|The Improvement of Tropical and Subtropical Rangelands (BOSTID)|
|Criteria for plant selection|
Plant improvement consists for the most part of developing genotypes that are superior when compared with the average of the species or genetic entity. Although the selection of superior types is the most expeditious way to obtain improved plants, by crossing genotypes that have particularly valuable features, a new entity is created that combines the desired features of the original parents into one. Thus, the plant breeder is able to produce plants that are more specialized than those that previously existed. However, this degree of specialization can limit the range of adaptation of the improved genotype, unless its selection and performance testing are done in an environment similar to that of the sites where the improved plant is to be established.
Ideally, a species to be used in range seedings needs to possess vigor, as well as considerable variation, in order to fit into the diverse niches and environments of rangeland sites. Although it is desirable to have uniform and highly productive species, it is also important to have sufficient genetic diversity within a population that some individuals survive and prosper in various rangeland situations. In range plant improvement, it is possible to make selections from the diverse gene pool of existing grasses, fortes, trees, and shrubs in order to identify the superior types available. As newer techniques in biotechnology become available, they can also be applied to the challenge of unlocking the great reservoir of variability and useful features in arid and semiarid plant populations. The breeding of more specialized varieties, as has been done for pasture grasses and legumes, serves as an example of what can be done for tropical and subtropical rangelands.
An example of a successful plant improvement program is the extensive study, selection, and breeding of plants in the genus Agropyron by the U.S. Department of Agriculture's Agricultural Research Service group at Logan, Utah. Considerable progress has been achieved in the development of new varieties of wheatgrasses and crosses with closely related species (Dewey, 1983). Asay and Knowles (1985) documented the release of seven improved cultivars of Agropyron and Elymus for rangeland improvement. Because of their importance for forage and conservation use, many other taxonomic groups of grasses are the subject of efforts in genetic improvement. They would include Bouteloua (Heizer and Hassell, 1985), Panicum, Sorghastrum, and Tripsacum (Vogel et al., 1985), as well as Cenchrus, Eragrostis, Digitaria, Leptochloa, Muhlenbergia, Setaria, and Sporobolus (Voigt and Oake, 1985).
The improved palatability of range plants is often mentioned as a desirable plant improvement goal, but according to a review by Voigt (1975), the factors that influence palatability must first be understood. Plant chemical composition is a major palatability factor, but morphology, succulence, disease resistance, and stage of growth are also involved. The selection of superior plants with these characteristics is necessary to produce plants with improved palatability, and to relate palatability to management strategies.
One of the most critical needs is for grasses with a superior ability to germinate and establish themselves in stressful environments. Wright (1975) has reviewed the work of plant breeders in developing species with special traits useful in rangeland situations. He cited results in the improvement of Panicum antidotale for increased drought tolerance, Bouteloua eriopoda for improved seed set, B. curtipendula for greater seed dormancy to avoid excessively early germination, and P. obtusum for improved seed viability.
Heinrichs (1975) pointed out that in many rangeland areas, the scarcity of adapted legumes appears to be a problem for maintaining the nitrogen level in the soil adequate to support good plant growth. Most of the research work in the United States has been done on legume fortes such as Trifolium, Medicago, and Vicia. According to Rumbaugh and Townsend (1985), more than 25 improved cultivars of legume species adapted to humid and semiarid rangelands are now available, but the greatest challenge is for an improved legume cultivar adapted to range sites receiving less than 250 mm of precipitation annually. There is also a need to develop improved varieties of leguminous shrubs. The International Livestock Centre for Africa,
The International Center for Agricultural Research in the Dry Areas, the Central Arid Zone Research Institute (India), and the Commonwealth Scientific and Industrial Research Organization (Australia) are among the growing number of organizations engaged in research into fodder legumes in the tropics and substropics.
McKell (1975) emphasized that inclusion of adapted fortes and shrubs in range improvement projects was a means of increasing the productivity of harsh sites. By including selected ecotypes of droughtand salinity-tolerant shrub species, such as Atriplex, Ceratonia, Acacia, and Purshia, the stress-tolerant biotypes within the populations would be the ones to establish themselves and survive. Release of "Rincon," a new variety of Atriplex canescens, exemplifies the opportunity to improve range plants by making selections from a breeding population of superior individuals. However, to be successful, such improved varieties must be genetically flexible for long-term survival (Stutz and Carlson, 1985). Interest in improving shrubs for range revegetation was stimulated by an international shrub symposium (McKell et al., 1972) and by the Shrub Research Consortium, a group of universities and natural-resource agencies in the western United States. A symposium at the 1985 meeting of the Society for Range Management featured reports of some of the research groups. Monson and Davis (1985) pointed out that the family Rosaceae has a high potential for superior selections from species favored for their value as browse in range improvement programs and with certain genera, for symbiotic fixation of nitrogen. Other families of shrubs favored in improvement programs are the Compositae (McArthur et al., 1985), for their aggressive habit, persistence, and potential for industrial chemicals, and the Chenopodiaceae, for their protein content and tolerance to salinity (Stutz and Carlson, 1985).
Many leguminous trees and shrubs are important sources of feed for animals and for range improvement in East Africa (Pratt and Gwynne, 1977). Based upon nutritional and ecological observations, Bogdan and Pratt (1967) recommended species for reseeding in Kenya. During the 1940s and 1950s, Bogdan assembled a collection of tree and shrub legumes at the National Grassland Research Station in Kitale, Kenya, for the purpose of selecting those with superior growth, adaptation, and productivity, but much remains to be done with this valuable genetic resource. Further information regarding the role of trees and shrubs in African livestock production systems is contained in the proceedings of the International Symposium on Browse in Africa (le Houu, 1980).
In summary, there is serious need for the development of improved plant materials for rangeland revegetation use. Within the diversity of plant species native to arid and semiarid rangelands, there is ample variability from which to select in developing varieties suitable for the stressful conditions characteristic of rangeland environments. Traditional methods, in concert with newer techniques in biotechnology, hold promise for developing improved plant materials needed to meet the challenges of renewing degraded rangelands.