Index | Search | Home | Table of Contents

Burton, G.W. 1990. Grasses: New and improved. p. 174-177. In: J. Janick and J.E. Simon (eds.), Advances in new crops. Timber Press, Portland, OR.

Grasses: New and Improved

Glenn W. Burton



Of all plants, the grasses are of the greatest use to humans. The cereal grasses supply over three-fourths of our food energy requirements and over half of the protein (Burton 1968). Together with the forage grasses, they furnish the bulk of the feed for livestock, protect the soil from erosion, carpet the lawns, cover the athletic fields, and beautify the environment.

The grass family, Poaceae-Gramineae, is the third largest of the flowering plants (Chase 1948). In number of individuals, it exceeds all others. Grasses occupy more of the earth's surface than any other family of plants and are found growing wherever higher plants can grow.

Although the grasses developed late in the evolutionary chain of higher plants, they cannot be considered "new" to planet earth. Webster's Unabridged Dictionary states that "new" grasses could be "those having been seen or known for a short time although perhaps existing before." This definition can apply to most of the important grasses that serve humans in the United States today. They had not been seen by the American Indian until someone brought them from another continent. The less than 500 years that these grasses have been in the U.S. is a very "short" part of the time they have been on earth.

For convenience, I'll divide the major grasses that we use into three groups: the cereal grasses grown primarily for grain; the forage grasses that provide feed for livestock; and the turf grasses that blanket out lawns, athletic fields, cemeteries, and parks. The "new" grasses will be those species introduced into the U.S. from other continents after 1492 and "improved" grasses will be named cultivars that are better than a standard in one or more characteristics.


The cereal grasses, as a group, are the most important food plants for the human race. They are rice, Oryza sativa L.; wheat, Triticum aestivum L.; maize, Zea mays L.; rye, Secale cereale L.; barley Hordeum vulgare L.; oats, Avena sativa L.; sorghum, Sorghum halepense L. and millets, Pennisetum glaucum L. (the most important species). Of these all but maize were new to the American Indian when they arrived (rice from the orient, sorghum and millet from Africa and the others from Eurasia) less than 500 years ago. Within the last 100 years, extensive plant breeding programs have developed many improved cultivars for all of these species. Cereal grasses will be discussed in another section in the symposium.


'The forage grasses are those species that produce the bulk of the feed for our cattle, horses, sheep, and goats. They may be grazed or harvested and fed as greenchop, hay or silage. In his Manual of Grasses of the United States, Hitchcock (1935) describes 1100 species grouped into 159 genera. Of these, 44 genera and 151 species were introduced, mostly from the Eastern hemisphere, and may be considered "new" according to our definition. Less than 20 of these foreign grasses have become important sources of forage for livestock in the U.S.

Forage grasses are usually grouped according to the season when they grow best. Cool-season grasses are those that produce seed and grow best under moderate temperatures, usually in the spring. They are the grasser, brought to the northeastern U.S. by the farmer immigrants from Europe in the 17th and 18th centuries. Among the species they brought that have become important forages are bluegrass, Poa pratense, L.; orchardgrass, Dactyli glomerata L.; timothy, Phleum pratense L.; redtop, Agrostis alba; tall fescue, Festuca arundinacea Schreb.; meadow fescue, F. elatior L.; reed canarygrass, Phalaris arundinacea L.; smooth bromegrass, Bromus inermis Leyss, and annual ryegrass, Lolium multiflorium Lan.

Most of the cool-season grasses have been improved. Generally germplasms, often landraces bearing the name of the locale where grown for many years, have been collected and tested in small plots for adaptation and yield. The best in these tests were increased, named and distributed as the first product of an improvement program. These cultivars were often improved by several cycles of mass selection of plants with greater pest resistance, better adaptation, etc.

The next stage of improvement usually consists of developing synthetics produced by intermating several superior clones. The clones for a synthetic are plants that have ranked at the top of populations screened for yield, pest resistance, greater cold or drought tolerance, etc. The populations may be unselected landraces, one or more landraces blended together and improved by phenotypic selection or F2s of crosses between clones made to combine their desirable characteristics. Many of the cool-season grass cultivars are advanced generations of synthetics involving very few clones. Generally, the clones used to develop a synthetic have not been retained. An interesting exception is `Polar' smooth bromegrass that has been consistently superior in yield and winterhardiness in Alaska (Hodgson et al. 1971). The 16 clones selected primarily for yield and winterhardiness used to develop 'Polar' have been retained in a replicated polycross nursery where they intercross naturally to produce breeders seed. Two generations of increase are permitted from breeder seed an- d one each of foundation and certified classes.

Warm-season grasses are those species that grow best during the warm summer months. They are less winterhardy than cool-season grasses and are restricted to the southern part of the U.S. However, cold tolerant ecotypes occur in some warm-season species. Many warm season forage grasses fix solar energy by the C4 process whereas most cool season grasses use the less efficient C3 process.

In spite of the limited support for forage grass breeding, by 1985 there were 75 cultivars of 24 western warm-season grasses (Voight and MacLauchlan 1985). Of these, 32 were U.S. ecotypes that had been screened in small plots often in different environments, named and released. Another 27 were native grass cultivars developed by Agricultural Research Service and State Experiment Station grass breeders from superior ecotypes often supplied by Soil Conservation Service Plant Materials Centers. The remaining 16 cultivars were foreigners, most of them were apomictic and only two, `Neuces' and `Llano' buffelgrass Cenchrus ciliaris L. were the product of additional grass breeding (Bashaw 1980). A sexual female plant crossed with an apomictic male released the variability that made it possible to increase yield and winterhardiness in these two cultivars by 25 to 35%.

'Pathfinder' switchgrass, Panicum virgatum L., is an example of cultivar development among the western warm-season grasses. Starting with a collection of adapted ecotypes, Newell polycrossed "f" type clones, tested their progenies, selected the 12 clones producing the best progenies and polycrossed them in isolation (Newell 1968). From the polycrossed progeny of these 12 clones, Newell selected 192 plants that were intercrossed to produce breeder seed of 'Pathfinder' switchgrass.

Warm-season forage grasses used in improved pastures in the South were foreigners and may be considered "new." Of the many new species tested, only 10 are being used and three of them, carpetgrass, Axonopus affinis Chase; johnsongrass, Sorghum halapense L. Pers.; and the obligate apomictic dallisgrass, Paspalum dilatatum Poir. have not been improved.

The tetraploid bahiagrasses, Paspalum notatum Flugge from South America are obligate apomicts and breed true. Most of the many introductions tested have been adapted in the deep South but have lacked the winterhardiness needed to be dependable north of Florida. 'Argentine', an introduction tested, increased, named and released by a SCS Plant Materials Center is widely grown in south Florida.

The sexual tetraploids required to improve the apomictic bahiagrasses were created by doubling the chromosomes in the sexual diploid `Pensacola' bahiagrass (Burton and Forbes 1960). These sexual tetraploids used as females when crossed with apomictic males, produced highly variable progenies that contained both sexual and apomictic plants. Additional breeding efforts produced the apomictic 'Tifton 54' cultivar that is, superior to apomictic introductions in yield and winterhardiness.

'Pensacola' bahiagrass is the most widely grown seed-propagated perennial warm-season grass in the South. Compared with the tetraploid bahiagrasses, the diploid sexual 'Pensacola' cultivar yields more forage and seed, is more winterhardy and easier to establish. `Pensacola' bahiagrass was first improved by finding two self-sterile cross-fertile clones that could be planted vegetatively in alternate strips in a seed field. All ,seed harvested from these fields produced 'Tifhi l' and 'Tifhi 2', selected F1 hybrids, that yielded 25% more forage when clipped and 17% more liveweight gain when grazed.

In 1960, when vegetatively establishing seed fields became too labor-intensive to interest seed producers, population improvement by phenotypic selection was begun. The initial plan involved intermating the top 200 plants in a space-planted population of 1000 plants measured for forage yield for two years. Three years were required per cycle. Modifications in procedures to improve the precision of the screen for yield and the efficiency of the intermating system, now permit one cycle per year with the same 16% increase in space plant yield per cycle (Burton 1982). The modifications that have produced the breeding method called recurrent restricted phenotypic selection have in 9 cycles produced cultivar `Tifton 9', released in 1987. Compared with `Pensacola' bahiagrass, 'Tifton 9' is more vigorous in the seedling stage, is easier to establish and in a 3-year seeded-plot clipping test produced more seed and 47% more forage.

Common bermudagrass, Cynodon dactylon (L.) Pers., an Africa native is the most widely distributed warm season grass in the South. In the 1930's, common bermudagrass was the worst weed in most cultivated fields. It was brought there, no doubt, in the digestive tracts of the animals used to till the soil. The first cultivar developed, named 'Coastal' for the Coastal Plain Experiment Station where it was bred, was the best of 5000 F1 hybrids between a unique common bermudagrass and two excellent introductions from South Africa (Burton 1947). Compared with common bermudagrass in repeated tests, `Coastal' was more disease resistant, more efficient in fertilizer use, grew tall enough to cut for hay, yielded twice as much in good years and 6 times as much during the severe drought of 1954, but had to be propagated vegetatively

Since 'Coastal's' widespread use across the South, 11 other vegetatively propagated cultivars have been developed and used. The winterhardy `Tifton 44', the best of 3500 F1s between 'Coastal' and a winterhardy common bermuda found in Berlin, Germany, and `Tifton 78', (an F1 hybrid between `Tifton 44' and `Callie') that ,produced 15% better average daily gains (ADGs) and 36% more liveweight gain (LWG) than `Coastal' in a 3-year replicated grazing trial are the most recent releases.

Pangolagrass, Digitaria decumbens Stent, introduced from South Africa in 1935 is one of the most widely grown grasses in south Florida. It is completely sterile and all attempts to improve it by hybridization have failed.

Napiergrass, Pennisetum purpureum Schumach, introduced from Africa in 1913, is a robust perennial bunch grass that resembles sugar cane and may reach a height of 6 m. In the tropics and subtropics, it is cut and fed to livestock. Breeding napiergrass in 1936 produced a superior vegetatively propagated F1 hybrid named 'Merkeron'. It was the best of many hybrids between an outstanding tall selection No. 1 and a very leafy dwarf No. 208. Selfed progeny of `Merkeron' released a number of leafy dwarfs, one of which gives outstanding animal performance when grazed. It has been called 'Mott' (Hanna 1987).

Sudangrass and forage sorghums [Sorghum bicolor (L.) Moench] are robust, annual warm-season grasses from Africa. They are used primarily for temporary summer grazing and grow best on the heavier soils having a pH above 6.0. Improved cultivars such as 'Texas sweet', 'Tift' and 'Georgia 337' have been replaced with a number of higher yielding sorghum sudangrass F1 hybrids produced by commercial sorghum breeders.

Pearl millet, Pennisetum glaucum L., is a robust warm-season annual grass from Africa where it is grown primarily for its grain. In the U.S., it is the best summer annual grazing crop on the sandy soils of the Coastal Plain. Pearl millet is free of the prussic acid glucosides which occur in sorghums that may poison livestock. Pearl millet is also more drought tolerant than the sorghums and will grow on more acid soils. `Starr', the first cultivar released, was a synthetic developed by intermating selected F3 progenies of a single cross that carried the increased leafiness and later maturity desired (Burton 1983). In small replicated plots cut to simulate rotation grazing, 'Starr' produced 25% more forage than 'Cattail', the pearl millet grown prior to its release. 'Tiflate', a blend of seed from 54 short day introductions from Nigeria and Upper Volta is a synthetic that will not mature seed in the continental U.S. before frost. It has been a useful cultivar in tropical countries such as Mexico and Brazil.

Hybrid vigor was first used to improve forage yields in pearl millet with the development of cultivar `Gahi 1' (Burton 1983). 'Gahi 1' seed containing 75% of hybrid seed and 25% of selfs and sibs was produced by, harvesting all seed from fields planted to a mixture of four inbred lines that gave high yielding hybrids in all combinations. The more vigorous hybrid seedlings crowd out the selfs and sibs of the 4 inbred parents enabling 'Gahi-1' to yield as well as the double cross involving the 4 parents—50% more than the cattail check.

The discovery of cytoplasmic male sterility in pearl millet in 1954 and the development of `Tift 23A1' and its maintainer `Tift 23B1'permitted the production of singlecrosses yielding at least 50% more forage than cattail millet. The first of these cultivars was `Gahi 3', an F1 between `Tift 23d2A1' (reduced 50% in height with the d2 gene to facilitate seed harvest) and `Tift 186'. Because `Tift 186' lacks the R gene for fertility restoration and carries dominant genes for immunity to leafspot [Pyricularia grisea (Che.) Sacc.],'Gahi 3' is immune to leafspot and sets no seed. Its failure to produce seed keeps it vegetative longer and reduces its weed potential. Dwarfing `Tift 186' with the d2 gene produced `Tift 383'. `Tift 23D2A1' x `Tift 383' produces cultivar `Tifleaf 1' that is only half as tall as 'Gahi 3', yields less dry matter but more leaves and produces better ADGs and LWGs/ha. 'Tifleaf 2', similar to 'Tifleaf 1', but immune to rust (Puccinia substriate var. indica) and leafspot is a cross between `Tift 85D2A1' and `Tift 383'. Its freedom from rust and leafspot results from dominant genes for immunity to these diseases transferred from P. glaucum subspecies monodii (Maire) Brunken with repeated backcrosses (Hanna et al. 1987).


The principal cool-season turfgrasses are the bluegrasses, Poa ssp.; fescues, Festuca spp.; ryegrasses, Lolium spp.; and bentgrasses, Agrostis spp.; all introductions from Europe. Cultivars of all species have been developed. Most are named ecotypes selected in old lawns and athletic fields or synthetics developed from them.

The warm-season turfgrasses are the bermudagrasses, Cynodon spp. from Africa; zoysia grasses, Zoysia spp. and centipedegrass, Eremochloa ophiuroides from the orient, St. Augustinegrass, Stcnotaphrum secundatum Walt Kuntze; and carpetgrass, Axonopus afinis Chase from the West Indies and bahiagrass, Paspalum notatum Flugge from South America. Named cultivars have been selected and named for all of these species except carpetgrass. With the exception of the seed propagated apomictic tetraploid bahiagrasses, the improved turfgrass cultivars are propagated vegetatively. The Tif bermudagrass cultivars are controlled interspecific hybrids or mutants from them. Most of the other cultivars are ecotypes selected from old lawns and athletic fields.


Last update February 20, 1997 by aw