in the sod, compared to broadcasting it on the surface. Seeding-year stands of red clover in Exp. 3 were better when drilled into quackgrass sods compared to broadcasting the seed on the surface. This stand advan- tage was still evident a year later in the 4-cut and herbicide- suppression treatments (Table 5). Trefoil seedings in all experi- ments were equally good when the seed was drilled in the sod or broadcast on the surface (Tables 3, 5, and 7). The early broadcast seeding of 26 Mar. was made with better moisture and provided a greater growth advantage towards the com— petitive early grass growth. Drilled seedings were made four weeks later on 21 Apr. when grasses were beginning to grow and the advan- tage of seed placement in the soil was reduced by greater grass competition. These findings are consistent with previous work showb ing the lack of differences in stand establishment between early (31 Mar.) broadcast and later (15 Apr.) drilled seedings of trefoil (Holland, 1980). Red clover benefited from the early-spring rains equally as 89 well as trefoil but the poorer stands of the 26 Mar. seedings than of the later 21 Apr. drilled seedings were likely due to freezing injury to the red clover seedlings. Trefoil stands were similar from both seedings dates which is consistent with data (Holland and Tesar, 1983) showing that trefoil will tolerate freezing temperatures likely to be encountered in March more readily than red clover. Red clover stands were not expected to decline sharply in the year after seeding as stand depletion of this species is not usually noticeable until the third growing season. Averaged over seeding and suppression treatments, the stand decline was 52% over all sown grass species (Table 1) with a decline of 36, 57, and 61% for seedings in bromegrass, reed canarygrass and orchardgrass sods, respectively. This relative percentage decline was a clear indicator of the differ- ing competitive growth of the grass species toward the introduced clover seedlings. Averaged trefoil stand decline in the same period was 33% over all sown species with a decline of 32 and 33% for reed canarygrass and orchardgrass sods, respectively (Table 3). Red clover stands declined 64% when seeded into quackgrass sods, which was a 12% greater decline than from seedings in sown grasses (52%). Red clover declined 65% in Exp. 3 (Table 5) and 63% in Exp. 4 (Table 6). Trefoil stand decline of 24% in quackgrass was 9% less than the stand decline in sown grasses (33%). Trefoil declined 16% in Exp. 3 (Table 5) and 33% in Exp. 4 (Table 7). Yields Yield of red clover in the year after seeding (1981) was 9O equally good when the sown grasses were suppressed with a herbicide or by 4 cuts (Table 2). Legume yield averaged over seeding methods and grasses clearly showed the effectiveness of the suppression treatments. Compared to O cuts, 564 and 755% more red clover was produced when the grass competition was reduced by 2 and 4 cuts, respectively. These yields showed much more accurately the differ- ences between the grass suppression treatments than did the seeding year counts. Reduced competition from the grasses produced larger, more vigorous legume plants and consequently higher yields. Average red clover yields in the second year after seeding (1982) were 48% better from the herbicide treated swards (3.7 t ha-l) than in the 4-cut suppression treatment (2.5 t ha-l). The yield difference in the first year after seeding was only 5% better with the herbicide treatment but total forage (grass plus red clover) produced in both years from either treatment did not differ significantly (Table 2). This reduction in yield in the third year of growth provides strong evidence for the recommendation of seeding red clover, a short-lived perennial, into pastures every second year in order to maintain a satisfactory stand. Herbicide suppression of a quackgrass sod (Exp. 3) which was less dense than in Exp. 4 did not increase red clover yields more than the cutting treatments (Table 6) but it did reduce the greater quackgrass competition in Exp. 4 to produce substantially higher yields (Table 8). Moisture was not limiting on the heavier soil of Exp. 3 but, on the extremely drought-prone, coarse-textured soil of Exp. 4 grasses suppressed with 4 cuts competed for water at the expense of high legume yields. In this trial, both broadcast and 91 drilled trefoil seedings were made on the same date because melting snow caused flooding along one edge of the experimental area during the earlier date. Even though the land slope was less than 2%, greater soil moisture retention and subsequent availability to the trefoil seedlings was suspected to have helped produce stands and yields that were equally good by each seeding method. Average dry matter yield of trefoil in sown grasses was low (2.7 t ha-l) in the year after seeding, but more than doubled the following year (5.5 t ha-l) (Table 4). This is in contrast to red clover which declined 48% in yield over the same period as it is less perennial in nature than trefoil. Herbicide suppression helped pro- duce better yields of trefoil than 0 or 2 cuts in all cases except in broadcast seedings into an orchardgrass sod. The reduced effec— tiveness of pronamide on orchardgrass suppression has been shown (Holland, 1980; Triplett et al., 1977) and did not adequately suppress this grass for a high legume yield. A distinct two-year yield advantage for trefoil was obtained by reducing the grass compe- tition by cutting twice (8.9 t ha-l) compared to no cuts (4.6 t ha-l). Trefoil yield increases obtained from suppression treatments of the sown grasses in comparison to 0 cuts were similar: 2 cuts—-l93%; 4 cuts-207%; and herbicide-230%. Legume yields were excellent from all suppression treatments when seeded in a quackgrass sod on the fine-textured soil of Exp. 3 (Table 6). At the initiation of this study, 80% of the sod was quackgrass. Even with this high quackgrass content, the sward was less vigorous than anticipated and much more easily suppressed by a herbicide or cutting than similar stands at other locations. This 92 was demonstrated by the lack of yield differences obtained between 2 and 4 outs and zero suppression treatments in both 1981 and 1982. Yields from seedings made in the coarse-textured soil of Exp. 4 bene- fited much more by complete sod suppression with a herbicide (Table 8). Two cuts did not suppress the quackgrass sufficiently more than 0 cuts to produce better yields. Red clover yields were 264 and 407% higher from the 4-cut and herbicide-suppression treatments, respectively, when compared to 0 cuts. For the same treatments, trefoil yields were 356 and 511% better than with 0 cuts. Yield decline of both legumes in the second year after seeding was attributed to the droughty soil conditions of this site. As a lowbcost method of pasture improvement with sown grasses, broadcasting red clover or birdsfoot trefoil seed on the surface in early spring and reducing competition during the seeding year by adequate mowings or rotational grazing of the grasses is a very satisfactory method for stand establishment and subsequent good yields. With greater competition from a quackgrass sod than from any sown grasses, the need for better, more complete suppression increases. Vigorous quackgrass swards were suppressed more effectively by pronamide than by the most frequent defoliation treatment of 4 cuts. Suppression by cutting particularly in more droughty soils of a quackgrass sod, is not adequate for good establishment of red clover or birdsfoot trefoil. SUMMARY AND CONCLUSIONS Four field experiments were conducted by sod seeding red clover and birdsfoot trefoil into bromegrass, reed canarygrass, orchardgrass, and quackgrass swards. Seedings were made by broad- casting the seed on the surface on 26 Mar. or drilling it into the sod on 21 Apr. Competition from.the grasses to the introduced legume seeds was reduced by the fall-applied herbicide pronamide or by 0, 2, or 4 cuts of the grasses at various times after seeding. It was determined from this study that: 1. Suppression of sown-grass sods by cutting, was as effec- tive as herbicide suppression in stand establishment and subsequent good yields when sod seeding red clover and birdsfoot trefoil. 2. Sown grasses were suppressed more consistently by 4 cuts than by pronamide for good legume establishment. 3. Dense, vigorous quackgrass sods required suppression by a herbicide for successful introduction of legume species but compe- tition from a less dense quackgrass sod was controlled successfully by four defoliations for good legume establishment. 4. Initial red clover stand establishment was benefited by drilling the legume seed in the grass sod, but yields and stand perisistence were similar from either broadcast or drilled seedings. 5. Birdsfoot trefoil seedings were equally good when broad- cast early on 26 Mar. or drilled four weeks later on 21 Apr. 93 10. LITERATURE CITED Ackley, J.W. 1975. The evolution and potential of the powr-till concept. p. 53-71. No Till. Forage Sym. Columbus, Ohio. Ahlgren, H.L., M.L. Wall, R.J. Muckenhirn, and F.V. Burcalow. 1944. Effectiveness of renovation in increasing yields of permanent pastures in Southern Wisconsin. J. Amer. Soc. Agron. 36:121-131. Baker, B.S. 1980. Yield, legume introduction, and persistence in permanent pastures. Agron. J. 72:776-780. Barnhart, S.K., and W.F.'Wedin. 1981. Reduced-tillage pasture renovation in the semihumid temperate region of the U.S.A. p. 545-547. Proc. 14th Int. Grassl. Congr. Lexington, Kentucky. Burcalow, F.W., D.W. Smith, and L.F. Graber. 1940. The duration of the effects of renovation in the control of weeds and white grubs (Phyllophaga 8p.) in permanent bluegrass pastures. J. Amer. Soc. Agron. 32:15-22. Cullen, N.A. 1966. Pasture establishment on unploughable hill country in New Zealand. p. 851-855. Proc. 10th Int. Grassl. Congr. Helsinki, Finland. . 1970. The effect of grazing, time of sowing, fertil- izer and paraquat on the germination and survival of oversown grasses and clovers. p. 112-115. Proc. 11th Int. Grassl. Congr. Queensland, Aust. Decker, A.M., and F.R. Dudley. 1976. Minimum tillage establish- ‘ment of five forage species using five sod seeding units and two herbicides. p. 140-145. Proc. Int. Hill Lands Sym. W. Virginia Univ., Morgantown, W. Virginia. , H.J. Retzer, M.L. Sarvna, and H.D. Kerr. 1969. Permanent pastures improved with sod-seeding and fertilization. Agron J. 61:243-247. , and F. G. Swain. 1964. Improved soil openers for the establishment of small-seeded legumes in sod. Agron. J. 56: 211- 214. 94 11. 12. 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 95 Dowling, P.M., R.J. Clements, and J.R. McWilliams. 1971. Establishment and survival of pasture species from seeds sown on the soil surface. Aust. J. Agric. Res. 22:61-74. Fuelleman, R.F., and L.F. Graber. 1938. Renovation and its effect on the populations of weeds in pastures. J. Amer. Soc. Graber, L.F. 1927. Improvement of permanent bluegrass pastures with sweet clover. J. Amer. Soc. Agron. 19:994-1006. Harris, D.A. 1975. The role of equipment. p. 42-52. Proc. No Till. Forage Sym. Columbus, Ohio. Holland, Clive. 1980. Establishment of alfalfa (Medicago sativa L.) and birdsfoot trefoil (Lotus corniculatus L.) in various grass sods as affected by date and method of seeding, and herbicide application. M.S. Thesis, Mich. State Univ. East Lansing, Mich. , and M.B. Tesar. 1983. Establishment of three forage legumes as influenced by date of seeding and freezing tempera- tures. Unpublished research (chap. 2 this manuscript) Mich. State Univ. East Lansing, Mich. Love, R.M. 1944. Preliminary trials on the effect of manage- ment on the establishment of perennial grasses and legumes at Davis, California. J. Amer. Soc. Agron. 44:699-703. Olsen, E.J., J.H. Jones, and J.J. Patterson. 1981. Sod-seeding forage legumes in a tall fescue sward. Agron. J. 73:1032-1036. Roberts, I.P. 1910. The Roberts pasture. p. 385-391. In the Pastures in New York. Cornell Univ. Agric. Exp. Stn Bull. 280. Robinson, G.S., and MQW. Cross. 1960. Improvement of some New Zealand grassland by oversowing and overdrilling. p. 402-405. Proc. 8th Int. Grassl. Congr. Reading, England. Smith, E.M., T.H. Taylor, J.H. Casada, and W.J. Templeton, Jr. 1973. Experimental grassland renovation. Agron. J. 65:506-508. Sprague, M.A. 1952. The substitution of chemicals for tillage in pasture renovation. Agron. J. 44:405-409. . 1960. Seedbed preparation and improvement of unplow- able pastures using herbicides. p. 264-266. Proc. 8th Int. Grassl. Congr. Reading, England. Suckling, F.E.T. 1976. A 20-year study of pasture development through phosphate and legume oversowing on NOrth Island hill country of New Zealand. p. 367-381. Proc. Int. Hill Lands Sym. W. Virginia Univ., Morgantown, W. Virginia. 25. 26. 27. 28. 29. 30. 31. 32. 96 Taylor, R.W., and D.W. Allinson. 1983. Legume establishment in grass sods using minimum-tillage seeding techniques without herbicide application: forage yield and quality. Agron. J. 75:167-172. Taylor, T.H., E.M. Smith, and W.C. Templeton, Jr. 1969. Use of minimum tillage and herbicide for establishing legumes in Kentucy bluegrass (Pba pratensis L.) swards. Agron. J. 61:761—766. Tesar, M.B. 1980. Sod seeding birdsfoot trefoil and alfalfa. Mich. State Univ. Ext. Bull. E-956. Triplett, G.B., R.W. Van Keuren, and V.H. watson. 1975. The role of herbicides in pasture renovation. p. 29-41. No Till. Forage Sym. Columbus, Ohio. , and J. D. walker. 1977. Influence of 2, 4-D, pronamide, and simazine on dry matter production and botanical composition of an alfalfa-grass sward. Crop Sci. 17:61-65. Van Keuren, R.W. 1976. Hill land improvement in Eastern United States. p. 77-90. Proc. Int. Hill Lands Sym. W. Virginia Univ., Morgantown, W. Virginia. , and G.B. Triplett. 1970. Seeding legumes into estab- lished grass swards. p. 131-134. Proc. 11th Int. Grassl. Congr. Queensland, Aust. White, J.G.H. 1970. Establishment of lucerne (Mbdioago sativa L.) in uncultivated country by sod-seeding and oversowing. p. 134-138. Proc. 11th Int. Grassl. Congr. Queensland, Aust. 97 Table 1. Stand density of red clover 3 and 15 months after sod seeding into 3 grass awards that were suppressed with a herbicide (check) or different cutting frequencies (Exp. 1). Gras Reed 3 Bromegrass canarygrass Orchardgrass Average Stand loss Suppression 1980 1980 1981 1980 1981 1980 1981 1980 1981 .1980 to 1981 -2 Plants m 2 Broadcast 26 Mar. 1980 0 cuts 104 61 104 31 82 20 97 37 62 2 cuts 98 70 132 62 102 46 111 59 47 4 cuts 136 87 169 79 130 65 145 77 47 Herbicide 126 79 182 86 115 51 141 72 49 Average 115 74 147 65 107 46 123 62 50 Drilled 21 Apr. 1980 0 cuts 84 61 123 38 122 26 110 42 62 2 cuts 138 85 170 65 108 45 139 65 53 4 cuts 148 99 230 100 155 63 178 87 51 Herbicide 166 95 216 103 134 59 172 86 50 Average 134 85 185 77 130 48 150 70 53 Average of Broadcast and Drilled O cuts 94 61 114 35 102 23 103 40 61 2 cuts 118 78 151 64 105 46 125 63 50 4 cuts 142 93 200 90 143 64 162 82 49 Herbicide 146 87 199 95 125 55 157 79 50 Average 125 80 166 71 119 47 137 66 52 LSD(0.05 1980 1981 Between Within 61 38 Grasses The same seeding method and level of suppression 33 NS Seeding methods The same grass and level of suppression 26 26 Suppression tmts The same grass and seeding method 98 m.m N.N 0.0 o.m o.m o.o m.¢ m.m o.o m.m H.N H.o owouo>¢ o.mH N.q N.N w.HH N.N m.w o.MH m.¢ H.m w.MH m.o m.w ovHoHnuom m.0H w.N H.m n.0H N.N N.N H.HH H.N N.m o.HH H.N m.m mono c m.m H.N N.o o.m m.N H.o N.m m.m a.m m.m o.m m.o mono N m.m N.N o.H o.c H.N m.H o.m m.N N.o N.N m.N N.o muss o 83 .34 8 e335 o.m m.N o.o 0.5 H.N n.m m.w N.N m.o H.m w.N N.o owouo>< N.HH N.N m.w n.m N.N m.m o.NH H.N m.w m.mH H.H m.w ooHoHouom m.OH H.N ¢.w N.m N.H m.m N.HH q.N m.m n.oH N.N m.w mono q q.w N.N H.o N.w H.N H.o o.w m.N H.o N.N N.N 0.0 muse N o.m m.N H.H N.N m.N w.o N.N q.N m.o o.¢ m.N m.H moon 0 ome .uoz 0N uomovooum o; u HouoH NmmH meH Houpu Nme HmmH HouoH Non HmmH Houoa Nme meH ome GOHoooummsm owouo>< moouwuuozouo mmouwmuoaoo boom mmouwoaoum mmouu OAH. ogmv ooHoaonvoum wcHuuoo uoouoNMHw up so Axoonov ooHoHnuo: o nuHs vomoouoeso ouoa uonu mpuosm oousu ouoH mouov can so voooom pom AmHmoausouoe oH mooum msHa uo>oHo oouv uo>oHo no“ mo vHoH» .N oHooH 99 voguoa waHvooo coo moouw oaoo onH ouau sonmoumoom N.H H.H sonooueeoo mo Ho>oH new moon» oaom oeH ovosuoa waHvoom o.H mz conooueeso mo Ho>oH poo ponuoa wanoom osoo oea oomoouo m2 m2 sHsqu soosuom Nme HmmH Amo.onmH As.oHV AH.mv Am.mv As.oHVAm.mv AH.wv Aw.mHVAe.ev Au.mv Am.eHVAs.wv Am.wv m.w m.N o.m N.N o.N w.m N.m H.N N.o m.m H.N N.c owouo>< Am.eHv Am.wv An.mv Am.NHVAs.wV Am.mv AN.NHVAH.mv Ae.mv Ao.mHVAm.mV Am.aV H.NH N.N N.N w.0H N.N o.w N.HH N.N o.m o.mH N.H m.w ouHoHnuom Am.eHv Aw.ev Am.aV HH.NHVA5.NV as.mv Aa.oHVA~.eV An.av A~.wavao.mv As.sv N.OH m.N N.N m.m N.N @.N q.HH m.N w.w N.HH N.N m.w mono q AN.~HV Aw.mv Am.mv Am.eHvaH.aV Aw.mv A~.5HVAs.mv Am.mv AH.wHVAm.wV Au.mv m.w N.N N.o o.w m.N H.o m.w m.N o.e a.» N.N m.o muoo N Ae.~Hv Am.ev HH.mv Aw.~HVAo.wV Am.qv Am.HHVAw.oV an.ev Ao.mHVA~.hv Am.mv N.N N.N H.H o.¢ N.N N.H m.m N.N w.o o.m m.N H.H muoo o poHHHuv moo umoovoouo mo owouo>¢ .usoo N oHooH 100 Table 3. Stand density of birdsfoot trefoil 3 and 15 months after sod seeding into two grass swards that were suppressed with a herbicide (check) or different cutting frequencies (Exp. 2). Grass Reed canarygrass Orchardgrass Average Stand Loss Suppression 1980 1980 1981 1980 1981 1980 1981 1980 to 1981 -2 Plants m % Broadcast 26 Mar. 1980 O cuts 116 54 95 51 106 53 50 2 cuts 120 93 128 96 124 97 22 4 cuts 187 116 150 114 167 115 31 Herbicide 171 124 115 78 143 101 29 Average 148 97 122 85 135 91 33 Drilled 21 Apr. 1980 O cuts 109 57 102 41 106 49 54 2 cuts 128 83 126 92 127 88 31 4 cuts 153 118 133 99 143 109 24 Herbicide 137 94 143 93 140 94 33 Average 132 88 126 81 129 85 34 Average of Broadcast and Drilled O cuts 113 56 99 46 106 51 52 2 cuts 124 90 127 94 126 92 27 4 cuts 170 117 142 107 156 112 28 Herbicide 154 109 129 89 142 99 30 Average 140 93 124 83 132 88 33 LSD(0.05) 1980 1981 Between Within NS NS Grasses The same seeding method and level of suppression NS NS Seeding methods The same grass and level of suppression 43 48 Suppression tmts The same grass and seeding method 101 Table 4. Yield of birdsfoot trefoil (trefoil plus grass in parenthesis) sod seeded on two dates into two grass awards that were suppressed with a herbicide (check) or different cutting frequencies (Exp. 2). Grass Reed canarygrass Orchardgrass Average Suppression 1980 1981 1982 Total 1981 1982 Total 1981 1982 Total t ha-1 Broadcast 26 Mar. 1980 0 cuts 0.5 3.3 .8 0.7 5.0 5.7 0.6 4.2 4.8 2 cuts 2.6 5.9 .4 3.5 6.4 9.8 3.1 6.2 9.3 4 cuts 3.0 5.4 .4 3.9 6.9 10.8 3.5 6.2 9.7 Herbicide 4.7 5.7 10.4 3.6 5.6 9.2 4.2 7.5 11.7 Average 2.7 5.1 7.8 2.9 6.0 8.9 2.7 5.6 8.3 Drilled 21 Apr. 1980 0 cuts 0.5 3.9 4.3 0.4 3.9 4.3 0. 3.9 4.4 2 cuts 2.2 6.4 8.7 2.5 6.4 8.9 2. 6.4 8.4 4 cuts 3.4 6.1 9.4 3.3 5.7 9.0 3. 5.9 9.3 Herbicide 4.6 4.7 9.4 3.7 5.7 9.4 4.2 5.2 9.4 Average 2.7 5.3 7.9 2.5 5.4 7.9 2.6 5.4 8.0 Average of broadcast and drilled 0 cuts 0.5 3.6 4.1 0.6 4.5 5.0 0.6 4.1 4.6 (3.3) (6.2) (9.5) (3.3) (7.8) (11.1) (3.3) (7.0) (10.3) 2 cuts 2.4 6.2 8.6 3.0 6.4 9.4 2.8 6.3 8.9 (5.6) (8.6) (14.2) (5.9) (9.3) (15 2) (5.7) (8.9) (14.6) 4 cuts 3.2 5.8 8.9 3.6 6.3 9.1 3.5 6.1 9.5 Herbicide 4.7 5.2 9.9 Average 2.7 5.2 7.9 (6.4) (8.8) (15.2) 4.2 6.4 10.6 (6.6) (8.9) (15.5) 2.7 5.5 8.2 (5.5) (8.4) (13.9) LSD(0.05 1981 1982 Between Within NS NS Grasses The same seeding method and level of suppression 0.8 NS Seeding methods The same grass and level of suppression 0.9 1.5 Suppression tmts The same grass and seeding method 102 Table 5. Stand density of red clover and birdsfoot trefoil 3 and 15 months after sod seeding into an 80% quackgrass sward that was suppressed with a herbicide (check) or different cutting frequencies (Exp. 3). Birdsfoot Grass Red clover trefoil Suppression 1980 1980 1981 1980 1981 Plants mfz Broadcast 26 Mar. 1980 0 cuts 93 62 104 92 2 cuts 125 80 98 85 4 cuts 131 94 126 107 Herbicide 113 88 133 100 Average 115 81 115 96 Drilled 21 Apr. 1980 O cuts 131 81 95 92 2 cuts 168 96 105 90 4 cuts 201 139 144 115 Herbicide 196 113 145 117 Average 174 107 122 103 Average of Broadcast and Drilled 0 cuts 112 72 100 92 2 cuts 147 88 102 88 4 cuts 166 117 135 111 Herbicide 155 101 139 109 Average 145 94 119 100 LSD(0.05) 1980 1981 Between Within NS NS Legumes The same seeding method and level of suppression 35 22 Seeding methods The same legume and level of suppression 29 19 Suppression tmts The same legume and seeding method 103 Table 6. Yield of red clover and birdsfoot trefoil sod seeded on two dates into an 80% quackgrass sward that was suppressed witha herbicide (check)om’different cutting frequencies (Exp. 3). Grass Red clover Birdsfoot trefoil Suppression 1980 1981 1982 Total 1981 1982 Total t ha-l Broadcast 26 Mar. 1980 0 cuts 8.8 7.7 16.5 7.2 9.2 16.4 2 cuts 9.6 7.1 16.7 8.0 8.6 16.6 4 cuts 9.6 7.5 17.1 7.5 8.1 15.6 Herbicide 9.9 7.7 17.6 8.4 9.3 17.7 Average 9.5 7.5 17.0 7.8 8.8 16.6 Drilled 21 Apr. 1980 O cuts 9.1 7.7 16.9 7.9 9.4 17.3 2 cuts 9.6 7.9 17.5 7.9 8.5 16.4 4 cuts 10.2 8.0 18.2 7.5 8.8 16.3 Herbicide 10.0 8.3 18.3 8.8 9.0 17.8 Average 9.8 8.0 17.7 8.0 9.0 17.0 Average of Broadcast and Drilled 0 cuts 9.0 7.7 16.7 7.5 9.3 16.9 2 cuts 9.6 7.5 17.1 8.0 8.6 16.5 4 cuts 9.9 7.8 17.7 7.5 8.5 16.0 Herbicide 10.0 8.0 18.0 8.6 9.2 17.8 Average 9.7 7.8 17.4 7.9 8.9 16.8 LSD(0.05) w 1981 1982 Between Within 0.9 NS Legumes The same seeding method and level of suppression NS NS Seeding methods The same legume and level of suppression 0.8 NS Suppression tmts The same legume and seeding method 104 Table 7. Stand density of red clover and birdsfoot trefoil 3 and 15 months after sod seeding into a 90% quackgrass sward that was suppressed with a herbicide (check) or different cutting frequencies (Exp. 4). Birdsfoot Grass Red clover trefoil Suppression 1980 1980 1981 1980 1981 Plants m.-2 Broadcast 1980 26 Mar. 21 Apr. 0 cuts 3 21 2 cuts 15 4 cuts 23 18 57 42 Herbicide 78 59 40 26 Average 29 22 33 22 Drilled 21 Apr. 1980 O cuts 29 15 17 ll 2 cuts 58 33 30 17 4 cuts 60 34 66 37 Herbicide 111 57 66 50 Average 65 35 45 29 Average of Broadcast and Drilled 0 cuts 19 9 19 10 2 cuts 32 19 23 13 4 cuts 42 26 62 40 Herbicide 95 58 53 38 Average 47 29 39 26 LSD(0.05) 1980 1981 Between Within NS NS Legumes The same seeding method and level of suppression 36 NS Seeding methods The same legume and level of suppression 29 27 Suppression tmts The same legume and seeding method 105 Table 8. Yield of red clover and birdsfoot trefoil sod seeded on two dates into a 90% quackgrass sward that was suppressed with a herbicide (check) or by different cutting frequencies (Exp. 4). Grass Red clover Birdsfoot trefoil Suppression 1980 1981 1982 Total 1981 1982 Total t ha”1 Broadcast 1980 26 Mar. 21 Apr. 0 cuts 1.3 0.2 1.5 0.9 0.3 1.2 2 cuts 2.0 0.3 2.3 1.9 1.5 3.4 4 cuts 3.6 0.3 3.9 3.6 2.4 6.0 Herbicide 6.3 2.0 8.3 3.4 2.3 5.7 Average 3.3 0.7 4.0 2.5 1.6 4.1 Drilled 21 Apr. 1980 0 cuts 1.5 0.3 1.8 0.8 0.1 0.9 2 cuts 3.1 0.4 3.5 1.0 0.7 1.7 4 cuts 3.7 0.3 4.0 2.8 3.0 5.8 Herbicide 5.0 1.9 6.9 4.9 3.5 9.2 Average 3.3 0.7 4.0 2.4 1.9 4.3 Average of Broadcast and Drilled 0 cuts 1.4 0.3 1.7 0.9 0.2 1.1 2 cuts 2.6 0.4 3.0 1.5 1.1 2.6 4 cuts 3.7 0.3 4.0 3.2 2.7 5.9 Herbicide 5.7 2.0 7.7 4.6 2.9 7.5 Average 3.3 0.7 4.0 2.5 1.8 4.3 LSD(0.05) 1981 1982 Between . Within NS NS Legumes The same seeding method and level of suppression NS NS Seeding methods The same legume and level of suppression 1.9 1.7 Suppression tmts The same legume and seeding method CHAPTER 4 SURVIVAL OF THREE FORAGE LEGUMES SEEDED IN CONTACT WITH PHOSPHORUS AND POTASSIUM ABSTRACT Pasture improvement with the introduction of high producing legumes into an existing grass sward requires excellent management during establishment. Fertilizer applied to promote maximum early vigorous growth must be placed, when sod seeding, in contact with seeds in a single slit in the sod. Injury from fertilizer when placed in contact with seeds of corn (Zea mays L.) and small grains has been well documented but injury to forage legume seeds has not been well studied. Greenhouse trials were conducted with alfalfa (Medficago sativa L.), red clover (Trifblium pratense L.) and birdsfoot trefoil (Lotus corniculatus L.) seeded in contact with various rates of P and K fertilizer to determine injury. Phosphorus reduced seedling survival more severely than sim- ilar rates of K. Extremely low fertilizer-band pH was considered the primary cause of poor stand survival from P in contact with the seeds. To apply similar quantities of P and K, 2.5 times more commercial P fertilizer was required. This larger bulk of material produced a 106 107 more constant distribution pattern of fertilizer which also likely contributed to the lower seedling survival when P was applied. Legume stands were not reduced by rates of K lower than 35.5 kg ha- except for red clover with K at 11.5 and 23.0 kg ha-l. Legume yields were reduced by both fertilizers at rates higher than those normally recommended for field seedings. A sharper decline in yield, with increased fertilizer application, was obtained with P compared to K. Additional index words: Mbdicago sativa L., Trifblium‘pratense L., Lotus corniculatus L., pasture renovation, sod seeding, poor germina- tion, seedling injury, fertilizer injury. INTRODUCTION Of the estimated 42 million hectares of permanent pastures in the United States, almost half can be improved significantly simply by proper grazing, fertilization and weed control (Baylor, 1975). The cropland used for hay and pasture approximates another 30 million hectares according to the USDA conservation-needs inven- tory. A large portion of these grassland areas could be maintained at a more productive level by pasture renovation. This involves the introduction of a more productive forage legume into a grass sward without the growing of an intervening crop. Complete tillage, fol- lowed by reseeding, has been the most widely accepted and successful method of pasture improvement (Tesar and Hildebrand, 1975). However, in hilly, erodible areas, reduced tillage or sod seeding of forage legumes has successfully prevented erosion and greatly increased pasture productivity. Reducing competition from the grasses to the introduced legumes has been the most important factor in making sod seedings. Fertilization of small-seeded legumes at seeding has also been recog- nized for many years as being important for good early growth (Cook and Millar, 1944; Tesar et al., 1954). When fertilizer placement studies were started in the 19203, according to Mortvedt (1976) the problems were much different from.those of today. Low analysis fertilizers were applied at rates up to 330 kg ha.1 close to, or in 108 109 contact with the seed without causing any problems. Today, the average nutrient content of fertilizers and rates of application are higher. Mortvedt maintains that the likelihood of damage from cur- rent high analysis fertilizers placed too near or in contact with the seed is much higher now. Duell (1976) reviewed the literature on P fertilization for forage establishment and found general agree- ment that legume seedlings are usually less able than grasses to obtain P from low soil-P concentrations associated with broadcast fertilizer applications. Seedling growth of both grasses and legumes is enhanced by placing P in concentrated bands directly underneath the seed row (Sheard et al., 1971; Tesar et al., 1954). These researchers found that if the fertilizer band was 2 or 3 cm to the side of the seed row, early growth of alfalfa seedlings was signifi- cantly reduced. Pasture renovation with seedings drilled in the sod has necessitated the placement of fertilizer with the seed for maximum stimulation and early growth of legume seedlings. Machinery adapted to place seed and fertilizer separately into untilled soils has not been commercially available and the physical firmness of these seedbeds has added to the difficulties of separate placement of seed and fertilizer. Considerable research has been conducted on the effects of various types and quantities of fertilizer placed in contact with seeds of corn, small grains and various other crops. The overall hesitancy of workers to conclude that seed in contact with fertilizer is beneficial and without problems was well summarized by Mortvedt (1976). He stated that if the resulting fertilizer placement was too close to the seed, reduced seedling emergence may result. He further 110 added that placement of small amounts of fertilizers in the seed row is not widely used because of possible delays or decreases in seed germination and decreased stands. Cook (1957) emphasized that young seedlings should be well fertilized to promote rapid, vigorous growth, but to avoid injury to most crops, fertilizers should not be placed in contact with seeds. In giving recommendations to producers on how to successfully sod seed birdsfoot trefoil and alfalfa, Tesar (1980) sug- gested, based on unpublished field trials, that 22 kg ha.1 of P could be added in contact with the seed without injury. Duell (1963), how- ever, concluded from'his work that both alfalfa and trefoil showed reduced emergence when seeds were close to concentrations of soluble fertilizers while moisture was low. This study was conducted under closely controlled conditions in the greenhouse to determine how injurious various rates of the most commonly used commercial P and K fertilizers would be to three forage legumes when placed in contact with the seed at the time of seeding. MATERIALS AND METHODS .Alfalfa (Medficago sativa L.), red clover (Trifblium pratense L.), and birdsfoot trefoil (Lotus corniculatus L.) were seeded in contact with various rates of P and K fertilizer under controlled conditions in the greenhouse. Treatments were replicated four times in a split-plot, randomized, complete block design. Pots were moved twice weekly within each block in a directed rotation to minimize light effects. A Brookston loam (fine-loamy, mixed mesic Typic Agriaquolls) soil of pH 7.3 was placed in 22.5-cm-diameter pots and prepared for seeding by forming a single crease 0.5 cm deep across each pot (Fig. 2). A specially designed vacuum-operated seeder placed 22 seeds 1 cm apart in the pre-formed creases in the soil in each pot. Commercial-grade fertilizer, (0-46-0 and 0-0-60), without additional grinding or pulverizing, was distributed evenly in the soil crease. Rates of fertilizer were calculated based on the surface area of the pots and adjusted to equal applications in kg ha"-1 in rows spaced at 17.5 cm. Rates of P were 0, 12, 24, 36, 48, 60, 72, and 96 kg ha-l. Rates of K were 0, 11.5, 23.0, 34.5, 46.0, 69.0, 92.0, and 138.0 kg ha-l. The fertilizer was added after seeding to prevent the deflection and uneven spacing of the seeds by the fertil- izer granules and crystals. The legume seed was inoculated prior to seeding with a slurry of appropriate rhizobia inoculum. Additional dry inoculum was added in the soil crease at seeding. Measured 111 112 quantities of water containing a fungicide were added every second day for the first 20 days to facilitate uniform fertilizer dilution and minimize damping off disease due to Eythium spp. Determinations of pH were made 3 and 10 days after seeding in situ on the fertilizer band in the pots containing red clover. A Beckman 3560 digital meter with combination electrode was used to obtain pH values by inserting the electrode into the area of banded seed and fertilizer immediately after applying a measured amount of de-ionized water to the surface of the pot. The lowest, stable pH value averaged from.two readings per pot was obtained. Seedling counts were made every 5 days up to 25 days after seeding with a final determination at 90 days. At this time, all top growth was harvested, dried with forced air at 65 C for 48 hours, and weighed. Yields are reported in grams of dry matter per pot. RESULTS AND DISCUSSION Seedling survival increased up to 10 to 15 days after seed- ing in all except the four highest rates of P, and then showed no further increase (Tables 1 and 2). Red clover was somewhat more tolerant of being in contact with P than alfalfa or trefoil but was less tolerant than alfalfa or trefoil when in contact with K. The lowest rate of P (12 kg ha-l) placed in contact with the seed significantly reduced 90-day seedling survival of all three legumes (Table 1). Stand survival was best when no P was added. Averaged over legumes the survival decreased with increases in rates of p in kg ha-l: 0—932 > 12—7474 > 24—36% > 36—17%. Only a few seedlings survived when in contact with 48 and 60 kg of P; none survived the 82- and 96-kg rates. At similar rates, K in contact with these forage legume seeds did not reduce seedling survival nearly as much as P. Legume stands were equally good at 0, 11.5, 23.0, or 34.5 kg of K ha.1 except at 11.5 and 23.0 kg where the red clover stand was significantly reduced in comparison to that obtained with no fertilizer (Table 2). Averaged over legumes, seedling sur- vival decreased with increased rates of K in kg ha-l: 0-93% > 11.5 —85% > 46—77% > 69—59% > 92—1874 > 138—7%. The salt index of KCl fertilizer per unit of nutrient is over nine times higher than that of concentrated superphosphate according to Radar et a1. (1943). This clearly indicates that the lower survival rate of legumes in 113 114 contact with P compared to those in contact with K was not caused by fertilizer salt injury. Rates of P up to 48 kg ha-1 are recommended as safe to apply in a band below the seed in most Michigan soils when seeding forage legumes (Warncke and Christenson, 1980). Almost twice this amount is the maximum P recommended for broadcast application at one time on soils with the lowest amount of P. This contrasts with maximum.recome mended rates of K at 46 kg ha-1 in a band under the seed and 370 kg K ha.1 broadcast, for high production on K-deficient soils. The Michi- gan State University Soil Testing Laboratory tested 24,631 soil samples during 1982. Fifteen percent of these samples were tested for producers requesting recommendations for suitable fertilizers to be applied when making new alfalfa seedings (Meints, 1983). Only 45% of all the samples tested for alfalfa establishment required the addition of P. Twenty percent required P at 12 kg ha-l, 15% needed 36 kg ha-l, and 10% required rates higher or equal to the highest "safe" recommended rate. If producers had made sod seedings of alfalfa and added P in contact with the seed according to their soil test results, based on data from this greenhouse experiment, they would have sustained a stand loss of 19, 57, and 76% when applying 12, 24, and 36 kg ha-1 of P, respectively (Table l). A 20% stand reduction would not be detected readily in field seedings and even a 50% stand loss would be difficult to detect without having comparable non-fertilized strips in each seeding. Field seedings on a fine-textured soil have been made, however, with rates of P as high as 24 kg ha.1 in contact with legume seeds (Tesar, 1978) but it was not reported how many seedlings were killed 115 by P. Figure 3 shows placement of seed and fertilizer granules in field seedings made by a commercial drill. The speed of the seeding equipment produced a scattered distribution of seed and fertilizer and likely avoided serious seedling injury through lack of contact. In firmer, more root-bound sods, the slit where seed and fertilizer are placed together in the soil is usually more closed, particularly when the soil is moist and high in clay, thereby producing a greater likelihood of injury. Blaser and Kimbrough (1972) state that applications of K for forage establishment should be low to avoid interference with germina- tion from high soluble salts and indicate ratios of 1P:lK have given good results. In the experiment reported here, reduced stand sur- vival when K was palced in contact with the legume seeds was expected because of the high fertilizer salt content but stand depletion was no greater with similar rates of K than that from the P fertilizer. Most soils, however, require three to four times the amount of K to P fertilizer for maximum growth of legumes and recommendations . normally reflect much higher rates of K than P being applied when seeding legumes. Guttay (1957) concluded from his work with fertilizers in contact with seeds of wheat and oats that the P content of fertil- izers was just as important as N and K in producing seed injury, but he could not identify the major constituent of P fertilizers respons- ible for the seed injury. Readings of pH in the fertilizer band obtained 3 and 10 days after seeding in the trial reported here showed that an extremely acid micro-environment was formed around the seeds with P fertilizer, but not with K (Fig. l). A 1:1 water— 116 fertilizer solution, equilibrated for one hour, produced pH readings of 3.0 for P, and 6.3 for the K fertilizers used in this trial. The saturated solution from a moistened granule of P fertilizer has been shown to be as acid as pH 1.48 (Lindsay et al., 1959). Lindsay and Stephenson (1959) found that when P was banded in a soil of pH 5.5, the pH of the soil samples from 0 to 10 mm from the band remained below 3.0 for at least 6 weeks. At this extremely low pH, these workers indicated that large amounts of Al, Fe, and Mn were brought into solution. High levels of any of these, particularly Al, for even a short period would have toxic effects on sensitive germinating seeds or seedlings. At similar rates of P and K, 2.5 times more com? mercial P fertilizer used in this study was added in comparison to the K fertilizer. This greater concentration of total fertilizer would have given a more even distribution of P and contributed to the over- all stand decline with less probability for spaces between P granules than between K crystals. If both fertilizers had been pulverized before application, seedling injury and stand reduction from K would likely have been increased considerably. It was determined from addi- tional control pots that many of the seedlings that survived high K rates grew between large fertilizer crystals. Fertilizers were not ground in order to simulate seeding and fertilizing under field condi- tions. Yields of the three legumes 90 days after seeding were reduced at the higher rates of P and K.which were greater than those recom- mended for making field seedings (Tables 3 and 4). A sharper decline in yield with increased fertilizer application was obtained with P compared to K. The higher yield from red clover was similar with both 117 P and K and was likely due to a greater tolerance of red clover than alfalfa or trefoil to reduced levels of light in the greenhouse (Gist and Mott, 1957). The benefits of fertilizer, particularly P, to legume seed- lings have been well documented. Faster and stronger root develop— ment and better initial growth are the most important benefits of adequate P to young seedlings. Producers of alfalfa are advised that even if the soil test indicates P is not necessary, the use of a starter fertilizer containing 12 kg P ha.1 placed 2 to 4 cm under the seed will help strong seedling development (Tesar, 1978). Data obtained here indicate, however, that when P at this rate was in con- tact with forage legume seeds as in sod seeding, stands were reduced by 19%. When the P rate was doubled from 12 to 24 kg ha-l, however, survival decreased 38%-from 74 to 36%-a very severe stand loss. Stand injury from P, therefore, can be expected on Michigan loam soils and very likely, on more coarse-textured soils if P is placed in contact with legume seeds. If P is recommended for seedling establishment when sod seeding in the field, better stands would be obtained by broadcasting this fertilizer before or after seeding rather than in contact with the seed. SUMMARY AND CONCLUSIONS Alfalfa, red clover, and birdsfoot trefoil were seeded under controlled conditions in the greenhouse in contact with various rates of P and K fertilizer. Seedling survival was more severely reduced by P than similar rates of K. From.this study with commercial grade P (0-46-0) and K (0-0-60) fertilizer, it was determined that: 1. Phosphorus placed in contact with the seed reduced seed- ling survival at all levels of application. 2. Poor germination and survival when P was placed in con- tact with the seeds was caused by the extremely 1ow fertilizer-band pH. 3. Potassium in contact with the seeds did not reduce legume stands at rates less than 34.5 kg ha-l. 4. Red clover was more tolerant of P but less tolerant of K in contact with the seed than alfalfa or trefoil. 5. Phosphorus should not be applied in contact with forage legume seeds when sod seeding. 118 10. 11. LITERATURE CITED Baylor, J.E. 1975. The need for renovation. p. 12-14. In No Till. Forage Sym. Columbus, Ohio. Blaser, R.E., and E.L. Kimbrough. 1972. Potassium nutrition of forage crops with perennials. p. 423-445. In V.J. Kilmer, S.E. Younts, and N.C. Brady (eds.). The role of potassium in agriculture. Amer. Soc. Agron. Madison, Wis. Brown, B.A. 1959. Band versus broadcast fertilization of alfalfa. Agron. J. 51:708-710. Cook, R.L. 1957. Fertilizer placement for better crops. Better Crops with Plant Food. 41-6:14-22. , and C.E. Millar. 1944. Fertilizers for legumes. Mich. Agr. Exp. Sta. Spec. Bull. 328. Duell, RJW. 1964. Fertilizer-seed placement with birdsfoot trefoiltZotus corniculatua L.) and alfalfa (Mbdficago sativa L.). Agron. J. 56:503-505. . 1976. Fertilizing forage for establishment. p. 67-93. In D.A. Mays (ed.) Forage fertilization. Amer. Soc. Agron. Madison, Wis. Gist, G.R., and G.C. Mott. 1957. Some effects of light intensity, temperature, and soil moisture on the growth of alfalfa, red clover and birdsfoot trefoil seedlings. Agron. J. 49:33-36. Guttay, J.R. 1957. The effect of fertilizer on the germination and emergence of wheat and oats. Mich. Agric. Exp. Sta. Quarterly Bull. 40:193-202. Lindsay, W.L., J.R. Lehr, and R.F. Stephenson. 1959. Nature of the reactions of monocalcium phosphate monohydrate in soils: III. Studies with metastable triple—point solution. Soil Sci. Soc. Amer. Proc. 23:342-345. ' , and R.F. Stephenson. 1959. Nature of the reactions of monocalcium phosphate monohydrate in soils: II. The solution that reacts with the soil. Soil Sci. Soc. Amer. Proc. 23:12-18. 119 12. 13. 14. 15. 16. 17. 18. 19. 20. 120 Meints, V.W. 1983. Unpublished statistical summary of the annual activity of the M;S.U. Soil Testing Laboratory. Mich. State Univ. East Lansing, Mich. Mortvedt, J.J. 1976. Band fertilizer placement-how much and how close? NOv.-Dec. Fert. Solutions Mag. Radar, L.F., Jr., L.M. White, and G.W. Whittaker. 1943. The salt index-a measure of the effect of fertilizers on the con- centration of the soil solution. Soil Sci. 55:201-218. Sheard, R.W., C.J. Bradshaw, and D.L. Massey. 1971. Phosphorus placement for the establishment of alfalfa and bromegrass. Agron. J. 63:922-927. Tesar, M.B. 1978. Good stands for top alfalfa production. Midh. State Uhiv. Ext. Bull. E-1017. . 1978. Unpublished data from field trials of forage legumes sod seeded in contact with fertilizer. Mich. State Univ. East Lansing, Mich. . 1980. Sod seeding birdsfoot trefoil and alfalfa. MiCh, State miv. Ext. BUllo E-956o , and S.C. Hildebrand. 1975. Re-establishment of pastures and hay fields in one year. Mich. State Univ. Ext. Bull. E-527. Warncke, D.D., and D.R. Christenson. 1980. Fertilizer recom- mendations-—vegetable and field crops in Michigan. Mich. State wiv. Ext. BUllo E‘SSO. 121 Table 1. Percent survival of three forage legumes grown in the greenhouse after seeding in contact with various kg ha'1 of P. Days After Seeding Legumes 5 10 15 20 25 90 2 Alfalfa 90 86 88 88 88 88 Red clover 80 90 93 95 95 95 Trefoil 71 77 94 95 95 95 Average 80 84 92 93 93 93 .12 Alfalfa 51 76 74 74 74 75 Red clover 58 84 79 77 83 76 Trefoil 43 68 70 69 69 70 Average 51 76 74 73 75 74 , .2_4 Alfalfa 15 39 39 34 33 34 Red clover 13 51 47 45 45 44 Trefoil 6 19 28 29 27 30 Average 11 36 38 36 35 36 gg_ Alfalfa 6 20 22 21 18 16 Red clover 5 28 26 21 20 20 Trefoil 3 5 17 17 l7 14 Average 5 18 22 20 18 17 l._8_ Alfalfa 3 12 10 8 8 8 Red clover 11 24 18 17 17 17 Trefoil 0 0 4 5 S 6 Average 5 12 ll 10 10 10 ‘29 Alfalfa 0 3 3 3 3 0 Red clover 0 7 5 4 4 3 Trefoil 0 0 0 0 0 0 Average 0 3 3 2 2 1 .12 Alfalfa 0 0 0 0 0 0 Red clover 0 l 1 0 0 0 Trefoil 0 0 0 0 0 0 Average 0 0 0 0 0 0 ‘26 Alfalfa 0 0 0 0 0 0 Red clover 0 0 0 0 0 0 Trefoil 0 0 0 0 0 0 Average 0 0 0 0 0 0 LSD(0.05) Legumes 16 22 NS NS NS NS Fertilizer l4 l4 13 13 14 12 Average 8 8 7 7 8 7 122 Table 2. Percent survival of three forage legumes grown in the greenhouse after seeding in contact with various kg ha” of K. Days After Seeding Legumes S 10 15 20 25 90 2 Alfalfa 90 86 88 88 88 88 Red clover 80 90 93 95 95 95 Trefoil 71 77 94 95 95 95 Average 80 84 92 93 93 93 11.5 Alfalfa 87 95 91 88 89 92 Red clover 100 91 82 80 79 79 Trefoil 79 94 89 86 85 85 Average 89 93 87 85 84 85 L 23.0 Alfalfa 96 100 100 100 100 100 Red clover 82 90 82 81 80 76 Trefoil 80 100 98 96 95 94 Average 86 97 93 92 92 90 34.5 Alfalfa 51 92 91 88 89 93 Red clover 31 96 86 84 84 84' Trefoil 17 90 9O 9O 90 90 Average 33 93 89 87 88 89 46.0 Alfalfa 49 91 88 88 88 90 Red clover 24 76 70 67 64 61 Trefoil 9 85 82 81 81 81 Average 27 84 80 79 78 77 gm Alfalfa 9 59 58 58 59 62 Red clover 13 66 59 55 53 53 Trefoil 6 59 60 61 61 61 Average 9 61 59 58 58 59 92.0 Alfalfa l 25 23 22 22 23 Red clover 0 16 11 ll 11 ll Trefoil 0 15 22 21 20 19 Average 0 19 19 17 18 18 1 38 O 0 Alfalfa 0 13 13 13 13 13 Red clover 2 l l l 1 1 Trefoil 0 3 7 7 7 7 Average 1 6 7 7 7 7 LSD(0.05) Legumes 16 NS NS NS 21 22 Fertilizer 14 14 13 13 14 12 Average 8 8 7 7 8 7 123 Table 3. Yield of three forage legumes obtained in the greenhouse 90 days after seeding in contact with various kg ha"1 of P. P kg ha.1 Alfalfa Red clover Trefoil Average 8/Pot 0 13.4 26.5 13.6 17.8 12 13.1 25.8 12.6 17.2 24 10.1 24.0 9.3 14.5 36 5.9 16.5 3.4 8.6 48 3.5 17.3 1.4 7.4 60 0.0 4.9 0.0 1.6 72 0.0 0.0 0.0 0.0 96 0.0 0.0 0.0 0.0 average 5.8 14.4 5.0 8.4 LSD<0005) Between legumes at the same fertilizer rate 1.3 Between fertilizer rates within the same legume 3.8 124 Table 4. Yield of three forage legumes obtained in the greenhouse 90 days after seeding in contact with various kg ha"1 of K. kg :a-l Alfalfa. Red clover Trefoil Average s/pot 0.0 13.4 26.5 13.6 17.8 11.5 14.9 21.9 12.4 16.4 23.0 12.3 24.6 12.7 16.5 34.5 11.5 25.9 13.5 17.0 46.0 11.8 22.8 12.6 15.7 69.0 11.4 20.3 12.3 14.7 92.0 9.0 17.9 7.6 11.5 138.0 7.3 3.3 3.9 6.2 average 11.5 20.4 11.1 14.3 LSD(O.05) Between legumes at the same fertilizer rate 1.3 Between fertilizer rates within the same legume 3.8 125 705 j V 1 fi j j j fi' 1 1 1 j 1 7 9 t f - w an; ‘ I 5.5 r- ' - . n. 3 DRYS n 60' . ‘ Z a: m 5.5 - , a: U 51 5.0 P - H .t’. .- ‘05 ' 1 (2 U h. ‘aar G 3.5 r . 3" J j l l l 1 I. l j l 1 J l B 23 46 69 92 115 138 Pomsswn (kg ha“) 7 s I U V 1 V U T j U 7.0 " «I I 6.5 . 1 1 Cl .. . 2 6.0 E 5.5 p 1 0! #3 H 5" ’ la :3an ‘ z.’ .— 405 ’ q m U 3 DHYS 3.5 - w 3" l J l J A l j J l a 12 24 38 48 SB 72 84 96 PHOSPHORUS (kg ha“) Fig. 1. Fertilizer-band pH sampled in situ 3 and 10 days after placing various rates of K (top) and P (bottom) in a Brookston loam soil. 126 was usmswomaa .Hocmom woumquOIsbaom> nonaawuumw van vomm MOM Anwov so m.ov ammouo HHom .N .mE 127 .vom vomwwHAASmlwwHoHnumz a coma mafivmom com cos? unwanfisvw Hmauwoseoo hp HoNHHHuHam M can boom «mamew mo unmamomam .n .wE 3. 4. 5. 10. 11. 12. 13. GENERAL BIBLIOGRAPHY Brown, B.A. 1959. Band versus broadcast fertilization of Christenson, D.R., D.D. Warncke, and R. Leep. 1981. Lime for Michigan soils. Michigan State Univ. Ext. Bull. E-47l. Decker, A.M., T.H. Taylor, and C.J. Willard. 1973. Establish- ment of new seedlings, p. 384-395. In M.B. Heath, D.S. Metcalf, and R.F. Barnes (eds.) Forages, the science of grassland agri- culture. Iowa State Univ. Press. Ames, Iowa. Graber, L.F. 1936. Renovating bluegrass pastures. Wisconsin Agric. Exp. Sta. Circ. 277. Olsen, F.J., J.H. Jones, and J.J. Patterson. 1981. Sod-seeding forage legumes in a tall fescue award. Agron. J. 73:1032-1036. Roberts, R.W. 1960. Chemical renovation of hill land in the north of Scotland. Trans. Royal Highland Agric. Soc. Scotland p. 1.13. Scholl, J.M., M.F. Finner, A.E. Peterson, and J.M. Sund. 1970. Pasture renovation in Southwestern Wisconsin. Wis. Agric. Exp. Sta. Res. Rep. 65. Smith, Dale. 1975. Forage management in the North. Kendall/ Hunt Pub. Co. Dubuque, Iowa. Taylor, T.H., E.M. Smith, and W.C. Templeton, Jr. 1969. Use of minimum tillage and herbicides for establishing legumes in Kentucky bluegrass (Pba pratensis L.) awards. Agron. J. 61:761-766. Tesar, MgB. 1980. Sod seeding of birdsfoot trefoil and alfalfa. Michigan State Univ. Ext. Bull. E-956. , and S.C. Hildebrand. 1975. Re-establishment of pastures and hay fields in one year. Mich. State Univ. Ext. Bull. 3527. , K. Lawton, and B. Kawin. 1954. Comparison of band seeding and other methods of seeding legumes. Agron. J.46:l89-194. White, J.G.H. 1966. Introduction of lucerne into acid soils. p. 104-113. Proc. New Zealand Grassl. Assoc. New Zealand. 128 1.‘ . ‘I'l¢\‘ l‘. I AN STATE UNIVERSI MICHIG IHIH “WWlilHlliIllWWlflm 3 3015 4321 1293 0