Historic, Archive Document Do not assume content reflects current scientific knowledge, policies, or practices. United States Department of Agriculture Agricultural Research Service UO. If S*>' f 3 c ^ Proceedings of the 39th Southern Pasture and Forage Crop Improvement Conference May 23-26, 1983 Oklahoma City, Oklahoma JT r x LT) r rn -- r Z& Xo w ro o «x> tji O — - 1 1 ISSN 0193-6425 Proceedings of the 39th Southern Pasture and Forage Crop Improvement Conference May 23-26, 1983 Oklahoma City, Oklahoma Sponsored by the Agricultural Experiment Stations of Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi. North Carolina, Oklahoma, Puerto Rico, South Carolina, Tennessee, Texas, and Virginia and the Agricultural Research Service U.S. Department of Agriculture Agricultural Research Service U.S. Department of Agriculture 1983 This publication is available from John D. Miller, Forage and Turf Research Unit, Agricultural Research Service, Georgia Coastal Plain Experiment Station, Tifton, Ga. 31793. Proceedings of the Southern Pasture and Forage Crop Improvement Conference, 39th, May 23-26, 1983, Oklahoma City, Oklahoma. Issued October 1983. Published by Agricultural Research Service (Southern Region) , U.S. Department of Agriculture, P.0. Box 53326, New Orleans, La. 70153, from camera-ready copy supplied by the authors, who accept responsibility for any errors in their papers. The opinions expressed by the authors are not necessarily those of the U.S. Department of Agriculture. Mention of pesticides does not constitute a recommendation for use by USDA, nor does it imply that the pesticides are registered under the Federal Insecticide, Fungicide, and Rodenticide Act as amended. The use of trade names does not constitute a guarantee, warranty, or endorsement of the products by USDA. ii CONTENTS The Forage-Livestock Industry in Oklahoma Oklahoma’s livestock industry Robert Totusek 1 Pasture-livestock management systems W. E. McMurphy 8 Forage resources of Oklahoma P. W. Santelmann 11 Forage breeding programs at the Oklahoma State University C. M. Taliaferro and J. L. Caddel 21 Range research at the Southern Plains Range Research Station P. L. Sims 26 Forage Plant Resources Buffelgrass germplasm research for the southern Great Plains E. C. Bashaw and C. W. Johns 31 New sources of genetic variability in dallisgrass and other Paspalum species Byron L. Burson, Paul W. Voigt, and Wayne R. Johnson 35 Clover and special purpose legume germplasm resources for the future Gary A. Pederson and William E. Knight 42 Forage attributes for improved animal performance H. Lippke 56 Panel Discussion: Data Required Before Releasing Forages. What Kind and How Much? The need for animal trials D. A. Sleper, F. A. Martz , A. G. Matches, and J. R. Forwood 61 USDA's practice at Tifton, Ga. Glenn W. Burton and Warren G. Monson 67 Forage quality assessment: Important factors for plant breeders to consider S. W. Coleman 71 Grazing management and utilization research prior to release of pasture cultivars Carl S. Hoveland 85 State agricultural experiment station policies W. C. Godley 89 iii Page Composition and ruminal availability of sulfur in cool- season grasses B. P. Glenn and D. G. Ely 93 Kochia — forage or weed? L. M. Rommann 96 No-till forage establishment Harlan E. White 98 Techniques using electronic communications Clement E. Ward 102 Computerized hay marketing Gerrit W. Cuperus 106 Effect of fertilizer application and grazing management on grazed New Zealand hill country M. Greg Lambert and David A. Clark 108 Recent progress in forage production and utilization in Scotland Thomas David Alexander Forbes 115 Contributors 125 IV The Forage-Livestock Industry in Oklahoma OKLAHOMA'S LIVESTOCK INDUSTRY Robert Totusek Oklahoma State University INTRODUCTION Oklahoma is livestock country, with two-thirds of the total farm income in the state resulting from livestock. It has not always been so. Fifty years ago, only one-third of the state's farm income was attributed to livestock. During the last half- century there has been a gradual shift from the production of cash crops to the production of forage to support the major portion of the state's livestock industry. Today, three- fourths of the state's 44 million acres are utilized to a great er or lesser degree by grazing animals. Although Oklahoma's livestock industry is predominantly a graz- ing livestock industry, the non-grazing types will also be men- tioned briefly because they do impact directly or indirectly on the total animal industry in the state. SWINE The swine industry in Oklahoma was at one time much larger than today. In 1945 the state swine population totaled 1.2 million, compared to 400 thousand today. The nature of the swine indus- try has also changed, shifting from a family-farm type of pro- duction with several sows on each farm, to large, highly spe- cialized operations with several hundred sows each, in confine- ment or semi-confinement. There is great potential for increas ed swine production in the state. Oklahoma is a pork deficit state, producing only about one-half of its needs and has sev- eral inherent assets such as a mild climate. Consequently, it has been projected that hog numbers in Oklahoma could increase as much as 50% by 1990 and perhaps double by the end of the century. POULTRY Many people are surprised to learn that poultry production 1 ranks 5th in Oklahoma among all agricultural commodities, rank- ing only behind beef cattle, wheat, dairy cattle and hay. Per- haps one reason is that the poultry industry in Oklahoma is al- most totally located in the eastern tier of counties and is not apparent to the casual traveler in many parts of the state. The poultry industry in Oklahoma is primarily one of broiler produc- tion; the number of broilers produced in the state increased from 3 million in 1969 to 36 million in 1980. The nature of the poultry industry has also changed drastically, from one of farm- flock production with some poultry on every farm, to a highly integrated, highly automated industry with relatively few but very large producers. It has been estimated that by 1990 per capita production of poultry will exceed that of beef, with a demand for 44% more broilers and 34% more layers. Consequently, it is likely that Oklahoma’s poultry industry will continue to increase. HORSES Oklahoma is horse country! Although not considered food pro- ducing animals in the culture of the United States, the many horses in Oklahoma do compete vigorously for available forage and must be considered relative to forage usage. Changes in horse numbers in Oklahoma have been similar to those of the United States. The national horse herd peaked at 20 million in 1920, decreased to 3 million in 1960, and is expect- ed to reach 20 million again by 1985. There are likely about 500 thousand horses in Oklahoma, with more horses per square mile than in any state in the country. In 1920 and before horses were used primarily for work and transportation, and only somewhat incidentally for pleasure; today about 97% of the horses are used for pleasure. Although many horses do serve an important role on working cattle ranches in Oklahoma, the majority are used for rodeoing, trail riding, exhibition, youth projects and simply for pleasure riding and as companion animals. The horse industry in Oklahoma is big business! It has been estimated that each horse generates $1,000 in business for the general economy, which means that in Oklahoma the horse indus- try is a $500 million business. Now, with the advent of para- mutual racing, it is anticipated that the state's horse popula- tion will increase further, with a projected increase to 600 thousand by the end of the century if not much sooner. SHEEP Sheep numbers in Oklahoma have followed the same pattern as national and even world numbers, with a decrease from 268 thou- 2 sand in 1940 to 72 thousand in 1977. Since the low point was reached, there has been a gradual increase to 105 thousand in. 1982. Because sheep are very complementary to wheat production, more than one-half of the total sheep in Oklahoma are found in the major wheat producing area of north-central Oklahoma. How- ever, they can be produced very satisfactorily throughout the state. Sheep offer some important advantages. They produce a choice product without grain, produce both food and wool, facilitate optimum range and pasture utilization when grazed with cattle, are adapted to small farms, and perhaps most important have tended to return a profit even in those years when beef cattle production has been unprofitable. However, there are some serious constraints to the expansion of sheep numbers. There is a general lack of knowledge about sheep, and some general sociological constraints in the minds of many people who prefer to produce cattle. Although predators and parasites present fewer problems than in the past, sheep are seasonal breeders, and there is the perennial "vicious circle" of low numbers and low consumption, with the low consumption caused not only by low numbers, but also the high price of lamb. At the present rate of increase, sheep numbers could double in Oklahoma by 1990 and more than quadruple by the end of the century. How- ever, the constraints mentioned above will likely dampen the projected increase in sheep numbers, especially if beef cattle production is at all profitable. DAIRY CATTLE Dairy cattle rank as the third most important commodity in Oklahoma agriculture, in spite of the fact that the dairy cow numbers have decreased by 50% during the past 20 years. How- ever, as is true nationally, the great increase in production per cow has allowed the maintenance of a stable milk supply with a considerably lower dairy cow population. In addition, dairy production units have become much larger and more spe- cialized; the average milk produced per farm in Oklahoma doubled from 560 thousand in 1960 to 1.2 million pounds in 1980. The very high level of milk production per cow found in some herds today is a vivid testimony to the development and appli- cation of technology, particularly in the areas of nutrition, progeny testing and artificial insemination. A need for 35-40% more milk has been projected by the 2000. In that event, recognizing Oklahoma’s inherent tages of a relatively mild climate and a geographical close to the sunbelt areas of high population growth, cattle industry will increase in Oklahoma. year advan- location the dairy 3 BEEF CATTLE Oklahoma is obviously beef cattle country, a situation which has developed over the past 60 years: CASH RECEIPTS FROM YEAR CATTLE & CALVES, % 1929 14 1949 25 1963 41 1980 56 More specifically, Oklahoma is beef cow country. Again, the in- crease over the past 60 years has been very dramatic, from 200 thousand cows in 1920 to 2.3 million in 1980. Today, there are more beef cows per square mile in Oklahoma than in any state. There is also a significant cattle feeding industry in Oklahoma, largely centered in the Panhandle, where almost 600 thousand head are fed each year. The nature as well as the size of the beef cattle industry in Oklahoma has changed markedly in the past 100 years, from the time of the Longhorn cattle (in some respects we have come "full circle" because today Longhorn bulls are being frequently used on first-calf heifers to minimize calving problems) . Un- til about 1950 commercial beef cattle production in Oklahoma consisted largely of straight-bred British breeds which had been bred up from the original Longhorn base. However, as knowledge developed about the benefits of heterosis, producers began crossbreeding and today the majority of calves coming to market are crossbreds, and often carry some blood of the conti- nental breeds which for the most part were imported about 1970. In very recent years we have seen Brahman cattle moving north- ward in Oklahoma from the breed's traditional stronghold in the southeastern part of the state. OTHER CHARACTERISTICS OF OKLAHOMA'S LIVESTOCK INDUSTRY At least three additional traits help to characterize the live- stock industry in Oklahoma. Seed Stock Industry Oklahoma has historically been important in the genetic improve- ment and seed stock production of livestock, in some cases far out of proportion to the importance of the commercial industry in the state. In the case of beef cattle, of course, it is no surprise that Oklahoma ranks in the top five in the production of purebred beef cattle in seven breeds, with four additional breeds ranking in the top nine. Likewise, it is not surprising that three separate breeds of horses rank in the top three na- tionally. It is a bit surprising, however, to find that Okla- homa has six breeds of hogs ranking anywhere from fourth to ninth nationally and similarly, three breeds of sheep ranking 4 fourth and another breed 7th. Performance Testing Many of Oklahoma’s seed stock producers were pioneers in per- formance testing. That emphasis on genetic improvement in traits of economic importance is still evident today, with a central boar test station and central bull test station among the largest and most prominent in the country. For example, approximately 700 bulls are tested each year at the Oklahoma Beef, Incorporated facility at Stillwater alone. Oklahoma seed stock producers and Oklahoma State University have a somewhat unique arrangement wherein the University provides land on a long-term lease basis plus supervision, and the breeders pro- vide the facilities and all costs of the performance testing. Youth Livestock Projects Oklahoma has historically placed heavy emphasis on youth live- stock projects, including 4-H and FFA involvement in livestock shows. The spring Junior Livestock Show in Oklahoma City, for example, is billed as the largest junior livestock show in the country. Some people are critical of shows, and there are ways in which they could be improved, but among other assets they serve to maintain the interest of young people in animal agri- culture. The contention that young people will not be able to adapt to "real life situations" due to the impracticality of livestock shows is without foundation. Young people are very astute and are able to maintain their perspective and conse- quently have no problem in adapting to commercial livestock production at the appropriate time. Actually, the most promi- nent carry-over from show involvement is that the young people tend to transfer the desire to excel to livestock production. THE FUTURE OF THE LIVESTOCK INDUSTRY Everyone agrees that animal agriculture in the future faces some constraints which are largely applicable nationally and which include at least the following: 1. High production costs 2. Low profits 3. Misinformation (about animal foods) 4. Decreased demand for animal foods 5. Animal rights (and welfare) issue 6. Regulations 7. Waste management On the other hand, there are some real opportunities ahead in animal agriculture: 1. Demand for food 2. Technology 3. Domestic markets 4. Foreign markets 5. Producing to meet demand 5 6. Geographic location 7. Climate Again, most of the opportunities apply nationally, with only the last two being primarily applicable to Oklahoma. (Okla- homa's advantages are likely offset by other advantages in other areas of the country.) There will be a great increase in demand for food by the end of the century, which obviously bodes well for animal agriculture. By the year 2000 there will be a need for 75% more milk, 80% more beef and 90% more sheep and goats. Technology represents the most important opportunity over which the individual producer has control. For example, considerable research in Oklahoma and elsewhere has shown that with existing technology most forage-producing operations could double, or even triple and yes, even quadruple forage production, given adequate economic reward. Or, on the animal side, research has shown that beef production can be increased 10% through the use of two- breed crossbreeding, 20% through the use of three- breed crossbreeding, 30% if Brahman are included in the cross- breeding program, 40% if a growthy breed is included in the crossbreeding program, and 50% if a heavy milking breed is in- cluded. The list could go on and on. As we look ahead, we would agree that "we ain't seen nothing yet" when we think of opportunities which have been projected through the use of new technology in such areas as genetic engineering. KEYS TO SURVIVAL What does the producer need to do as he looks ahead, to strengthen his operation, and indeed just to survive? He needs to do two things: First, he needs to adopt all available tech- nology that is applicable and profitable in his operation. It is rather revealing to consider the rates of technology adop- tion in various animal industries. Percentage of Industry Technology Adopted Poultry 90 Dairy 75 Swine 75 Beef Cattle 40-50 Sheep 40-50 These figures certainly provide a vivid explanation of the past and certainly provide some warnings for the future. It is quite apparent that the rate of technology adoption must be increased with our forage producing animals. 6 Second, the producer must become involved. Historically, the livestock producer has been a rugged individualist, and has been proud of it. This has been commendable but the livestock producers of the future cannot afford such luxury. They must unite to solve problems, such as those relating to market de- velopment, public relations and orientation of decision-makers (political impact) . CHALLENGE OF ANIMAL AGRICULTURE Although this applies more to Oklahoma than some areas, we have two challenges in terms of the agriculture economy. One is to revitalize the beef cattle sector, primarily through the appli- cation of existing technology and the development of new tech- nology as possible and necessary. Second is to diversify ani- mal agriculture through the enhancement and growth of such areas as dairy, poultry, swine, sheep and perhaps even in some cases, horses. We have, in Oklahoma, essentially a two-commo- dity agriculture (beef cattle and wheat) , and in those years when prices of beef cattle and wheat are both depressed, the economy of the state suffers. OKLAHOMA STATE UNIVERSITY'S ROLE What do we at Oklahoma State University intend to do about the constraints, the opportunities and the challenges relating to animal agriculture? We intend to do three things essentially: (1) Through research we will develop more technology. The difficult questions ahead will require both more research and more sophisticated research, both applied and basic research, and both biological and economic research. (2) We will become more effective in our extension programming to facilitate a higher rate of technology adoption, through such innovations as educational TV, satellite communications, continuing educa- tion, expanded use of total mass media, and home video courses just to mention a few possibilities. (3) In our teaching pro- grams we will need to "produce" more graduates to meet the in- creasing needs of high technology production in many areas of the livestock industry and related agribusiness, and we will need to do a better job of training the students through empha- sis not only on the art and science of livestock production, but on the business aspect as well. Productivity in animal agriculture in terms of output per fe- male (cow, sow, ewe) approximately doubled during the 50 years beginning in 1925. This was largely due to the cooperative efforts of our system of research and education, the hard working livestock producer and an effective agribusiness com- plex. With the same kind of team work in the future, we have every reason to think that we can more than double productivity per unit in the next fifty years. 7 The Forage-Livestock Industry in Oklahoma PASTURE-LIVESTOCK MANAGEMENT SYSTEMS W. E. McMurphy Oklahoma State University Precipitation is a controlling factor in the forage production in Oklahoma. The eastern one third of the state has an annual precipitation of over 40 inches and will produce an abundance of forage, both cool and warm season. The western one third of Oklahoma receives about one half the annual precipitation that eastern Oklahoma receives, but the distribution of mois- ture strongly favors warm season species. Average monthly precipitation is about one inch from November through March, five months. The small grains with mostly wheat pasture are the only viable cool season forages for this western area. Fescue toxicosis has not been a big problem in Oklahoma, pos- sibly because not enough fescue is grown. Tall fescue is the only cool season perennial grass reasonably well adapted to eastern Oklahoma. However, tall fescue is unreliable as a source of winter forage. No fall growth was produced in four of six years in a study near Pawhuska. Ranchers are thus unwilling to purchase necessary N fertilizer for fall growth for an unreliable situation. The rough rocky wooded lands of the Ouachita and Ozark High- lands Resource Areas that currently produce blackjack and post oak have great forage potential. Herbicides will control the woody species. Burning will prepare a seedbed for tall fescue. Aerial application of seed and fertilizer have been proven techniques in developing this potential. Unfortunately the economics of this practice are not practical at the present time. Arrowleaf clover is not the major pasture legume of eastern Oklahoma. This species has three important characteristics vital to its success: (1) it is a prolific reseeding annual, (2) it has hard dormant seed, and (3) it grows tall. Perennial pasture legumes often die during summer drought. The hard 8 dormant seed characteristic provides seed for another crop when early fall precipitation causes germination followed by drought which can be lethal to all seedlings. The tall growth charac- teristic enables it to survive spring grazing mismanagement if grasses get too tall. Other pasture legumes in use are hop clover, red clover, white clover, crimson clover, and subter- ranian clover. Bermudagrass is the most important introduced grass and occupies at least six million acres. Many of the pastures in eastern Oklahoma that are dominated by broomsedge and weeds appear to be rangeland because that is the way they are being managed. However, these areas that have bermudagrass present can be quickly converted to very productive pasture. Mowing in early June removes the dormant cool season annual grasses and controls many broadleaf weeds. Then an application of N fertilizer plus P and K fertilizer if needed will quickly convert these seem- ingly low productive lands into bermudagrass pastures within a month. Winter hardiness has always been a problem with any new bermuda- grass varieties. The Midland and Hardie varieties are the best adapted ones for Oklahoma. A five-year grazing test with steers at Perkins, Oklahoma compared Midland and Hardie bermudagrass . A three paddock rotation was used with the objective to graze grass that was between two and three weeks of age. A split application of N fertilizer was used with 50 lb of N per acre applied three times each season. Average daily gain was 1.80 lb for Hardie and 1.61 lb for Midland. Stocking rates were adjusted with the put and take method to use available forage and averaged 2.4 steers per acre for Hardie and 2.3 steers per acre for Midland. Total beef production per acre was 636 lb for Hardie and 492 lb for Midland. The value of an improved variety was apparent. Native range that dominates the grassland resources of Oklahoma requires a different management philosophy than that of most introduced grass pastures. The goal of range management for cattle production is to promote plant succession to the point of the tallest climax native grasses the site will support. The goal of pasture management is to prevent plant succession. Plant succession in rangeland is promoted best by permitting the grasses to grow, preferably all during the growing season with grazing done in the dormant season. This deferment when com- bined with herbicides or fire in special situations is very effective in promoting plant succession. With season long grazing on range, the rule of "take half and leave half" must be followed. It is very necessary to practice this moderate use of range during the growing season to maintain the necessary root carbohydrate reserves for plant vigor and competitive ability. However, these practices do not apply to introduced pasture species and would be a wasted effort. We prevent plant succes- sion in bermudagrass pastures by mowing and fertilizing, 9 followed shortly by grazing. On rangeland the practices of mow- ing, fertilizing, then grazing would be disastrous to the native climax grasses. The range manager must be concerned with adjusting the stocking rate, time of grazing, and degree of use, because he has little control over the quantity of forage production. Stocking rate flexibility is necessary. The pasture manager can adjust the quantity of forage produced with N fertilizer and has more control over timing of that production through selection of species planted and timing of the fertilizer application. Native range grasses have a slow rate of physiological maturity, and grasses deferred from grazing from May 1 to July 1 are still good quality forage. This is not true of the introduced forage species because they have a much faster rate of physiological maturity with the corresponding decline in forage quality. The native tall grasses of the True Prairie region have very slow regrowth following herbage removal after July 1. This is the result of evolutionary selection pressures. These species are very palatable, they evolved with grazing, and slow regrowth is a survival mechanism. There is a vast genetic diversity within these tall grass species, but any ecotype which evolved which had rapid regrowth would have been vulnerable. Rapid regrowth occurs at the expense of the root carbohydrate reserves, regrowth is very palatable to herbivores, and such ecotypes probably disappeared from the ecosystem. With such slow regrowth the native ranges will require a much longer period of deferment in a rotation system than the introduced grasses. The pasture systems of Oklahoma are combined with the native range resource throughout the state. Management of each requires different techniques, but the greatest potential for expansion is with introduced forage species. The technology is available, but the present economical pressures upon the beef industry limit its expansion in Oklahoma. 10 The Forage-Livestock Industry in Oklahoma FORAGE RESOURCES OF OKLAHOMA P. W. Santelmann Oklahoma State University I appreciate the opportunity to present an overview of the plant resources of Oklahoma. Since it is difficult to talk about plants without discussing water and soil I would like to mention these resources also. The growing of plants for food, fiber, feed, fuel, conserva- tion, recreation, and esthetics is big business in Oklahoma. Plant agriculture is important not only as a livelihood for our farmers and ranchers but also for the well-being of our citi- zens. We still consider ourselves an agricultural state. However, the climate in Oklahoma is harsh for plant production. Rainfall and temperature vary quite widely across the state and the distribution at any one locality is highly uneven from year to year. Unseasonable cool temperature, frost, or hot dessi- cating winds frequently reduce plant growth and crop yields. The extensive types of crop and livestock production have dominated agricultural enterprises in Oklahoma since settlement (as contrasted to intensive). This type of production was best suited to the state's resources and climate. Forage production kept pace with the growing livestock industry. Oklahoma soil and climatic conditions coupled with the intense interest in livestock makes forage production well suited to the state. The land devoted to ranges, pastures, and forage crops exceeds land devoted to cultivated crops by a wide margin. The acreage in improved pasture has quadrupled in the last 25 years. An increase in forage production is anticipated - primarily through improved pasture and range management and the conver- sion of some ranges into improved pastures. Water. Certainly one element necessary for successful forage production is water. The sources of water for both livestock and man in Oklahoma include farm ponds, large reservoirs, flood water detention reservoirs, and major streams such as the 11 Arkansas, Cimarron, Canadian, and Red Rivers and their tribu- taries. In addition a few of our counties have underground water resources from the Oogalala water formation. We have over 100,000 farm ponds in Oklahoma. There are 13 large reservoirs throughout the state but most of the water in these is not available for agriculture use. Our annual precipitation varies from about 52 inches in the southeast to 16 inches in the northwest corner of our High Plains. Almost one million acres in Oklahoma is irrigated each year, but this is primarily not on forages. There are exceptions to this as some alfalfa and bermudagrass are irrigated. Most of the water goes for irri- gated crops such as cotton, peanuts, soybeans, wheat, and sorghum. Soils. There are about 44 million acres of land in Oklahoma. Our soils vary widely in the nature of their parent material, their topography, their age, and properties such as organic matter, pH, and cation status. In general organic matter content is low - in the area of 1% or less. Soil pH in most of the state is on the neutral or basic side, but does become acidic as you get into the higher rainfall areas in eastern Okl ahoma. The state is roughly divided into nine resource areas. Of course there are many different soil series within each resource area. Starting from the east the Ouachita Highlands in south- eastern Oklahoma is characterized by a series of parallel ridges running generally east and west. The rugged surface and sizable acreages of stoney, shallow soils are developed from weathering of sandstone and shale. This resource area contains nearly a half million acres in pasture and rangeland in Classes I through IV and even more in Classes V through VI. The Ozark Highlands in northeastern Oklahoma also has a variable surface relief and comprises about 1.6 million acres. Pasture and range make up only about 14% of this area. When we get into the northeast and southeastern corners of the state there are about 2\ million acres of rangeland in these mountainous areas. The average stocking rate in this area is about 40 acres per animal unit (AUY). The Forested Coastal Plains consist of about one and one-third million acres in southcentral Oklahoma. Most of these soils are sandy and are developed from beds of unconsolidated sands, clays and sandy clays. Pasture and range make up about 15% of this area. The Cherokee Prairies consist of 6i million acres of gentle and somewhat rolling land in eastcentral and northeastern Oklahoma. The annual precipitation varies from 35 to 45 inches per year. Low ridges of outcropping sandstone traverse the area and these soils are generally sandy, shallow, and non-arable . The 12 Bluestem Hills (Flint Hills) are included in the northwest part of this resource area. Pasture and range comprise over half of the acreage in this area. In these eastern prairies we find that the tall grasses are dominant - including switchgrass, Indiangrass, and the big and little bluestems. The ranges of this area (about 1.5 million acres) support an average stocking rate of 10 AUY. This area offers tremendous possibility of expanded forage production. The Cross Timbers comprise another six million acres through the central part of Oklahoma. The surface relief varies from gently rolling to hilly. The dominant soils are sandstone derived and under natural conditions support mainly a post oak and blackjack oak savannah type of vegetation. Soil is very shallow and has a lower stocking rate than some of the western lands, approxi- mately 45 AUY. The species prevalent in this 2.2 million acres of range area are primarily Indiangrass and the bluestems. The Cross Timber area is rapidly going to improved pasture and we hope to see this trend continue. The Grand Prairie in southern Oklahoma contains almost two million acres. The surface relief ranges from gently wavey to rolling and hilly. The soils were developed from limestone on shale under the cover of tall grasses. The soils are predomi- nately dark colored and heavy or clayey. About 70% of this area is in pasture and range and forage production can be improved considerably. The Reddish Prairie in westcentral Oklahoma occupies a wide belt through the state and contains about 8i million acres of wavey to gently rolling surface relief. The soils developed under a grass cover over weakly calcareous red shales and sandstones. This area has the highest concentration of cultivated cropland in the state, but still about 1/2 of it is devoted to pasture and range. In the Red Prairies there are about 2.3 million acres of range which has primarily big bluestem, side oats gramma, and little bluestem and a stocking rate of about 20 AUY. The Rolling Red Plains make up a large resource area of about 9£ million acres in the western part of the state. Like the rest of the state it tilts toward the southeast with elevation ranging from 1000 feet in the east to 3000 feet above sea level in the west. The surface is rolling with deep cut valleys and narrow strips of alluvial soils. Most of the 20 to 30 inches of precipitation occurs between April and September, but the distribution is irregular and droughts are common. About four million acres is in pasture and range and comprising primarily of sand and big bluestem, blue gramma, and little bluestem. Here the average stocking rate is about 30 AUY. The High Plains contain almost four million acres of land sloping from the southeast to a high point in the northwest almost 5000 feet above sea level. The rainfall is only 15 to 20 13 inches per year. These soils developed from outwash material imported from the higher elevations of the west. On the High Plains we find that the dominant forage species are blue gramma, buffalograss and little bluestem. The range area comprises about \\ million acres and has a stocking rate of about 40 AUY. Rangeland. Rangeland and forest range occupy about 20 million acres, or almost one-half of the land area in Oklahoma. It includes all lands on which the native vegetation is predomi- nately grasses, grass-like plants, forbes or shrubs suitable for grazing or browsing. These include land revegetated naturally or artificially to provide a forage cover that is managed like native vegetation. Rangelands in Oklahoma include natural grasslands, hay meadows, savannahs, shrub lands, abandoned cropland and areas originally planted to introduce pasture species but which have reverted to predominately native vegeta- tion because of a lack of proper management. Rangelands may also include many forest lands and grazable woodlands in Oklahoma. The Oklahoma resouces inventory indicates that about 65% of the rangeland in Oklahoma needs some type of conservation treatment to restore the land to its full potential. The primary needs are for brush and weed control, grazing management and other range improvement practices which increase range condition, herbaceous plant production and groundcover. The role of rangeland in the Oklahoma economy is difficult to measure because aggregate production data concerning livestock and other uses of rangeland are not available. However, exist- ing knowledge and technology applicable to Oklahoma rangeland could easily double current livestock and wildlife production if implemented thoughout the state. Too often the focus has been on range improvements as a cure for improper grazing management. Pests. Pests on Oklahoma rangeland include primarily weeds, brush , and insects. There are undesirable plants on most of the 20 million acres of rangeland and forest range. About 11 million of these acres have a serious woody plant problem. These plants are considered undesirable since they are not utilized by livestock and compete with desirable plants. The primary weed problems on rangelands and pastures are broomweeds, ironweeds, and the ragweeds. In the western half of the state the problem is dominated by western ragweed and the broomweed. In the eastern half the problem is dominated by western ragweed and lance-leaf ragweed. Both the common broomweed and the lance-leaf ragweed are annuals, and are particularly a problem following drought years. Most areas are overgrazed during periods of low production and this allows open spaces for the annuals to germinate and establish. Western ragweed is a perennial that spreads both by seed and vegetative underground stems. Once the plants become established the problem tends to increase each year. 14 Most of the herbaceous plants have about the same requirement for growth as the native desirable plants so that there is about one pound of desirable forage loss for every pound of weeds produced. Weed production on pastures and range varies consid- erably throughout the state but about 1000 pounds of weeds produced is very common and 2000 pounds is not uncommon. There are normally three control options for taking care of herbaceous weeds on our forage lands. Grazing is one option that is often overlooked. It can be very effective and an economical alter- native. It does require heavy stocking rates for a short period of time when the weeds are palatable and then removing the cattle to allow regrowth of desirable grasses. Fair results have been obtained with prairie threeawn, broomsage, sandbur, and western ragweed but results with other species such as western ironweed and common broomweed have been poor. A second option for controlling weeds is mowing, but it is primarily a cosmetic option the way that mowers use it. Although it can be effective in preventing weeds from producing seed most of the competition has already resulted. There is also loss of desirable forage from mowing. Burning can best replace mowing if done properly. The third and probably best all around option is the use of herbicides to control weeds. The primary herbicide use for weed control is 2,4-D-which is effective on many of the broadleaved weeds and is available in a number of formulations. Dicamba has been mixed with 2,4-D for specific weed problems. Atrazine is also approved for use on rangelands. Its primary advantage is its activity on annual grasses such as the annual bromes and prairie threeawn. Rain- fall after application is necessary to move the atrazine into the root zone. The primary brush problems on the rangelands and pastures of Oklahoma are the scrub oaks. This ranges from the blackjack and post oak complex as the dominant vegetation on the sandy soils of the Cross Timbers and in southeastern Oklahoma to the shin- nery oak-sand sage complex which is the dominant vegetation on sandy soils in the western part of the state. Native grass production on some of these areas is less than 500 lbs. per acre. Brush is constantly invading the Oklahoma grasslands since environmental conditions are favorable for brush. It is estimated that there are more acres of rangeland infested with brush now than at any time since statehood. Much of this is attributed to a large increase in eastern red cedar. Some ten years ago this was a problem on about one million acres, but today they are becoming a problem on more than 4i million acres of the Cross Timbers and Reddish Prairie lands. This increase is attributed to the lack of burning coupled with a large number of seed trees scattered throughout the state. In addition the major brush herbicides do not control eastern red cedar. Brush control options are available but limited. Mechanical clearing has become very expensive and most desirable sites have 15 already been converted. The requirement that these sites be "farmed" for two seasons to control resprouts puts a severe limitation on this option. Mowing is possible on level areas but is non-effective on most woody species, with the possible exception of small eastern red cedar trees. In fact mowing will often increase the number of stems of some species. There is also a decrease in top to root ratio and this results in less effective control with any follow up foliar sprays. Burning has essentially the same limitation as mowing. However both mowing and burning can be effective on trees that don't resprout. In fact burning may be the only economical control option available for cedar control . Herbicides are the most selective and in most cases the most economical brush control option available. The major limitation is that there are currently only a few herbicides that have label clearances for use on rangeland. For 35 years 2,4,5-T has been the major chemical but its economic advantage may be coming to an end. As recently as 1970 a standard application would cost only $6 to $7 per acre. However, the cost has tripled in the last 10 years. For many of the oaks treatments for two consecutive years are needed - which also drastically increases the cost. Graslan first received label clearance in Oklahoma and Texas in 1979. It has proven to be an excellent herbicide for blackjack oak, post oak, and winged elm control on shallow sandy soils. However, it is very expensive to use, particularly when one considers the current cattle prices since it takes at least eight acres of brush converted rangeland to provide enough forage for one cow per year. The amount of grass release after spraying depends on the amount of desirable grass in the treated area, the amount of brush control, the productivity of the site and the amount of effec- tive moisture available for plant growth. The highest yield of grass obtained two or three years after spraying for brush control was about 4000 pounds per acre and this represented a four-fold increase in grass production. The actual advantage of brush control on range usually results in enough increased grass production to allow a doubling of the carrying capacity in addition to an increase in the calving percent and the weaning weight of calves. Several problems confront range managers and scientists who are attempting to meet the separate demands of ranchers and society. Since rangelands inherently have a low production protential and there is great variability associated with the weather in the range area the capital investment requires long periods for benefits to be realized, and are usually not cost effective when capital costs are high. Higher producing rangelands still continue to occasionally be converted to cropland and other land uses while marginal croplands are being allowed to revert to rangeland. Increases in the densities and the encroachment of brush species on rangeland reduces their production potential 16 and yet cost effective environmentally acceptable methods of controlling brush are not generally available for use. Exten- sive management systems are generally called for on rangeland but are often overlooked or ignored because they usually do not result in immediate or sizable increases in production. Pastures and Forages. Approximately 8.5 million acres in the state are devoted to pasture, hay, and other tame forage pro- duction. Forage in this state provides 80% of the nutrients for beef production and 65% of the nutrients for milk production. Beef production in Oklahoma has more than doubled in the past 20 years because slaughter weights have remained relatively con- stant and cattle in feed lots have less than doubled during this period. Beef production from rangeland and forages has more than doubled. Significant gains in productivity played a major role in the total increase in forage production. The average yield for all hay increased from 1.45 tons per acre in 1958 to 2.12 tons per acre in 1979. Similar productivity gains have been realized for pastures. A number of factors have enabled Oklahoma farmers and ranchers to increase productivity. Development and utilization of more productive grass varieties and introduction of commercial fertilizers contributed to the increased production for pas- tures. On-going research in these areas as well as renewed efforts and variety improvement will contribute to production in the future. A number of grass varieties introduced by the Oklahoma Agricultural Experiment Station has significantly raised pasture productivity in the State. Notable varieties include Midland bermudagrass (1953), Morpa lovegrass (1969), Plains bluestem (1970), and Hardie bermudagrass (1974). Most recently Brazos bermudagrass, WW Spar bluestem, and Guymon bermudagrass have been released by various agencies at least partially as the result of the efforts by Oklahoma grass breed- ers. Introduced warm and cool season annual and perennial grasses are used extensively in Oklahoma either to supplant or supplement native vegetation. The principle introduced annual grasses used for pasture or forage are the cool season cereals (particularly wheat), the warm season sorghum, and the millets. A significant portion of the winter wheat seeded in Oklahoma each year is grazed by livestock. Wheat pasture provides a significant forage support to the beef cattle industry of the state. However, gains of wheat pasture Stockers are frequently reduced by 1) inadequate fall or winter forage and 2) snow or ice cover of wheat pasture. Stability of the wheat pasture Stocker enterprise could be increased by improved agronomic practices such as earlier planting dates to increase the amount of fall and winter forage. While planting dates are influenced by climatic conditions the optimal seeding date for grain production (early October) is too late for production of fall 17 forage for winter grazing. However, the extreme variability of Oklahoma's climate often will not permit early planting or provide sufficient winter rainfall for best forage production - which increases the risk factor in buying stocker cattle. Because of the large amount of the wheat forage that is produced in the spring there is some potential for increasing gains by extending the grazing period beyond the traditional March 10-15 cut-off date it grain is to be harvested. The principle introduced perennial grasses used in the state are bermudagrass , weeping lovegrass, yellow bluestem, and tall fescue. The annual and perennial species are often distin- guished on the basis of their cultural requirement by the terms cultivated forages and tame pastures respectively. The annual warm season grasses have high productivity capability and the forage is of intermediate nutritive value. The major con- straints associated with their use include high production costs, insect and disease pests, and in the case of the sorghum species the potential for hydrocyanic acid poisoning of grazing animal s. The introduced perennial grasses used in the state are charac- terized by high production potential and low energy value. The principle constraints associated with their production include establishment difficulties, high nitrogen fertilizer require- ment, and intensity of management necessary to optimize yield of digestible nutrients on a consistent basis. The perennial warm and cool season grass species are further characterized by great genetic variability and consequently very significant genetic improvement potential. Much additional progress is needed in quantifying the effects of management variables on perennial pasture species and in the development of systems models to guide decisions for maximizing of their net economic returns. Although introduced grasses are harvested mainly by grazing animals, sizable amounts are processed as hay and sold on the open market. On a per unit basis many of these forage crops because of their high production potential can compete with grain crops in the production of protein. Energy efficient processing methods must be developed however, before this potential can be fully realized. Nitrogen fertilizer will continue to rise in cost in the future because of its natural gas base. Although nitrogen fertilizer is the most dependable means of increasing forage quantity and quality, its use on pastures in Oklahoma would depend on the prices paid to producers for meat and milk. The best alternative for improved production is more wide spread use of grass-legume mixtures for grazing. Winter hardiness, growth during the winter, and drought tolerance are not always satis- factory for legumes introduced into the state, although some legumes developed in the southeast produce quite well in Oklahoma. Legume use will spread to central and western 18 Oklahoma as productive adapted legumes are developed. The grazing management essential to maintain perennial legumes in a sward or to optimize production must be developed for Oklahoma producers to remain competitive. Alfalfa. Alfalfa production is an essential part of the beef, horse, and dairy industries of the southern plains and its importance will continue to increase with higher costs of energy. It is valued for both the fact that it produces a high quality forage without annual seeding costs and because it increases nitrogen levels in the soils, which is valuable if rotated with other crops. Probably the two greatest problems with alfalfa production at present are related to harvest management and to pest control . Losses due to pests represent the greatest limitations to increased alfalfa production at the present time in Oklahoma. A variety of insects, pathogens, and weeds cause reduced production in alfalfa. The most important forage insect pests include the spotted alfalfa aphid, the blue alfalfa aphid, alfalfa weevil, and the pea aphid. In addition seed production insect pests such as lygus bugs and the alfalfa seed chalcid are important. Many weeds are found in our alfalfa fields but henbit, winter annual grasses, and pigweed are often among the more dominant species. Alfalfa stands are not as long lived as we would like to see in Oklahoma and we suspect that phytophthora rootrot is a significant cause. Alfalfa hay is an important part of the income of many southern Oklahoma alfalfa producers. Four to five cuttings per summer is the normal procedure in Oklahoma with average yields of 3.3 tons per acre. Producers generally are able to produce more alfalfa than they can easily sell at a good price. If marketing pro- blems can be resolved they will probably be receptive to new production practices and varieties. Marketing this hay through the new "Haymarket" program will be of significant help to our growers. Alfalfa seed production at one time was important to the state. This industry shifted to western states using improved produc- tion practices. The apparent constraints to seed production are primarily the lack of sufficient numbers of insect pollinators and the knowledge of their management along with problems caused by insects that feed on alfalfa seeds before harvesting. Research is being conducted to try to solve these problems. Presently pest controls emphasize solutions for individual problems as they occur. These controls are often quite expen- sive with little consideration for the most efficient use of the resources available. For example, use of resistant varieties in alfalfa production would save producers millions of dollars annually. Possible interactions between control measures and the various pest complexes are usually not considered and the adverse effects of pest regulation practices on non-target organisms are often ignored. One of the most effective means 19 for controlling many of these pests would be the use of resis- tant varieties, for which we have an alfalfa breeding program. Recent research in Oklahoma suggests that harvest management may be different from that in some of the eastern and northern alfalfa growing areas. We are finding that the first harvest can be made well before the first signs of bloom when only a few small flower buds are observed. Harvesting at this early stage of growth can help to control weeds by reducing their seed set, can help control insects by removing the forage and exposing them to sunlight, and produces a high quality feed without reducing stand longevity or yield. Similar research has recently shown that harvesting established stands of alfalfa at any fall date has little or no effect on spring forage yields and stand persistence. Alfalfa in Oklahoma may never go completely dormant and consequently green leafy material remain- ing after the last fall harvest is present for photosynthesis and may provide for the plants winter and early spring energy needs . There are many problems associated with the production of forages in Oklahoma. There are also many opportunities in both our research and our extension activities related to forages. In spite of these problems we are optimistic about the potential for Oklahoma to provide its share of the expected increase and demand for range and tame forages in the United States. 20 The Forage-Livestock Industry in Oklahoma FORAGE BREEDING PROGRAMS AT THE OKLAHOMA STATE UNIVERSITY C. M. Taliaferro and J. L. Caddel Oklahoma State University Presently there are two forage breeding projects at the Oklahoma State University dealing, respecti vely , with grasses and alfalfa. The purpose of this report is to provide an overview of the work that is in progress in each of these two projects . The broad objectives of the grass breeding project are to: 1) develop new cultivars that are superior to existing ones in such characteristics as adaptation, yield, and forage quality, 2) evaluate new accessions and selections of forage plants to determine their adaptation and potential value in Oklahoma agriculture and 3) investigate the reproductive mechanism, breeding behavior and improvement potential of important and potentially important forage species. Breeding and/or selec- tion work is presently underway with four grass species: bermuda ( Cynodon spp.), introduced (Old World) bluestems (Bothriochl oa spp. ) eastern gamagrass (Tripsacum dactyloides) , and kleingrass (Panicum coloratum). A brief description of the principle areas of endeavor in each of these species follows. The bermudagrass breeding program has been underway for several years and has been primarily concerned with the improvement of nutritive value in vegetatively propagated varieties. This work encompasses the interspecific hybridization of high quality but nonwi nterhardy plants belonging primarily to Cynodon nlemfuensi s varieties nlemfuensi s or robustus with well-adapted but relatively low quality plants belonging primarily to the taxon Cynodon dactyl on var. dactyl on . The initial interspecific crosses were made in the 1960's and over the years a modified recurrent selection program has been used in an attempt to increase the frequency of genes enhancing forage quality and other physiological and morphological characteri sties related to yield and adaptation. Progress is being made in combining the desirable attributes of the 21 parental species. We presently have in our breeding nurseries progeny selections that are relatively winterhardy and signif- icantly higher in dry matter digestibility than check cultivars such as Midland. Although most of the selections of this type are deficient in one or more performance characteristics (forage yield, disease resistance, or establishment character- istics), they serve as parents of progeny populations in which further selection is practiced. Another objective of the bermudagrass breeding program is the development of seed-propagated cultivars for forage and turf use. In the early stages of the bermudagrass program, some accessions from the germplasm collection were found to have relatively good seed set and apparent high seed production capability. Subsequent testing showed that excellent seed yields could be produced from fields planted to a mixture of two such self-incompatible, cross-compatible clonal plants. In a 3-year study (1974-1976) conducted at the Southwestern Livestock and Forage Research Station near El Reno, Oklahoma, an average seed yield of 743 kgs/ha was produced. The parental plants possessing relatively good fertility and their progeny populations are winterhardy but do not possess the fineness of texture desired in a turf cultivar nor the yield potential desired in a forage cultivar. Hence, we initiated a restricted recurrent phenotypic selection program for fertility and plant type. In this program the first level of selection is made for plant type, i.e., forage versus turf, and then within each of these categories, selection is practiced for fertility as expressed by percent of open-pollinated seed set. The second cycle of selection is presently underway in this program. The introduced bluestems possess a number of desirable attri- butes which, in our opinion, will insure their continued use as pasture grasses in the southern Great Plains. They are easily estaolished, aggressive, persistent, productive, and have the ability to tolerate such stresses as drought and overgrazing but still retain a stand. The Plains bluestem variety was released in 1972 by the Oklahoma Agricultural Experiment Station and has been enthusiastically accepted in a large geographical area on the southern Great Plains. The obligate and facultative modes of apomictive reproduction found within the genus make hybridization difficult, but some crossing can be done between facultative parents and between obligate and facultative parents where the obligate apomictic parent is used as the male. Some hybridizations have been made between Bothriochloa i schaemum and Bothriochloa intermedia in an attempt to combine the superior winterhardiness of the former with the greater vigor of the latter species. The majority of the improvement effort, however, has been directed toward the selection of existing superior biotypes within the germplasm col lection. 22 Eastern gamagrass is native to much of the eastern half of the United States and is generally regarded as a "high quality" grass because of its superior palatabil ity. Its exceptionally good palatabil ity to all classes of livestock is attested to by the fact that it has been eliminated from much of its native habitat by overgrazing and presently is found only in areas protected from continuous grazing. Our basic objective with eastern gamagrass is to elucidate its potential for use as a grazed or stored forage and to determine the extent of genetic variation for traits of agronomic importance and the breeding behavior of the species. Present indications are that there is a wide array of genetic variation for most of the important agronomic traits such as yield and quality of forage, and seed production and its components. However, the most important and yet unanswered question relates to the basic nutritive value of the species. The expense and difficulty of establishment and the necessity for a high level of management that will be necessary for sustained high yield of gamagrass detract from its potential and dictate that it must be outstanding in some other characteristics. That characteristic most logically should be nutritive value. A cooperative experiment with dairy scientists was recently completed in which lactating dairy cows were fed gamagrass and alfalfa hays. The gamagrass hay was comprised of initial spring growth cut at the boot stage and 5-week old regrowth. The alfalfa hay was also comprised of the first and second cuttings with each cutting being made at approximately 10% bloom. Cows fed the gamagrass hay had significantly less dry matter intake per day (19.20 versus 20.07 kgs/day) than did the cows consuming the alfalfa hay. Cows consuming gamagrass hay also produced significantly less milk than did the cows on alfalfa (22.93 versus 24.06 kgs/day). These results suggest gamagrass may not be too different in forage quality from other grasses, particularly the Sudan grasses and sorghum-sudan hybrids. However, data are needed comparing the performance of grazing animals on gamagrass and other suitable grass controls in order to more firmly establish its basic nutritive value. Kleingrass is a warm-season, perennial, bunch grass indigenous to Africa. It possesses drought tolerance, is a prolific seed producer, is easily established, is a little higher in forage quality than bermudagrass cultivars such as Coastal and Mid- land, and is valuable as a wildlife habitat for quail and other wild game birds which thrive on its seeds. Kleinarass is presently used most extensively in western Texas north to about 34th parallel. Its marginal winterhardiness does not allow it to be used reliably north of this line and consequently it cannot be used anywhere in Oklahoma presently. Our sole objective with kleingrass is to develop a more winterhardy cultivar that can be used in the southern half of Oklahoma. A restricted recurrent phenotypic selection program is underway which encompasses the selection and intercrossing of the most 23 winterhardy field grown plants. Two cycles of selection have been completed to date. The basic objective of the alfalfa breeding program is the incorporation of multiple pest resistance into cultivars that are well-adapted to Oklahoma. Oklahoma lies in a transition zone in terms of the amount of winterhardiness needed in alfalfa cultivars. Cultivars developed for states north of Oklahoma tend to go dormant earlier in the fall than most producers would like, while those cultivars developed for southern and southwestern states, with less severe climatic conditions than Oklahoma's, do not have enough winterhardiness to persist under Oklahoma conditions. We are testing germplasm pools possessing a wide array of dormancy to identify the least dormancy required for persistence. Oklahoma has most of the major alfalfa insect and disease pests found in other alfalfa producing states. These include the alfalfa weevil, the spotted, pea, and blue alfalfa aphids, potato leaf hoppers and various foliage feeding caterpillars. A sizable amount of alfalfa seed is also produced in the state, and the alfalfa seed chalcid and the lygus bug are major insect pests of that enterprise. Important alfalfa diseases found within the state include anthracnose, downy mildew, lepto leafspot, spring and summer blackstem, and Phytophthora , Fusarium and Phymatotrichum root rots. Damage from these pests in reduced forage yield varies from season to season, but all will normally be important sometime during the life of a stand. Emphasis is on the development of germplasm pools containing high levels of genetic resistance to as many of these insect and disease pests as is possible. Presently the bulk of the effort is being spent on the incorporation and/or increase in the level of resistance to spotted and blue aphids and to anthracnose and Phytophthora root rot. Conventional greenhouse procedures are used to screen for resistance to aphids and anthracnose. In these procedures seedling plants grown in flats of sterilized soil are infested with the pest organism and the surviving plants are polycrossed to produce a progeny population in which subsequent screening and recurrent selec- tion can be practiced. Phytophthora root rot screening is presently done primarily under field conditions using irriga- tion to maintain saturated soil conditions necessary for disease development. Most of the germplasm pools used in the Droqram have satisfactory levels of resistance to the oea aphid and bacterial wilt. All practical efforts are made to maintain resistance to these pests. Efforts are also made in some of the breeding populations to retain, or to incorporate, resis- tance to the alfalfa weevil while increasing resistance to the other pests. The goal of this work is, as indicated, to develop germplasm pools which contain high levels of resistance to multiple insect and disease pests and from which cultivars well-adapted 24 to Oklahoma and surrounding areas can be extracted. This project was initiated in 1977. Progress to date has been the development of broad gene base germplasm pools tracing to six different source populations: Dormants (from northern Great Plains), semi-dormants (from southern plains and upper south), moderately dormant types (Bel tsvil le-6) , Oklahoma Commons (general adaptation), and Starnes Farm material (for resistance to weevil and anthracnose) . To each of these pools we have increased the frequency of genes for pest resistance through one or two cycles of selection for one to four pests. 25 The Forage-Livestock Industry in Oklahoma RANGE RESEARCH AT THE SOUTHERN PLAINS RANGE RESEARCH STATION P. L. Sims U.S. Department of Agriculture The Southern Plains Range Research Station (SPRRS) is located at Woodward, Oklahoma, in the northern part of the Southern Plains. This research center has direct responsibility for range and range livestock research in the High Plains, Rolling Plains and Breaks, Sandstone and Flint Hills, and the Reddish Prairie resource areas. The long-term average annual precipitation at the research station is 58 cm, with a range of 25 to 107 cm over 100 years. Approximately 70% of this precipitation occurs during the summer months. Summer temperatures average 20 C compared to 7 C in the winter, with a range from -32 to 45 C. MISSION The mission of the SPRRS is to increase the efficiency of red meat production and range resource utilization through inte- grated management of energy flow, nutrient cycling, and hydro- logic dynamics in forage-animal production systems in a manner consistent with perpetuation of the range resource. This mission is based on the "ARS National Research Program [for] Improved Vegetation and Management Practices for Range" (1976), which addresses three broad technological objectives: 1) germplasm enhancement, 2) development of range improvement practices, and 3) the development of grazing management systems. The station mission also supports the "Agricultural Research Service Program Plan" (1983a, 1983b). PROGRAM INTEGRATION Five research program areas established for the station include 1) forage germplasm enhancement, 2) forage plant physiological ecology, 3) rangeland soil-plant relationships, 4) range manage- ment and plant-animal interaction, and 5) rangeland systems analysis. Each of these areas is supported by a series of ob- jectives for research, and approaches vary from process-oriented studies to integrated interdisciplinary studies. The research is conducted by a team of scientists composed of an agronomist, 26 plant physiologist, soil scientist, range scientist, and systems ecologist. The following is a brief synopsis of some aspects of each of the station's research programs. Germplasm Evaluation and Enhancement Old World bluestems. Extensive research with Old World bluestems (BothriocEloa spp.) has shown that they have signi- ficant potential for use not only in the Southern Plains but throughout many regions of the United States. These grasses can contribute significantly to beef production and soil conserva- tion. Presently, the primary use of Old World bluestems is to reclaim marginal croplands interspersed among the region's rangeland. Research at the SPRRS helped to elucidate the agronomic, physiological, ecological, and animal utilization characteris- tics of the Old World bluestems. For example, basic studies have shown measurable differences in carbon dioxide exchange rate, water-use efficiency, and turgor maintenance between various accessions of Old World bluestems. These characteris- tics may be related to performance during drought and perhaps to other stress conditions (Coyne et al. 1982). Such results indicate that the Old World bluestems comprise a broad germplasm reservior that can be tapped for specific uses in forage-based beef production systems in the Southern Plains and throughout the southeast United States. These grasses have produced as much as 200 pounds of beef to the acre under semiarid to subhumid conditions (Sims and Dewald 1982), approximately four times what can be expected from well- managed native range. 'Plains' bluestem has been interseeded into an overgrazed mixed-grass prairie site. During subsequent years total forage production on the site was significantly increased in comparison to the check plots with no 'Plains' bluestem (Berg and Sims, in press). Many grasses such as the Old World bluestems have "chaffy seed," which is difficult to harvest and process. A seed harvester has been developed that combines both the flailing and vacuum principles into one unit for stripping seeds of this kind (Dewald and Beisel 1982). Work is underway to develop technology to process the chaffy seed to bare caryopses and thus facilitate planting. Eastern gamagrass. Eastern gamagrass (Tripsacum dactyloides) is a highly productive, extremely palatable grass that has received sporadic attention from researchers over the last 100 years. This grass has produced 3 to 4 metric tons of dry matter per hectare under irrigation at the SPRRS. This grass under optimum conditions has rapid regrowth potential and provides significant forage production on relatively fertile lands. A study of seasonal vegetative establishment and shoot reserves of eastern 27 gamagrass (Dewald and Sims 1981) showed that planting of shoot bases during the dormant period resulted in excellent stands. Some recently acquired accessions of eastern gamagrass germplasm have markedly greater seed production potential than previously acquired accessions. There appears to be sufficient genetic diversity in this species to support further germplasm development (Taliaferro et al. 1982). Weeping lovegrass. Weeping lovegrass (Eragrostis curvula) has been studied for more than 30 years. This research has led to wide usage of this species on sandy lands throughout the Southern Plains (Shoop et al. 1976, Voigt et al. 1970). Weeping lovegrass is a highly productive, drought-tolerant, warm-season species which can be used alone or in complement with native rangeland or other forages to optimize beef production on some fragile lands that are subject to wind erosion. Other species. Work has been conducted on an array of native species for over 40 years; species studied include sideoats grama (Bouteloua curtipendula), blue grama (Bouteloua graci 1 is), sand bluestem (Andropogon gerardi var. hallii), and an assortment of cool-season species (Mcllvain and Savage 1954). Currently, a collection of Andropogon gerardi accessions from a number of habitats is being gathered and grown in a common garden for detailed physiological and ecological studies that should lead to a greater understanding of the physiology and the ecology of this plant species and, ultimately, to an improved germplasm of this major component of native ranges in the Great Plains . Plant-Animal Interaction A short-duration grazing system is now being evaluated with yearling stocker steers on native sandhill range. The current effort includes a 16-pasture system stocked at 120 kg of live animal per hectare compared to a normal stocking rate of 76 kg of live animal per hectare for continuously grazed control pastures. The short-duration system steers are rotated at a 3.5-day interval throughout the year with adjustment for changing rates of grass growth. Preliminary results indicate that gains for the short-duration grazing system averaged 62 kg/ha compared to about 35 kg/ha for the continuously stocked pastures. In this study, forage production, species composi- tion, litter cover, soil water infiltration, and soil water are some of the site parameters measured. Cow-calf research leading to improved forage-livestock manage- ment systems has been an integral part of this station's activities since 1940. A native range-complementary annual farmed forage system developed at the SPRRS has consistently resulted in weaning weights of around 320 kg from crossbred cows (Dewald and Mcllvain 1975). This study has been expanded to include Brahman-Hereford females along with Angus-Hereford 28 cows bred to Simmental sires. These animals and their calves are being evaluated on the native range-complementary farmed forages system (4.4 and 0.4 ha, respectively) and on a native range program (7 ha per cow-calf unit). Cows on the native range system will be bred to calve in early spring while those on the native range-complementary forage system are bred for fall calving. REFERENCES Berg, W. A., and Sims, P. L. Herbage yields and water-use efficiency on a loamy site as affected by tillage, mulch and seeding treatments. J. Range Manage. (In press). Coyne, P. I.; Bradford, J. A.; and Dewald, C. L. 1982. Leaf water relations and gas exchange in relation to forage production in four Asiatic bluestems. Crop Sci. 22:1036-1040. Dewald, C. L., and Beisel, V. A. 1982. The Woodward Flail-Vac seed stripper for harvesting chaffy seeded grasses for range improvement. Society for Range Management, 35th Annual Meeting Abstracts, Calgary, Alberta, Canada, p. 21. Dewald, C. L., and Mcllvain, E. H. 1975. Forage fed 700-lb. weaner calves today— 1000-lb. weaners by 1980. Proceedings Forage and Livestock Conference — with economic considerations for the producer. Noble Foundation, Ardmore, Okla. [Dewald, C. L., and Sims, P. L.] 1981. Seasonal vegetative establishment and shoot reserves of eastern gamagrass. J. Range Manage. 34:300-304. Mcllvain, E. H., and Savage, D. A. 1954. Progress in range improvement. Adv. Agron. 6:1-65. Shoop, Marvin; Mcllvain, E. H.; and Voigt, P. W. 1976. Morpa weeping lovegrass produces more beef. J. Range Manage. 29:101-103. Sims, P. L., and Dewald, C. L. 1982. Old World bluestems and their forage potential for the Southern Great Plains. A review of early studies. U.S. Agric. Res. Ser. Agric. Rev. Man. South. Ser. No. 28, 15 pp. Taliaferro, C. M.; Dewald, C. L.; and Bush, L. J. 1982. Tripsacum dactyloides-germpl asm variability and potential for forage use. Agronomy Abstracts, 1982 Annual Meeting, American Society of Agronomy, p. 154. Madison, Wise. U.S. Agricultural Research Service. 1976. ARS national research program. NRP No. 20110. Improved vegetation and management practices for range. 45 pp. The Service [Washington, D.C.]. 1983a. Agricultural Research Service program plan. U.S. Dep. Agric. Misc. Publ . 1429, 73 pp. 1983b. Agricultural Research Service program plan. 6-Year 29 implementation plan. 1984-1990. 34 pp. The Service [Washington, D.C.]. Voigt, P. W.; Kneebone, W. R.; McIVvain, E. H.; Shoop, M. C.; and Webster, J. E. 1970. Palatibility, chemical composition, and animal gains from selections of weeping lovegrass, Eragrostis curvul a (Schrad.) Nees. Agron. J. 62:673-676. 30 Forage Plant Resources BUFFELGRASS GERMPLASM RESEARCH FOR THE SOUTHERN GREAT PLAINS E. C. Bashaw and C. W. Johns U.S. Department of Agriculture and Texas A&M University Buffelgrass (Cenchrus ciliaris L.) is a drouth resistant, perennial, warm-season species used as a range and pasture grass in arid areas with mild winter temperatures. It is apparently native to South Africa, where maximum variation exists, and extends north into India. Most natural ecotypes are obligate apomicts and obligate (completely) sexual plants have never been found in the native habitat. Fortunately the species is polymorphic, allowing for selection and use of superior introductions. Some apomictic accessions are being used for forage in arid regions of several countries and one strain, introduced into the USA about 1950, is responsible for over 90% of the revegetation in southwest Texas and north Mexico. Unfortunately, accessions received in the past have not had sufficient hardiness for consistent survival north of San Antonio, Texas and at high altitudes. Significant improvement in winter hardiness and adaptation to a wider range of soils could result in extensive use of this grass in the arid Southwest. This report reviews progress in buffel- grass improvement and the present status of germplasm research. Chance discovery of a sexual '’mutant" in a seed production field provided the first opportunity for improvement of buffelgrass. This plant proved to be heterozygous for method of reproduction and cross-compatible with apomictic ecotypes. When crossed with apomicts, both sexual and true-breeding apomictic F^ hybrids were produced. Genetic studies showed that method of reproduction was rather simply inherited in buffelgrass and that obligate apomixis could be manipulated effectively in a breeding program (Taliaferro and Bashaw, 1966). Until recently, improvement of buffelgrass was based on hybridization of the sexual mutant with apomictic accessions and selection and evaluation of the best obligate apomictic hybrids. Crossing the sexual plant with rhizomatous 31 apomicts resulted in some apomictic hybrids with strong rhizomes and improved vigor, yield and earliness. Protection afforded by the rhizomes enhanced both winter survival and persistence under grazing. True tissue resistance to cold was not achieved in these hybrids but two cultivars, 'Nueces' and 'Llano' (Bashaw, 1981), derived from them are able to survive about 100 miles further north than any other buffelgrass because of their rhizomes. As with most introduced apomictic grasses, lack of adequate native germplasm and sexual plants for use in hybridization have been serious problems in buffelgrass improvement. In 1976 we conducted an exploration in South Africa seeking new buffelgrass sources and acquired over 800 ecotypes with diverse characteristics and adaption to a wide range of soils. Some strains with exceptional drouth tolerance were recovered in a 4 to 8 inch rainfall area south of the Kalahari desert. Preliminary evaluations of the African collection for drouth and cold tolerance, vigor, seed production and digestibility were completed in 1982 and the results indicate that we have promising germplasm for the milder areas. Facultative apomictic plants were found among the progeny of 73 of the African accessions and the remaining 733 acces- sions were obligate apomicts (Table 1). No obligate sexual plants were recovered from the native habitat. Embryo-sac studies showed that individual ovules of facultative apomictic plants may contain: (1) only nuceller sacs, (2) nucellar sacs and a sexual sac, or (3) a single sexual sac with no evidence of nucellar development (Johns and Bashaw, 1980). The percentage of ovules with a sexual sac was low in most facultative plants (2 to 20%), but there were plants in which more than 60% of the ovules had a sexual embryo-sac. However, the percentage of variant offspring was always much lower than the apparent sexual potential indicating that many of the sexual sacs did not function in reproduction. Some obligate sexual plants have been identified among the progeny of 22 facultative plants. A few of the sexual plants are healthy and highly fertile and represent a valuable source of new sexual germplasm. However, most were weak and low in fertility or completely sterile. Over half of the sexual plants were aneuploid (37 or 38 chromosomes) with several lagging chromosomes , and embryo-sac studies revealed high levels of female abortion. Buffelgrass is apparently a segmental al lotetraploid (2n=4x=36, typically 2 IV + 14 II at diakinesis) and two or three lagging chromosomes are common at anaphase I. Sterility and low vigor of the sexual plants probably resulted from structural hybridity and accumulated aberrations in the apomictic parents and from inbreeding depression. These factors along with environmental effects and grazing pressure probably account for failure to find sexual plants in the 32 Table 1. — Mode of reproduction and chromosome number of African buffelgrass accessions Mode of Chromosome No . of reproduc t ion numbe r accessions Obligate apomixis 36 102 — 576 40 1 42 2 44 43^ 45 48 2 54 4 Facultative apomixis 36 676i 45 Obligate sexual — 0 Cold tolerant ecotypes with an alien genome of 9 univalent s . natural habitat. The derived sexual plants offer new prospects for hybridization and expanding the genetic base for breeding. However, considerable research is needed to identify the best derived sexual plants for use as germplasm. Forty-nine nonrhizomatous accessions obtained in a mountainous area around Beaufort West in the lower Cape Province proved to be more winter hardy than any previous buffelgrass. These accessions, representing at least 31 different ecotypes, survived temperatures of 10 to 20 0 F that destroyed all but the most rhizomatous strains during the winters of 1978 and 79 at College Station and Temple, Texas. In subsequent tests at 7 locations in north Texas , 6 of the accessions have survived without winter injury for three years. Further tests are needed to determine if any of the accessions possess sufficient winter hardiness to be released as cultivars. Since dormant tissue of these plants is relatively cold tolerant compared to the other nonrhizomatous buf felgrasses , they represent a valuable source of new germplasm. Cytological studies were conducted to determine mode of reproduction and chromosome number and behavior of the cold tolerant accessions. Six of the accessions were facultative apomicts and 43 were obligate apomicts. All 49 cold tolerant 33 accessions were found to be pentaploid with 36 buffelgrass chromosomes and an alien genome of 9 chromosomes that behave as univalents. The 18 buffelgrass chromsomes pair normally but the univalents show no tendency to pair among themselves or with buffelgrass chromosomes. They lag as a group of 9 at anaphase I and usually divide precociously before telophase. As a result most diads receive the whole alien genome. We assume that these plants probably were derived from fertilization of an unreduced egg in various tetraploid apomicts and that cold tolerance is conditioned by genes on the nine univalents. We hypothesized that significant improvement in winter hardiness might be achieved if the cold tolerance of these accessions could be combined with the strong rhizomes of some of our breeding lines. Two of the most vigorous cold tolerant strains were used as male parents in crosses with sexual buffelgrass and 21 hybrids were recovered. All hybrids contained the normal buffelgrass complement (36) plus one to nine of the univalents. Twenty hybrids were sexual, and one containing all 9 univalents was an obligate apomict. Fertility in the sexual hybrids was low (5 to 20% seed set) but the apomictic hybrid set 55% seed. These results suggest that further hybridization should result in hybrids with significantly better winter hardiness. We have just initiated studies to identify the most sexual and highly fertile facultative plants among the cold tolerant strains for use as female parents in crosses with highly rhizoraatous apomictic breeding lines. In addition to the usual type of hybrids we will be looking for plants derived by fertilization of an unreduced egg in the aposporous embryo sacs. This happens frequently in apomictic buffelgrass and allows for simultaneous combination of all the genomes of the female with the reduced complement of the male. With a wealth of new germplasm including new sexual parents and two potentially useful mechanisms for winter survival we are optimistic that significant advances can be made in expanding the adaptation of buffelgrass. REFERENCES Bashaw, E. C. 1981. 'Nueces' and 'Llano' buffelgrass. Texas Agr . Exp. Sta. Leaflet L-1819. Johns, C. W. , and Bashaw, E. C. 1980. New sources of sexuality in apomictic buffelgrass. Agron. Abstr. p. 58 Taliaferro, C. M. , and Bashaw, E. C. 1966. Inheritance and control of obligate apomixis in breeding buffelgrass. Crop Sci . 6:473-476. 34 Forage Plant Resources NEW SOURCES OF GENETIC VARIABILITY IN DALL IS GRASS AND OTHER PASPALUM SPECIES Byron L. Bur son and Paul W. Voigt, U.S. Department of Agriculture, and Wayne R. Jordan, Texas A&M University Paspalum is a large diverse genus with more than 400 species (Chase 1929). Common dallisgrass, P. dilatatum Poir. , one of the more economically important species in the genus , is native to eastern Argentina, Uruguay, and southern Brazil. It was introduced into the U.S. in the 1840’s (Chase 1929) and has spread throughout the southeastern U.S. where it is a valuable forage grass. Dallisgrass produces good quality forage, is highly palatable, and persists under heavy grazing. Its most serious problem is low seed fertility and quality. The grass is susceptible to ergot, Claviceps paspali Stevens & Hall, and this undoubtedly contributes to the low seed fertility. Common dallisgrass has 50 chromosomes that pair at metaphase I of meiosis as 20 bivalents and 10 univalents (Bashaw and Forbes 1958) . This suggests it is a natural hybrid with chromosomes from three different sources. The unbalanced chromosome consti- tution has been preserved because common dallisgrass is an obligate apomict (Bashaw and Holt 1958). However, apomixis has prevented any improvement through conventional breeding method- ology because of the lack of variability in the species (Burton 1962, Bennett et al. 1969). Plant breeders have utilized several different approaches during the past 50 years in an attempt to obtain or create variability within common dallis- grass and/or to circumvent the apomictic barrier. These include selection of plants from naturalized populations in the south- eastern U.S., acquisition of plant introductions from South America, radiation to create desirable mutations, chromosome doubling, and intra- and interspecific hybridization. For different reasons, these approaches to obtain or create varia- bility have been unsuccessful in producing improved forms of common dallisgrass. For the past several years, our primary thrust has concentrated on ways to circumvent apomixis . Some variable forms of dallisgrass have been obtained from these efforts. 35 PLANT INTRODUCTIONS On three occasions since 1975, the senior author collected PaspaZum germplasm in South America. Plant exploration was concentrated in the region where dallisgrass is considered to have originated as well as in adjoining areas. These include southern Brazil, Uruguay, northern Argentina, Paraguay, and southern Bolivia. Six different dallisgrass biotypes have been found in the area considered to be the center of origin of the species (table 1) . The common, prostrate and erect yellow-anthered bio types have been available to plant breeders in the U.S. for many years. However, a large number of phenotypically different ecotypes of common dallisgrass representing variability not previously available have been collected since 1975. Only one or two accessions of the variable yellow-anthered biotype from a small region in the northeast part of the state of Rio Grande do Sul in Brazil were available prior to 1975. The recently acquired accessions were collected from a broader area of their native habitat and are considerably more variable than the previously available accessions. There is a wide integradation of types ranging from plants similar to vaseygrass, P. UTVillei Steud., to the more typical yellow-anthered dallisgrass . It appears that natural hybridization is occurring between vaseygrass and yellow-anthered dallisgrass. These accessions and their progeny provide valuable sources of new germplasm and sexuality to use in the hybridization program. Two biotypes, the Torres and Uruguaiana dallisgrasses , had never been available through the U.S. Plant Introduction program. Both biotypes are considerably different from common dallisgrass (table 1) and the Uruguaiana biotype appears to have potential as a forage grass. A forage evaluation test was conducted at Temple, Texas, from 1978-1980 to determine the forage potential of 17 different biotypes including common dallisgrass. Forage yields and IVDMD values for the accessions are presented in table 2. The erect Uruguaiana biotype produced more forage than common dallisgrass. The four common dallisgrass ecotypes which were phenotypically different than typical common dallisgrass produced less forage than common, and the Torres biotype produced the least amount of forage. The IVDMD values for all the accessions were low because the samples were collected when the plants were mature. Common dallisgrass did not differ significantly from the other acces- sions in IVDMD (table 2) , but there were significant differences among the Uruguaiana types in IVDMD. Because there was not a significant correlation (r = -0.28) between yield and IVDMD for the Uruguaiana accessions, we believe that we may be able to select a more productive plant without adversely affecting IVDMD, but more detailed studies are needed. In 1982 the seven 36 Table 1. — Biotypes of Paspalum dilatatum from South America G o 44 -H O 44 4-4 4-» 44 44 O o o i— 1 i — 1 O O G 3 •H •H G G •rH •H P P S 0 G G 0 0 O O O o X X o O S M PM pm G G a a pm <3 C CO CO <3 <3 •H »* rG N CO Pi fX Ptf r G N G 44 G O /—N , — v G G G G 54 G <44 i-H 44 i — 1 44 rH G •H 54 •H O pq G O •H O •H O •H O 4-> pq 4-4 c o G N N N G •H G G G PQ r G G G G G G G 4-1 G G G r 54 44 54 44 54 44 54 toO G toO 54 bO G 44 G G pq G PQ G PQ G P 54 G 54 G rG O G G 44 '> — ' 44 v <■ 44 •>— ' 54 •H <3 rG <3 G 44 •H G CO CO CO P 5-4 4-1 toO G * 44 toO rH r— 1 i — 1 4-» C G G G O > G G G G G G C G G CO 54 O 54 54 CO O G 54 54 CO 54 CO 54 CO 54 •H 0) co G P 54 toO P G G G G P 4-4 4J r Pm 54 44 O 44 O 44 o 44 CO CO G >> < G CO P CO P CO p CO >4-4 G >> G 54 G G 54 G G G G O G G G G G 54 G G G G G £ G £ P G rG P toO G CO rG rG P rG P rG P rG G 4J toO 44 44 G 44 G 44 44 C 44 G 44 G 44 g 54 G 54 54 54 CO 54 G 54 G 54 G G G 54 O 54 o O G G •H o O 54 O 54 O 5 o <3 25 P 55 !5 Pm W <3 CO 55 O 2 O CO O !5 G G G 44 44 G 44 G G i — 1 G rG 54 54 44 54 4-4 4-4 | 4-1 44 44 G 44 44 5 *H •H CO CO a •H CO G O P 0 O O G 54 o G 54 G G 54 54 54 G 54 54 O P co cm PM W > PM W G G 44 P P G G 1 G 1 G G G 54 £ 54 £ 54 CO •H O 44 O G O G G G G g CO t — 1 rG i — 1 rG 54 G 0 g o r-H 44 i — 1 44 54 toO G o 54 G G G C O 54 55 CJ Pm 5>4 G >m G H P 37 Table 2. — Performance of new dallisgrass ecotypes and biotypes Accession Type Forage yield Mean IVDMD kg /ha % 461 Uruguaiana 8300 a 47.8 bed 455 II 8300 a 47.6 bed 426 II 7600 ab 48.0 bed 554 II 7600 ab 49.5 abc 458 II 7500 abc 47.5 cd 460 II 7400 abc 46.9 d 555 II 7300 abc 50.1 ab 427 II 6800 bed 48.0 bed 432 II 6500 b-e 47.3 cd 459 II 6500 b-e 48.0 bed 544 II 6300 c-f 50.5 a Mean II 7300 48.3 Common Common 6300 c-f 48.6 a-d 552 Common off type 5900 d-g 49.7 abc 543 II II II 5500 efg 49.0 a-d 547 11 II II 5200 fgh 47.4 cd 351 II II II 5000 gh 48.0 bed Mean ft II II 5400 48.5 178 Torres 4000 h 48.4 a-d lean of 3 years. 38 best accessions and common dallisgrass were planted into forage tests at Temple and Angleton, Texas, for further evaluation. Because previous plant breeders and geneticists have used plant introductions in an attempt to improve the species, it may seem unusual that this approach is more successful today. Essen- tially all Paspatum introductions obtained prior to 1975 were collected by taxonomists from the countries where the plants were growing. Because they were unaware of the objectives of the breeding programs and did not understand the problems asso- ciated with breeding apomictic grasses, their collections did not comprise the best available germplasm. The senior author was familiar with these problems, and therefore was able, with the assistance of South American scientists, to identify valu- able new ecotypes and biotypes not included in previous collec- tions. This clearly demonstrates the necessity that the geneticist working with a particular species to explore the center of origin of the species and collect the diverse germ- plasm that will be of value to the improvement program. Accessions of four other Paspalum species are also being evalu- ated at Temple. The species are P. nicovae _, P. plioatulim j P. denticulatwn , and P. unispicatum. INTRA- AND INTERSPECIFIC HYBRIDIZATION Numerous interspecific Paspalum hybrids have been made during the past several years. Many of these have involved the erect sexual yellow-anthered dallisgrass biotype as the female parent. Most hybrids were made for cytogenetic analysis to determine the phylogenetic relationships among various species (Burson 1983) . Even though some hybrids have potential as productive forage grasses, they all lack sufficient fertility to be propagated by seed. Attempts to restore fertility in these hybrids have not been successful. Because of sterility in the interspecific hybrids, intraspecific hybridization appears to have the most potential in producing improved germplasm. A fertile intraspecific hybrid has been produced between the yellow-anthered and common dallisgrass biotypes (Bennett et al. 1969). However, because of the insta- bility of the 10 common dallisgrass univalent chromosomes in the Fj hybrid, apparently some important genetic material is lost when the 10 univalents are eliminated in the subsequent genera- tions. Recently, this yellow-anthered X common Fj hybrid was backcrossed to common dallisgrass, and some variable purple- anthered plants have been recovered. This could provide a means of stabilizing some of the univalents as backcrossing to common dallisgrass is continued. In the future the sexual yellow- anthered plants will be crossed with the Uruguaiana biotype in an attempt to recover improved types. 39 TISSUE CULTURE Variation is observed frequently in tissue culture-derived plants. When small segments (3-5 mm in length) of young common dallisgrass inflorescences or stems are placed on an agar based medium, callus tissue often develops. After the callus has increased to a sufficient size, it is transferred to other media with appropriate auxin-cytokinin content to stimulate shoot and root differentiation. These young plantlets are transplanted into peat pots in the greenhouse and eventually planted into a field nursery. More than 300 plants have been established in a field nursery. Differences in plant size, leaf width, leafiness, and pubescence have been noted. Because we are interested in increasing drought tolerance in dallisgrass, a study was initiated to determine if there were differences in heat tolerance among the tissue culture-derived plants. Using a heat tolerance test (Sullivan 1972) , 250 plants were tested under both drought stressed and non-stressed conditions. When drought stressed, the percent cellular damage from a heat challenge of 50°C for 45 min ranged from 20 to 70% with the largest number of plants occurring in the 50% damage category. The common dallisgrass control was also in the 50% damage category. Plants were visually rated for their susceptibility or resistance to stress when under drought conditions. The visual ratings and the cellular damages from the heat tolerance test were negatively correlated (r = -0.32**). When the plants were sampled under non-stressed conditions, the percent damage ranged from 60 to 100% with the most frequent category being 80%. Common dallis- grass was in the 80% category. These findings suggest that differences occur among the tissue culture-derived plants in their ability to acclimate to heat or drought stress . Because common dallisgrass is an obligate apomict, this amount of varia- bility would not normally be expected and, in fact, was not observed among the common dallisgrass checks. Therefore, it appears that real differences do occur in the tissue culture- derived plants. Research is underway to further evaluate the physiological and genetic bases for these differences under controlled conditions. Variants produced in this manner may be important in extending the range of adaptation of common dallis- grass into drier regions. 40 references Bashaw, E. 1958. Bashaw, E. 1958. Bennett, H. 1969. Bur son, B. 19-83. Burton, G. 1962. Chase, A. 1929. C., and Forbes, I., Jr. Chromosome numbers and microsporogenesis in dallis- grass Paspalum dilatatum Poir. Agron. J. 50: 411-445. C. , and Holt, E. C. Megasporogenesis , embryo sac development and embryo- genesis in dallisgrass Paspalum dilatatum Poir. Agron. J. 50: 753-756. W.; Burson, B. L.; and Bashaw, E. C. Intraspecific hybridization in dallisgrass, Paspalum dilatdtum Poir. Crop Sci. 9: 807-809. L. Phylogenetic investigations of Paspalum dilatatum and related species. In J. A. Smith and V. W. Hays (eds.), Proc. 14th Int . Grassl. Congr., Lexington, Ky., 15-24 June, 1981, pp. 170-173. Westview Press, Boulder, Colo. W. Conventional breeding of dallisgrass Paspalum dilatatum Poir. Crop Sci. 2: 491-494. North American species of Paspalum. Contr. U.S. Natl. Herb. 28(1): 1-310. Sullivan, C. Y. 1972. Mechanisms of heat and drought resistance in grain sorghum and methods of measurement. l!n N. G. P. Rao and L. R. House (eds.). Sorghum in the Seventies. Oxford and IBH Publishing Co., New Delhi, India. 41 Forage Plant Resources CLOVER AND SPECIAL PURPOSE LEGUME GERMPLASM RESOURCES FOR THE FUTURE Gary A. Pederson and William E. Knight U.S. Department of Agriculture INTRODUCTION Between 1950 and 1970, research on clovers and special purpose legumes was greatly reduced in the U.S.; however, in the last ten years, interest has rekindled in the utilization of legumes in pastures and in conservation tillage systems. The energy crisis stimulated part of this interest as well as a need perceived by livestock producers for higher quality forage with better seasonal distribution. Recent economic conditions in the livestock industry have resulted in erodable land being placed in row crops. Increased erosion clearly shows the need for better conservation tillage systems. Legumes should be an integral part of these systems. Presently, there are 75 legume species (excluding alfalfa) in public improvement programs in the Southeastern U.S. This paper will summarize the state and the U.S. Department of Agriculture programs involved in improvement of these 75 legume species. Also, examples of current research that will result in improved legume germplasm for the future will be cited. AREAS FOR IMPROVEMENT There are a number of areas where significant improvement can be made in the germplasm of the future: 1) Increased forage yield distribution and quality. 2) Resistance to fungal pathogens. 3) Resistance to legume viruses. 4) Nematode resistance. 5) Insect resistance to primary pests as well as to vectors of viruses . 6) Improved N2 fixation by developing a more efficient symbio- tic relationship with improved Rhizobium strains. 42 7) Increased drought tolerance and winterhardiness to improve persistence and increase adaptation of legumes to South- eastern U.S. climatic conditions. 8) Improved seed production to enable producers to establish initial stands and improved reseeding in annuals for establishment of subsequent stands. 9) Development of interspecific hybrids to utilize traits from other legume species. 10) Plant introduction and evaluation to broaden the gene base and reduce genetic vulnerability. WHITE CLOVER White clover, Trifolium repens L. , is a perennial clover widely adapted for use in pastures of the Southeastern U.S. Alabama, Florida, Louisiana, Mississippi, North Carolina, and South Carolina are presently involved in white clover improvement (table 1). The standard white clover cultivars are the ladino types, 'Regal' and 'Tillman', and the intermediate types 'Louisiana S-l' and 'Nolin's Improved'. Regal is a five-clone synthetic released by the Auburn University Agricultural Experi- ment Station in 1962 (Johnson et al. 1970) while Tillman, a six-clone synthetic, was released by the South Carolina Agricul- tural Experiment Station and the U.S. Department of Agriculture in 1965 (Gibson, Beinhart, and Halpin 1969). Louisiana S-l, a five-clone synthetic released by the Louisiana Agricultural Experiment Station in 1952 (Hollowell 1958), and Nolin's Improved, a naturalized cultivar, behave as reseeding annuals in the Southeast. Objectives of white clover research include improvement of per- sistence and development of resistance to viruses, root dis- eases, and nematodes. Prior to retirement. Dr. P. B. Gibson led an extensive and productive multidisciplinary program on white clover at Clemson, S.C. He led a research team that produced 11 interspecific hybrid combinations, eight for the first time, among five species of clover (Chou and Gibson 1968; Gibson and Beinhart 1969; Gibson et al. 1971; Gibson and Chen 1975) . The Clemson team developed a technique to screen white clover plants for resistance to root-knot nematodes (Baxter and Gibson 1959) and released SC-1 root-knot nematode resistant germplasm in 1972 (Gibson 1973). A forthcoming virus- resistant germplasm release was developed through a five-state cooperative project coordi- nated by Dr. P. B. Gibson. Plants were screened by mechanical inoculation, aphid inoculation, and field tests for resistance to alfalfa mosaic virus, peanut stunt virus, and clover yellow vein virus. Further field screening resulted in 44 clones being found resistant to these viruses in all tests. A germplasm re- lease of this material will be made shortly. Other promising white clover germplasm includes Florida XP-1 and XP-2 developed by the Florida Agricultural Experiment Station and the Brown Loam germplasm developed by the Mississippi Agricultural Experi- ment Station and the U.S. Department of Agriculture. 43 Table 1. — Forage legumes other than alfalfa in public improvement programs in the Southeastern United States^ M 3 & • CO 3 /—N e o 3 6 /•—N 3 1 — 1 •H 3 /^*N 3 4-1 • • 3 5-4 4-1 s i — 1 3 /- — \ u CO T3 3 3 •H 3 • •H 3 3 3 3 H •H a CO 05 3 5-i 6 3 3 a a 05 a X 3 3 •H 4-1 3 C0 > CO 05 O 3 i — 1 3 3 05 i — 1 3 50 3 6 3 e 5-1 3 •H ■H H 3 3 H PO a • •H > — ' •H , S •H t— i • i — 1 £ H H 1 ^ 6 H| O M— 1 '■w' 44 cd •H 05 H| 1 6 3 H H |H 3 5-i i — 1 ''w' 3 O 4-/ s — * •H 3 H S 3 CO H •H ' — ' o r— 1 CO 6 3 3 3 3 !-i i — 1 3 •H 3 4-1 3 CO 5-1 cd 3 3 3 •H 3 3 S-i < PQ PQ C_> S a Pi CO PQ n 3 3 5-i 3 a 5-i 3 a) a co c a • 3 -1-1 H| -U 5-i "■ — ^ CU *H 5-1 rC 6 a d • o 3 3 • h h a> 5-1 Hi'-' -H ,£3 3 as a 3 *h 3 m 'd co g 5-1 4-1 •sc -sc -sc H H H a pd a •SC -sc * •SC t-H C\] I— 1 CN CN CN Pi a a a Pi Pi • «■. •SC *> ■SC *> »• • PQ CO PQ CO CO CO PQ PQ • CO CO CO CO CO CO CO CO CO CO CO PQ • CO CO CO CO CO CO CN Ptf CN CN CN rH • • t—l Pi Pi Pi Pi • t—l • • Pi • Pi • • rv CO PQ CO PQ • • • CO • CO •SC -5C rH i — 1 t—l ptf Pi Pi •SC * »> PQ PQ PQ B B i — 1 3 3 • ✓ — N •H 3 3 3 3 Ou 4-4 4-1 3 3 •H •H H /-v 3 *H 6 50 3 • 3 •H i— 1 a) 3 3 3 4-1 3 O • CO 3 /—N 6 3 3 3 4-4 4-4 3 O 3 a *h N H 3 •H 3 3 3 3 4-1 3 H • 3 33 4-1 t — 1 3 a •H 3 3 H a 3 3 3 a a r3 3 X> 3 3 6 •H a 3 50 t—l 6 3 3 3 •H 3 3 3 4-1 4-1 H 3 H| 3 3 3 3 ,H| 1 10 3 3 3 3 O 3 3 H | w 5-1 5-4 50 3 3 3 •• 4-1 4-1 N 3 X -H 50 N 3 -3 *H 3 Pi CO CO CO S N 3 05 a CO 05 . • cd h| ail 05 •H 5-i 05 3 H|' O I — I O 05 a 3 • co 05 Cd PQ 3 44 Table 1. — Forage legumes other than alfalfa in public improvement programs in the Southeastern United States — continued x H w cn m co w w w PS pS PP PP >1 s-j >| PO g •H O pq O > CO w OJ g co co co co co co cd Cl d d cd CO cd o cd u •H g i — i PC o •H d d i — i Ci •H CO cd •H •H cd X •H cd Cl d u — - CO i — i cu g o cu cd d C d u 4-1 d) cd Cl C o cd g d cu £ > cu x> •H d cd ds 0) cd •H > CO cd Cl •H ,d g 4-) 4-1 o cd cd g o d o g o cd CO •H rH rd M— 1 cu d •H 4-J d d u •H CO u 1 1 CO o PO d) *H X i — i d cd cd o CO Cl > dd cd d cu CU > •H Cl cd •H d O o 4-4 u CO O cd CO 4-1 d- > g CO X d CO d o CO d) • • rH cd cd (U • Ci < •X r~H X < o X r-H X pi p-l CO CO *> CO X X < X 'rl X x • CO 3 bO • cd 3 X X • 4-1 co X 3 S 3 3 4-J X CO 3 3 X 3 5-4 3 3 , V 3 •H •H 3 S 3 X 3 • n • a, CO 3 bC O /"-'N 3 3 3 X 3 • cd 3 Q X 3 O 3 5-1 X v — / 3 •H 3 •<— > 03 3 X O 3 3 <4-4 X /— s 3 3 3 a 3 a X o 6 3 X 3 I — 1 5-4 3 X 2 X 0) bO 3 X 3 3 3 3 < v ' 3 X 6 •H ■H X i — O X T— I 3 X 3 3 3 T3 5-i 3 X i — 1 X r— 1 3 o pj 3 3 60 3 3 3 X 3 X 3 X •H O 3 X 3 3 X 3 4-) W 3 X 3 3 X X i — 1 3 3 3 — i 3 X 03 'w' O * X s O 3 3 i — 1 3 3 X 3 P, *H 5-1 3 CO X 3 3 3 3 3 •H X 3 X O O 3 o 3 X 3 3 3 b0 pL, 3 O ' <■ 3 X bO 3 3 X 3 CO 3 X 3 O X ' — * a 3 3 CO •H X a X 3 i — 1 3 6 5-i X 3 CL, XI 6 X 3 X o 0 X 3 rH X 3 X 3 3 3 c! X CO O 3 i — i i — 1 3 3 3 3 Cl, 3 •H •H 3 4-J o •H 3 3 3 X 3 X •H 3 4-> 3 3 Q O X 3 X X X 3 X 3 3 O o CO •H 1! 3 X 3 3 o 3 3 3 3 O X 3 5-4 >> a N— " •H X 3 O X 3 O bO 3 3 3 CL, i— i •H X 3 1 — 1 3 X 3 6 3 X X bO O cd 3 3 >•> I — 1 3 o 3 • •H 3 3 3 X 3 o X X X X X X s S| s s X CX P-4 Ph PL| I P-l I 46 scandens tetragonolobus Table 1. — Forage legumes other than alfalfa in public improvement programs in the Southeastern United States — continued x H WWW o CO o S3 CO CO CO CO CO c3 pq 6 P CL) 5-1 P CO a p 5-1 p q> cd a i — i o P !-i a 4-1 CO •H cd co •H P cu i — 1 P 6 XI •H O q) •H e o •H O qJ cd i — 1 5-1 si cd CO 1 — 1 •H X CO cd p p r— 1 4-) ,P CO •H a) cd p cd O. P cu CO o rH o o O cd -P p •H P hJ o cd P i — l p cd* cu cd P o w p i — i .. o p p p P CU p cd o o cd • cd cd cd s> CO cu rH o (U p S 2 cn S ■H cd P .a •H 4-4 p o cd P u P p O cd P cd 1— 1 rP 00 CO CO p 4-4 P 5-1 >> CU 00 •H p P •H CO CO CO co |co|co|co|H > >1 Pj cd (P cu co cx 5-i cd (U *H P P P P •H O 00 P •H X) cu a) P pq w II II i— I CN pq co p3 Pi 47 Evaluation and selection. Released germplasm or cultivar. Release probable within 5 years. RED CLOVER Red clover, Trifolium pratense L., is a perennial clover which is more widely used in other areas of the U.S. other than the Southeast. Florida, Kentucky, and Louisiana are involved in improvement programs on red clover (table 1). For the Southeast, the standard cultivars are ’Kenland' and 'Kenstar'. These two cultivars were developed by the Kentucky Agricultural Experiment Station and the U.S. Department of Agriculture in 1951 (Hollowell 1951) and 1973 (Taylor and Anderson 1973), respectively. Objectives of red clover improvement programs include increasing persistence in the Southeast, developing fundamental cytogenetic knowledge on Trifolium species, and increasing resistance to viruses, root diseases, and nematodes. Dr. N. L. Taylor and associates at the Kentucky station have established an extensive red clover improvement program. Recent red clover germplasm re- leases include 11 gene markers (Taylor 1982) and 9 generations of bulked plant introductions (Taylor 1979) . The Kentucky pro- gram has also been instrumental in interspecific hybridization of Trifolium species and has released germplasm of Trifolium sarosiense x TL alpestre and Trifolium medium x T_. sarosiense in 1978 (Taylor and Quesenberry 1978a, 1978b) . ARROWLEAF CLOVER Arrowleaf clover. Trifolium vesiculosum Savi. , is a winter an- nual clover that has received a great deal of interest in the Southeast recently. Florida, Georgia, Louisiana, Mississippi, and Texas are involved in arrowleaf clover improvement (table 1) . The standard cultivars of arrowleaf clover are ' Amclo ' , 'Yuchi' , and 'Meechee' . Amclo, an early-maturing cultivar, was released by the Georgia Agricultural Experiment Station and the U.S. Department of Agriculture (Beaty et al. 1965). Yuchi, a later- maturing type, was released by the Auburn University Agricultural Experiment Station (Hoveland 1967) . Meechee, the latest maturing cultivar with the most winterhardiness, was released by the Mississippi State University Agricultural Experiment Station and the U.S. Department of Agriculture (Knight et al. 1969). The objectives of arrowleaf clover research are to obtain early emergence with uniform stands, increase late spring persistence, improve drought tolerance, and increase resistance to viruses, root diseases, and nematodes. Virus diseases are a major pro- blem not only in arrowleaf but in many of the other true clovers. Viruses can reduce the clover stand by direct effect or by in- creased susceptibility to secondary pathogens through reduced plant vigor. Arrowleaf clover is more susceptible to nematode damage than other annual clovers (Nichols et al. 1981). Nema- tode-resistant germplasm should be forthcoming from the Florida breeding program. 48 CRIMSON CLOVER Crimson clover, Trifolium incarnatum L., is a winter annual le- gume with a distinctive bright crimson flower. Florida, Kentucky, Louisiana, Mississippi, and Texas have active im- provement programs in crimson clover (table 1) . The standard cultivars are ’Dixie’, ’Chief', ’Autauga’, and ’Tibbee’. Dixie was released in 1946 by the Georgia Agricultural Experiment Sta- tion and the U.S. Department of Agriculture (Hollowell 1953). Chief and Tibbee were released by the Mississippi Agricultural Experiment Station and the U.S. Department of Agriculture (Hollowell 1960 and Knight 1972) . The objectives of crimson clover research are to improve N£ fix- ation, improve fall growth, reduce seed shatter and lodging, and increase resistance to the clover head weevil, viruses, root diseases, and nematodes. The clover head weevil is a major in- sect problem on crimson clover. Feeding by this insect reduces the reseeding ability of crimson clover and may result in poor stands in subsequent years. A screening program is presently underway in Mississippi to develop resistant germplasm. SUBTERRANEAN CLOVER Subterranean clover, Trifolium subterraneum L., is a winter an- nual clover which only recently has been extensively grown in the Southeast. Florida, Georgia, Louisiana, Mississippi, and Texas are involved in the improvement of subclover (table 1) . The standard subclover cultivars are ’Mt. Barker’ , ’Nangeela’ , ’ Woogenellup ’ , and ’Meteora’ , which are imported from Australia. Objectives of subclover programs include evaluation of Australian material for adaptation to the Southeastern U.S., increasing forage yield, improving reseeding ability, and developing resistance to viruses and mildew. A subclover germ- plasm with increased adaptation to the Southeast has been de- veloped by the Mississippi Agricultural Experiment Station and the U.S. Department of Agriculture. The Mississippi ecotype subclover has persisted for over 30 years from an original seed- ing of the Australian cultivars Mt. Barker, ’Bacchus Marsh’ , and ’Tallarook’. This subclover will be released as germplasm or as a cultivar in the near future. BERSEEM CLOVER Berseem clover. Trifolium alexandrinum L., is a winter annual clover that is in the improvement programs of Florida, Louisiana, Mississippi, and Texas (table 1). Presently, there are no standard cultivars for the Southeast except Florida since no cultivars have adequate winterhardiness. The main ob- jective of berseem clover improvement is to increase the winter- hardiness of the species. Secondary objectives include deve- loping resistance to leaf diseases and improving recovery after clipping. A winterhardy berseem clover has been developed in 49 Mississippi from plants of the Italian cultivar 'Sacromonte' that survived field temperatures as low as 5° and 8° F. A re- lease of this material will be made shortly. PERSIAN CLOVER Persian clover, Trifolium resupinatum L., is widely adapted in the South. Producers' fear of bloat has caused this species to fail to reach its potential as a forage crop. A wide range of plant material, representing wide variations in maturity, forage yield, and recovery after clipping, is available through plant introduction (Massey 1966) . Weihing (1962) applied selec- tion pressure for hard seed to desirable agronomic types from three plant introductions to develop the improved, hard-seeded cultivar 'Abon'. Reluctance to use Persian clover should di- minish with present knowledge and use of poloxalene in blocks and in molasses mixtures. Interest in Persian clover has increased, and Abon is being eva- luated in regional tests. BALL CLOVER Interest in ball clover, Trifolium nigrescens L., has increased in Alabama, Mississippi, and Louisiana. The Alabama Experiment Station includes this species in its improvement program. Breeding objectives in this program are 1) increased forage and seed yield, and 2) improved pest resistance. Recently a farmer variety, 'Segrest' was added to the regional variety tests, and the Alabama station plans to make a germplasm release in the near future. VETCH Vetches are used most commonly as winter cover crops in the Southeast, with 75% of the total acreage in Oklahoma, Arkansas, Texas, and Louisiana. However, vetch makes hay, silage, and pasture of high quality. Hairy vetch, Vicia villosa Roth, accounts for 85% of the vetch acreage in the U.S. In 1959, the Alabama Agricultural Experiment Station released 'Warrior', a variety of common vetch, Vicia sativa L. (Donnelly 1965a) . Warrior vetch is resistant to the vetch bruchid and three species of root-knot nematode and produces high forage and seed yields . The Louisiana Seed Company is distributing seed of four proprie- tary varieties from the Alabama program (Donnelly 1979) . These recent releases are 'Vantage', 'Cahaba White', 'Nova II', and 'Vanguard' . At the Kentucky Agricultural Experiment Station, a locally adapted strain of bigf lower vetch, Vicia grandif lora var. kitaibeliana W. Koch, has shown promise as a pioneer legume in 50 pasture renovation research conducted by Templeton and Taylor (1975). 'Woodford' has been released as a new variety of big- flower vetch from this work. LESPEDEZA Both perennial and annual lespedezas have been under evaluation for use in the Southeast. Most of the current research is on the perennial sericea lespedeza, Lespedeza cuneata (Dum.) G. Don, rather than the two annual lespedezas — striate, Ij. striata (Thunb.) Hook & Arn. , and Korean, L^. stipulacea Maxim. Alabama, Kentucky, Louisiana, and North Carolina are involved in sericea lespedeza improvement (table 1) . The main objectives of sericea lespedeza research have been to develop a lespedeza low in tannin and to increase nematode resistance. Until recently, the stan- dard cultivars of sericea lespedeza were 'Arlington', 'Serala', and 'Interstate'. Serala and Interstate were released by the Auburn University Agricultural Experiment Station in 1962 and 1969 (Donnelly 1965b and 1971) . In 1978, three cultivars were released. 'Serala 76' and 'Interstate 76' released by the Alabama and Georgia Agricultural Experiment Stations and the U.S. Department of Agriculture contained nematode resistance and other improvements not in the original cultivars (Donnelly and Minton 1979). 'Appalow' , the first prostrate lespedeza, was released by the Kentucky Agricultural Experiment Station and the U.S. Department of Agriculture (Henry and Taylor 1981). In 1980, 'AU Lotan' was released by Dr. E. D. Donnelly of the Alabama Agricultural Experiment Station (Donnelly 1981) . This cultivar is low in tannin content and has greater nematode re- sistance than previous sericea lespedeza cultivars. LUPINE Lupine species that are under evaluation in the Southeast are blue lupine (Lupinus angus tifolius L.), white lupine (L. albus L.), and bicolor lupine (L^. hispanicus spp. bicolor , Merino, J. S. Gladstones). The objectives of lupine improvement are to reduce the alkaloid content, increase seed shattering resis- tance, and increase winterhardiness. Georgia and Louisiana are involved in lupine improvement. A number of recent cultivar and germplasm releases have been made by Dr. John D. Miller, Dr. Homer D. Wells, and others of the U.S. Department of Agri- culture and the Georgia Agricultural Experiment Station. The standard cultivars of blue lupine have been 'Richey' , 'Borre' , 'Rancher', 'Blanco', and 'Frost'. Blanco and Rancher were released by the Georgia Agricultural Experiment Station and the U.S. Department of Agriculture in 1960 (Forbes et al. 1964 and Forbes and Wells 1967) . Frost was released by the Georgia and Florida Agricultural Experiment Stations and the U.S. Department of Agriculture (Wells et al. 1980a) . This cultivar showed more seed-shattering resistance than previous cultivars. Also, Georgia and the U.S. Department of Agriculture released a 51 winterhardy germplasm, WH-1, of blue lupine in 1980 (Wells and Miller 1981) . The standard cultivar of white lupine was ’Hope' released by the Arkansas Agricultural Experiment Station in 1970 (Offutt 1971) . In 1980, the Georgia Agricultural Experiment Station and the U.S. Department of Agriculture released ' Tifwhite-78 ' white lupine, which has a low alkaloid content, improved winter- hardiness, and decreased seed shattering (Wells et al. 1980b). In 1982, a bicolor lupine germplasm, Bicolor-1, was released by the Georgia Agricultural Experiment Station and the U.S. Depart- ment of Agriculture (Miller and Wells 1983a) . BIRDSFOOT TREFOIL Interest in birdsfoot trefoil, Lotus corniculatus L., has in- creased in the South. New releases are being considered in Alabama and Georgia. Kentucky recently released the cultivar ’Fergus'. These cultivars have been developed for 1) improved forage yield, 2) improved seed yield, 3) improved persistence, and 4) improved pest resistance. ’AT-P’ birdsfoot trefoil will be released by the Alabama Agricultural Experiment Station if adequate seed production can be obtained. OTHER LEGUMES Many other legume species are involved in selection and improve- ment programs in the Southeastern U.S. Though not all of the species can be covered in this paper, some of the recent and forthcoming releases will be cited. The Alabama Agricultural Experiment Station is planning germplasm releases of T_. purpureum and T_. mutabile . The Georgia Agricultural Experiment Station and the U.S. Department of Agriculture released Tift-1 hyacinth bean (Lablab purpureus (L.) Sweet) germplasm in 1982 (Miller and Wells 1983b) and Tifhardy-1 Desmodium canum (J. F. Gmel.) Schinz and Thell. germplasm in 1981 (Miller and Wells 1981) . The Florida Agricultural Experiment Station released ’Florida’ Desmodium heterocarpon (L.) DC. in 1979 (Kretschmer et al. 1982). The Kentucky Agricultural Experiment Station re- leased KY M-l zigzag clover (Trifolium medium L.) germplasm in 1982 (Taylor et al. 1982). In Florida, Dr. A. E. Kretschmer has an extensive evaluation program involving over 4,000 accessions of tropical legumes. Some of the tropical legumes are being evaluated at the Iberia Livestock Experiment Station in Louisiana by Dr. C. C. Shock. Although some of these species may be adapted and contribute to forage systems in the tropical part of the region, their use over much of the region will be limited by climatic conditions. This paper covers primarily the public improvement programs for legume species other than alfalfa. However, the contribution of industry to the development, promotion, and distribution of 52 legume cultivars should be recognized. Private industry has been actively involved in red clover improvement with the release of many cultivars, including 'Florie' by Northrup King and Company, ’Redland II’ by North American Plant Breeders, and 'Redman' by FFR Cooperative (Buker et al. 1979) . 'Arcadia' is an improved ladino clover distributed by Northrup King and Company. Cal/ West Seeds has a white clover improvement program with several experimentals in the regional evaluation test. The Louisiana Seed Company has increased and distributed four proprietary cultivars from the Alabama vetch breeding program (i.e., Vantage, Cahaba White, Nova II, and Vanguard) as well as Tibbee and Chief crimson clovers released by the U.S. Department of Agriculture and the Mississippi Station. A wide range of diverse legume germplasm is available for gen- eral and for special purpose use in the Southeastern U.S. How- ever, the large number of species involved will increase the breeder's challenge to provide the public with improved legume germplasm resources in the future. REFERENCES Baxter, L. W. , and Gibson, P. B. 1959. Effect of root knot nematodes on the persistence of white clover. Agron. J. 51:603-604. Beaty, E. T.; Powell, J. D.; and Young, W. C. 1965. Amclo arrowleaf clover. Crop Sci. 5:284. Buker, R. J.; Baluch, S. J.; and Sellers, P. A. 1979. Registration of Redman red clover. Crop Sci. 19:928. Chou, Meei-chih, and Gibson, P. B. 1968. Cross compatibility of Trifolium nigrescens with diploid and tetraploid Trifolium occidentale . Crop Sci. 8:266-267. Donnelly, E. D. 1965a. Registration of Warrior vetch. Crop Sci. 5:605. 1965b. Serala sericea. Crop Sci. 5:605. 1971. Registration of Interstate sericea lespedeza. Crop Sci. 11:601-602. 1979. Registration of Cahaba White, Vantage, Nova II, and Vanguard vetch. Crop Sci. 19:414. 1981. Registration of AU Lotan sericea lespedeza. Crop Sci. 21:474. Donnelly, E. D., and Minton, N. A. 1979. Registration of Serala 76 and Interstate 76 sericea lespedeza. Crop Sci. 19:929. Forbes, Ian, Jr.; Burton, Glenn W. ; and Wells, Homer D. 1964. Registration of Blanco blue lupine. Crop Sci. 4:448. Forbes, Ian, and Wells, Homer D. 1967. Registration of Rancher blue lupine. Crop Sci. 7:278. Forbes, Ian; Wells, H. D. ; Edwardson, J. R. ; Burns, R. E.; and Dobson, J. W. 1970. Registration of Frost blue lupine. Crop Sci. 10:726. 53 Gibson, P. 1973. Gibson, P. J. T. 1971. Gibson, P. 1969. Gibson, P. 1969. Gibson, P. 1975. Henry , D . S 1981. Hollowell, 1951. 1953. 1958. 1960. Hoveland, C 1967. Johnson, W. 1970. B. Registration of SC-1 white clover germplasm. Crop Sci. 13:131. B.; Barnett, 0. W. ; Chen, Chi-Chang; and Gillingham, Interspecific hybridization of Trifolium unif lorum L. with T”. occidentale D. Coombe and with _T. repens L. Proc. Assn, of So. Agric. Workers. (Abstr.) Feb. 1-3, Jacksonville, Fla. B., and Beinhart, G. Hybridization of Trifolium occidentale with two other species of clover. J. Hered. 60(2): 93-96. B.; Beinhart, George; and Halpin, J. E. Registration of Tillman white clover. Crop Sci. 9: 522. B., and Chen, Chi-Chang. Registration of SC-2 and SC-3 clover germplasms. Crop Sci. 15:605-606. . , and Taylor, N. L . Registration of ’Appalow' sericea lespedeza. Crop Sci. 21:144. E. A. Registration of varieties and strains of red clover II. Agron. J. 43:242. Registration of varieties and strains of crimson clover, I. Agron. J. 45:318-320. Registration of varieties and strains of white clover (T. repens) . Agron. J. 50:692. Registration of varieties and strains of crimson clover, II. Agron. J. 52:407. . S. Registration of Yuchi arrowleaf clover. Crop Sci. 7:80. C. ; Donnelly, E. D.; and Gibson, P. B. Registration of Regal white clover. Crop Sci. 10:208. Knight, W. E. 1972. Registration of Tibbee crimson clover. Crop Sci. 12:126. Knight, W. E.; Ahlrich, V. E; and Byrd, Morris. 1969. Registration of Meechee arrowleaf clover. Crop Sci 9:393. Kretschmer, A. E., Jr.; Brolmann, J. B.; Snyder, G. H.; and Coleman, S. W. 1982. Registration of ’Florida’ carpon desmodium. Crop Sci. 22:158-159. Massey, J. G. 1966. Preliminary evaluations of some introductions of Persian clover (Trifolium resupinatum L.). Ga. Agric. Exp. Stn. Bull. (N.S.) 180, 14 pp. Miller, John D., and Wells, Homer D. 1981. Registration of Tifhardy-1 Desmodium canum germ- plasm. Crop Sci. 21:476. 54 1983a. Registration of Bicolor-1 lupine germplasm. Crop Sci. 23:189. 1983b. Registration of Tift-1 hyacinth bean germplasm. Crop Sci. 23:190-191. Nichols, R. L.; Minton, N. A.; Knight, W. E.; and Moore, W. F. 1981. Meloidoyne incognita on arrowleaf clover. Nemo- tropica ll:No. 2. 191-192. Offutt, M. S. 1971. Registration of Hope white lupine. Crop Sci. 11: 602. Taylor N. L. 1979. 1982. Taylor, N. 1973. Taylor, N. 1982. Taylor, N. 1978a. 1978b. Templeton, 1975. Weihing, R. Registration of red clover introduction bulk germ- plasm. Crop Sci. 19:564. Registration of gene marker germplasm for red clover. Crop Sci. 22:1269. L., and Anderson, M. K. Registration of Kenstar red clover. Crop Sci. 13: 772. L.; Cornelius, P. L.; and Sigafus, R. E. Registration of KY M-l zigzag clover germplasm. Crop Sci. 22:1278-1279. L., and Quesenberry, K. H. Registration of Trifolium medium x T^. sarosiense hybrid germplasm. Crop Sci. 18:1102. Registration of Trifolium sarosiense x 4X _T. alpestre hybrid germplasm. Crop Sci. 18:1102. W. E., Jr., and Taylor, N» L. Performance of bigflower vetch seeded into bermuda- grass and tall fescue swards. Agron. J. 67:709-712. M. 1962. Selecting persian clover for hard seed. Crop Sci. 2:381-382. Wells, Homer, D. ; Forbes, Ian, Jr.; Burns, Robert; and Miller, John D. 1980a. Registration of Tifblue-78 blue lupine. Crop Sci. 20:824. Wells, Homer D.; Forbes, Ian; Burns, Robert; Miller, John D.; and Dobson, Jim. 1980b. Registration of Tifwhite-78 white lupine. Crop Sci. 20:824. Wells, Homer D. , and Miller, John D. 1981. Registration of WH-1 blue lupine germplasm. Crop Sci. 21:992. 55 Forage Plant Resources FORAGE ATTRIBUTES FOR IMPROVED ANIMAL PERFORMANCE H . Lippke Texas Agricultural Experiment Station Most forage species will provide at least moderate levels of nutrition to ruminants during brief periods of the growing sea- son. The objective of breeding and management research is to improve those characteristics that will increase both rate of animal production and the length of time when a high rate of production can be achieved. Of the six classes of nutrients required in balanced supply by animals, carbohydrates and pro- teins are of major economic concern for forage diets. Minerals, vitamins, and water can be supplemented at relatively low cost. Essential lipids, like the B vitamins, appear to be provided by rumen microorganisms in sufficient quantity to all but the youngest ruminants. Attention, therefore, centers on the sup- ply of energy-yielding compounds, primarily carbohydrates, and nitrogenous compounds, primarily protein, and the influence of plant structure on their concentration and availability. The work of Weston and Hogan (1968) indicates that the concen- tration of dietary protein needed by rumen microorganisms for maximum rumen function may be as low as 6%, a value much lower than that needed by the host animal when producing at even mod- erate levels. In the great majority of forages, however, it is energy availability and not protein that first limits animal performance. The exceptions include stored forages having a high grain content . Figure 1 outlines the characteristics of plant structure that bear upon digestible energy intake. Rate of fermentation in the rumen and rate of passage of undigested residues from the rumen regulate intake through their effect on chemical satiety and gut fill (Ellis, 1978; Forbes, 1980; Lippke, 1980). Extent of digestion has its obvious effect on gut fill and, therefore, on intake. The relationships between extent of digestion and rate of fermentation or rate of passage are usually positive. When plant structure is such that normal processing (i.e., grazing and chewing) results in smaller particles, rate of fermentation increases more than rate of passage and extent of 56 INTAKE EXTENT OF DIGESTION 57 digestion increases. Mechanical grinding, on the other hand, commonly increases rate of passage more than rate of fermenta- tion, and extent of digestion declines (Hogan and Weston, 1967). Rate of fermentation is strongly influenced by the proportion of soluble carbohydrates in forages and the composition of structural carbohydrates, also called cell wall constituents (CWC) . Fermentation of neutral detergent solubles is very rapid (Gray et al., 1967) and almost complete (Van Soest, 1967). CWC, represented analytically by neutral detergent fiber (NDF) (Van Soest and Wine, 1967), have a much slower fermentation rate that is influenced primarily by the degree of structural fragility and, hence, particle size (Laredo and Minson, 1973). Paradoxically, rate of fermentation of solubles can sometimes be too fast, as with very immature ryegrass, resulting in a disruption of rumen function, and severely reduced intake (Lippke, 1982). A minimum of about 10% indigestible fiber appears to be needed in the diet for maximum digestible energy intake (Lippke, 1982), presumably to help buffer the rapid release of acids that accompany a high fermentation rate (Van Soest , 1982) . The structural components of forages appear to be fermented at variable rates. Akin and Burdick (1975) reported that both mesophyll and phloem tissues were rapidly and totally degraded by in vitro rumen fermentation within 12 hr while outer bundle- sheath cells and noncutinized portions of epidermis were de- graded more slowly. Sclerenchyma and lignified vascular tissue were not degraded. Akin et al. (1975) also reported that acid detergent removed approximately the same tissues from Coastal bermudagrass leaves as did in vitro fermentation. These obser- vations agree with the report by Lippke (1980) indicating a high positive correlation between hemicellulose : cellulose and NDF intake, where hemicellulose = (NDF - ADF) . A higher pro- portion of rapidly digested fiber, as indicated by a higher hemicellulose : cellulose ratio, should correlate well with in- creased rate of passage from the rumen and increased NDF intaka The polyphenolic polymers, commonly referred to as lignin, within the plant cell wall appear to be responsible for the limitation of intake of most forages, both by their effect on extent of digestion and on fragility. Lignin has commonly been thought to exert its inhibitory action by preventing the physical attachment of rumen bacteria to the cell wall (Dehority and Johnson, 1961). Alternatively, Bacon (1979) has suggested that the structure of indigestible fiber is composed of a highly substituted polysaccharide chain having covalent linkages to lignin and that digestive enzymes cannot proceed past such linkages along the chain. Work is underway now to determine whether or not lignin has an inhibitory action on rumen microorganisms independent of its direct effect on utili- zation of plant fiber (D.E. Akin, personal communication). 58 The "ideal" forage for high animal performance can be summariz- ed as having 34% neutral detergent solubles (half of that pro- tein) , 54% potentially digestible fiber having a fermentation rate of .4 hr-1, and 12% indigestible fiber, which includes 3% lignin. For warm-season grasses, significant progress toward this ideal can be made by bringing NDF down to 60% and then selecting for increased mesophyll and phloem within the struc- tural carbohydrates. Some improvements might also be made in many cool-season species by increasing NDF while reducing lig- nin. Researchers in forage improvement should be encouraged by the substantial gains in animal performance that can be achiev- ed with relatively small changes in structure of most forage species . While not included in this paper, such agronomic factors as growth habit and persistence under grazing are regarded as being at least of equal importance to those topics discussed. REFERENCES Akin, D. E.; Barton, F. E., II: and Burdick, D. 1975. Scanning electron microscopy of Coastal bermuda and Kentucky-31 tall fescue extracted with neutral and acid detergents. J. Agric. Food Chem. 23:924. Akin, D. E., and Burdick, D. 1975. Percentage of tissue types in tropical and temper- ate grass leaf blades and degradation of tissues by rumen microorganisms. Crop Sci. 15:661. Bacon, J. S. D. 1979. Plant cell wall digestibility and chemical struc- ture. Studies in Anim. Nutr. and Allied Sciences. Rowett Research Institute. 35:99. Dehor ity, A., and Johnson, R. 1961. Effect of particle size upon the in vitro cellulose digestibility of forages by rumen bacteria. J. Dairy Sci. 44:2242. Ellis, W. C. 1978. Determinants of grazed forage intake and digesti- bility. J. Dairy Sci. 61:1828. Forbes, J. M. 1980. A model of the short-term control of feeding in the ruminant: Effect of changing animal or feed characteristics. Appetite 1:21. Gray, F. V.; Weller, R. A.; Pilgrim, A. F.: and Jones, G. B. 1967. Rates of production of volatile fatty acids in the rumen. V. Evaluation of fodders in terms of volatile fatty acid produced in the rumen of sheep. Aust. J. Agric. Res. 18:625-634. 59 Hogan, J. 1967. Laredo, M. 1973. Lippke, H. 1980. Lippke, H. 1982. Van Soest, 1967. Van Soest, 1982. Van Soest, 1967. Weston, R. 1968. P., and Weston, R. H. The digestion of chopped and ground roughages by sheep. II. The digestion of nitrogen and some carbohydrate fractions in the stomach and intes- tines. Aust. J. Agric. Res. 18:803-819. A., and Minson, D. J. The voluntary intake, digestibility, and retention time by sheep of leaf and stem fractions of five grasses. Aust. J. Agric. Res. 24:875-888. Forage characteristics related to intake, digesti- bility and gain by ruminants. J. Anim. Sci. 50:952. Indigestible fiber and diet selection by yearling cattle. Tex. Agric. Expt. Sta. PR-3949. P. J. Development of a comprehensive system of feed analysis and its application to forages. J. Anim. Sci. 26:119. Peter J. Nutritional Ecology of the Ruminant . 0 & B Books, Inc., Corvallis, Oregon, p. 38. P. J., and Wine, R. H. Use of detergents in the analysis of fibrous feeds IV. Determination of plant cell-wall constituents J. Assoc. Official Anal. Chem. 50:50. H., and Hogan, J. P. Factors limiting the intake of feed by sheep. IV. The intake and digestion of mature ryegrass. Aust J. Agr. Res. 19:567. 60 Panel Discussion: Data Required Before Releasing Forages. What Kind and How Much? THE NEED FOR ANIMAL TRIALS D. A. Sleper and F. A. Martz, University of Missouri, and A. G. Matches and J. R. Forwood, U.S. Department of Agriculture One of the limitations in many forage breeding programs is the inability to identify, during the course of cultivar develop- ment, plant materials that would lead to improved animal pro- ductivity • Many forage breeding programs simply do not have the resources to evaluate selected materials with animals. Those forage improvement programs that have the resources for animal evaluation will often use animal evaluation as a last step prior to release. It is desirable to have animal perfor- mance data prior to release of the new cultivar. PHASES OF EVALUATION At the University of Missouri, we have made a commitment to evaluate new tall fescue and orchardgrass synthetics with graz- ing animals prior to release. The forage grass breeding pro- cedure is directed into several phases : Phase I. Plant introduction. Plants are obtained from various sources such as plant introduction stations, plant breeders, local collections, crosses, and cultivars. Phase II. Identify desirable traits and genetic variability. Objectives of what trait (s) is (are) desirable in the new cul- tivar is (are) established. The amount and kind of genetic variability is determined to make an intelligent decision re- garding the choice of the breeding procedure to incorporate the traits . Phase III. Plant development. The breeding procedure selected is used to improve the trait (s) chosen in Phase II. The new synthetic (s) is (are) developed. Phase IV. Small plot evaluation. During this phase we can determine what the influence of environmental variables such as fertility, locations, weather, insects, diseases, management. 61 compatibility with legumes, etc., have on the new synthetics. If more than one synthetic is evaluated, the best ones are chosen for animal evaluation trials. Phase V. Seed increase. It is necessary to increase the amount of seed of the new synthetic so that grazing trials can be established. Phase VI. Animal evaluation. Animal performance is used to evaluate the new synthetic (s) . At the University of Missouri, animal performance is usually evaluated with grazing trials and in some instances feeding trials with hay. Phase VII. Release and increase of seed. Do all potential forage cultivars need to have animal data prior to release? The answer to this question is not immediately apparent. For example, one might conclude in developing a cul- tivar for only higher seed yields that animal evaluation may not be necessary. This is a dangerous assumption since the breeder may not be aware of changes in plant composition such as antiquality factors, mineral concentrations, etc., that may have an influence on animal performance. Also, if resis- tance for pathogens and insects is bred into a cultivar, the factors responsible for conditioning this resistance may also influence animal performance. For example, phenolic compounds have been reported to be synthesized by the host in response to fungal infection (Swain et al., 1979) and have been reported to be antibacterial agents. It is conceivable that these antibac- terial agents could influence animal performance. Certainly, if the breeding objective for a cultivar is the improvement of a particular quality parameter, animal performance data are essential . ANIMAL EVALUATION TRIALS WITH TALL FESCUE Under grazing, differences in animal gain among forage treat- ments are a function of the amount of herbage available, its nutritive value, and the amount consumed by animals (Matches et al. , 1983) . To evaluate the forage quality of potential cultivars, herbage allowance per animal must be uniform among all treatments. A grazing experiment was initiated in 1974 to compare a new University of Missouri tall fescue synthetic (experimental 1-96 and later released as 'Missouri-96') with four other tall fescue strains. The primary objectives of the experiment were to determine if differences in animal performance could be de- tected among tall fescue synthetics using relatively small pas- tures, and to identify factors associated with differences in animal performance. Five tall fescue synthetics, namely ' Kentucky- 31 ' , 'Kenmont', 'Fawn', 'Kenhy', and Missouri-96, were seeded into 0.47-ha pas- 62 tures at the University of Missouri's Southwest Research Center. The experimental design was a randomized complete block with three replications. Hereford heifers which averaged 225 and 204 kg live weight at the start of grazing in 1974 and 1975, respectively, were used as testers. Grazing periods were approximately of 40 days' dur- ation each for separate evaluations during the spring, summer, and fall. Pastures were grazed by three tester animals in the spring and fall and by two testers during the summer. Individ- ual testers grazed the same cultivar all season, and between periods of evaluation, they grazed the same cultivar in a re- serve pasture. All pastures were strip grazed and stocked at the same grazing pressure. Therefore, the amount of herbage allowance was equaL The same herbage allowance was maintained by adjusting with electrical fences the area grazed dependent upon the amount of herbage available. The animals were allowed to graze each strip 7-10 days. Cattle were alloted (disregarding sward growth) a daily amount of herbage dry matter equivalent to 2.5% of their body weight. Weekly pasture samplings recorded the amount of herbage available before grazing and the amount of residue herbage remaining after grazing. No grain was fed in these pastures, but mineral and salt were available. Significant differences were found (P < 0.05) in heifer average daily gains (ADG) among cultivars in the spring and fall for both years (Table 1) . However, no significant differences were detected among cultivars for the summer period. The ADG's on Kenhy and MO-96 were over 40% greater than ADG on Ky-31 which is the cultivar most commonly grown in the southern corn belt. Herbage yield was significantly different (P < 0.05) among cul- tivars on a per ha basis in the spring and fall, but not during the summer. Table 1. — Mean average daily gain (ADG) of heifers and estimated herbage intake on tall fescue cultivars in 1975-76 „ Spring Summer Fall Cultivar - — ADG Intake ADG Intake ADG Intake kg/day Kenhy 0.74 3.95 0.34 5.88 0.65 6.06 Kenmont 0.50 4.02 0.27 5.54 0.44 5.95 MO- 9 6 0.71 4.04 0.24 6.00 0.62 5.75 Fawn 0.48 4.11 0.30 5.68 0.45 5.90 Ky-31 0.51 3.96 0.18 5.54 0.45 5.76 Significance "k : k ns ns ns * ns Std. error 0.04 0.14 0.05 0.13 0.06 0.17 *, ** Significant at the 5 and 1% level of probability, respectively. 63 Intake was estimated by the difference between the herbage available at the start of grazing less the amount of residue remaining after grazing. In no case were there differences (P <0.05) for intake among cultivars, and correlations between ADG and intake were low (r = 0.11 to 0.57) . Weekly sward meas- urements confirmed that there were no differences among culti- vars for rate of growth. We do not know for certain why Kenhy and MO-96 have given su- perior ADG's. Quality analysis has shown in most cases that in vitro dry matter digestibility (IVDMD) is about the same for all cultivars tested. Occasionally, Kenhy will have higher IVDMD than Ky-31 or MO-96; however, it usually is not statisti- cally significant. Neutral detergent fiber, acid detergent fiber, hemicellulose, lignin, cellulose, ash, and silica deter- minations have given little insight as well in explaining the ADG differences. Intake studies using hay showed slight in- creases for MO-96 as compared to Kenhy, Fawn, Kenmont, and Ky- 31 (Martz et al. , 1975). ANIMAL EVALUATION TRIALS WITH ORCHARDGRASS Two orchardgrass synthetics selected for general resistance to the stem rust pathogen Puccinia graminis Pers. f . sp. dactyli- dis Guyot de Massinot were evaluated in replicated grazing trials similar to what was described for tall fescue earlier. The grazing trial was conducted in the fall of 1982 and lasted 42 days. Average daily gains ranged from 0.68 to 1.02 kg/day (Table 2). MO-I and MO-II had the highest ADG's and gains/ha. The coefficient of variation for both ADG and gain/ha was 10.4%. Histological studies conducted by Edwards et al. (1981) may ex- plain the differences in ADG of the orchardgrass synthetics. Figure 1 contains the cross section of an orchardgrass leaf blade infected by ]?. graminis . The cross section was stained Table 2. — Mean average daily gains (ADG) initial weights, final weights and gain/ha of steers on orchardgrass synthetics evaluated in the fall of 1982 Synthetics74 Measurement MO-I MO-II Hallmark Potomac Sterling Initial weight (kg) Final weight 209 a 209 a 204 a 209 a 208 a (kg) 250 a 252 a 240 ab 244 ab 237 b ADG (kg/day) 0.98 ab 1.02 a 0.85 abc 0.83 be 0.68 c Gain/ha (kg) 310 ab 325 a 268 abc 268 abc 212 c Numbers followed by the same letter are not significantly different at the 0.05 level as evaluated by the Duncan's Multiple Range Test. 64 Fig. 1. — Cross section through a pustule of P_. graminis f. sp. in an undigested orchardgrass leaf blade. in saf ranin-fast green. Urediospores showed a positive stain- ing for lignin and easily identified the underlying plant tissues in which mycelium occurred. The epidermal layer had ruptured during sporulation and was separated intact from the underlying mesophyll cells. When the infected tissue was placed in rumen fluid and later examined histologically, there was a lack in the amount of tissue digested as compared to the non-inf ected control (Fig. 2) . The lack of digestion was not limited to tissues adjacent to uredia, but could be observed in tissues some distance away. This gives good evidence that this infected tissue is not digested and one may speculate that this would give lower animal performance. It appears that the magnitude of ADG can be attributed to the level of infection by P_. graminis . MO-I and MO-II had the least amount of infection while Sterling had the most. In summary, animal trials are necessary before the release of forage grass cultivars. The kind and amount of animal data depends upon the species and upon the objectives for releasing a cultivar. Our objective was to release a cultivar with im- proved forage quality. Therefore, conducting grazing trials where the grazing pressure was constant was essential. We also feel that more than one location is desirable and that grazing experiments should be conducted for 2 or 3 years. 65 REFERENCES Edwards, M. T. , Sleper, D. A., and Loegering, W. Q. 1981. Histology of healthy and diseased orchardgrass leaves subjected to digestion in rumen fluids. Crop Sci. 21:341-343. Martz, F. A., Bell, S. , Mitchell, M. , Matches, A. G. , and Sleper, D. A. 1975. Advanced fescue evaluation studies--a team research project. p. 12-17. University of Missouri South- west Research Center Report 177, University of Missouri, Columbia, MO. Matches, A. G. , Martz, F. A., Sleper, D. A., and Belyea, R. L. 1983. Grazing techniques for evaluating quality of forage cultivars in small pastures, p. 514-516. Proc. XIV Inter. Grassld. Congr. , Lexington, KY. Swain, T. , Haeborne, J. B. , and Van Sumere, C. F. 1979. Biochemistry of plant phenolics. In F. A. Lowens and V. C. Runeckles (ed. ) . Recent advances in phytochemistry. Vol. 12. Plenum Press, New York. Fig. 2. Cross section through a pustule of IP. graminis f. sp. dactylis in an orchardgrass leaf blade after 48 hours of digestion. Note lack of digestion in area associated with the infection. 66 Panel Discussion; Data Required Before Releasing Forages. What Kind and How Much? USDA’S PRACTICE AT TIFTON, GA. Glenn W. Burton and Warren G. Monson U.S. Department of Agriculture At Tifton we have usually required a new cultivar to be equal to those in use in most traits and superior in one or more important traits such as yield or quality. This requires that selectionswith potential must be compared in various tests with checks of known performance. Their breeding behavior must be known. Such selections must be tested in the same form as the cultivar will be when it reaches the farm. For example, a 4-clone synthetic should not be tested in the Syn-1 generation unless like Gahi 1 pearl millet it can reach the farm in the Syn-1. Information relative to a known check that we like to have about a selection before it is released as a cultivar may be outlined as follows: 1. Area of adaptation 2. Dependability 3. Ease of establishment 4. Persistence when utilized for hay or pasture 5. Pest resistance 6 . Management requirements 7. Yield of dry matter and cow matter 8. Quality (palatability and digestibility) Selecting the top forage requires precise effective screens for forage and seed yield, cold and drought tolerance, and disease and insect resistance. Estimates of forage quality that correlate well with animal performance are needed and have been supplied in our work with Monson’ s IVDMD test. We believe some grazing experience is desirable for every new cultivar that differs appreciably from a known check. Grazing experiments can supply this experience and may give differences that are statistically significant if they are large enough. Release of new cultivars should not be held up until complete 67 adaptation information is available but enough information relative to a check of known performance should be obtained to suggest where a new cultivar can be successfully grown. Because potential new cultivars are frequently little better than those in current use and because the breeder's greatest asset, his credibility, is at stake when a new cultivar is released, he must do everything possible to improve the preci- sion of his experimental tests. Soil heterogeneity in a test field can nullify the results of any cultivar evaluation test. To avoid this problem such tests should be placed in uniform fields based on uniform crop performance. We have found that uniform pre- cropping with legumes such as velvet beans or soybeans for grass tests and with grasses such as pearl millet or small grains for legume tests is good prac- tice. Perfect stands of the precrops is a must - otherwise they can do more harm than good. We have found that methyl bromide fumigation is a good practice for spaced plant and small plot tests. It is expensive! Com- mercial application at Tifton is costing $1,000 per acre in 1983. However, it eliminates practically all weeds and greatly reduces soil borne pests. The reduced weeding costs probably cover most of the fumigation cost. More important, however, is the fact that fumigation makes the soil environ- ment more uniform, permits rapid establishment of seeded plots and makes first year results meaningful. The year saved pays big dividends on the investment. Machine fertilization can be more uniform than hand applica- tion if the best machines available are used properly. We have found that lattice square designs can increase the precision of some yield trials a great deal. The 9x9 balanced lattice square test that we have used for testing the yield potential of experimental pearl millet hybrids has on the average doubled the precision of our tests. With the com- puter analysis of the data, such tests involve no more work on the part of the breeder than randomized block experiments. Increased yield of animal product is usually the bottom line in any forage breeding program. Increased pest resistance, if significant, must ultimately result in increased yield of cow matter and probably dry matter. If yield can be increased in the presence of the pests, the higher yielding plants will probably be more resistant if the pests are affecting yield appreciably. The forage breeder must, therefore, continually search for new and better screens for forage yield. Usually spaced plants with no replication must be screened first. This is the screen that must identify the top few plants in a population. Improving the uniformity of every 68 step from the seed to the field will pay dividends. We have our best technician transplant every seedling from flats to 2- inch clay pots. The 2-inch pots set in sand on a greenhouse bench give each seedling a uniform environment in which to develop. When large enough to be transplanted without loss of stand, the plants are removed from the pots and carefully set in our most uniform fields that have been precropped with velvet beans and have been uniformly prepared, fertilized, etc. With all of this and more, it is impossible to select the highest yielding plant in such plantings. Hopefully, a number of the highest yielding plants can be selected for replicated small plot tests. With our perennial grasses, we have established small plot tests with clonal material that we usually start in pots in the greenhouse in the winter. These are then transplanted in sufficient numbers to small plots to give a coverage of a 2-foot wide strip before the first season is over. In some of our bahiagrass research, we are interested in com- paring the performance of different 2-clone hybrids in clipped plots. To produce enough seed for seeded plots would require at least one additional year and much extra work. Twelve seed- lings established in the greenhouse in the winter and set 1-foot apart in the center of a plot will make a sod 2-feet wide before the season is over. Sixty seedlings of a cross can give 5 replications and producing so few hybrid seeds is easy. We have just concluded a 3-year clipping test with 10 entries with seeded plots and potted plant plots side-by-side with 5 replications. In this test, the potted-plant-plots established faster, yielded more the first 2 years, and when correlated with seeded plots gave r values of +.67, +.73, +.83 and +.93 for the 1st, 2nd, 3rd, and 3-year average yields. We believe such potted-plant-plot tests can be satisfactory for preliminary screening. Replicated small plot tests from direct seeding will be preferred at a later date when more seed is available. We like the spaced-plant population progress (SPPP) test for assessing progress made in population improvement such as our RRPS breeding program with Pensacola bahiagrass. This test as we have developed it consists of single-spaced-plant plots replicated 100 times. The test is established from 100 potted seedlings chosen at random from each cycle of RRPS. The spaced seedlings are set far enough apart (.9 x .9 m) to per- mit them to express their yield potential the first year with- out competition. Yields are taken twice. Two years of using this test indicate that it maximizes the precision per plant tested and establishes significant mean differences not evi- dent in other tests. It also gives the relative estimate of the genetic variance left in each population, information of value to the breeder and minimizes the land and resources required for the test. 69 At Tifton promising spaced plants of b.ahiagrass and bermuda- grass are tested in replicated small plots for 3 years. Plot size for bahiagrass is 3 x 16 feet and for bermudagrass is 9 x 16 feet. Tests are replicated 5 or more times. Usually N at 200 lb/A/yr. + P and K are applied annually. Green yields are taken from 2 x 14 foot plots with a sickle bar mower usually 5 times per season. Forage from each plot is weighed green and is sampled for dry matter percentages. These samples are later ground and analyzed for IVDMD. The following outline lists steps that we consider very impor- tant in measuring forage quality. A failure at any step can produce misleading results worth less than none at all. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. At Tifton, Georgia, the best entries in clipped plots have usually been grazed in replicated 2-acre pastures with 150 lbs/N/A/yr. plus P and K for 3 years before release to certified growers. Region of adaptation information has been obtained by agrono- mists in other locations. Outstanding cultivars at Tifton have usually performed well at similar or warmer latitudes. Uniform plant growing environment Uniform management Uniform age Representative sam Uniform drying (70 Active rumen fluid Precise laboratory procedures Careful laboratory technicians Replication Known performance checks gling C) and grinding 70 Panel Discussion: Data Required Before Releasing Forages. What Kind and How Much? FORAGE QUALITY ASSESSMENT: IMPORTANT FACTORS FOR PLANT BREEDERS TO CONSIDER S. W. Coleman U.S. Department of Agriculture INTRODUCTION The importance of forage quality assessment in plant breeding and evaluation programs has been readily recognized in the past decade, although such thoughts may have been somewhat heretical in the early 1960's (Johnson 1969). Progress has been made in what is now commonly called "mul tidiscipl ined" research where agronomists, animal scientists, pest managers and economists form teams to attack problems. Hodgson (1981) suggested four questions to be asked concern- ing breeding and evaluation programs. They are: (a) the reliability of relatively simple assessment procedures as predictors of the potential value of plant material for animal production, (b) the relevance of the measurements to farming practice, (c) the most effective way of incorporating the alterna- tive procedures into a selection program, and (d) the resources to be committed to the various stages . Selection criteria usually include total production, nutri- tive value of the forage produced, the ability to establish easily, to withstand climatic stresses, to adapt to different management systems, to resist or tolerate pests and diseases, and to readily produce material for propagation (Burton 1970). A successful breeding program will incorporate many of the attributes listed above. The complexity clearly il- lustrates the impossibility of selection on "increased pro- duction" alone. Certain desirable characteristics are known to be limiting in certain species, such as digestibility in most warm-season perennial grasses. Breeding for improvement of such can improve animal performance. However, concentra- tion on a particular trait can create its own problems such as lack of persistence under grazing among more digestible 71 varieties . The emphasis of this paper involves animal consumption and utilization of nutrients contained in forages. Though defi- ciency of any one of the many nutrient factors required by animals may limit growth, lactation or reproduction to some degree, energy is the nutrient most frequently found lacking in forages. Even when deficient, most other nutrients can usually be easily and economically supplemented. FORAGE OUALITY The ultimate assessment of forages is animal production, ei- ther production per animal or production per unit of land. The nutritive value of a feed is the product of intake, di- gestibility and utilization (Raymond 1968). The common for- ages fed to ruminants have been evaluated much more exten- sively for energy content, digestibility and even utilization than for intake (Waldo 1969). Heaney (1969) suggested that combining digestibility and intake into a single index pro- vides a means of evaluating the feeding value of forages more effectively than the evaluation of either alone. This index, if effective, should be highly correlated with average daily gain. Data are limited concerning this relationship though a few trials give reason for optimism (Crampton 1957, Lippke 1980). When using digestible energy intake (DEI) as an index for forage quality, the relative contribution of intake and di- gestibility are not the same. Milford and Minson (1965) found that digestible dry matter intake (DDMI) of tropical grasses was more correlated with intake of dry matter than its digestibility. Crampton et al . (1960) reported that var- iations in intake accounted for 70 percent of the variability in the Nutritive Value Index. Crampton (1957), Osbourn et_ al . (1970) and Ventura et aj_. (1975) agreed that intake is Tfie more important factor in determining quality, but intake of a given forage is more variable between animals than is digestibility (Blaxter et al . 1961, Minson et__a]_. 1964, Heany _et_ _al_. 1968). This variation among animals may be due to (1) animal weight (Heaney 1969); (2) fatness (Bines et aJL 1969); (3) physiological rumen volume (Purser and Moir T966 ) ; and/or (4) retention time of organic matter in the rumen (Campling et al_. 1961, Hungate 1966). The importance of voluntary intake however does not imply that digestibility is not important in determining DEI or other expressions of quality. Blaxter _et aj_. (1961) calculated that under ad 1 i bi turn feeding conditions, a change in digestibility of DM from 50 to 55 percent resulted in 100 percent increase in weight gain. Much work has been done attempting to relate chemical compo- 72 sit ion of forages to forage quality. Low protein content has been considered the limiting factor in controlling feed in- take (Milford and Minson 1965). However, this generally oc- curs only when crude protein (CP) content of the forage falls below 6-7% of the DM (Minson and Milford 1967). Above this "critical level", rumen fill is considered the primary deter- minant of intake in ruminants (Campling et al . 1961, Conrad 1966) especially with lower quality forages (Fig. 1). When digestibility reaches 65-70%, then control of intake by rumen capacity yields to chemostatic or thermostatic controls (Montgomery and Baumgardt 1965, Blaxter — «” 03 c c: c c c • X z: 2: s z: i — CD 03 to X X X X X CD X c e c: c c X O to 03 03 03 03 03 CD c CD QJ O o ' O O O X to i- on oo XJ OO X oo X oo x oo O oo • r— cd cd CD r-^ CD 0^ CD 0^ CD X- 4— c 03 V- 03 V- 03 U- 03 V- 03 X 03 i- aj 03 i— 1 03 r— 1 03 i— 1 03 X 03 X O i — I o Od —I X - — X ' — 1 ' — z — CD X 03 E OO to LO x— H OO oo LO ' X uo uo to to CXJ CXJ CXI to LU CXI UO o CXI X XI CXI O o CXJ oo 03 03 Qd CXI O o o LO X X CXI oo oo oo r— 1 oo r^. l 1 S- CD X i- S- •i — CD CD 4- X X •1 — •i — X 4— 4- C CD 4-> X 03 C c i- X O) CD CD ai CD cn X i_ s_ S- CD 03 CD aj X s_ X x X CD CD , — 03 c XI X X 03 c c X •I— o S_ • !— a X X X X C c z: i- •1 — •r— 3 03 03 o 03 a; c_> o CD • r— > CD > 0 C 0 c= CD CD CD co i- r— JO -0 0 to 0 CM •1— CD a) CD CO CD CO CD CO Cl Q. TO 00 s_ r^- 4- O CD C- c C- c Q_ CL V- CD L- CD CD L- •>- 03 -i- 03 -r- •1— ^ r-H O 1 — 1 oc 1— CD _ 1 2: — 1 s: _l 3 ■— 4— O CD i_ +-> O 03 CD C_ E CO cm 00 CD CO ■i— . . . I . . UJ +-> I — 1 OO CD LO 1 CD r"» CO • LlJ TO OO CO CO CD CD CD CM CM CM co CD CD O CD ce: * * * * 03 00 CD CD o 1 1 . . 1 1 oo '3- 1 1 CD CD 1 1 UO r^ c LD r^ r-^ . 03 • • • • . 1 1 3 a) CD CO 00 1 t 1 — 1 1 1 \ ^f- LO LO CD cd JO -JJ CO CD 03 CD CL Q. Cl CL CL 1 — CL +-> •1— CD CD CD CD CD -t-> CD c O CD CD CD CD CD +-> CD 1 — 1 CD JC -C JC -C JC 03 JC Q. 00 OO 00 00 00 CJ OO OO Tf- 0 0 CM CM i — 1 CD r-H co co 1 — 1 rH CM 1"- s- c CD 0 JO •r— i_ • 1 — +-> CD 4- 03 JO 0 •r— +-> •I— 4- C +J >> CD (/) CD +-> CD fO 5- C L- TO TO E CD CD CD CD CD C +-> CD -l-> V- V- r— CD •1 — L_ CD 03 03 «T3 CO CD TO i- c_ • 1— CD c +-> 4- 4- a c CD CD 1 — C c ■a •1— • 1— TO TO 03 •1 — •f— 0 4- TO L- sz •1— c TO 4-> S- 4-> +-) • 1— r— •r— =3 03 03 CD C- C- 03 O CD CD CD A APHIS PLAN r P R OTS C r I N F S3 LDS LIBRARY 8 3 UARA1TIK R)0 M 5 63 A HY AITS V IL L r M3 Z(j’ . L