LIBRARY Michigan State University This is to certify that the thesis entitled Evaluation of Medicago sativa spp. fa/cata in Michigan presented by Elysia A Berry has been accepted towards fulfillment of the requirements for the MS _ degree in Crop and Soil Science M Major Professor’s Signature fi _ 57/ 5/2 a 0 7 Date MSU is an Affirmative Action/Equal Opportunity Employer \——\‘ PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 5/08 K:IProj/Acc&Pros/ClRC/DatoDue.hdd EVALUATION OF Medicago sativa spp. falcata IN MICHIGAN By Elysia A Berry A THESIS Submitted to Michigan State University In partial fulfillment of the requirements For the Degree of MASTER OF SCIENCE Crop and Soil Science 2009 ABSTRACT EVALUATION OF Medicago sativa spp. falcata IN MICHIGAN By Elysia A Berry Grazing livestock is an important practice in Michigan to reduce feeding costs of harvested forage. Alfalfa and grass mixtures are commonly used for pastures in the north-central region. Falcata alfalfa (Medicago sativa spp. falcata) possesses several desirable characteristics for pastures in this region including winter hardiness and prolonged forage quality. This study evaluated forage yield, grazing preference, forage quality, and stand persistence of binary combinations of Medicagofalcata cv.Yellowhead, Medicago sativa cv.ZG9830, and Birdsfoot trefoil (Lotus corniculatus) cv. Noreen seeded with one of the following perennial, cool season grass species: meadow fescue, orchardgrass, or timothy. Three stands were sown in May 2005: one in the Upper Peninsula and two in the Lower Peninsula of Michigan. Over the three year data collection period, an average of four grazing events occurred each year. M. sativa mixed with each of the grasses produced the greatest forage yield. M. sativa averaged almost 2.2 metric tons per hectare of dry matter over M. falcata and almost 4.5 metric tons per hectare of dry matter over birdsfoot trefoil. Overall, the cattle have shown no grazing preference for falcata over sativa, but a slight trend was seen for the birdsfoot trefoil. Significant differences were found in forage quality, both within and between the three locations. Stand persistence (plant density) varied each year due to sampling error. ACKNOWLEDGMENTS I would like to begin by thanking Dr. Richard Leep for this great opportunity to study with him and enhance my knowledge of forages. I would also like to thank my graduate committee of Dr. Donald Penner and Dr. Steven Rust for their help and wonderful suggestions to improve this research trial. My deepest thanks go to Tim Dietz and Stephanie Peck, who started this project in 2005 and collected the data for me before I started my graduate studies. I would also like to thanks Doug Carmichel and his crew at Lake City Experiment Station for their help in doing grazing preference ratings. Also, I want to thank Christian Kapp for collecting and analyzing almost all of my data from the Upper Peninsula Experiment Station. Finally, I would like to thank my family and friends for all of their help, support, and encouragement through my time at Michigan State University. iii Table of Contents Page LIST OF TABLES ............................................................................ v LIST OF FIGURES ........................................................................... vi LIST OF ABBREVIATIONS ............................................................... ix 1. INTRODUCTION .......................................................................... 1 11. MATERIALS AND METHODS ........................................................ 4 III. RESULTS AND DISCUSSION ....................................................... 10 Weather Records a) Chatham .......................................................................... 10 b) Lake City ......................................................................... 11 c) East Lansing ...................................................................... 12 Forage Yield Results a) Chatham .......................................................................... 14 b) Lake City ......................................................................... 15 c) East Lansing ..................................................................... 16 Forage Quality Results a) Crude Protein ..................................................................... 17 b) Acid Detergent Fiber ............................................................ 21 c) Total Digestible Nutrients ...................................................... 24 d) Relative Feed Value ............................................................. 28 e) Net Energy for Lactation ........................................................ 32 Palatability a) Chatham ........................................................................... 35 b) Lake City .......................................................................... 36 0) East Lansing ...................................................................... 37 (1) Total ............................................................................... 38 Stand Persistence a) Chatham ........................................................................... 39 b) Lake City ......................................................................... 40 c) East Lansing ...................................................................... 41 IV.CONCLUSION ............................................................................ 43 V. APPENDIX .................................................................................. 46 VI. LITERATURE CITED ................................................................... 50 iv LIST OF TABLES Table Page 1 AN OVA output for yield ................................................................ 9 2 Precipitation Data for Chatham, M1 (2006-2008) (cm) ............................... 10 3 Temperature Data for Chatham, M1 (2006-2008) (°C) .............................. 11 4 Precipitation Data for Lake City, M1 (2006-2008) (cm) ............................. 11 5 Temperature Data for Lake City, M1 (2006-2008) (°C) ............................ 12 6 Precipitation Data for East Lansing, M1 (2006-2008) (cm) ......................... 13 7 Temperature Data for East Lansing, MI (2006-2008) (°C) ......................... 13 8 Forage Nutritive Values for a grazing dairy cow .................................... 17 9 East Lansing trial plot map. (The area in grey indicates a seeding error) ......... 46 2005. 10 Lake City trial plot map. 2005 .......................................................... 47 1 1 Chatham trial plot map. 2005 ........................................................... 47 12 Average Accumulative forage yield at each location (metric tonnes dry ......... 48 matter per hectare) 13 Stand persistence data at Chatham trial location across four years post- ........... 48 seeding. 2009. 14 Stand persistence data at Lake City trial location across four years post- ......... 48 seeding. 2009. 15 Stand persistence data at Lake City trial location across four years post- ......... 49 seeding. 2009. Figure LIST OF FIGURES Page Total forage yield per combination at Chatham trial location across three ...... l4 harvest years in metric tons dry matter per hectare. LSD0,05= 0.88. Combinations with the same letters are similar. (bft=birdsfoot trefoil, fal=falcata, sat=sativa, fes=meadow fescue, or=orchardgrass, tim=timothy) Total forage yield per combination at Lake City trial location across three. . . ..1 5 harvest years in metric tons dry matter per hectare. LSD0,05= 0.99. Combinations with the same letters are similar. Total forage yield per combination at East Lansing trial location across ....... 16 three harvest years in metric tons per dry matter hectare. Standard Error 005 = 2.6. Combinations with the asterisk are similar. Total weighted percent crude protein for all locations across three harvest. . 1 8 years. Chatham LSD0_05=1.16. Lake City Standard Erroro,05=3.6. East Lansing Standard Erroro,05=6.4. Crude Protein for maintenance diet for Dairy cows is indicated by the dashed line (18%). Weighted percent crude protein at Chatham trial location across three ......... 18 harvest years. LSD0,05=1.16 . Combinations with the same letters are similar. Crude Protein for maintenance diet for dairy cows is indicated by the dashed line (18%). Weighted percent crude protein at Lake City trial location across three ....... 19 harvest years. LSD0_05=3.6. Combinations with the same letters are similar. Crude Protein for maintenance diet for dairy cows is indicated by the dashed line (18%). Weighted percent crude protein at East Lansing trial location across three. . . . 20 harvest years. LSDo,05=6.4(0t=0.05). Combinations with the same letters are similar. Crude Protein for maintenance diet for dairy cows is indicated by the dashed line (18%). Total acid detergent fiber for all three locations across three harvest ........... 21 years.Chatharn LSD0_05= 1.9. Lake City LSD0,05= 8.2. East Lansing LSD0_05 is not significant. Dashed line indicates minimum ADF required for lactating dairy cow (30%). Total acid detergent fiber at Chatham trial location across three harvest ...... 22 years. Chatham LSD0,05= 1.9. Combinations with the same letters are similar. Dashed line indicates minimum ADF required for lactating dairy cow (30%). vi 10 11 12 13 14 15 16 17 l8 19 Total acid detergent fiber at Lake City trial location across three harvest 22 years. Lake City LSD0,05= 8.2. Combinations with the same letters are similar. Dashed line indicates minimum ADF required for lactating dairy cow (30%). Total acid detergent fiber at East Lansing trial location across three ........... 23 harvest years. East Lansing LSDaos is not significant. Dashed line indicates minimum ADF required for lactating dairy cow (3 0%). Percent total digestible nutrients across all three locations for three ............ 24 harvest years. Chatham LSD0,05=4.2. Lake City LSD0_05=14.5. East Lansing LSDoos is not significantly different. Dashed line is the percent TDN needed for a lactating dairy cow (61 %). Percent total digestible nutrients at Chatham trial location across three ....... 25 harvest years. LSD0_05#.2. Combinations with the same letter similar. Dashed line is the percent TDN needed for a lactating dairy cow (61%). Percent total digestible nutrients at Lake City trial location across three ...... 26 harvest years.LSD0_05=14.5. Combinations with the same letters are similar. Dashed line is the percent TDN needed for a lactating dairy cow (61%). Percent total digestible nutrients at East Lansing trial location across ......... 28 three harvest years. LSD0,05 is not significantly different. Dashed line is the percent TDN needed for a lactating dairy cow (61%). Weighted relative feed value for each location across the three trial ............ 29 years. Chatham LSD0_05=9.6, Lake City Standard EI'I‘OI'ofls =3.3, and East Lansing Standard EITOI‘ons =4.3. Dashed line is the minimum RFV needed to maintain the diet of a medium producing dairy cow (100). Weighted relative feed value at Chatham trial site across three harvest. . . . . . ...30 years. LSDo_05=9.6. Combinations with the same letter are similar. Dashed line is the minimum RF V needed for a medium producing dairy cow (100). Weighted relative feed value at Lake City trial site across three harvest ....... 30 years. Standard Erroro_05=3.3. Combinations with the same letter are similar. Dashed line is the minimum RFV needed for a medium producing dairy cow (100). Weighted relative feed value at East Lansing trial site across three ............. 31 harvest years. Standard Erroro,05=4.3. Combinations with the same letter are not significantly different. Dashed line is the minimum RF V needed for a medium producing dairy cow (100). vii 20 21 22 23 24 25 26 27 28 29 30 Weighted net energy for lactation for each location across the three trial ...... 32 years. Chatham LSD0.05=0.05, Lake City Standard Error0.05 =0.l3, and East Lansing Standard Error0.05 =0.22. Dashed line is the minimum nutrient NEL for a legume-grass pasture during summer grazing (1.38 Meal/kg). Weighted net energy for lactation at the Chatham trial location across the. . .. 33 three trial years. Chatham LSD 0.05=0.05. Dashed line is the minimum nutrient NEL for a legume-grass pasture during summer grazing (1.38 Mcal/kg). Combinations with the same letter are similar. Weighted net energy for lactation at the Lake City trial location across ........ 34 the three trial years. Lake City Standard Erroro,05=0.13. Dashed line is the minimum nutrient NEL for a legume-grass pasture during summer grazing (1.38 Meal/kg). Combinations with the same letter are similar. Weighted net energy for lactation at the Chatham trial location across ........ 35 the three trial years. East Lansing Standard Erroro_05=0.22. Dashed line is the minimum nutrient NEL for a legume-grass pasture during summer grazing (1.38 Mcal/kg). Combinations with the same letter are similar. Grazing preference rating at Chatham trial location. Scale is 1 (0-20% of. 36 the plot consumed) to 5 (80-100% of the plot consumed). LSD0,05 = 0.34. Combinations with the same letters are similar. Grazing preference rating at Lake City trial location. Scale is 1 (0-20% of. . . .37 the plot consumed) to 5 (SO-100% of the plot consumed). LSD0_05= 0.76. Combinations with the same letters are similar. Grazing preference rating at East Lansing trial location. Scale is l (0-20%....37 of the plot consumed) to 5 (80-100% of the plot consumed). LSDOOS = 0.54 Combinations with the same letters are similar. Grazing preference rating for all locations across the three harvest years ...... 38 Scale is 1 (0-20% of the plot consumed) to 5 (80-100% of the plot consumed). LSD0,05= 0.33. Combinations with similar letters are similar. Stand persistence of combinations across four years post-seeding at ........... 40 Chatham trial location. 2009. Stand persistence of combinations across four years post-seeding at .......... 41 Lake City trial location. 2009. Stand persistence of combinations across four years post-seeding at ........... 42 East Lansing trial location. 2009. viii ADF BFTFES BFTOR BFTTIM CP DM EL F ALFES FALOR FALTIM LC NEL NDF RF V SATFES SATOR SATTIM TDN UPES LIST OF ABBREVIATIONS Acid Detergent Fiber Birdsfoot Trefoil and Meadow Fescue Birdsfoot Trefoil and Orchardgrass Birdsfoot Trefoil and Timothy Crude Protein Dry Matter Michigan State University Dairy, East Lansing, MI Medicago sativa spp. falcata and Meadow Fescue Medicago sativa spp. falcata and Orchardgrass Medicago sativa spp.fa1cata and Timothy Lake City Experiment Station, Lake City, MI Net Energy for Lactation Neutral Detergent Fiber Relative Feed Value Medicago sativa spp. sativa and Meadow F escue Medicago sativa spp. sativa and Orchardgrass Medicago sativa spp. sativa and Timothy Total Digestible Nutrients Upper Peninsula Experiment Station, Chatham, MI ix I. INTRODUCTION Alfalfa is often preferred over other forage legumes in feeding ruminants, due to the high feed value, high forage yields, and drought resistance in areas of livestock production (Dietz, 2003). In 2001 , alfalfa represented about 2.5% of the total agricultural hectarage in the USlwith approximately half of the alfalfa hectarage in the Upper Midwest and northern Great Plains (Riday and Brummer, 2002). In many cases, alfalfa is included in grazing systems to not only increase animal production (Campbell, 1963 and Barker et al., 1999), but to increase pasture and rangeland stand persistence (Berdahl et al, 1986), increase soil organic carbon (Bliss, 2003) in the pasture and rangeland and to provide enough nitrogen to sustain the pasture and rangeland for more years (Mortenson etaL,2004) Yellow-flowered alfalfa (Medicago sativa spp. falcata; hereafter falcata) was a relatively unknown legume in America until its rediscovery on the Norman “Bud” Smith Ranch in South Dakota in the early 1960’s (Smith, 1997). Since then forage plant geneticists, especially Dr. Arvid Boe of South Dakota State University, have been interested in testing falcata to learn how this legume can be incorporated in to today’s grazing and forage production. Many cow/calf operations in the North Central region of the United States function on a narrow profit margin resulting in the need for forage management systems‘with little costs past those involved in the initial input. There are many positive characteristics of purple flowered alfalfa (Medicago sativa L.; hereafter sativa) that are exhibited by falcata; even so, additional benefits of prolonged forage quality, greater pest resistance, and increased winterhardiness indicates falcata has good potential for use as a pasture legume in the North Central region of the United States (Boe et a1, 1998). F alcata alfalfa maintains forage yield (Boe et al.,1994) and quality longer, as well as being more persistent and shown to be more resistant to potato leafhopper (Empoascafabae) (PLH) than sativa cultivars in the North-central plains (Bortnem et al., 1993). F alcata shows higher grazing tolerance and stand persistence due to the branching roots and prostrate growth habit of the alfalfa (Berdahl et a1, 1989, Hendrickson and Berdahl, 2003). Even though sativa forage quality diminishes rapidly after the first bloom due to reduction in vegetative growth, lignification and leaf loss, falcata has been shown to maintain forage quality well-beyond flowering. Because falcata and sativa are the same species, it is convenient to produce germ crosses of these two subspecies, trying to get the best of both in one plant. The falcata- sativa crosses commonly produce greater forage yields and increased stand persistence compared to falcata or sativa on its own (Berdahl et al., 1986; Riday and Brummer, 2002). This characteristic of falcata could increase the flexibility of grazing and stockpiling in pastures for cow/calf and dairy producers reducing the need and added cost of hay production. Even with falcata’s known benefits, there are a few characteristics which may pose problems in grazing situations. One negative characteristic of the falcata gerrnplasm has been slower regrowth, dormancy during summer drought, and a more decumbent growth habit when stressed by the environment or heavy grazing conditions (Berdahl et al., 1989; Riday and Brummer, .2002). The slower regrowth may actually be attributed to falcata utilizing carbohydrate root reserves to maintain greater stand persistence following grazing (Berdahl et al., 1989). Seed production of falcata is a major issue due to problems with seed production and supply. Falcata has indeterminate growth habit, thus, each plant in a stand has the ability to produce the sickle-shaped seed pods at different times, causing some pods to be ready before others, increasing the risk of pods shattering while trying to harvest the seed (Boe et al., 1998). Falcata could be stockpiled in a field after September and fed to cattle during the winter months, if forage yield, quality (which will be lower than non-stockpiled material), and grazing palatability can be maintained. This study will evaluate the forage yield, forage quality, stand persistence, and palatability of falcata, sativa, and birdsfoot trefoil to provide information to farmers on the pasture potential of falcata compared to the sativa and birdsfoot trefoil varieties represented in the market. Hypothesis and Objectives The hypothesis that falcata will produce comparable forage yield, forage quality, and palatability to sativa (birdsfoot trefoil being the most palatable) and will have equal or greater stand persistence under rotational grazing compared to sativa or birdsfoot trefoil was the basis for this study. The objectives of this study were: 1) To evaluate palatability of falcata under grazing practices as compared to sativa and birdsfoot trefoil in three locations in Michigan. 2) To determine forage yield and stand persistence of falcata as compared to sativa and birdsfoot trefoil under the same grazing practices. 3) To determine forage quality and palatability of falcata compared to sativa and birdsfoot trefoil under free-choice grazing by dairy and beef COWS. II. MATERIALS AND METHODS Site Description Alfalfa grazing research trials were established at three locations across Michigan in spring 2005: Michigan State University Dairy in East Lansing (EL), Ingham County, Michigan (42°72’ N, 84°49’W) on a Marlette fine sandy loam; Lake City Experiment Station in Lake City (LC), Missaukee County, Michigan (44°35’ N, 85°18’W) on a Nester soil (fine sandy loam, mixed Typic Eutroboralfs); Upper Peninsula Experiment Station in Chatham (UPES), (46°35’ N, 86°92’W) on an Eben soil (sandy loam, very cobbly, mixed, fiigid Alfic Haplorthodes). The Lake City Experiment Station is located 140 miles north of East Lansing. The Upper Peninsula Experiment Station is located 370 miles north of East Lansing. Soil pH at the beginning of the trial in East Lansing was 7.3. Soil pH at the beginning of the trial in Lake City was 6.3. Soil pH at the beginning of the trial in Chatham was 7.4. Legume-Grass Combinations The three legumes chosen for this study were Medicago sativa spp. falcata (var. Yellowhead, developed by AG and Agri-Food Canada, Swift Current, Saskatchewan (Hendrickson et al., 2008, hereafter referred to as falcata), Medicago sativa spp. sativa (var. ZG9830, a grazing-type alfalfa, hereafter referred to as sativa) and Lotus corniculatus (var. Norcen, hereafter referred to as birdsfoot trefoil). Each legume was assigned to three plots: one plot contained the legume mixed with timothy grass (var. Dolina), one plot contained the legume mixed with meadow fescue (var. Laura), and one plot contained the legume mixed with orchardgrass (var. tekapo). Each plot was duplicated in a split-plot design with the legume as the main plot and the grass combination as the subplot. Each plot consisted of two separate passes with the seeder so destructive sampling (plant digs for stand persistence and grazing palatability) could be obtained from one-half of the plot, while forage quality and forage yield was collected from the undestroyed half of the plot (Appendix Table 1-3). Each legume-grass combination was replicated four times at each trial location. Establishment and Harvesting A single soil test (pH, phosphorus, potassium, magnesium, calcium, and cation exchange capacity) at each location was obtained from a subsample from 20 cores (20-25 cm depth) randomly located within the trial area. Soil amendments, if necessary, were made prior to establishment. Conventional tillage (moldboard plowing and fitting with a disc or a drag, followed by cultipacking) was used to prepare the soil for the seeding after weeds were killed using glyphosate. Prior to seeding, alfalfa seed was inoculated with Sinorhizobium meliloti and birdsfoot trefoil with Rhizobium Ioti (Urbana Lab., Urbana, IL). A 0.9 m-wide Carter self-propelled nursery cone seeder (Carter Manufacturing, Brookston, IN) was used to seed the plots. All of the trial locations were seeded in spring 2005, while the East Lansing plots were reseeded in August 2005 due to accidental herbicide application which killed the grass in the plots. Seeding rates were as follows: falcata at 7.26 kg/acre, sativa at 5.44 kg/acre, birdsfoot trefoil at 2.72 kg/acre, meadow fescue at 3.63 kg/acre, orchardgrass at 1.81 kg/acre, and timothy at 0.91 kg/acre. These seeding rates were used following the recommendations for each of these species in actual planting for pastures. The plots were harvested, starting in spring 2006, using a rotary flail harvester (Carter Manufacturing Co. Inc., Brookston, IN) to harvest the 0.9 by 7.6 m plots at a cutting height of 7 to 9 cm from the soil surface. Data Analysis Dry Matter Forage Yield, Dry matter content of harvested alfalfa was determined by collecting a subsample of harvested biomass which was weighed wet, dried at 60°C for 72 h, and weighed again. Dry matter was determined as: DM content (%) = Dry (g)/ Wet (g) x 100. Forage Quality Samples of alfalfa used for nutritive evaluation were collected at the time of harvest by clipping ~ 250 g of alfalfa from each plot prior to harvest or by capturing chopped biomass from the harvester. Samples were dried at 60°C for 48 h, and ground to pass through 1 mm screen in a Christy-Turner Lab Mill (Ipswich, Suffolk, UK). A minimum subsample of 20 g was retained for nutritive analysis. Each sample was scanned with a 6500 near-infrared spectrophotometer (N IRS, FOSS NIRSystems, Inc., Eden Prairie, MN) with wavelengths between 800 and 2500 nm. Reflected wavelengths were recorded. Crude protein (CP), acid detergent fiber (ADF), and neutral detergent fiber (N DF) were predicted from equations developed by the NIRS Consortium (Madison, WI) and the MSU Forage Lab. A randomly selected subset of samples was compiled based on the neighborhood and Global H statistic to validate the NIRS prediction of crude protein (CP), neutral detergent fiber (N DF), and acid detergent fiber (ADF). The Goering and Van Soest (1970) method was used for NDF and ADF determination with the addition of 1 mL of alpha-amylase to the neutral detergent solution for the breakdown of starch. Dry matter (DM) content was determined by drying 0.5 g of sample in ceramic crucibles at 100°C for 24 h. The samples were ignited in a muffle furnace at 500°C for 6 h to determine ash content. Total digestible nutrients for the NIRS were calculated using the equation: TDN: dNFC+dCP+2.25*dFA+dNDF-7 where digestible Non-fiber Carbohydrate= 0.98 (100-[(NDF-NDCIP)+CP+EE+Ash])*PAF dFA= digestible Crude Fat digestible Crude Protein= CP“ exp[-1.2*(ADICP/CP)] digestible Neutral Detergent Fiber= 0.75* (NDFn-L)* [1 -(L/NDFn)0'°°7] (National Research Council, 2001 p. 14) Relative Feed Value (RFV) for the NIRS was calculated using the following formula: (DDM*DMI)/ 1.29, where DDM is the digestible dry matter and DMI is the dry matter intake, estimated from ADF and NDF values, respectively. (Garcia et al., 2003). Net Energy for Lactation (N BL) is used to describe the amount of energy used by a cow for body maintenance, mild production, growth, and reproduction measured in megacalories per kilogram (Garcia et al., 2003). It can be calculated from TDN values using the following equation: 0.0245*TDN-0.12 (National Research Council, 2001, pg. 13). Stand persistence Stand persistence was determined after by digging and counting plants in two 0.093 m2 areas of the “destructive sampling plots” in the fall and spring of each year. In the spring, plots were visually rated for winter kill (scale: 1 to 10, 10-100% of the plot killed). Palatability Palatability was rated by using two people standing at opposing ends of each plot in the alleys between replications to assess the percentage of the plot consumed by a scale of 1 to 5 (1=0-20% of the plot consumed; 5:81-100% of the plot consumed). The length of each plot averaged 7.6 m. Statistical Analysis All data collected was tested for normality and unequal variances using PROC UNIVARIATE based on the Shapiro-Wilk statistic and data sets not normally distributed will be transformed. Analysis of variance was performed on all data with the PROC GLM procedure software version SAS 9.1 (SAS Institute, 2009) using the Kenward- Roger method for determining degrees of freedom. Means of forage yield, palatability, and weighted means of forage quality were separated by Fisher’s Protected Least Significant Difference (LSD) test at the 5% level of significance. If LSD was not able to be obtained due to missing data points, standard error was used instead to determine significant differences. Normality was not significantly different for any of the tested data once the cutting interactions were taken out of the equations. There were significant interactions between the sub plots, but not between the main effects. The cuttings caused a significant interaction due to the variability between East Lansing (13 cuttings total) and Lake City and Chatham (12 cuttings total each). Once normality was obtained, variances for forage quality were not statistically significant (a=0.05) by year or location. Differences in forage yield totals (number of cuttings) between the locations caused the forage yield data to be significantly different by location, resulting in data being run separately by location for forage yield. Table 1. ANOVA output for yield. Type 3 Analysis of Variance Error Source DF F Value Pr > F trt 592 5.01 <.0001 Loc 592 60.78 <.0001 Rep 592 0.82 0.4834 Year 592 7.70 0.0005 Cut 592 5.55 0.0009 III. RESULTS AND DISCUSSION Weather Records Chatham, MI The amount of precipitation received between April and October each year was less than the 30 year average, during the three harvest years in Chatham. Chatham averaged 4.1 cm of precipitation less than the 30 year average of 60.3 cm, during the growing season. The total precipitation for the 2006, 2007, and 2008 growing seasons was 51.6 cm, 59.4 cm, and 57.45 cm, respectively (Table 4). Table 2. Precipitation Data for Chatham, MI (cm). Chatham Norm“ 2006 2007 2008 Deviation Apr 6.3 5.1 16.3 16.0 6.2 May 8.0 14.5 5.5 9.3 1.8 June 9.2 2.2 5.2 9.4 -3.6 July 9.0 5.4 4.4 4.5 -4.3 Aug 9.0 8.4 2.0 2.7 -4.76 Sept 10.6 6.5 12.7 9.6 -1.0 Oct 8.2 9.6 13.3 5. 9 1.4 Total I 60.3 I 51.6 I 59.4 I 57.4 I -4.1 *30 year average During the course of the growing season (April to October), Chatham had slightly warmer than normal weather for 2006 and 2007, having a total average of 128°C and 144°C, respectively (Table 5). The grazing season for 2008 was only slightly below average at 123°C. The 30 year average temperature for Chatham (April to October, 1971-2000) was 12.6 °C. 10 Table 3. Temperature data for Chatham, MI (°C). Chatham Norm* 2006 2007 2008 Deviation April 3.9 6.7 2.8 3.9 0.6 May 10.6 10.6 12.8 7.2 -0.4 June 15.6 14.4 17.2 13.9 -0.4 July 18.3 20.6 18.9 17.8 0.7 Aug 17.8 17.8 18.9 18.3 0.6 Sept 13.3 12.8 16.7 15.6 1.7 Oct 8.3 6.7 13.3 9.4 1.5 Average 12.6 12.8 14.4 12.3 0.6 * 30 year average Table 4. Precipitation Data for Lake City, MI (cm). Lake City Norm* 2006 2007 2008 Deviation Apr 7.3 9.3 8.8 8.9 1.7 May 6. 8 14.1 5.6 4.5 1.3 June 7.9 12.7 9.5 18.2 5.6 July 8.3 5.5 3.1 10.0 -2.1 Aug 7.7 11.6 6.3 5.5 0.2 Sept 8.3 8.3 4.8 7.5 -1.4 Oct 6.7 11. 9 6.2 6.8 1.6 TotalI 52.9 I 73.4 F443 I 61.4 I 6.9 *30 year average Lake City, MI The amount of precipitation received between April and October in 2006 and 11 2008 was above the 30 year average, while 2007 was a below average year for precipitation in Lake City. The 30 year average for Lake City is 52.9 cm. In the 2006 and 2008 growing seasons, the amount of precipitation equaled 73.4 and 61.4 cm, respectively. In 2007, the amount of precipitation Lake City received was 44.3 cm. Overall, the average rainfall during the course of this study was above the 30 year average, having an average of 6.9 cm of precipitation more each year (Table 6). The average temperature during the course of the growing season (April to October), Lake City had slightly warmer than normal weather for 2006 and 2007, having a total average of 14.1°C and 14.4°C, respectively (Table 7). The grazing season for 2008 was only slightly below average at 130°C. The 30 year average temperature for Lake City (April to October, 1971-2000) was 13.3°C. Table 5. Temperature Data for Lake City, MI (°C) Lake City Norrn“ 2006 2007 2008 Deviation April 5.0 7.2 4.4 7.2 1.3 May 11.7 13.3 13.9 9.4 0.6 June 16.7 17.2 17.8 16.7 0.6 July 19.4 21.1 18.9 18.9 0.2 Aug 18.3 19.4 18.9 17.2 0.2 Sept 13.9 13.3 15.6 14.4 0.6 Oct 7.8 6.7 11.1 7.2 0.6 Average 13.3 14.1 14.4 13.0 0.5 * 30 year average East Lansing, MI The amount of precipitation received each year was an average of 11.73 cm greater than the 30 year average precipitation received from April to October, during the three harvest years in East Lansing. The 30 year average was 50.6 cm. The 2006, 2007, and 2008 growing season precipitation amounts were 68.4 cm, 60.7 cm, and 57.8 cm of precipitation, respectively (Table 8). The average temperature during the course of the growing season (April to October) for East Lansing was slightly warmer than normal for 2006, 2007, and 2008. 12 The total averages for each growing season were 15.7°C, 168°C, and 161°C respectively (Table 9). The 30 year average temperature for East Lansing (April to October, 1971-2000) was 15.2°C. Table 6. Precipitation Data for East Lansing, MI (cm) East Lansing Norm“ 2006 2007 2008 Deviation Apr 7.1 5.9 4.7 5.5 -1. 8 May 6.9 14.3 10.6 3. 5 2.5 June 9.0 7.4 14.1 12.2 2.3 July 7. 7 9.4 1.3 9. 5 -1.0 Aug 7.9 14.3 13.2 1.3 1. 7 Sept 6.4 7.5 5.3 21.4 5.1 Oct 5.6 9.7 11.5 4.6 3.0 TotalI 50.6 I 68.4 I 60.7 I 57.8 I 11.7 *30 year average Table 7. Temperature Data for East Lansing, MI (°C) East Lansing Norm“ 2006 2007 2008 Deviation April 7.2 10.0 6.7 10.0 1.7 May 13.9 14.4 16.1 12.8 0.6 June 18.9 18.9 20.6 20.0 0.9 July 21.1 22.8 21.1 21.7 0.7 Aug 20.0 21.1 21.7 20.6 1.1 Sept 15.6 15.0 17.8 17.8 1.3 Oct 10.0 7.8 13.9 9.4 0.4 Avera e 15.2 15.7 16.8 16.1 1.0 * 30 year average 13 Forage Yield Results Chatham Four grazing events took place in the 2006, 2007, and 2008 grazing seasons. The harvest plots were harvested with the Carter flail harvest within two days of the grazing event. The sativa-timothy combination resulted in the greatest forage yield, with a 3-year total of 20 metric tonnes of DM/a. The significantly lowest yielding combinations across the three years were birdsfoot trefoil-meadow fescue and birdsfoot trefoil-orchardgrass combinations with total forage yields averaging 11 and 10 metric tonnes of DM/a, respectively (Figure 1). The falcata-orchardgrass and falcata-meadow fescue combinations were significantly lower yielding than sativa-timothy combinations, but the falcata-meadow fescue combination was higher than the falcata-orchardgrass, but not different than the sativa-meadow fescue or sativa-orchardgrass combinations. The 21 , 1? 3184* I: S [“15” O .512 a, , E E9 ‘2“ b6 D 3 0 L bftfes bftor bfttim falfes falor faltim satfes sator sattim Figure 1. Accumulative forage yield per treatment across three harvest years in metric tonnes of dry matter per hectare. LSD0,05= 0.88 Treatments with the same letters are similar. (bft=birdsfoot trefoil, fal=falcata, sat=sativa, fes=meadow fescue, or=orchardgrass, tim=timothy) birdsfoot trefoil-grass combinations were significantly lower yielding than the falcata- grass and sativa-grass combinations (or=0.05). The average total forage yields ranged from 16 metric tonnes of DM/a (falcata-orchardgrass) to 20 metric tonnes of DM/a (sativa-meadow fescue). Lake City Four grazing events took place in the 2006, 2007, and 2008 grazing seasons. The mechanically harvested plots were harvested with the Carter flail harvest within two days of the grazing event. The legume-grass combination with the highest forage yield across E1 2008 El 2007 I 2006 Dry Matter (Metric Tonnes/a) J bftfes bftor bfttim falfes falor faltim satfes sator sattim Figure 2. Accumulative forage yield per treatment across three harvest years in metric tonnes of dry matter per hectare. LSD0_05= 0.99. Treatments with the same letters are similar. the three grazing years was the sativa-orchardgrass with a total 3—year forage yield of 28 metric tonnes of DM/a; however, this was not significantly different than the sativa- timothy combination. The combinations which had significantly lower total average forage yield were the birdsfoot trefoil-meadow fescue and birdsfoot trefoil-timothy combinations with total 3-year forage yields of 19.8 and 20.2 metric tonnes of DM/a, respectively (Figure 2). Other than the sativa-orchardgrass and sativa-timothy combinations, the other sativa-grass, falcata-grass, and birdsfoot trefoil-orchardgrass combinations were not significantly different (or=0.05) with an average total forage yield ranging from 23 metric tonnes of DM/a (birdsfoot trefoil-orchardgrass) to 28 metric tonnes of DM/a (sativa-meadow fescue). Dry Matter (Metric Tonnes/Hectare) E bftfes bftor bfttim falfes falor faltim satfes sator sattim Figure 3. Accumulative yield per treatment across three harvest years in metric tonnes of dry matter per hectare. Standard Error = 2.6 with a degree of freedom for error= 8. Combinations with an asterisk are similar. East Lansing Five grazing events took place during the 2006 and 2007 grazing seasons and three data collections were taken for the 2008 grazing season. The harvest plots were harvested with the Carter flail harvest within two days of the grazing event. Due to incomplete data collection during the fifth grazing events in 2006 and 2007, Least Significant Differences were not able to be analyzed, but standard error was used to analyze significant differences across the average total forage yields for the three years. Birdsfoot trefoil—timothy was the lowest yielding combination; an average total forage yield of 25 metric tonnes of DM/a was harvested. This combination is significantly different from the falcata-timothy, sativa-meadow fescue, and sativa-timothy combinations; average total forage yields of 33, 32.5, and 32 metric tonnes of DM/a, respectively (Figure 3). Using the standard error, the sativa-orchardgrass combination was significantly higher yielding (35.5 metric tonnes of DM/a) and the lowest yielding combination (birdsfoot trefoil-timothy) was significantly lower yielding at 25 metric tonnes of DM/a. The rest of the legume-grass combinations were not significantly different (or=0.05). Forage Quality Results The recommended forage nutritive values for this section are based upon a small framed grazing dairy cow. Table 7 shows the recommended forage nutritive values to maintain the diet of a small framed grazing dairy cow producing 27 kg milk a day. Table 8. Forage Nutritive Values for a grazing dairy cow (Amaral-Phillips et a1.) Acid Crude Detergent Net Energy Protein Fiber for Lactation 22% 19% 1.67 Mcal/kg_ Crude Protein The recommended minimum crude protein percentage for a maintenance diet for dairy cows is between 150 and 190 g/kg (National Research Council, pg. 50). As seen in 17 Figure 4, all of the legume-grass combinations (except the birdsfoot trefoil-grass combinations in Chatham) are above the required level for a maintenance diet of a dairy §§§§ Crude Protein (glkg) 8 O bflfeslilnr littimfafesfdorfdtimsafesselasatfim Figure 4. Total Weighted percent crude protein at all locations across three harvest years. Chatham LSD0,05=1.16. Lake City Standard Error=3.6. East Lansing Standard Error=6.4. Crude Protein for maintenance diet for dairy cows is indicated by the Badman ElLdeCity ‘IEastLatsingi ’e 3%. e dashed line (150 g/kg). cow for all of the locations across the three grazing years. None of the legume-grass combinations were significantly different from one another between locations, however 8 0 § Crude Protein (g/kg) § p—tl LII 0 U1 O bftfes bftor bfttim falfes falor faltim satfes sator sattim Figure 5. Weighted percent crude protein at Chatham trial site across three harvest years. Chatham LSD0,05=1.16. Treatments with the same letters are similar. Crude Protein for maintenance diet for dairy cows is indicated by the dashed line (150 g/kg)- statistical significance was seen within each location (a=0.05). The results of the Chatham crude protein are given in Figure 5. The averages of the weighted percentage for the birdsfoot trefoil-grass combinations are significantly lower than the other legume-grass combinations, but still near the minimum requirements for a small framed dairy cow; this may be due to the limited number of birdsfoot trefoil plants in this particular trial and their inability to provide sufficient nitrogen to the grasses, lowering the crude protein percentage for the birdsfoot trefoil-grass combinations. The sativa-grass and falcata-grass combinations are greater than the birdsfoot trefoil-grass combinations with sativa-meadow fescue and sativa-timothy resulting in the significantly higher crude protein percentage at 199 g/kg each, respectively. Figure 6 shows the weighted crude protein percentages for Lake City. Standard error was used to determine significant differences (or=0.05). The only birdsfoot trefoil- grass combination that was statistically lower than the sativa-grass and falcata-meadow 200 a b ’55 —_ a 150 ‘" r“ E ._‘ 0 § 100 . D- ..... g 7‘ ‘5 50 r :..T O bftfes bftor bfttim falfes falor faltim satfes sator sattim Figure 6. Weighted percent crude protein at Lake City trial site across three harvest years. Lake City Standard Error0_05=3.6. Treatments with the same letters are similar. Crude Protein for maintenance diet for dairy cows is indicated by the dashed line (180 g/kg). fescue combinations was the birdsfoot trefoil-orchardgrass combination. Of the sativa- grass and falcata-grass combinations, the falcata-orchardgrass and falcata-timothy combinations were statistically lower than the sativa-grass and falcata-meadow fescue combinations. The sativa-orchardgrass combination was also statistically lower than the sativa-meadow fescue and sativa-timothy combinations. In each of the legume-grass combinations, the orchardgrass combinations were each statistically lower than the other grass combinations. This may be due to the orchardgrass maturing slightly faster than the other grasses used in this study, lowering the crude protein percentage at the Chatham trial. The weighted crude protein percentages for East Lansing show that only the birdsfoot trefoil-orchardgrass and birdsfoot trefoil-timothy combinations were significantly lower in weighted crude protein percentages, with values of 175 g/kg and 182 g/kg, respectively. Birdsfoot trefoil-meadow fescue and all falcata-grass and sativa- grass combinations were not significantly different from each other (or=0.05). 250 a a a a. 200 in .5 150 e :3, 100’ a U 50 0 7 bftfes bfior bfttim falfes 7 falor faltim satfes sator sattim7 Figure 7. Weighted percent crude protein at East Lansing trial site across three harvest years. East Lansing Standard En0r0,05=6.4. Treatments with the same letters are similar. Crude Protein for maintenance diet for dairy cows is indicated by the dashed line (18%). 20 Acid Detergent Fiber The recommended maximum percent of acid detergent fiber from forage easy for a lactating dairy cow is near 30% ADF (Garcia et al., 2003); after 30%, the cow needs to expel energy to break down the ADF. As seen in Figure 8, all of the legume-grass combinations are near the required minimum, even for a maintenance diet for grazing dairy cows, let alone the minimum for a lactating dairy cow at peak production. Both Chatham and Lake City had statistical differences within the acid detergent fiber percentages at their locations, but East Lansing did not have statistical difference. There was no statistical significance seen between the locations (0t=0.05) (Figure 8). Acid Detergent Fiber (%) W O N N O U- 1 .— LII ._. O IJI O bftfes bftor bfttim falfes filor faltim satfes sator sattim Figure 8. Total weighted acid detergent fiber at all three locations across three harvest years. Chatham LSD0,05= 1.9. Lake City LSD0,05= 8.2. East Lansing LSD0_05 is not significant. Dashed line indicates minimum ADF required for lactating dairy cow (30%). Percent acid detergent fiber for Chatham is given in Figure 9. All of the birdsfoot trefoil-grass combinations were statistically different from the sativa-grass and falcata- grass combinations. The sativa-grass combinations and falcata-grass combinations were more similar statistically than the birdsfoot trefoil-grass combinations (or=0.05). The sativa-timothy and sativa-meadow fescue combinations resulted in the lowest percent 21 weighted ADF with 29.8% and 29.5%, respectively. The birdsfoot trefoil-orchardgrass combination resulted in the highest percentage of weighted ADF at 33.6% ADF. This may be due to the earlier maturity of the orchardgrass variety compared to timothy and meadow fescue. U) U! v-‘NNbJ UIOUIO Acid Detergent Fiber (%) O U. 7 7 bftfes Pflor bfttim 938537,, faloL faltim _ 7 satfes 77 sator 7 isattim Figure 9. Total weighted acid detergent fiber at Chatham trial site across three harvest years. Chatham LSD0_05= 1.9. Treatments with the same letters are similar. Dashed line indicates minimum ADF required for lactating dairy cow (3 0%). Acid Detergent Fiber (%) bflfes bftor bfttim falfes falor faltim satfes sator sattim Figure 10. Total weighted acid detergent fiber at Lake City trial site across three harvest years. Lake City Standard Erroro_05= 8.2. Treatments with the same letters are similar. Dashed line indicates minimum ADF required for lactating dairy cow (3 0%). 22 The Lake City results are given in Figure 10. The birdsfoot trefoil- meadow fescue, birdsfoot trefoil-timothy, sativa-meadow fescue, and sativa-timothy combinations were statistically significant from the rest of the combinations with percent ADF values of 29.98%, 29.06%, 29.08% and 28.89%, respectively (or=0.05). Each of the falcata- grass combinations, sativa-orchardgrass, and birdsfoot trefoil-orchardgrass combinations were not statistically different from one another. All combinations resulted in nearly ideal ADF content for lactating dairy cows. East Lansing results are given in Figure 11. There were no significant differences between combinations in percentage of weighted ADF (Least Significant Difference or=0.05). The percentage of the weighted ADF ranged from 31.02% (birdsfoot trefoil- meadow fescue) to 34.27% (falcata-orchardgrass; Figure 11). The range of ADF content for all locations was near or slightly higher than the ideal 30% ADF values needed for lactating dairy cows. Orchardgrass-legume combinations resulted in the highest ADF compared to all other treatment combinations. I t——~ .— La) 0 N U! N O .— (II _. C) 1 Acid Detergent Fiber (%) Ur bftfes bfior bfitim falfes falor faltim satfes sator sattim Figure l 1. Total Weighted Acid Detergent Fiber at East Lansing trial site across three harvest years. East Lansing LSD0,05 is not significant. Dashed line indicates minimum ADF required for lactating dairy cow (30%). 23 Total Digestible Nutrients Total Digestible Nutrients were determined using the following formula: TDN= dNFC+dCP+(2.25*dFA)+dNDF-7 (National Research Council, p. 14). Total Digestible Nutrients takes into account the digestible nonfibrous carbohydrate concentration, digestible crude protein concentration, and the digestible neutral detergent fiber concentration to better understand the amount of energy available to the grazing cow. The recommended minimum percent of total digestible nutrients required to maintain the diet of a lactating dairy cow producing 40 kg of milk per day is 61% (National Research Council, p. 16). Results of TDN are given in Figure 12. Both Lake City and East Lansing resulted in legume-grass combinations near or above 61%; however, the combinations at Chatham did not. This may be attributed to the fact that the Chatham location had a 30 plus-day rest period between grazing events and with the cooler temperatures, the plants 1°\°‘ 80 " 7 r i 7 r r r 3 70 ' 5 ,, , '5 6° ‘ Scream 1 = i z ,, I 2 5° emkecny I .D '5 40 a .East I I a 30 Wire 3 ‘ g 20 - Percent E" TDN I 10 needed for ‘ W 7 = 1 lactating ‘ 0 V ‘ dairy cow bflfes bftor bfttim falfes falor faltim satfes sator sattim Figure 12. Weighted percent total digestible nutrients at all three locations for three harvest years. Chatham LSD0,05=4. 2. Lake City LSD0,05=14. 5. East Lansing LSDoos is not significantly different. 24 matured much faster than the other two locations, decreasing the crude protein and increasing the NDF fiber content. Lake City was on a 30-day rest period between grazing events, resulting in higher crude protein and lower NDF percentages. East Lansing, overall, had higher TDN values due to the fact that a grazing event occurred every 3-4 weeks (shorter rest period), keeping combinations in a less mature growth stage, resulting in the higher crude protein and lower NDF values. There was no statistical difference between the weighted TDN percentages for each location; however, there was statistical difference within the Chatham and Lake City combinations (a=0.05). None of the legume-grass combinations at Chatham met requirements for maintaining the diet of a lactating dairy cow. Perhaps, due to the fact that Chatham had a longer rest period between grazing events, the plants may have remained in a mature state for longer time, decreasing the crude protein and increasing NDF percentages. Another w o I I I I I I I I I I I I I Total Digestible Nutrients (%3 N A o o ' I I I I I | I I I I I I I I 1' I I I I I I I —n O I I I I 0 ' , . e. . . a bftfes bftor bfttim firlfes falor faltim satfes sator sattim 7. I I Figure 13. Weighted percent total digestible nutrients at Chatham trial site across I three harvest years. LSD0,05=4.2. Combinations with the same letter similar. Dashed line is the percent TDN needed for a lactating dairy cow (61%). 25 factor may have been a lower legume content compared to the amounts of legume found at the Lake City and East Lansing trial locations. Each of the birdsfoot trefoil-grass combinations were below 30%, with the lowest TDN percentage for Chatham being the birdsfoot trefoil-orchardgrass combination (23.43%) (Figure 13). The sativa-grass and falcata-grass combinations were all statistically similar, except for the falcata- orchardgrass combination with a TDN percentage of 37.3%. The highest TDN percentage was the sativa-timothy combination at 49.4%. The Lake City plots had a 30-day rest period between grazing events, resulting in slightly lower crude protein and lower NDF percentages; these lower percentages resulted in higher TDN percentages across each of the combinations at Lake City compared to Chatham. Since LSD was not able to be calculated due to missing data points, standard error was used. Standard error indicates all combinations at Lake City 90 b b ab ': '— 60 sip n—L— -!- ___ -—,,—,! q,— A! L—i-L— E —7— as 1‘,- a g— n-fihulg an, —é—' — .- n-g'jui n: c I W-p-Ur COO Total Digestible Nutrients (%; N O I 7 bftfes bftor 7 bfltim7 falfes 7 falor faltim satfes sator 7 sattim Figure 14. Weighted percent total digestible nutrients at Lake City trial site across three harvest years. LSD0,05=14.5. Combinations with the same letter are similar. Dashed line is the percent TDN needed for a lactating dairy cow (61%). 26 are above the line of required percent TDN to maintain the diet of a lactating dairy cow. Again, the birdsfoot trefoil-grass combinations were statistically lower than the sativa- grass and falcata-grass combinations with TDN percentages of 47.63% (birdsfoot trefoil- meadow fescue), 52.42% (birdsfoot trefoil-orchardgrass), and 49.62% (birdsfoot trefoil- timothy), respectively (Figure 14). The falcata-grass combinations and the sativa- meadow fescue combination were not statistically different from one another; however, the sativa-orchardgrass and sativa-timothy combinations were statistically greater than any of the other combinations at Lake City. The TDN percentages for sativa- orchardgrass and sativa-timothy were 69.55% and 66.76%, respectively. Even though the falcata and sativa combinations were statistically similar, only the sativa combinations had TDN percentages above the 61% requirement for maintaining the diet of a lactating dairy cow. East Lansing forage combinations were grazed on a shorter schedule than the other two locations, allowing only 3-4 weeks of rest before the combinations were grazed again. This may help attribute to the higher TDN percentages seen at East Lansing than the other two locations. By grazing the forage combinations more frequently, the plants did not have as much time to mature, resulting in higher crude protein and NDF percentages and, consequently, higher TDN percentages. There were not any statistically significant differences between any of the forage combinations at East Lansing (or=0.05). The lowest TDN percentages were evident in the birdsfoot trefoil-orchardgrass and birdsfoot trefoil-timothy combinations at 55.35% and 56.24%, respectively (Figure 15). The two highest TDN percentages were obtained with falcata-meadow fescue and sativa- timothy combinations at 73.66% and 74.35%, respectively. 27 90 w 4:- UI 0" \l O O O O 0 Total Digestible Nutrients (%_ N O bftfes 77bfior7 bfttiln falfes 7falor7 faltim satfes sator sattim Figure 15. Weighted percent total digestible nutrients at East Lansing trial site across three harvest years. LSDoos is not significantly different. Dashed line is the percent TDN needed for a lactating dairy cow (61%). Relative Feed Value Relative Feed Value is an index by which hay and haylages are ranked based in calculations using digestible dry matter and dry matter intake, where (DDM*DMI)/1.29=RFV (Garcia et al., 2003). Legumes that are in full bloom will have an RFV of 100; legumes not in bloom will have a higher RFV rating, resulting in better quality hay and haylage. For this paper, an RFV of 100-120 will be used to show the amount of RF V needed to maintain the diet of a medium producing grazing dairy cow (Weiss et al., 1999). Relative feed value results are given in Figure 16. Each of the locations produced legume-grass DM that was at or above the RFV needed to maintain the diet of a medium producing grazing dairy cow. At each location, the legume-orchardgrass combination had the lowest scoring RFV. This may be due to the orchardgrass maturing faster than the other grasses and legumes, reducing the ADF and NDF values, resulting in 28 I160 140 I 120 7' W 7* . . El Chatham I 100 - I CI Lake City 80 . I East I 60 Esme, L I 40 — Minimum . RFV for I 20 Medium producing I 0 dairy cow bftfes bftor bfltim falfes falor faltim satfes sator sattim I Figure 16. Weighted relative feed value for each location across the three trial years. Chatham LSD0_05=9.6, Lake City Standard Error0.05 =3.3, and East Lansing Standard Erroroos =4.3. Dashed line is the minimum RFV needed to maintain the diet of a medium producing dairy cow (100). lower DDM and DMI values. RFV was similar among locations, but there were differences within each location. All of the combinations met the minimum RF V needed to maintain the diet of a medium producing dairy cow (Figure 17) at the Chatham trial location. Within each legume-grass combination, the legume-orchardgrass combination ranked lowest with RFV of 113 (birdsfoot trefoil-orchardgrass), 121 (falcata-orchardgrass), and 127 (sativa- orchardgrass), respectively. This may be due to the orchardgrass maturing a little earlier than the legumes and other grasses, resulting in lower DDM and DMI values. The highest ranking legume-grass combinations were the falcata-meadow fescue, falcata- tirnothy, sativa-meadow fescue, and sativa-timothy combinations with RFV of 142, 141, 146, and 146, respectively. 29 160 ‘1407777 w fiM' ._ e , ,7, e %120 _ , L ’i—F - —~ 100 , —- -—e: —I- —;_,_. L- -— -- g —- ,—,,--— -- -:— —- —,1— — 80- _ re 1 r ' , ' L . 60 , , i e» e L. ~ e I 40 - , LL, - , - I 20 — —— ~ ~~ I— , v 0 bftfes bflor bfttim falfes falor faltim satfes sator sattim Figure 17. Weighted relative feed value at Chatham trial site across three harvest years. LSD0_05=9.6. Combinations with the same letter are similar. Dashed line is the minimum RFV needed for a medium producing dairy cow (100). Relative feed values at the Lake City trial location also were above the minimum RFV for a medium producing dairy cow, as seen in Figure 18. In each legume-grass I80 , ,1 , , - 60* ,, ,, e ,, ,, I407 , , .7 , , ,,, ,,, 4- I20 ,, , 0 bftfes bftor bfttim falfes falor faltim satfes sator sattim Tguragwgghted relative feed Value at Lake City trial site across thrieeiharvesf 7 years. Standard Erroro_05=3.3. Combinations with the same letter are similar. Dashed line is the minimum RFV needed for a medium producing dairy cow (100). 30 combination, the orchardgrass produced the lowest RFV with values of 124 (birdsfoot trefoil-orchardgrass), 127 (falcata-orchardgrass), and 128 (sativa-orchardgrass), respectively. This may, again, be due to the orchardgrass maturing earlier than the other grasses or legumes. The highest ranking RFV at Lake City were the birdsfoot trefoil- timothy, falcata-meadow fescue, sativa-meadow fescue, and sativa-timothy combinations with RFV of 146, 144, 144, and 148, respectively. Figure 19 shows that the relative feed values for the East Lansing trial location were also all above the minimum RFV to maintain the diet of a medium producing dairy cow. The East Lansing trial also had the legume-orchardgrass combinations rank lowest with RFV of 103 (birdsfoot trefoil-orchardgrass), 110 (falcata-orchardgrass), and 112 (sativa-orchardgrass), respectively. The highest ranking RFV at the East Lansing trial bftfes bftor bfttim falfes falor faltim satfes sator sattim Figure 19. Weighted relative feed value at East Lansing tlial site across three harvest years. Standard Error0_05=4.3. Combinations with the same letter are similar. Dashed line is the minimum RFV needed for a medium producing dairy cow (100). 31 were falcata-meadow fescue, sativa-meadow fescue, and sativa-timothy with values of 131, 135, and 132, respectively. Net Energy for Lactation Net Energy for Lactation (NEL) is used to describe the amount of energy used by a cow for body maintenance, milk production, growth, and reproduction measured in megacalories per kilogram (Garcia et al., 2003). It can be calculated from TDN values using the following equation: 0.0245*TDN-0.12 (National Research Council, 2001, pg. 13). For a mixed, mostly grass pasture, during the summer production months, the average NEL value is 1.38 Mcal/kg (Amaral—Phillips et al., 1997) so that is the value used to compare the combinations with in this study. The total weighted Net Energy for Lactation for each location is shown in Figure 20. Chatham and Lake City had all legume-grass combinations above the minimum ts a ._0 Batman E IIdtthy IEast 8 E Net Energy for Lactation (Mcal/kg 5‘ St Figure 20. Weighted net energy for lactation for each location across the three trial years. Chatham LSD0,05=0.05, Lake City Standard Error0.05 =0.l3, and East Lansing Standard Error0.05 =0.22. Dashed line is the minimum nutrient NEL for a legume- grass pasture during summer grazing (1.38 Meal/kg). 32 nutrient NEL for legume-grass pastures; however, East Lansing did not have allcombinations above the common nutrient NEL. This may have been due to the shorter rotational period between grazing events, resulting in the plants not being able to mature as much as the plants at Lake City and Chatham. The grass which had the lowest NEL was orchardgrass at each location with each legume combination. This may have been due to the orchardgrass maturing faster than the other grasses and legumes. All of the combinations at the Chatham trial location were above the minimum nutrient NEL for a legume—grass pasture during the summer grazing months (Figure 21). In each legume-grass combination, the grass with the lowest NEL was orchardgrass, no 1.52 1.50 1.48 7 1.46 1.44 1.42 7 1.40 1.38 1.36 1.34 Net Energy for Lactation (Mcal/kg bfifes bftor bfttim falfesf alor faltim satfes sator sattim Figure 21. Weighted net energy for lactation at the Chatham trial location across the three trial years. Chatham LSD0,05=0.05. Dashed line is the minimum nutrient NEL for a legume-grass pasture during summer grazing (1 .38 Mcal/kg). Combinations with the same letter are similar. matter the legume combination. The NEL values for the birdsfoot trefoil-orchardgrass, falcata-orchardgrass, and sativa-orchardgrass combinations were 1.41 Meal/kg, 1.43 Mcal/kg, and 1.45 Meal/kg, respectively. The highest NEL values belonged to the falcata-meadow fescue, sativa-meadow fescue, and sativa-timothy combinations with NEL values of 1.49 Meal/kg each, respectively. 33 The Lake City trial NEL values are given in Figure 22. The lowest legume-grass combinations for nutrient NEL are those combinations with orchardgrass. The NEL values for the birdsfoot trefoil-orchardgrass, falcata-orchardgrass, and sativa- orchardgrass combinations are 1.41 Meal/kg, 1.42 Meal/kg, and 1.43 Meal/kg, 1.52 ‘a: Van D EEEQ 1 .42 Net Energy for Lactation (Meal/kg) QQE bftfes bftor bflfim fdfes fala faltim safes saor sattim Figure 22. Weighted net energy for lactation at the Lake City trial location across the three trial years. Lake City Standard Error 0.05:0-13- Dashed line is the minimum nutrient NEL for a legume-grass pasture during summer grazing (1.38 Meal/kg). Combinations with the same letter are similar. respectively. The rest of the combinations, except for the falcata-timothy combination (NEL=1.47 Meal/kg), have NEL values near 1.49 Mcal/kg. Two of the East Lansing legume-grass combinations were below the minimum nutrient NEL for grass-legume pastures (Figure 23). This may be due to the shorter resting period between grazing events in East Lansing. The legume-grass combinations did not have as much time to regrow and mature in East Lansing, leading to less energy acquired from the plant when grazed. Even so, looking at each legume-grass combination, the grass with the lowest NEL values is orchardgrass. Even with the shorter resting period, it appears that the orchardgrass still had time to mature, leading to the smaller NEL values. The NEL values for the birdsfoot trefoil-orchardgrass, falcata- 34 Net Energy for Lactation (Meal/kg bftfes bftor bfttim fal fes falor faltim satfes sator sattim Figure 23. Weighted net energy for lactation at the East Lansing trial location across the three trial years. East Lansing Standard Error 0.05:0.22. Dashed line is the minimum nutrient NEL for a legume-grass pasture during summer grazing (1.38 Meal/kg). Combinations with the same letter are similar. orchardgrass, and sativa-orchardgrass combinations are 1.34 Meal/kg, 1.34 Meal/kg, and 1.40 Meal/kg, respectively. The legume-grass combinations with the greatest nutrient NEL were falcata-meadow fescue and sativa-meadow fescue with NEL values of 1.42 Meal/kg and 1.43 Meal/kg, respectively. Palatability Chatham Each legume-grass combination was palatable based on the grazing preference rating scale used where 1 equaled 0-20% of the plot consumed and 5 equaled 80-100% of the plot consumed. The legume-grass combinations that were most preferred were birdsfoot trefoil-orchardgrass, birdsfoot trefoil-timothy, and sativa-meadow fescue with average grazing preference ratings of 3.42, 3.4, and 3.38, respectively. The legume-grass combinations that were least preferred were sativa-timothy and falcata-meadow fescue with average grazing preference ratings of 2.67 and 2.66, respectively (Figure 16). These 35 bftfes bftor bfitim falfes falor faltim satfes sator sattim Preference Scale (amount of plot consumed) Figure 24. Grazing Preference Rating at Chatham trial site. Scale is 1 (0-20% of the plot consumed) to 5 (SO-100% of the plot consumed). LSD0,05= 0.34. Combinations with the same letters are similar. two may have been least preferred due to having fewer legumes in each of the combinations. Having fewer legumes may have deterred the cows from eating these combinations. All of the other legume-grass combinations were not statistically different (or=0.05). Lake City Each legume-grass combination was palatable based on the grazing preference rating scale used where 1 equaled 0-20% of the plot consumed and 5 equaled 80-100% of the plot consumed. The legume-grass combination that was most preferred was the sativa-orchardgrass combination with an average grazing preference rating of 3.94. The legume-grass combination that was least preferred was the falcata-meadow fescue combination with an average grazing preference rating of 3.09 (Figure 17). All of the rest of the legume-grass combinations were not statistically different (01:0.05). 36 Preferance Scale (amount of plot consumed) bftfes bfior bfttim falfes falorfa ltim satfes sator sattim Figure 25. Grazing Preference Rating at Lake City trial site. Scale is 1 (0-20% of the plot consumed) to 5 (SO-100% of the plot consumed). LSD0.05= 0.76. Combinations with the same letters are similar. East Lansing All legume-grass combinations were palatable based on the grazing preference rating scale used where 1 equaled 0-20% of the plot consumed and 5 equaled 80—100% of the plot consumed. The legume-grass combination that was most preferred was the 4.5 L 2.5 Preference Scale (amount of plot consumed) C. I o L . bftfes bftor bfttim falfes falor faltim satfes sator sattim Figure 26. Grazing Preference Rating at East Lansing trial site. Scale is 1 (0-20% of the plot consumed) to 5 (SO-100% of the plot consumed). LSDolos= 0.54. Combinations with the same letters are similar. 37 birdsfoot trefoil-meadow fescue combination with an average grazing preference rating of 4.62. The legume-grass combination least preferred was the falcata-orchardgrass combination with an average grazing preference rating of 3.75 (Figure 18). All other legume-grass combinations were not statistically different (ot=0.05) except the falcata orchardgrass combination which was less palatable than two of the sativa grass combinations and two of the birdsfoot trefoil grass combinations. Total Palatability based on the grazing preference rating scale used where 1 equaled 0- 20% of the plot consumed and 5 equaled 80-100% of the plot consumed resulted in legume-grass combinations that were most preferred being birdsfoot trefoil-orchardgrass and birdsfoot trefoil-timothy combinations with average grazing preference ratings of 3.59 and 3.55, respectively. The legume-grass combination least preferred was the bftfes bftor bfitim falfes falor faltim satfes sator sattim Figure 27. Grazing Preference Rating at all locations across the three harvest years. Scale is 1 (0-20% of the plot consumed) to 5 (80-100% of the plot consumed). LSD0,05= 0.33. Combinations with the same letters are similar. 38 falcata—meadow fescue combination with an average grazing preference rating of 2.91 (Figure 19). The remainder of the legume-grass combinations were not statistically different (or=0.05). In each of the legume-grass treatments, the meadow fescue combinations were statistically lower than the other grasses offered. This lower grazing preference may be due to the hairless leaves of the meadow fescue; the cows may nto like to feel of the grass as they eat. These grazing preference ratings show that birdsfoot trefoil-grass combinations were slightly preferred over the falcata-grass and sativa-grass combinations, but these two combinations were not significantly different fi'om one another (or—70.05). Stand Persistence “Stand persistence is a complex trait affected by a large number of factors, including grazing, mechanical harvesting equipment, intensity of harvest management, diseases and pests, cold weather, inadequate dormancy, and inter- and intraspecies plant competition (Riday and Brummer, 2006).” This was evident in this study, including sampling error, which resulted in extremely varied stand persistence data, resulting in several data points discarded at each location. Statistics were not performed on remaining data; however, this section shows the trend of estimated stand persistence at each location across the three grazing years. Chatham The stand persistence of the combinations from fall 2006 and spring 2009 are shown in Figure 20. There is a trend over the course of three grazing seasons with the number of legumes per two-0.093 m2 sampling areas having decreased. The sativa-grass and falcata-grass combinations had initial plant populations ranging from 79 (sativa- 39 timothy) to 36 (sativa-orchardgrass) legumes per two-0.093 m2 area. At the end of the three grazing seasons, falcata-grass and sativa-grass combinations were more similar in numbers of legumes per two-one foot2 sampling areas with the range being from 45 falcata-orchardgrass) to 26 (falcata meadow fescue) legumes per two-0.093 m2 areas. hi 7 _____ ”Th __ _ _ _ ' T— _ '-'"I I 80 70~~ — Number of Legumes per 0.093 ni Fall 2006 Spl'l_'rlg2009 Figure 28. Stand Persistence of treatments across four years post-seeding in Chatham. The birdsfoot trefoil-grass combinations had the lowest stand persistence legume numbers ranging from 12 (birdsfoot trefoil-orchardgrass) to 30 (birdsfoot trefoil-timothy) legumes per two-0.093 m2 areas at the start of the study; however, the birdsfoot trefoil did not lose as many legumes over the course of the 3 years of grazing, having decreased by only 1 (birdsfoot trefoil-orchardgrass), 2 (birdsfoot trefoil-meadow fescue), and 19 (birdsfoot trefoil-timothy) legumes per two-0.093 m2 areas. Lake City The stand persistence of the combinations from fall 2006 and spring 2009 are shown in Figure 21. Some of the legume-grass combinations declined, but others 40 increased dramatically, which is likely a result of the sampling error. At the start of the study (spring 2006), there was no defined difference between each of the legume—grass combinations, with a range from 9 (birdsfoot trefoil-orchardgrass) to 33 (sativa-timothy) \ o bftfes \ . N 30 . I! bftor 3:: 2' 25 bfttim 2 er , _ ~ . E 2 0 . falfes E” . - falor ,5 15 h . W. -- faltim 2 g 10 I , I —-satfes Z 5 —sator 0 r 1 —sattim F all 200" Spring 2008 Figure 29. Stand Persistence of treatments across four years post-seeding in Lake City. legumes per two-0.093 In2 areas. At the end of the three grazing seasons, the range was from 9 (birdsfoot trefoil-orchardgrass) to 36 (sativa-orchardgrass) legumes per two-0.093 m2 areas with no defined difference between the stand persistence of the three legumes. East Lansing The stand persistence of the combinations from fall 2006 and spring 2009 are shown in Figure 22. There is a trend over the course of three grazing seasons with the number of legumes per two-0.093 m2 sampling areas having decreased. Again, each legume-grass combination showed significant variance after the three grazing seasons, but were similar at the start of the study, with a range from 29 (falcata-meadow fescue) to 40 (birdsfoot trefoil-meadow fescue) legumes per two-0.093 m2 areas. At the start of the 41 study, birdsfoot trefoil-grass combinations had the greatest number of legumes; however, after the second year of grazing, the birdsfoot trefoil-grass combinations decreased rapidly in legume numbers. The sativa-grass and falcata-grass combinations were initially similar to the birdsfoot trefoil-grass combinations, but after the entire three years of grazing, the legume mortality was not as great as birdsfoot trefoil. At the end of the 45 - 40 e a O bfifes ; ~ bftor a. bfttim g ~+ "falfes ED falor g 7 faltim 3 15 " -6-satfes E 10 W W W “sator z 5 _ —sattilll 0 I l i 1 Fall 2006 Spring 2009 Figure 30. Stand Persistence of treatments across four years post-seeding in East Lansing. study, the content of sativa per two-0.093 m2 area was greater than birdsfoot trefoil, with l (birdsfoot trefoil-orchardgrass) to 33 (sativa-timothy) legumes per two-0.093 m2 area. The falcata-grass and sativa-grass combinations were similar, except for the falcata- orchardgrass combination, which was more similar to the birdsfoot trefoil-grass combinations at the end of the three grazing seasons. 42 IV. CONCLUSION The results of this study support the hypothesis of this study: falcata will produce comparable forage yield, forage quality, and palatability to sativa (birdsfoot trefoil being the most palatable) and will have equal or greater stand persistence under rotational grazing compared to sativa or birdsfoot trefoil. Forage yields from each location indicated that falcata-grass combinations were not statistically different than sativa-grass combinations, except for the sativa-timothy combination at the Chatham and East Lansing trials and the sativa-orchardgrass combination at the Lake City trial. In each location, the birdsfoot trefoil-grass combinations yielded less forage than the falcata-grass or sativa-grass combinations. The total weighted crude protein data show that combinations with lower than nutritionally required value of 18% were birdsfoot trefoil-grass combinations at Chatham. When analyzed by location, statistical differences were shown between birdsfoot trefoil- grass combinations; birdsfoot trefoil-grass combinations being significantly lower in dry matter forage yield than sativa or falcata-grass combinations. However, there was no statistical difference between the falcata-grass and sativa-grass combinations. The total weighted ADF results show that all combinations were above the 30% required value for the maintenance diet of dairy cows. There were no statistical differences between locations. When analyzing locations, there were statistical differences at Chatham and Lake City, but not at East Lansing with differences between the falcata-grass and sativa-grass combinations. Analysis of total weighted TDN results showed most of the combinations at Lake City and East Lansing were above the value required to maintain the diet of a dairy cow 43 (61%), but at Chatham, the forage combinations were below this value. This was likely due to a longer rotation period between grazing events than at the other two locations. The falcata-grass and sativa-grass combinations showed no statistical difference within each trial location. Relative feed values showed that all the combinations at each location had an RFV above the minimum requirement for a medium milk producing cow (100 RFV). At each location, the legume-grass combinations with the lowest RF V ratings were those legumes paired with orchardgrass. The lower RF V for orchardgrass may be due to the orchardgrass having been more mature at the grazing date than the other grasses and legumes. All of the net energy for lactation values at Chatham and Lake City were above the minimum nutrient NEL for legume-grass pastures (1 .38 Mcal/kg); however, East Lansing had two combinations below the common nutrient NEL. This may have been due to the shorter rotational period between grazing events, resulting in the plants not being able to mature as much as the plants at Lake City and Chatham. The grass which had the lowest NEL was orchardgrass at each location with each legume combination. This may have been due to the orchardgrass maturing faster than the other grasses and legumes. Each of the legume-grass combinations were related as palatable at each location. The most preferred combinations were the birdsfoot trefoil-orchardgrass and birdsfoot trefoil-timothy combinations with average grazing preference ratings of 3.59 and 3.55, respectively; the falcata-meadow fescue combination was the least preferred with an average grazing preference rating of 2.9. Stand persistence data was not analyzed due to 44 sampling error, but each of the legume-grass combinations showed a trend of decreasing legume numbers per two-0.093 m2 sampling areas. Birdsfoot trefoil had the least number of legumes after three grazing seasons, while falcata and sativa were more similar. The results of this study support the hypothesis and objectives stated before the start of the study. In summary, this study showed that Medicago sativa spp. falcata could be used as a legume in Michigan’s grazing systems without sacrificing forage yield or forage quality; however, this data does not show an advantage over Medicago sativa spp. sativa. 45 V. APPENDIX Table 9. East Lansing Plot Map (The light grey areas are seeding errors) N" vernal Vernal vernal vernal 101 falfes 201 Faltim 301 bftfes 401 faltim 101 falfes 201 Faltim 301 bftfes 401 faltim 102 falor 202 Bftor 302 falfes 402 bftor 102 falor 202 Bftor 302 falfes 402 bftor 103 faltim 203 Satfes 303 falor 403 satfes 103 faltim 203 Satfes 303 falor 403 satfes 104 bftfes 204 Sator 304 bfttim 404 sattim 304 bfttim 404 Sator 104 bftfes 204 sattim 105 Bftor 205 Bftfes 305 bftor 405 falfes 105 Bftor 205 Bftfes 305 bftor 405 falfes 106 bfttim 206 Sattim 306 faltim 406 sator 106 bfttim 206 Sattim 306 faltim 406 sator 107 satfes 207 Falfes 307 sator 407 falor 107 satfes 207 Falfes 307 sator 407 falor 108 sator 208 Bfttim 308 sattim 408 bftfes 108 sator 208 Bfttim 308 sattim 408 bftfes 1 09 sattim 209 Falor 309 satfes 409 bfttim 109 sattim 209 F alor 309 satfes 409 bfttim vernal Vernal vernal vernal combination Mume grass cede yellowhead Laura falfes yellowhead Dolina faltim yellowhead tekapo falor norcen Laura bftfes norcen Dolina bfttim norcen tekapo bftor ZG 9830 Laura satfes ZG 9830 Dolina sattim ZG 9830 tekapo sator 46 109 109 108 108 107 107 106 106 105 105 104 104 103 103 102 102 101 101 NA 101 101 102 102 103 103 104 104 105 105 106 106 107 107 108 108 109 109 border sattim sattim sator sator satfes satfes bfttim bfttim bflor bflor bftfes bftfes faltim faltim falor falor falfes falfes border vernal vernal falfes falfes falor falor faltim faltim bftfes bftfes bflor bfior bfttim bfttim satfes satfes sator sator sattim sattim vernal vernal Table 10. Lake City Plot Map 209 209 208 208 207 207 206 206 205 205 204 204 203 203 202 202 201 201 border falor falor bfttim bfttim falfes falfes sattim sattim bftfes bftfes sator sator satfes satfes bflor bftor faltim faltim border 309 309 308 308 307 307 306 306 305 305 304 304 303 303 302 302 301 301 border satfes satfes sattim sattim sator sator faltim faltim bflor bftor bfttim bfttim falor falor falfes falfes bftfes bftfes border Table 11. Chatham Plot Map 201 201 202 202 203 203 204 204 205 205 206 206 207 207 208 208 209 209 vernal vernal faltim faltim bflor bflor satfes satfes sator sator bftfes bftfes sattim sattim falfes falfes bfttim bfttim falor falor vernal vernal 301 301 302 302 303 303 304 304 305 305 306 306 307 307 308 308 309 309 47 vernal vernal bftfes bftfes falfes falfes falor falor bfttim bfttim bftor bflor faltim faltim sator sator sattim sattim satfes satfes vernal vernal 409 409 408 408 407 407 406 406 405 405 404 404 403 403 402 402 401 401 401 401 402 402 403 403 404 404 405 405 406 406 407 407 408 408 409 409 border bfttim bfttim bftfes bftfes falor falor sator sator falfes falfes sattim sattim satfes satfes bflor bftor faltim faltim border vernal vernal faltim faltim bftor bftor satfes satfes sattim sattim falfes falfes sator sator falor falor bftfes bftfes bfttim bfttim vernal vernal Table 12 Average Accumulative forage yield at each location (metric tonnes dry matter per hectare) (Metric tonnes DM Birdsfoot per hectare) trefoil Falcata Sativa Chatham 11.3 17.0 19.1 Lake City 20.9 23.6 26.6 East Lansing 26.5 31.8 33.9 Average 19.6 24.1 26.5 Table 13. Stand persistence data for Chatham across four years post-seeding. 2009. Fall Fall Spring Combination 2006 2007 2009 falfes 52 51 26 falor 65 40 45 faltim 47 42 41 bftfes 1 7 9 1 5 bftor 12 8 1 1 bfttim 30 15 1 1 satfes 47 27 28 sator 36 32 42 sattim 79 51 38 Table 14. Stand persistence data for Lake City across four years post-seeding. 2009 Fall F all Spring Combination 2007 2008 2008 falfes 22 19 25 falor 15 18 31 faltim 15 19 13 bftfes 22 1 1 25 bftor 9 19 9 bfttim 21 30 18 satfes 31 22 19 sator 26 28 26 sattim 33 , 32 24 48 Table 15. Stand persistence data for Lake City across four years post-seeding. 2009 Fall Fall Spring Fall Spring Combination 2006 2007 2008 2008 2009 falfes 29 16 18 29 19 falor 30 9 15 13 5 faltim 29 15 23 16 14 bftfes 40 1 1 8 10 3 bftor 34 4 3 7 1 bfttim 31 8 1 1 15 4 satfes 35 22 30 15 28 sator 30 14 26 20 18 sattim 30 15 29 22 33 49 VI. LITERATURE CITED Amaral-Phillips, DM, RW Hemken, JC Henning, and LW Turner. “Pasture for Dairy Cattle: Challenges and Opportunities.” University of Kentucky Extension Bulletin. April 1997 p.6,_ Barker, JM, DD Buskirk, HD Ritchie, SR Rust, RH Leep, and DJ Barclay. “Intensive Grazing Management of Smooth Bromegrass With or Without Alfalfa or Birdsfoot Trefoil: Heifer Performance and Sward Characteristics.” DE Professional Animal Scientist. June 1999 Vol. 15:2. p. 130-135. Berdahl, JD, AC Wilton, RJ Lorenz, and AB Frank. “Alfalfa Survival and Vigor in Rangeland Grazed by Sheep.” Journal of Range Management. Jan 1986 Vol. 39:1. p. 59-62. Berdahl, JD, AC Wilton and AB Frank. “Survival and Agronomic Performance of 25 Alfalfa Cultivars and Strains Interseeded into Rangeland.” Journal of Rangg Management. July 1989 Vol. 42:4. p. 312-316. Bliss, RM. “Interseeding Alfalfa in the Northern Plains: Flowering Alfalfa Breaks Barriers.” Agricultural Research. October 2003. Boe, A., R. Bortnem, and A. Kruse. 1994. Forage forage yield of stockpiled yellow- flowered and hay-type alfalfas. Proc. 34 N. Amer. Alfalfa Improvement Conf., p. 132. Boe, A., R Bortnem, KF Higgins, AD Kruse, KD Kephart, and S Selman. “Breeding Yellow-Flowered Alfalfa for Combined Wildlife Habitat and Forage Purposes”. SDSU Extension Bulletin. May 1998. Bortnem, R., A. Boe, K. Higgings, and A. Kruse. 1993. Evaluation of alfalfa gerrnplasms for combined wildlife habitat and forage purposes. Proc. 23rd Central Alfalfa Improvement Conf., p. 29. Campbell, JB. “Grass-Alfalfa versus Grass-Alone Pastures Grazed in a Repeated- Seasonal Pattern”. Journ_a1 of RangMagagement. March 1963: Vol. 16:2. p. 78- 81. Dietz, TS. 2003. Evaluation of Medicago sativa spp. falcata for Sustainable Forage Production in Michigan. Ph.D. Proposal. p.3. Garcia, Alvaro, Nancy Thiex, Kenneth Kalscheur, and Kent Tjardes. “Interpreting Hay and Haylage Analysis.” SDSU Extension Extra. August 2003. p.3. Hendrickson, JR and JD Berdahl. “Survival of 16 Alfalfa Populations Space Planted into Grassland”. Jourrlal of Range Management. May 2003: Vol. 56:3. p. 260-265. 50 Hendrickson, JR, MA Liebig, and JD Berdahl. “Responses of Medicago sativa and Medicagofalcata type alfalfas to different defoliation times and grass competition”. garladian Journ_al of Plant Science. 2008; Vol. 88. p. 61-69. Mortenson, MC, GE Schuman and LJ Ingram. “Carbon Sequestration in Rangelands Interseeded with Yellow-Flowering Alfalfa (Medicago sativa spp. falcata)”. Environmental Mmement. 2004, Vol 33: Supplement 1. p. 8475-8481. Riday, H and EC Brummer. “Forage Forage yield Heterosis in Alfalfa”. Crop Science. 2002; Vol 42. p. 716-723. Riday, H and EC Brummer. “Heterosis of Agronomic Traits in Alfalfa”. Crop Science. 2002; V0142. p. 1081-1087. Riday, H and EC Brummer. “Stand persistence and Forage yield Stability of Intersubspecific Alfalfa Hybrids”. Crop Science. 2006; V0146. p. 1058-1063. Smith, Norman G. “Yellow-Blossomed Alfalfa on Rangeland in South Dakota”. Rangelands. August 1997; Vol. 19:4. p. 24-25. National Research Council Subcommittee on Dairy Cattle Nutrition, Committee on Animal Nutrition, , Nutrient Requirements of Dairy Cattle. 7th revised edition. Washington, DC: National Academy Press, 2001. Weiss, WP, ML Eastridge, JF Underwood. “Forages for Dairy Cattle”. Ohio State University Extension Fact Sheet. February 1999. 51 IIIIIIIIIIIIIIIIIII 93 03082 8782 "'1 U E III1 A II ll 5 I 12 3 IIIIIIIIIIIIIII