September 1997 SCSB#386 Preservation and Utilization of Germplasm in Cotton 1981-1992 By Personnel Participating in Regional Research Project S-77 Southern Cooperative Series Bulletin #386 ISBN 1-58161-386-5 September 1997 Contact: Jack C. McCarty, Jr., Ph.D. USDA - ARS, Mississippi State, MS Phone: (662)323-2230 E-mail: jcm@ra.msstate.edu   Agencies and Personnel Participating in S-77 State Personnel Agency Alabama W.C. Johnson R.L. Shepherd   A.J. Kappelman, Jr. Agricultural Experiment Station (AES) U.S. Dept. of Agriculture/ Agricultural Research Service (USDA/ARS) USDA/ARS Arizona J.E. Endrizzi F.R.H. Katterman E.L. Turcotte C.V. Feaster R.G. Percy AES AES USDA/ARS USDA/ARS USDA/ARS Arkansas B.A. Waddle N.P. Tugwell Wayne Smith J.M. Stewart AES AES AES AES Louisiana J.E. Jones D.S. Calhoun AES AES Mississippi R.R. Bridge F.M. Bourland V.G. Meyer J.N. Jenkins J.C. McCarty W.R. Meredith W.L. Parrott AES AES AES USDA/ARS USDA/ARS USDA/ARS USDA/ARS New Mexico D.D. Davis N.M. Malm C. Hawkins J.R. Barrow AES AES AES USDA/ARS North Carolina L.L. Phillips D.T. Bowman J.A. Lee AES AES USDA/ARS Oklahoma L.M. Verhalen AES South Carolina T.W. Culp C.C. Green O.L. May USDA/ARS USDA/ARS USDA/ARS Texas D.W. Stelly G.A. Niles K.M. El-Zik D.W. Altman P.A. Fryxell A.E. Percival R.J. Kohel N.L. Trolinder J.E. Quisenberry USDA/ARS USDA/ARS USDA/ARS USDA/ARS USDA/ARS USDA/ARS USDA/ARS USDA/ARS USDA/ARS Project Coordinator: R.J. Kohel D.W. Stelly USDA/ARS Texas AES Administrative Adviser: Ken Tipton Louisiana AES CSRS Representative: Preston Jones   Table of Contents   Introduction Taxonomy Cotton Germplasm Collection Table 1. The Species of Gossypium L. Source of the Collection Plant Explorations Maintenance Distribution Evaluations Race Stock Conversion Germplasm Releases Qualitative Genetics Quantitative Genetics Combining Ability and Gene Action Heterosis Selection and Breeding Fiber Quality Table 2. List of Old and New Symbols and Phenotypes for Genes Regulating Trichomes Annotated Bibliography   Introduction Cotton is important to the economy of the southern and southwestern states, and the problems with which it is faced are well known. Regional Project S-77 and its predecessor S-1 were instruments through which Federal and State workers maintained Gossypium germplasm and coordinated research on the genetics, cytogenetics, and taxonomy of cotton. The research was designed to provide basic knowledge about the cotton plant and about specific traits of importance in cotton production. A better understanding of the diversity existing in Gossypium, of the heritable systems of the plant, and the systematics of the genus were researched. Progress under Regional Project S-77 and S-1 was summarized in Southern Cooperative Series Bulletin 47 which covered the period 1948-1955, Southern Cooperative Series Bulletin 139 which covered the period 1956-1967, and Southern Cooperative Series Bulletin 256 which covered the period 1968-1980. Each of these bulletins includes a comprehensive, annotated bibliography of the papers published by the members of the project for those respective periods. The present bulletin provides coverage from 1981-1992. Project S-77 was terminated in 1992 because its scope and objectives were considered too broad to meet the requirement for a Regional Project. Southern Regional Information Exchange Group - 61 (SRIEG - 61) was organized in 1993 and replaced many of the functions of S-77   Taxonomy Knowledge of the taxonomy of Gossypium and its closely related genera (the tribe Gossypieae) has advanced in several ways since 1980 in the work done under S-77. Seven new species of Gossypium have been discovered and described and others have been reinstated or recognized anew from among the previously known diversity of the genus. The new species include one species from Mexico (Gossypium schwendimanii) and six species from the Kimberley region of northwestern Australia (G. enthyle, G. exiguum, G. londonderriensis, G. marchantii, G. nobile, and G. rotundifolium), where a center of evolutionary diversity exists, with 11 species currently known from this relatively limited area. Allozyme analysis of a representative subset (153 accessions) of G. barbadense revealed geographic structure to the diversity within the species, a probable center of origin of the species in South America west of the Andes, probable patterns of diffusion, and evidence of introgression from G. hirsutum. Allozyme analysis did not support the recognition of infraspecific taxa such as var. braziliense. Greatest variability was noted in accessions collected in South America west of the Andes. Affinity among accessions of northeastern South America, the Caribbean, and Central America, and decreasing diversity in this series suggests a dispersal pathway from South America to Central America via the Caribbean. High variability in improved cultivars relative to regional variation was due in part to introgression from G. hirsutum. Little evidence of introgression was observed in regions of sympatry with G. hirsutum, leading to the speculation that introgression, where it persists, is largely due to human manipulation. An allozyme study of the Galapagos Islands endemic G. darwinii revealed surprising diversity. Although morphological evidence supports the hypothesis of interbreeding of some native Galapagos cottons with G. barbadense, and although it emphatically does not support the hypothesis of G. hirsutum introgression, nevertheless allozyme analysis has detected introgression of putative G. hirsutum alleles into G. darwinii. Accessions possessing introgressant alleles cluster on islands or areas that have seen the highest levels of human activity. The retention of G. darwinii in specific rank is supported. This new information about the diversity of Gossypium, including new information about relationships among the species derived in part from molecular studies in other laboratories, has been brought together into a new taxonomic monograph of Gossypium. This treatment includes a key for the identification of specimens, complete descriptions of the species, references to illustrations and distribution maps, and a classificatory breakdown into subgenera, sections, and subsections. One new section and one new species (from Africa) are described. The total number of species in the genus is thus enlarged to 50. Work being done with related genera primarily concerns interactions of these plants (especially in the genus Hampea) with the boll weevil. This work is ongoing, and it is premature to draw any conclusions. However, it can be said that some species of Hampea serve as natural hosts to anthonomid weevils in southern Mexico, and some do not. What factors mediate this difference are not yet clear, nor is it yet known if these weevils are forms of Anthonomus grandisor distinct (but undescribed) species of Anthonomus. It is hoped that a fuller knowledge of evolutionary and ecological relations of the boll weevil (or its nearest relatives) to its host plants will lead to a better understanding of the life cycle of the insect and thus to improved means of controlling it in an agricultural context.   Cotton Germplasm The 50+ presently recognized Gossypium species are listed in Table 1. J. O. Beasley established a cytological classification of genomes that is closely related to taxonomic affinities, and geographic distribution. However, in recent years opinions have been expressed for a reassessment of the present classifications within the genus. K. Vollesen has attempted to address this problem for the species native to Africa and the area of the Arabian Peninsula. However, it may be necessary to do this for the entire genus as recent explorations have uncovered new species and additional existing variability and relationships within species groupings. The present distribution of the species is as follows: A genome -two cultivated species from the Far East, Middle East, and Africa. B genome -eight wild species from Africa, and the Cape Verde Islands. C genome -16 wild species from Australia. D genome -13 wild species from Mexico, Peru, and the Galapagos Islands. E genome -four wild species from the Arabian peninsula and Northeast Africa. F genome -one wild species from East Central Africa G genome -one wild species from Australia. AD genome -five (two cultivated and three wild) species from Mexico, South America, the Hawaiian Islands, the Galapagos Islands, and Brazil (the cultivated two having recently attained worldwide distribution through cultivation).   Collection The US National Cotton Germplasm Collection resides in the Crop Germplasm Research Unit, Southern Crops Research Laboratory, Southern Plains Area, Agricultural Research Service (ARS), United States Department of Agriculture (USDA), in cooperation with the Soil and Crop Sciences Department, Texas A&M University, College Station, Texas; and regional coordination of its activities has been under the auspices of the Technical Committee of Regional Research Project S-77. It is maintained as a working collection with permanent storage at the National Seed Storage Laboratory, Fort Collins, Colorado. The collection is part of the National Plant Germplasm System (NPGS) and, as part of this system, all aspects for the preservation and use of the data information and physical germplasm are coordinated through the Cotton Crop Germplasm Committee (CCGC). The CCGC functions as an advisory group to provide expert advice to individuals and organizations such as the National Plant Genetics Resources Board (NPGRB), the National Plant Germplasm Committee (NPGC), ARS, State Agricultural Experiment Stations (SAES), and others, on technical matters related to cotton germplasm, its breeding, and effective utilization. Information on accessions maintained, and the evaluation information of these, is accessible through the Germplasm Resources Information Network (GRIN) computer system which is part of the Data Base Management System (DBMS), a part of the Plant Genetics and Germplasm Institute, Beltsville, Maryland. The collection maintains seed accessions of varieties, primitive race stocks, wild species, of the allotetraploids, and accessions of the cultivated and wild diploid species. TABLE 1. THE SPECIES OF GOSSYPIUM L. Date described Species Genome group Distribution 1763 G. hirsutum L. AD New World cultigen 1753 G. barbadense L. AD New World cultigen 1865 G. tomentosum Seem. AD US (Hawaii) 1907 G. darwinii Watt AD Galapagos Islands 1907 G. mustelinum Watt AD Brazil 1753 G. herbaceum L. A Old World cultigen 1753 G. arboreum L. A Old World cultigen 1860 G. anomalum anomalum Wawr. & Pevr. B Africa 1987 G. anomalum senarense Vollesen B Africa 1862 G. triphyllum (Harv. & Sond.) Hochr. B Africa 1950 G. capitis-viridis Mauer B Cape Verde Islands 1916 G. benadirense Mattei* - Africa 1987 G. bricchettii (Ulbri.) Vollesen* - Africa 1988 G. trifurcatum Vollesen* - Africa 1993 G. vollesenii Fryx.* - Africa 1958 G. longicalyx Hutch. & Lee F Africa 1863 G. sturtianum J. H. Willis C Australia 1875 G. robinsonii F. Muell. C Australia 1964 G. nandewarense (Derera) Fryx. C Australia 1858 G. australe F. Muell. - Australia 1863 G. costulatum Tod.* - Australia 1863 G. populifolium (Benth.) Tod. - Australia 1863 G. cunninghamii Tod.* - Australia 1923 G. pulchellum (C. A. Gard.) Fryx.* - Australia 1974 G. pilosum Fryx.* - Australia 1974 G. nelsonii Fryx.* - Australia 1992 G. enthyle Fryx.* - Australia 1992 G. exiguum Fryx.* - Australia 1992 G. londonderriense Fryx.* - Australia 1992 G. marchantii Fryx.* - Australia 1992 G. nobile Fryx.* - Australia 1992 G. rotundifolium Fryx.* - Australia 1910 G. bickii Prokh. G Australia 1824 G. trilobum (DC.) Skov. D Mexico 1853 G. klotzschianum Anderss. D Galapagos Islands 1854 G. thurberi Tod. D Mexico, US (Arizona) 1863 G. sturtianum J. H. Willis C Australia 1873 G. davidsonii Kell. D Mexico 1899 G. harknessii Brandg. D Mexico 1911 G. aridum (Rose & Standl.) Skov. D Mexico 1913 G. gossypioides (Ulbr.) Standl. D Mexico 1932 G. raimondii Ulbr. D Peru 1933 G. armourianum Kearn. D Mexico 1956 G. lobatum Gentry D Mexico 1972 G. laxum Phillips D Mexico 1978 G. turneri Fryx. - Mexico 1988 G. schwendimanii Fryx. - Mexico 1874 G. stocksii Mast. in Hook E Arabia 1895 G. areysianum (Defl.) Hutch. E Arabia 1904 G. somalense (Gurke) Hutch. E Arabia 1935 G. incanum (Schwartz) Hillc. E Arabia *Not available in cultivation.   Source of the Collection The collection presently maintains 5,500 seed accessions of the Gossypium spp. This material has been accumulated through the years and represents a significant accumulation of scientific capital from 76 countries and political jurisdictions. The material was obtained from planned explorations to various parts of the world, by donations from individual collectors, and by exchanges with other similar international collections, such as the Institut de Recherche du Coton et des Textiles Exotique, France; Central Institute for Cotton Research, India; Instituto Nacional de Investigaciones Agricolas, Mexico; Cotton Research Institute, Pakistan; Institute of Plant Industry, former Soviet Union; Germplasm Resources Research Division, Peoples Republic of China; and others. The collection makes available and preserves the broadest possible genetic base for cotton. It provides source material for basic studies in genetics, cytogenetics, taxonomy, and other disciplines, as well as applied studies in screening for resistance to pests and diseases, environmental stress, and in plant productivity. Seeds from the collection are available to cooperators for research studies of various kinds, within and outside of Regional Research Project S-77. However, activities that focus on maintenance and acquisition continue to be the primary objectives in order to preserve the natural variability of cotton as a resource for continued efforts to modify and improve cotton cultivars. Plant Explorations Cotton collecting expeditions have been taking place since the turn of the century. The early collections were for the most part to the purported center of variability of G. hirsutum, that is southern Mexico and Guatemala. These early expeditions were time consuming and difficult to arrange, because at that time there were only limited travel facilities in the areas explored. As interest in cotton germplasm collecting and preservation has increased, funding for this type of activity has become available, and today these collections are more easily arranged and carried out. Added interest in obtaining and preserving Gossypium germplasm has also increased the scope, not only of the geographic areas explored, but also of the material collected. Several of the plant explorations that have taken place during the previous decade are reviewed to give the reader an idea of the scope of these operations, and to establish a more readily available record of where some of these species may be found. Funds for these expeditions were provided by USDA, ARS and the United Nations, FAO, IBPGR. The information cited was gleaned from the exploration reports of the participating individuals. Simpson and Vreeland to Northern Peru, 1983 Stewart, Craven, and Fryxell to Australia, 1983 Schwendiman, Ano, and Percival to Ecuador, 1983 Fryxell and Burandt to Venezuela, 1984 Percival and Stewart to Southern Mexico, 1984 Stewart, Craven, and Fryxell to Australia, 1985 Schwendiman, Percival, and Belot to the Caribbean, 1985 Percival and Wilson to the Galapagos Islands, 1985 Percival and Stewart to Brazil, 1988 Percival and Stewart to Northwest Mexico, 1990 1. Simpson and Vreeland to Northern Peru, 1983 Beryl Simpson and James Vreeland, the University of Texas, Austin, had planned a botanical collection to Peru in the summer of 1983. Because they would be in the area where G. raimondii is endemic, and because this species was rumored to have become extinct since last being collected, these scientists were contracted to either verify the rumor or collect seeds of the species. Since herbarium collections of the species had been made in 1979 and 1980, it did not seem plausible that the species was extinct. Collecting during the summer of 1983 was difficult because of the excessive and unseasonable rainfall and flooding of 1982-1983 caused by the climatic phenomenon known as "El Nino." The roads in this part of the world were severely damaged, and many roads following rivers into the Andes were impassable. Nevertheless, they were able to visit known localities of G. raimondii, plus most localities illustrated in Boza and Madoo (1941) but from which specimens had never been collected, and several apparently previously unexplored neighboring areas. The first sightings of G. raimondii were from the Pan American Highway where it crosses the Chicama River. These plants, growing on the south bank of the river, were presumably in the same locale as reported by Phillips and Stephens in their technical report of 1966. From this highway westward along the old road to Cartavio, they found several patches of plants. The plant population exhibited variable age distribution. After working the lower Chicama, where they also found several populations of G. barbadense, they traveled from Casca NNW toward Santa Ana. Passing the crest separating the drainage of the Cascas and Santa Ana Rivers, they encountered extensive populations of G. raimondii in a region called Pampa Larga (ca. 950 m elevation). Hundreds of plants were seen growing along the valley and continued along the rocky rubble of the Santa Ana River. Descending to 800 m they also found a large population in the Santa Ana Valley. Plants became sparse as the elevation decreased. Short excursions were made up the Cupinsque River. Because elevations were never reached above 300 m, no evidence of G. raimondii was seen. Traveling to the Huertas River Valley, they found G. raimondii, with populations observed on both sides of the road, starting about 1 km south of Chilete and ending before the town of Huerta. A trip up the Zana River Valley proved fruitless. In the Department of Cajamarca in the Province of Hualgayoc, G. raimondii had been previously collected. Nanchoc, a small village that lies along the Zana River, was reached with the aid of a helicopter provided by the Peruvian Military. No G. raimondii was found, and their report indicates that the habitat is such that it is unlikely to grow in the immediate vicinity. 2. Stewart, Craven, and Fryxell to Australia, 1983 This expedition to western Australia, conducted by James McD. Stewart, Lyn A. Craven, and Paul A. Fryxell, was a collaborative project supported by USDA through ARS and the National Plant Germplasm Unit, and by the Commonwealth Scientific and Industrial Research Organization (CSIRO) through the Australian National Herbarium. The timing of the trip was set to correspond to the usual period when the Gossypium species of the area have matured some capsules but the plants have not desiccated. The primary purpose of this trip was to document, as far as possible, the extent of variation within and among Gossypium species of the region and to obtain seeds representative of that diversity for the US Germplasm Collection. Seed of G. hirsutum, G. australe, G. cunninghamii, G. pilosum, G. populifolium, and G. pulchellum were collected during this expedition. It was apparent to the participants that there is extensive diversity in the wet-dry tropics of Australia. The diversity has only begun to be measured because the remoteness of the area makes the logistics of collecting difficult. A previous trip to the area by Stewart suggested that the diversity of the cotton genus was greater in the Kimberley Region of Australia than was previously realized. Collections made on this trip confirmed that the taxonomic understanding of the Gossypium of the area is not complete. Specimens taken at 10 km intervals along the length of the Mitchell Plateau will be useful in deciphering an apparent cline that occurs there. The Gossypium collections north of the Carson River are distinctive and may represent undescribed species. However, the specimens do have similarities to known taxa and will require detailed study to determine their taxonomic position. At the very least, they represent previously unknown variation that will require accommodation in current species descriptions. The location of G. cunninghamii in the Northern Territory of Australia appears disjunct from the species of the Kimberley to which it is related. Quite likely, additional Gossypium diversity will be discovered in these areas once they are penetrated by botanists, as was the case for the Kimberley where each new area visited yielded something different. 3. Schwendiman, Ano, and Percival to Ecuador, 1983 Jacques Schwendiman and George Ano of the Institut de Recherches du Coton et des Textiles Exotiques (IRCT), France; and A. Edward Percival (USDA, ARS), USA, participated in a collecting expedition to Ecuador, including the Galapagos Islands, which was supported by the UN, FAO, IBPGR. They were joined by Andres Brando of the Instituto Nacional de Investigaciones Agropecuarias (INIAP), Ecuador, for part of the collecting. The first phase was collecting in continental Ecuador, and began with the intent to travel from Quito south to Puyo in Pastaza Province. This travel was not possible as the road leading there was temporarily blocked by landslides, due to the unseasonable weather mentioned above. The participants then headed straight south to Azuay and Loja Provinces, where the first cotton was collected in Loja. From southern Loja, travel was northwest to El Oro, then north to Guayas, Los Rios, and Manabi Provinces. In Loja, El Oro and Manabi, with few exceptions, only dooryard G. barbadense was collected. However, in Guayas, and Los Rios, large populations of endemic wild G. barbadense were found. One of the main problems encountered in all of the areas explored was that much of the cotton was not open. It was apparent that the unseasonable rains had greatly delayed boll maturity. On the Galapagos Islands, G. barbadense was found on San Cristobal. G. darwinii was found on Santa Cruz, Eden off of Santa Cruz, Floreana, Espanola, Gardner off of Hood, San Cristobal and Rabida. G. klotzschianum was found on Santa Cruz and San Cristobal, and as previously reported, it was found growing intermingled with G. darwinii in extensive populations of both species. The accessions collected have added to the germplasm diversity of the collections represented, and the large number of accessions collected from the Galapagos Islands should aid in clarifying questions that have been raised concerning the elevation of G. darwinii to a species level. 4. Fryxell and Burandt to Venezuela, 1984 Paul A. Fryxell (USDA, ARS) and Charles L. Burandt (Texas A&M Univ.) undertook this collection from January to February. A rough itinerary of the route followed was Maracaibo, Coro, Maracay, Barquisimeto, Guanare, Merida, Caracas, and back to Barquisimeto and Maracaibo. Collections of cotton seeds were made in natural vegetation, on roadsides, and in dooryards, from sea level to as high as 1,800 m elevation. Most of the samples collected were of G. hirsutum, but two dooryard G. barbadenses were also found. Considerable variability was found among the collections of G. hirsutum. Many samples were collected opportunistically as they were encountered; others were specifically sought out on the basis of prior information, especially the wild cottons occurring in natural vegetation, sometimes in remote places along the northern coast. The cottons collected were found to be in all stages of development. A few were in full foliage and in early stages of flowering with no mature fruits. Others were still flowering but with both green and open bolls, while still others were past flowering. Some plants were merely dry sticks lacking any foliage but with a mature crop of open bolls. The wild cottons observed in natural vegetation formed large but locally restricted populations. There was often one or a few parent plants of apparent great age in each population. Some of the cottons exhibited characteristics (e.g. short brown fiber, small flowers, and fruit) that set them apart from the dooryard and roadside cottons. In the opinion of the collectors, these wild cottons are an indigenous part of the vegetation and not escapes from cultivation. 5. Percival and Stewart to Southern Mexico, 1984 This cotton collection by A. E. Percival and J. McD. Stewart (USDA, ARS), during the month of September, was a collaborative project with the Secretaria de Agricultura y Recursos Hidraulicos, Instituto Nacional de Investigaciones Agricolas (SARH, INIA), Mexico, represented by Arturo Hernandez and Fernando de Leon. The rough itinerary followed was: Brownsville, Texas, south through the States of Tamaulipas and Veracruz, east to Tabasco, northeast and around the Yucatan Peninsula to Chetumal, Quintana Roo, south through Chiapas, west to the Isthmus of Tehuantepec, and back north to Texas. Seeds were collected of dooryard (one atypical) and wild strains of G. hirsutum, one G. barbadense, and one G. cf. aridum. The only truly wild G. hirsutum cottons collected were G. hirsutum var. yucatanense, from the northern coast of Yucatan. The distribution, growth habit, and morphology clearly indicated that these are wild and well adapted to the ecological niche where they were found. Interestingly, no dooryard cottons could be classified yucatanense. Likewise, the majority of feral cottons were associated with human settlement and were of types similar to the dooryard cottons. As important as the sites where cottons were found, were the observations in areas where cotton was not found. It is the classic story of germplasm loss. The town of Acala, Chiapas and the valley where it is located was specifically visited because it was the site of collections of the original germplasm that gave rise to the outstanding Acala cultivars. They found no cotton there, or at any locations near the road that runs along the length of the valley. One individual in Acala related that promoters from Tapachula, Chiapas tried to establish commercial cotton production in the area. When insects became a problem, the promoters recommended that all native cotton plants be destroyed to better control the insects. The commercial venture subsequently failed, and the collectors found no cotton being grown there today. 6. Stewart, Craven, and Fryxell to Australia, 1985 As with the 1983 expedition to Australia, the participants were James McD. Stewart, Paul A. Fryxell (USDA, ARS) from the USA, and Lyn Craven from Australia (Australian National Herbarium, CSIRO). This exploration was based on funding from IBPGR, USDA, and CSIRO. This plant exploration to central, northern, and northwestern Australia collected samples of most of the 12 currently recognized taxa from Australia and eight additional Gossypium variants. These findings have provided new germplasm for study and exploitation, and also made plain the need for taxonomic re-interpretation of the Kimberley cottons. This exploration significantly extended our knowledge of the geographic range of the Australian wild cottons and their range of variation, as was the case with the 1983 collection. The collectors found that in the arid zone of Central Australia, G. nelsonii occurs sympatrically with three other species of Gossypium, including G. australe, with which some workers confuse it. It was established that the species are indeed distinct in the field and that G. nelsonii occurs over a much wider geographical area than the one site previously reported. The wild Kimberley cottons had previously been allocated to five species: G. costulatum, G. populifolium, G. pilosum, G. pulchellum, and G. cunninghamii. Moreover, they previously were thought to have relatively isolated distributions within the region. It is clear from the results of this collection that this is not an adequate representation of the actual situation, and the descriptions of six new species resulting from this exploration are currently in preparation by the individuals named. In the Kimberley region, Gossypium was found to be far more widespread, abundant, and more variable than previously recognized. An exception is G. cunninghamii, which occurs on the Cobourg Peninsula, outside the Kimberley, and thus is isolated from the others. However, even this species was found to be more widespread and abundant than previously known. The climate where these species are found is tropical with alternating wet-dry seasons, and the plants are long-lived perennials that have adapted to a fire-mediated ecology by regrowing annual stems from woody rootstocks. In the absence of fire for one or more years, the stems occasionally survive the dry season and persist, especially in the erect-growing type of plants. The sample of variability among the many populations sampled appeared to be complex, with the morphological characters recombining in various ways. It seems clear that this group of wild cottons is in an early and active stage of speciation. Thus, this exploration provided materials to begin an analysis that will lead to a more satisfactory interpretation of the variability and the recognition of newly discovered species. 7. Schwendiman, Percival, and Belot to the Caribbean, 1985 This exploration, financed by IBPGR, included the same IRCT and USDA, ARS personnel of the 1983 exploration to Ecuador mentioned previously. It was conducted from the last of February to the first of April and included the following localities listed in the order collected. Trinidad and Tobago; Curacao, Bonaire and Aruba (Netherlands Antilles); Jamaica; Grand Cayman (British West Indies); South Florida (USA); The Dominican Republic; and Puerto Rico. The period for collecting seeds was optimal, as with few exceptions the cotton found was in the late flowering, open mature boll stage. Accessions of dooryard, feral, and wild G. hirsutum and dooryard G. barbadense were collected. The distribution, growth habit, and morphology of the wild cottons indicated that they are truly wild and adapted to the ecological niches where they were found. Wild types were found on Curacao, Bonaire, Jamaica, South Florida, the Dominican Republic, and Puerto Rico. The feral and dooryard types were found on all the islands and in Florida, and they were associated with human settlements or disturbances. With the exception of those populations that appeared to be of a truly wild nature on the islands of Curacao and Bonaire, the cottons found appeared to be plentiful and in no danger of being eliminated. However, the one wild population on Curacao and the one on Bonaire could be lost due to developments in the areas where they are established. 8. Percival and Wilson to the Galapagos Islands, 1985 This exploration by A. E. Percival and F. D. Wilson (USDA, ARS) was conducted to collect the western and northern islands of the Archipelago that had not been collected during the 1983 expedition to the these islands. This collection was a collaborative project with Instituto Nacional de Investigaciones Agropecuarias (INIAP), Ecuador, which was represented on the exploration by Gelasio Basante. The following islands were collected and explored during September: Santa Cruz (Indefatigable) -- Puerto Ayora, road from Las Gemelas to Baltra crossing and Turtle Beach; Marchena (Bindloe) -- Black Beach and Point Mejia; Pinta (Abingdon) -- Cape Chalmers, and north of Cape Chalmers; Isabela (Albemarle) -- Point Vincente Roca, Banks Bay, Black Cove, Tagus Cove, Urvina Bay, Elizabeth Bay, Iguana Cove, San Pedro Cove, and the road Villamil to Santo Thomas; and Fernandina (Narborough) -- two locations between Point Espinosa and Cape Douglas. Gossypium darwinii was collected from Santa Cruz, Marchena, and Isabela, and G. klotzschianum was collected from Santa Cruz and Isabela. The G. klotzschianum collected from Isabela was unique, as it had not been found on this island during previous expeditions. It was found at two locations, and in each case only a few small plants were seen growing and only a few seeds were gathered. It was not possible to determine if there might be other larger populations of the species on this island or whether the few plants growing resulted from being recently introduced. 9. Percival and Stewart to Brazil, 1988 This USDA, ARS funded exploration by A. E. Percival and J. McD. Stewart to northeast Brazil during the month of September was a collaborative project with the Centro Nacional de Recursos Geneticos, Empresa Brasileira de Pesquisa Agropecuaria (CENARGEN, EMBRAPA), Brazil. Antonio Miranda, Jose de Alencar, and Elusio Freire represented EMBRAPA. The area collected involved portions of the states of Bahia, Ceara, Pernambuco, Piaui, and Rio Grande do Norte. With the exception of a small area on the northern coast around Touros, Rio Grande do Norte, all of the area collected is a tropical semiarid region with a wet-dry season. Seeds of G. hirsutum, G. mustelinum, and G. barbadense were collected. The endemic allotetraploid wild species G. mustelinum was collected at four sites from where it had previously been reported (Pickersgill et al., 1975), and from two new sites. Except for variation in the ages of some of the plants at each site, little morphological variation was noted, and all of the sites were next to or near water drainage areas, indicating that the species has adapted to take maximum advantage of the limited rainfall of the area. It may have been indeed fortunate that this collection was made possible at this time. The area collected is an area almost exclusively devoted to the production of "Moco" type cotton, with limited corporate production. Moco cotton (G. hirsutum var. marie-galante) is morphologically variable, and has characteristics suggesting introgression from G. barbadense and G. mustelinum. Moco is grown as a perennial and plants are ratooned each season. Once fields are established, planting involves only replacement of plants that may have died. Moco growers are largely small farmers that grow the crop with limited, or no, technical agricultural input. Many of the fields are established on rocky hillsides, almost exclusively adaptable to a crop such as Moco cotton, which can survive the dry season, and where the plants can become established among the rocks and boulders. There are native insect pests, such as boll worms, that damage the crop, but not to the extent that it was not economical to grow. However, with the invasion to the area in 1985 by the boll weevil (Anthonomous grandis Boheman), production in parts of the area has been so reduced that it is no longer economical to harvest what little crop is produced. Breeding schemes were underway by EMBRAPA personnel to reduce the impact of this insect. Some of these involve developing early and/or resistant Moco type cultivars. Given the environmental conditions and sedentary nature of agricultural practices of cotton production in the area, it remains to be seen whether or not this will succeed. Regardless of the outcome of the breeding efforts to produce boll weevil-resistant and adapted varieties, the germplasm base of the material presently grown will change in the not-too-distant future. Moco cotton will either be eradicated in part or all of the area, or adaptable varieties will be successfully produced with introduced germplasm from other G. hirsutum types. In either case, this will permanently alter the present Moco germplasm base of the area. It is satisfying to note that this collection secured representative cotton germplasm from this area of the world. Some of this material may in future prove valuable as it appears to be variable for many lint quality and agronomic characters. 10. Percival and Stewart to Northwest Mexico, 1990 This collection was a collaborative project between USDA, ARS; The University of Arkansas; and Secretaria de Agricultura y Recursos Hidraulicos (SARH), Instituto Nacional de Investigaciones Forestales y Agropecuarias (INIFAP), Mexico. Lorenzo Perez Solis and Enrique A. Garcia Castaneda represented INIFAP. This collecting expedition to northwest Mexico was from mid-November to mid-December and covered parts of the states of Baja California Norte, Baja California Sur, Sonora, and Sinaloa. With the exception of portions of the state of Sinaloa, the areas explored were semiarid regions. Most of the areas visited have a wet-dry season, with precipitation of about 400 mm per year, and the coastal area of Sinaloa ranging to about 1,000 mm. The distribution of this rainfall can be variable within areas and from year to year. Seeds of Gossypium turneri were collected in Sonora, in the area of San Carlos above the city of Guaymas. G. thurberi was also collected in this state southeast of the town of Magdalena. G. armourianum was collected only on San Marcos Island in the Gulf of California. G. davidsonii and G. harknessii were collected in Baja California Sur. No seeds of G. aridum were collected as the period of mid December was early. Plants of this species were found in abundance, but they were only beginning to bloom.   Maintenance All of the accessions maintained in the collection are increased by self pollination at several locations. The U.S. Department of Agriculture and the National Cotton Council, in cooperation with the Instituto Nacional de Investigaciones Forestales y Agropecuarias (INIFAP), Mexico, maintains a Cotton Winter Nursery at Tecoman, Colima, Mexico. This nursery was moved to its present location from Iguala Guerrero, Mexico, where it had been in operation since 1950. This operation is located at a sufficiently southern latitude to make it satisfactory to grow accessions in the collection that are photoperiodic. This off season facility allows geneticist and breeders to grow an average of three generations every two years more efficiently, and at less coast, than if they did this in a greenhouse during the winter months. The non-photoperiodic varieties and stocks are increased at Weslaco, Texas. This location is maintained as it is proximal to the Mexican border, and from where other users of the Cotton Winter Nursery are able to send materials for seed increase during the off-season. The wild species, most of which are also photoperiodic, are increased in greenhouses at College Station, and Weslaco, Texas. Most of these accessions have unique growing requirements that can only be duplicated in a greenhouse.   Distribution Previous to 1985 seed stocks of cotton accessions maintained were distributed, for the most part, from in house available seed at various locations, or from the stocks kept at the National Seed Storage Laboratory. The permanent funding for the maintenance of The National Gossypium (Cotton) Collection at College Station, Texas, was secured in 1985. Since July of 1987 the collection has distributed seed of 9,691 of the accessions maintained. Request for seed has spanned the spectrum of uses from ornamental to biotechnology, with most coming from geneticist and breeders for crop improvement. Evaluations A primary purpose of maintaining a germplasm collection is to evaluate the accessions. Evaluations for agronomic characteristics are routinely conducted as this material is increased. However, specific evaluations can only be accomplished with the cooperation of the researchers in the various disciplines that screen this material seeking specific traits. Specific traits can and have been transferred. Wild and Diploid Germplasm. Nonfiber-producing cottons include most of the wild diploid species of Gossypium. Seeds of some of these species have hairs, but none bear usable or spinabe fiber. The seed hairs that may be present are too short and too firmly attached to the seed to be of any potential utility. Being diploids, these species are also too distantly related to cultivated allotetraploid cotton to be directly useful in conventional breeding programs. The fiber-producing cultivated Asiatic diploids also fall into this category. Nevertheless, they are potential sources of useful genes that have been, and can be, transferred to cultivated tetraploid cottons using special techniques. Wild and Allotetraploid Germplasm. The fiber-producing cottons include the two cultivated allotetraploid species G. hirsutum and G. barbadense. The other three allotetraploid species that are a potential source of germplasm are the wild G. tomentosum, G. darwinii, and G. mustelinum. The cultivated species have a wide range of variability in terms of cultivars, strains, feral types, and genetic mutants, followed by G. darwinii, which has less variability, limited to its geographic distribution on each of the Galapagos Islands. The other two species, G. tomentosum and G. mustelinum, have little observed variability, probably because few accessions of these have been collected, and because they have only been found in limited geographic locations. The transference of desirable characters between the allotetraploids is more straightforward, but it is also difficult. Hybrids between the allotetraploids break down in the F2 generation. The viable offspring tend to assimilate back to each of the two parent types, and the true recombinants are weak or unable to survive; thus, large populations are required in order to transfer the desired character. Germplasm Utilization. The introduction of desirable germplasm into agronomically acceptable cotton cultivars is an ongoing and dynamic enterprise in most cotton breeding programs. However, the transfer of desirable characters from exotic intraspecific and interspecific sources, though continuous, has primarily been done in state and federal breeding programs. Some examples of this are as follows: Cotton varieties that are hairy impart resistance to insects such as the jassids (Empoasca spp.), which are important pests in Africa and parts of Asia. The presence in the plant of the single major gene T1 is responsible for this desired phenotype. Conversely, the single major gene T1sm controls the smoothleaf character and would be beneficial to have in varieties where dense pubescence is not desirable. Varieties with smooth leaf characteristics help control insects that require plant hairs for egg laying. Okra leaf shape is conditioned by the gene L2o and is desirable in areas that have relatively high humid conditions as harvesting is approached. This leaf type has been found to reduce losses from boll rot organisms, and effects earlier maturity due to a more open plant canopy in varieties that have this leaf characteristic. Varying degrees of pest resistance and/or plant modification have been obtained using other monogenic inherited characters such as red plant color, bract genes, nectariless genes, leaf-shape genes, dwarf genes etc., and polygenic characters which control plant allelochemistry, fiber properties, water-use efficiency, nematode resistance, boll types, etc. A few examples of these are as follows: Bollworm/Tobacco Budworm Resistance. Numerous morphological traits have been determined to confer significant levels of resistance to the bollworm (Helicoverpa zea, Boddie)/tobacco budworm (Heliothis virescens, F.) complex (BW/TBW). Breeding work at the Louisiana Agricultural Experiment Station has been aimed at combining several of these resistance traits with high lint yield and acceptable fiber quality. Experimental strains, LA 850074 and LA 850075, are frego bract, smooth leaf, and their flower buds (including calyx lobes) have a high frequency of gossypol glands (HG). LA 850082 is frego bract and smooth leaf. LA 870210 and LA 870222 are HG. All five strains have suffered significantly less BW/TBW damage than commercial check varieties in the field, and have produced lint yields that are competitive with commercial varieties in numerous tests. All except LA 850082 also have acceptable fiber quality. Boll Weevil Resistance in Converted Races Stocks. Nine race-stock-derived, day-neutral strains were compared with the boll-weevil-susceptible cultivars, Stoneville 213 and Deltapine 41, and the resistant frego-nectariless breeding line, La.81-560FN, for relative field resistance to boll weevil and for anther number and mass per flower. Field resistance was confirmed in four strains (MT 109, MT 330, MT 763, and MT 1180) and identified for the first time in MT 323 and T 1219. The race-stock-derived strains had as many or more anthers per flower, but less than or equally as much anther mass per flower as the checks. All race-stock-derived strains were late and unproductive, but they provide sources of boll weevil resistance that should be used in cultivar development. Race Stock Conversion The utilization of the primitive race stocks has been limited because most require short days to initiate flowering and produce fruit. A program has been in progress for a number of years to incorporate day-neutral genes in the primitive race stocks. The program involves crossing short-day race stocks with a day-neutral donor line (commercial Delta-type cotton) at the Cotton Winter Nursery located at Tecoman, Colima, Mexico. The F1 generation is self-pollinated at the Winter Nursery and the F2 generation is grown at Mississippi State University where segregation for flowering response occurs. Large populations (about 1,000 plants) are grown because the number of factors controlling the short-day flowering habit is not known and varies among the race stocks. Equal numbers of open-pollinated bolls are harvested from each plant that sets fruit, and the seed are bulked for each population. These F3 seeds are increased for release and for research purposes. One plant that sets fruit at a low node and continues to fruit is selected from the F2 population. The F3 progeny from this plant are backcrossed to the race stock. The same procedure is followed for each backcross and is repeated for about four backcross cycles. The day-neutral donor parent is used as the female in all subsequent crosses. This permits the backcrossed material to be in an Upland cytoplasm during the conversion. In the ongoing conversion program, day-neutral selections have been made in more than 1,000 F2 populations. These represent more than 600 different race stock accessions. More than 300 of these accessions have been classified to race. Of the remaining accessions, about one-third have been classified as race latifolium. From 1-40 accessions from the other 6 races are in the conversion program. Germplasm releases involving more than 500 day-neutral lines have been made from this program. Seventy-nine of these have been backcrossed four times to the race stock accessions. Seed of the day-neutral lines have been supplied to commercial breeders and public research scientists in the United States. The converted race stocks should be useful for the diverse germplasm they contain. The day-neutral lines can be exploited by researchers in search of new traits and they can be used to expand the genetic base of cotton.   Germplasm Releases A major aspect of germplasm development and utilization concerns the formal release of breeding and commercial varieties at advanced stages of development. These stocks or varieties have particular characteristics and/or traits that are genetically stable, and of which larger amounts of seed are made available. The following releases have been made by individuals or programs that actively participated in S-77: Culp, T. W. 1981. Registration of Pee Dee 4548 germplasm line of cotton. Crop Sci. 21:992 Culp, T. W. 1984. 'PD-1,' a new cultivar with improved fiber strength, was released as a replacement for 'SC-1,' the first cultivar with extra-fiber strength genes from triple-hybrid origin that produced yield equal to commercial cultivars. The major advantages of PD-1 over SC-1 are higher yield, stronger fiber, and greater resistance to the fusarium wilt-rootknot nematode complex. PD-1 has higher lint percentage, smaller bolls, smaller seed, stronger and coarser fiber, lower elongation, and higher yarn strength. Culp, T. W. 1984. Seven germplasm with high-yield potential and improved fiber quality were released. Preliminary data suggest that a higher level of fiber strength has been reached through intermating and selecting but without a decrease in yield. The seven lines possess fiber properties that are beneficial to the textile industry. Culp, T. W., R. F. Moore, and J. B. Pitner. 1984. Simultaneous improvement of lint yield and fiber strength in cotton. South Carolina Agr. Exp. Stn. Tech. Bull. 1090. Culp, T. W., R. F. Moore, and J. B. Pitner. 1985. Registration of PD-1 cotton. Crop Sci. 25:198. Culp, T. W., R. F. Moore, and J. B. Pitner. 1985. Registration of seven germplasm lines of cotton. Crop Sci. 25:201-202. Culp, T. W. 1986. PD 6208, a germplasm line with high yield potential and fiber strength approaching the level of 'Acala SJC-1,' has been accepted for release as a commercial cultivar ('PD-3') for production in the Southeast. Culp, T. W., R. F. Moore, L. H. Harvey, and J. B. Pitner. 1988. Registration of 'PD-3' cotton. Crop Sci. 28:190 Feaster, Carl V. and E. L. Turcotte. 1983. Notice to growers relative to release of a commercial variety of American Pima cotton, 'Pima S-6.' USDA, and Ariz., New Mex., and Tex. Agric. Exp. Stn. Memo. 3 p. and Registration of Pima S-6 cotton. (Reg. No. 81). Crop Sci. 24:382. 1984 Pima S-6 was released as a replacement for 'Pima S-5' in a major portion of the Pima cotton belt. The advantages of Pima S-6 are earlier maturity and higher yield. Jenkins, J. N., J. C. McCarty, Jr., J. Fallieri, and J. F. Mahill. 1983. Release of 15 doubled haploid germplasm lines of Upland cotton with Gossypium hirsutum L. nuclear genes in G. barbadense L. cytoplasm. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 4 pp. and 1984 Crop Sci. 24:624-625. Jenkins, J. N., W. L. Parrott, J. C. McCarty, Jr., and W. H. White. 1983. Notice of release of MHR-1, a germplasm line of cotton with resistance to Heliothis virescens (F.). USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 5 pp. and 1984 Crop Sci. 24:625-626. Jenkins, J. N., W. L. Parrott, J. C. McCarty, Jr., and R. L. Shepherd. 1987. Notice of release of three noncommercial germplasms of Upland cotton tolerant to tobacco budworm. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. and 1988 Crop Sci. 28:869. Jenkins, J. N., W. L. Parrott, J. C. McCarty, Jr., and R. L. Shepherd. 1987. Notice of release of two noncommercial germplasms of Upland cotton tolerant to tobacco budworm. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. and 1988 Crop Sci. 28:870. Jenkins, J. N., W. L. Parrott, J. C. McCarty, Jr., and R. L. Shepherd. 1987. Notice of release of three noncommercial germplasms of Upland cotton tolerant to tobacco budworm and the tarnished plant bug. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. and 1988 Crop Sci. 28:869-870. Jones, J. E., J. I. Dickson, W. Aguillard, W. D. Caldwell, S. H. Moore, R. L. Hutchinson, and R. L. Rogers. 1991. Registration of 'LA 887' Cotton. Crop Sci. 31:1701. 'LA 887,' tested experimentally as LA 830887, was developed from a cross of LA 434-RKR x DES 11-9. LA 434-RKR is an experimental strain with superior fiber quality and resistance to root-knot nematode (RKN). DES 11-9 is an experimental strain obtained from R.R. Bridge, Delta Branch Experiment Station, Stoneville, MS. A selection (DES 11913) from DES 11-9 was subsequently released as 'DES 119.' LA 887 is characterized by premium fiber quality, resistance to the RKN/fusarium wilt complex, and high yield potential. Jones, J. E., J. P. Beasley, J. I. Dickson, and W. D. Caldwell. 1988. Registration of four cotton germplasm lines with resistance to reniform and root-knot nematodes. Crop Sci. 28:199-200. Lines included La. RN 4-4, La. RN 909, La. RN 910, and La. RN 1032. All were selections from LA 434-RKR. LA 434-RKR originated from a cross of Bayou 7769 x 'Deltapine 16'. Bayou 7769 is resistant to root-knot nematode (RKN) and was developed from a crossof 'Deltapine 15' x 'Clevewilt-6'. The germplasm lines were evaluated for nematode resistance in the greenhouse in RKN and reniform nematode (RN) infested soil, and in the field on natural RN-infested soil at Baton Rouge, LA. Jones, J. E., J. I. Dickson, E. Burris, D. F. Clower, W. D. Caldwell, J. G. Marshall, and S. J. Stringer. 1988. Registration of three insect resistant cotton germplasm lines. Crop Sci. 28:200. Lines included La. HG-063, La. HG-065, and La. HG-660 which combine resistance to bollworm/tobacco budworm (BW/TBW) with early maturity, good yielding ability, acceptable fiber quality, and reduced pubescence. BW/TBW resistance is attributed to a high frequency of normal-size gossypol glands (HG) located over the calyx (including lobes), ovary wall, and other plant parts. The lines were developed from a cross between two HG lines, La. HG 83-1-1546 x La. HG 1838-1497. The two parents were selected from an intercross population involving Louisiana advanced breeding lines, 'Stoneville 213,' and GT5A-10-15-2XG15. The strain, GT5A-10-15-2XG15, obtained from M. J. Lukefahr, was the original source of the HG trait. Mahill, J. F., J. N. Jenkins, and J. C. McCarty, Jr. 1982. Notice of release of a semigametic breeding line of cotton (Gossypium spp.) involving the G. harknessii Brandag. species cytoplasm with male sterility. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 4 pp. and 1983 Crop Sci. 23:403-404. Mahill, J. F., J. N. Jenkins, and J. C. McCarty, Jr. 1982. Notice of release of a semigametic breeding line of cotton (Gossypium spp.) involving cytoplasm of tetraploid species G. hirsutum L., G. tomentosum Seems, and G. barbadense L. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 4 pp. and 1983 Crop Sci. 23:403-404. Mahill, J. F., J. N. Jenkins, and J. C. McCarty, Jr. 1982. Notice of release of four semigametic germplasm breeding lines of cotton (Gossypium spp.) involving cytoplasm of diploid species G. herbaceum L., G. arboreum L., G. anomalum Wawr. and Peyr., and G. longicalyx Hutch. and Lee. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 4 pp. and 1983 Crop Sci. 23:403-404. Mahill, J. F., J. N. Jenkins, J. C. McCarty, Jr., and W. L. Parrott. 1983. Notice of release of four doubled haploid lines of cotton, Gossypium hirsutum L., with resistance to the tobacco budworm, Heliothis virescens (F.). USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 8 pp. and 1984 Crop Sci. 24:625. McCarty, J. C. Jr., J. N. Jenkins, and W. L. Parrott. 1981. Germplasm release of 23 BC2F4 and 33 BC1F4 noncommercial flowering lines of Upland cotton involving Gossypium hirsutum L. race accessions. Miss. Agric. and Forestry Exp. Stn. Res. Report 6(17). 4 pp. McCarty, J. C. Jr., J. N. Jenkins, and W. L. Parrott. 1982. Germplasm release of 6 BfC2F4 and 66 BC1F4 noncommercial flowering germplasm lines of Upland cotton involving Gossypium hirsutum L. race accessions. Miss. Agric. and Forestry Exp. Stn. Res. Report 7(16). 2 pp. McCarty, J. C. Jr., J. N. Jenkins, and W. L. Parrott. 1983. Germplasm release of 54 BC2F4 and 5 BC1F4 noncommercial flowering germplasm lines of Upland cotton involving Gossypium hirsutum L. race accessions. USDA and Miss. Agric. and Forestry Exp. Stn. Memo.5 pp. and 1984 MAFES Res. Report 8(15). 3 pp. McCarty, J. C. Jr., J. N. Jenkins, and W. L. Parrott. 1985. Notice of release of two germplasm lines of cotton with resistance to the boll weevil, Anthonomus Grandis Boh. USDA and USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. and Crop Sci. 26:1088. McCarty, J. C. Jr., J. N. Jenkins, and W. L. Parrott. 1985. Notice of release of 4 BC1F4 , 46 BC2F4 and 21 BC3F 4 noncommercial flowering germplasm lines of Upland cotton involving Gossypium hirsutum L. race accessions. USDA and Miss. Agric. and Forestry Exp. Stn Memo. 4 pp. McCarty, J. C. Jr., J. N. Jenkins, R. L. Shepherd, and W. L. Parrott. 1986. Notice of release of 38 BC2F4 and 29 BC3F4 noncommercial flowering germplasm lines of Upland cotton involving Gossypium hirsutum L. race accessions. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 4 pp. McCarty, J. C. Jr., J. N. Jenkins, R. L. Shepherd, and W. L. Parrott. 1988. Notice of release of 39 BC3F4 and 16 BC4F4 noncommercial flowering germplasm lines of Upland cotton involving Gossypium hirsutum L. race accessions. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 4 pp. McCarty, J. C. Jr., J. N. Jenkins, R. L. Shepherd, and W. L. Parrott. 1990. Notice of release of 22 BC2F4 and 10 BC4F4 noncommercial flowering germplasm lines of Upland cotton involving Gossypium hirsutum L. race accessions. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. McCarty, J. C. Jr., and J. N. Jenkins. 1992. Notice of release of 53 BC4F4 noncommercial flowering germplasm lines of Upland cotton involving Gossypium hirsutum L. race accessions. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 4 pp. Shepherd, R. L., J. N. Jenkins, W. L. Parrott, and J. C. McCarty, Jr. 1985. Notice of release of eight noncommercial nectariless-frego bract germplasm lines of Upland cotton, Gossypium hirsutum L. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. and 1986 Crop Sci. 26:1260. Shepherd, R. L., W. L. Parrott, J. C. McCarty, Jr., and J. N. Jenkins, . 1985. Notice of release of eight noncommercial okra leaf-frego bract germplasm lines of Upland cotton, Gossypium hirsutum L. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. and 1986 Crop Sci. 26:1260. Shepherd, R. L., J. C. McCarty, Jr., J. N. Jenkins, and W. L. Parrott. 1987. Notice of release of 12 root-knot-nematode-resistant, noncommercial, flowering germplasm lines of Upland cotton involving Gossypium hirsutum L. race accessions. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. and 1988 Crop Sci. 28:868-869. Shepherd, R. L., J. C. McCarty, Jr., J. N. Jenkins, and W. L. Parrott. 1989. Notice of release of 12 root-knot-nematode-resistant, nectariless germplasm lines of Upland cotton Gossypium hirsutum L. race accessions. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 3 pp. Shepherd, R. L., W. L. Parrott, J. C. McCarty, Jr., and J. N. Jenkins. 1989. Notice of release of nine root-knot-nematode-resistant germplasm lines of Upland cotton Gossypium hirsutum L. USDA and Miss. Agric. and Forestry Exp. Stn. Memo. 4 pp. Turcotte, E. L., Carl V. Feaster. 1985. Notice to plant breeders and geneticists relative to release of five noncommercial germplasm lines of Pima cotton. Ariz. Agric. Exp. Stn. and USDA Memo. 3 p. and Registration of five American Pima cotton germplasm lines (Reg. No. GP-225 to GP-259). Crop Sci. 26:206. 1986. Five germplasm lines of Gossypium barbadense L. incorporating the genetic traits okra leaf, fertility restoration, frego bract, glandless, and nectariless into Pima backgrounds were released. Turcotte, E. L., Carl V. Feaster, and E. F. Young, Jr. 1989. Notice to plant breeders and geneticists relative to release of six noncommercial germplasm lines of Pima cotton. Ariz. Agric. Exp. Stn. and USDA Memo. 5 p. and Registration of six American Pima cotton germplasm lines (Reg. No. GP-479 to GP-484). Crop Sci. 31:495. 1991. Six germplasm lines of Pima cotton, P45, P51, P53, P62, P66, and E15, representing a range of yield potential, plant height, earliness, tolerance to heat stress, boll and fiber properties, and spinning performance were released. Turcotte, E. L., R. G. Percy, and Carl V. Feaster. 1991. Notice of release of a commercial variety of American Pima cotton, 'Pima S-7.' USDA and Ariz. Agric. Exp. Stn. Memo. 3 p. and Registration of 'Pima S-7' American Pima cotton. (Reg. No. CV-101, PI 560140). Crop Sci. 32:1291 (1991). The advantages of Pima S-7 over Pima S-6 are earlier maturity, greater heat tolerance, higher yield potential at low and intermediate elevations, and earliness at high elevations of Pima cotton belt. Also slightly longer, 6 percent stronger, and slightly finer fiber. In processing, Pima S-7 gives 6 percent stronger yarns than Pima S-6.   Qualitative Genetics G. barbadense Golden veins is an incompletely dominant mutant that is characterized by golden-colored leaf veins and glossy leaf appearance. Homozygous dominant plants are extremely dwarfed and do not flower. The name golden veins and the gene symbol Gv are assigned to the mutant. Linkage tests between Gv and 20 Gossypium mutant genes were negative. Male-sterile-12 is a dominant male-sterile mutant, which is assigned the gene symbol Ms12. Linkage tests between Ms12 and 23 Gossypium marker genes were negative. Because dominant male steriles cannot be intercrossed, distinct loci designations are based on phenotypic differences, and gene designations have to be considered tentative. An incomplete dominant leaf trait was described in which heterozygous plants express several characteristics including small, yellow-green, and waxy leaves that are often rounded, modified narrow bracts, and normal branches arising from low mainstem nodes. Homozygous dominant plants express an extreme phenotype, and they rarely shed pollen, making them functionally lethal. The name Round leaf-3 and the gene symbol R13 is assigned to the mutant. Linkage tests between R13 and 23 other Gossypium mutant genes were negative. The monogenic recessive wrinkled leaf-2 mutant was described. The mutant expresses on successive leaves beginning with the first sympodial branch produced from nodes 6 through 8 on field-grown plants. The wrinkled leaf trait was not allelic with three other leaf mutant genes nor was it linked with 22 Gossypium mutant genes. The name wrinkled leaf-2 and the gene symbol wr2 were assigned to the mutant. The monogenic recessive virescent-21 mutant was described. Genetic studies showed that the virescent trait is not linked with 22 Gossypium mutant genes or allelic with v7 in G. barbadense L. or v1 and v2 in G. hirsutum L. The name virescent-21 and the gene symbol v21were assigned to the mutant. Kidney seed cottons are a distinctive type in which the seeds of each locule are conjoined into a single kidney-shaped mass. Genetic studies of kidney seed showed that it was inherited as a monogenic recessive, and that it was not linked with 21 Gossypium mutant genes. The gene symbol k was assigned to the kidney seed trait. Several kidney cottons in the G. barbadense germplasm collection are characterized by a complex of traits and are known collectively as G. barbadense var. braziliense. The present study expanded the trait complex associated with braziliense. G. hirsutum Ephemeral leaf, conditioned by a recessive gene, ep, is characterized by misshapen leaves from about nodes 6 to 15 on greenhouse plants and fewer nodes on field-grown plants in Arizona. The ep gene is not allelic to veins-fused or strap. Undulate leaf, also conditioned by a recessive gene, ul, is characterized by undulate leaf margins and a light-green color, caused by a reduced amount of chlorophyll. Pink filament, conditioned by duplicate, partially dominant factors, Pf1 and Pf2 shows a variable amount of anthocyanin coloring of the filaments and is hypostatis to petal spot, R2. Resistance to cotton leaf crumple virus is also conditioned by duplicate, partially dominant factors, C1 and C2. The complementary lethal factor of G. davidsonii was transferred to tetraploid cotton via a cross compatible G. barbadense stock ('15-4'). The lethality factor from G. davidsoni was designated Ledav. The cross compatible G. barbadense was le1le 1le2le 2. Cultivated tetraploids were identified as carrying two factors, Le1 and Le2, or were monomeric for the factors. Ledav in combination with Le1 and/or Le2 results in lethality. Linkage analyses established linkage of Ledav and Gl3dav with 26 percent RC. Further analyses established linkage of Gl2 with Le1 (28 percent RC) and Gl3 with Le2 (23 percent RC). Levels of tomentum on vegetative parts of cotton plants had been described by a series of genes that increased tomentum (H) and a series of genes that reduced or eliminated tomentum (Sm). Detailed genetic analysis established redundancy in allelic and loci designations, and the designations were revised to identify the five loci and their alleles that regulate the relative amounts of trichomes with the symbol (T). These relations are summarized in Table 1. A mutant called crinkle-yellow was assigned the gene symbol cy and was conditioned by homozygous recessive alleles at a single locus. The mutant plant is reduced in size with the most prominent feature being a crinkled condition of the leaves and a yellow flush of leaf laminal areas on seedlings and fall regrowth. Environmental stresses reduce or eliminate the mutant expression, so expression was not consistent. No linkage or chromosome associations were found. A spontaneous mutant in which the leaves tend to have a slight downward roll to the edges, prominent veins, and a yellowish color of the leaves is more noticeably associated with the veins. The degree of expression varied with seasonal variation. The mutant was named yellow- veins and assigned the symbol yv. Expression is conditioned by the homozygous recessive alleles. Linkage tests found associations with virescent-1 and Rugate to for linkage group XVII (Ru-37-yv-30-v1). In the effort to increase genetic mutant stocks, a concerted effort was made to increase virescent mutants. A series of spontaneous virescent mutants were obtained from various workers and locations. Several were allelic with the 11 known loci, and 5 new virescent mutants were identified. Four were simple recessives, v12, v13, v14, and v15, and one was a duplicate recessive, v16-v17. Virescent-14 tests suggested that it was a member of the homoelogous linkage groups II and VII, but not allelic with v5-v6. Additional virescent mutants in G. hirsutum have been identified by workers not associated with S-77; v18, v19 , v20, and v 21 have been identified in G. barbadense. The dominant glandless described in Egyptian cottons was analyzed in G. barbadense and G. hirsutum backgrounds. The mutant was determined to be a new allele at the Gl2 locus, Gl2e. Seedling classification determined it to be clearly expressed as incompletely dominant to normal. In combination with the duplicate recessive glandless gl2, gl2, gl3, gl 3, Gl2e enhances glandlessness of the heterozygous combinations. Numerous fiber color stocks exist in cotton germplasm. Variation exists for intensities of brown and green lint and fuzz variants. Fourteen lines, 12 brown and 2 green were tested with the known brown lint loci, Lc1, Lc2, and Dw, and green lint locus, Lg. The green fuzz variants were alleles of Lg, Lgf. The brown lint lines were found to be either allelic at known loci, or new loci. A dark brown line was conditioned by two new linked loci, Lc3 and Lc5. Two light brown lines were identified as Lc4 and Lc6. The lint color lines are incomplete dominants, but the intensity of the coloration determines whether or not the heterozygote or homozygote can be readily distinguished from normal background variation. A mutant was observed with multiple occurrences in cotton fields. Seedlings were observed in which the seedling became white and died. It was found that transplanting the seedlings to low-light conditions would allow some to turn green and set seed. This process allowed genetic analysis. It was established that the white seedling lethal condition was controlled by duplicate recessive genes, wht1, wht2. Few mutants are known that influence cotton fiber development. Two new mutants were found in production fields. One mutant was found in California in which the phenotype resembles bolls that have opened prematurely. This mutant was named immature fiber, and it is controlled by homozygous recessive alleles, im im. Since plants normally produce bolls that open prematurely, this mutation is often difficult to identify with certainty. Fibers of this mutant have reduced secondary wall development. The second fiber mutant was found in a Texas cotton field. This mutant produces normal plants but has short fibers that resemble Ligon lintless-1. It was found to be independent of Ligon lintless-1, and, like Ligon lintless-1, it is a complete dominant. It was called Ligon lintless-2 and assigned the gene symbol Li2. Based on the independence with Ligon lintless-1 and similar fiber phenotype, it is considered to be a possible homoeologue. An extremely debilitating mutant was found in a field of 'Rex' cotton. Mutant expression begins prior to flowering and increases in intensity, resulting in a very depauperate and unproductive plant. Leaves are misshapen and have a rounded appearance, plants have reduced hairs, stems often develop a corky epidermis, and bracts are small and misshapen to almost a frego-bract appearance. The mutant is conditioned by homozygous recessive genes, and it was named rex with the gene symbol rx. It was found to be a member of linkage group X with the gene order rx-17-Rg-21-rl1.   Quantitative Genetics Most characteristics of economic significance in cotton are inherited in a quantitative manner, including yield, fiber and seed quality, and resistance to pests and stress. Inheritance of these characteristics involves separation of environmental and genetic effects. The objective of increased knowledge of quantitative traits is an increase in breeding efficiency and generally involves two areas of research: identification by methodology or instrumentation of useful genetic variability and efficient use of that variability. Geneticists and breeders have been very successful in their efforts as evidenced by modern and obsolete cultivar comparisons which show yield has been increased by about 1 percent per year since 1960. Simultaneous improvements in pest resistance, earliness, and fiber quality have accompanied these yield increases.   Combining Ability and Gene Action Combining ability studies give general directions as to which crosses are likely to be superior in producing desirable segregates and the possibility of utilizing heterosis. A half-diallel analysis of tannin content, using 'Stoneville 213,' 'Tamcot CD3H1,' and breeding lines 86-E-3, 86-E-8, and 86E-20, which have elevated levels of tannins or have resistance to spider mites, was conducted in Texas. General combining ability was significant for tannin content of the mature leaf. No significant specific combining ability was detected. Stoneville 213 consistently had high general combining ability for tannin concentration while the other entries had low or inconsistent combining ability estimates. At Mississippi State University research indicated that plant height, first fruiting node number, height of first fruiting node, number of monopodial and sympodial branches, and nodes above white bloom were conditioned by general combining ability. Narrow sense heritabilities ranged from 0.5 to 0.85. Significant combining ability was not detected for any trait measured. Gene action for resistance to Phymatatrichum omnivorium was determined in a seven-parent half diallel mating scheme and analyses (P, F1, and F2 populations) at College Station, Texas. Additive and non-additive effects were significant. The magnitude of the dominance component was much larger than the additive component. Narrow sense and broad sense heritability estimated averaged 11 percent and 35 percent, respectively. Resistance to root rot was conditioned by polygenes with minor effects. Also, research at College Station showed that resistance to Pythium ultimums and Rhizoctonia solani was polygenetically inherited and conditioned by a complex of minor genes. Additive and epistatic effects were smaller in magnitude than the dominance effects. Average degree of dominance ranged from partial to over-dominance with different responses to both pathogens. Narrow sense heritabilities were low - ranging from 0.1 to 21 percent. General combining ability effects were important for the expression of resistance to R. solani and specific combining effects were important for resistance to P. ultimum. Resistance to reniform nematode, Rotylenchos reniformis is inherited in a quantitative and complex manner. Major resistance to root-knot nematode, Meloidogyne incognita, is inherited as a single-/or two-gene model.   Heterosis Progress in utilizing heterosis for yield and fiber quality has been made in two areas. First, the use of interspecific hybrids of G. hirsutum x G. barbadense, which was promoted by Dr. Dick Davis and associates at New Mexico State University, has produced exceptional yields, fiber strength, and fineness. However, interspecific hybrids result in large amounts of aborted ovules, neps, and fiber immaturity, which result in dyeing and fabric quality problems. Some improvements in yield and reduction in fiber quality problems have been achieved by use of short-statured, early-maturing, and coarse-fiber Pima lines. Hybrids produced with these lines exceed their parents for yield at low desert locations, but in high-elevation environments the hybrids produced significantly higher yields than their parents. The second area of practical use of heterosis involves the use of F2 hybrids as cultivars. Several studies indicate that F2 hybrids from upland x upland strains have good potential in producing higher yield-fiber quality cultivars than conventional cultivars. These studies also indicated fiber quality variability from F2 hybrids should not be a problem for the textile industry. Two methods of producing the F1's have been used: use of male gametocides and hand pollination. The major limiting factor for the successful use of F2 hybrids as cultivars is the practicality of working out the logistics of producing hybrids. The use of F2 hybrids has progressed to cultivar use by growers and several commercial companies are testing potential new F2 cultivars.   Selection and Breeding Germplasm enhancement for host-plant resistance, seed characteristics, and fiber quality continues to be an important objective for many breeders and geneticists. Screening for resistance to all major insect and disease pests is being conducted by numerous researchers who have developed improved techniques for detecting resistance. Stomatal conductance of Pima germplasm has been found to be highly correlated with yield (r = 0.95 - 0.99). Additive gene action for stomatal conductance predominated in wide crosses, but in narrower crosses the gene action was complex. Okra leaf has some major desirable characteristics (improved harvest index, earliness, less boll rot, and insect resistance) over normal leaf cottons. Okra leaf's primary limiting factor is its low leaf area index in the early development of the crop. Subokra leaf was found to result in an increase of about 5 percent in canopy photosynthesis and an increase in yield of about 5 percent over normal leaf in some cultivar backgrounds. Use of cluster analysis has proven to be an effective way of characterizing germplasm involving species introgressions into G. hirsutum or with commercial cultivars used in national tests. Results at the Pee Dee Experiment Station, South Carolina, indicate that the performance of F2 populations is a good indicator of a cross' potential to produce high-yielding and high-fiber-quality segregates. Individual plant selections for yield and quality parameters was not an effective method in selecting high-yield and high-fiber-quality progenies.   Fiber Quality Members of S-77 have conducted studies on fiber quality of both upland (G. hirsutum) and Pima cottons (G. barbadense). Investigations have evaluated breeding methods to improve fiber traits, methods of measuring fiber strength, effect of morphological traits on nonlint trash, effect of parental genotype on interspecific hybrid fiber length, and ovule abortion rates in upland, Pima, and interspecific hybrids. Progress from breeding for longer and stronger fiber without a reduction in lint yield continued to be made in the Pee Dee cotton breeding program. Germplasm lines possessing unusual combinations of fiber length, fiber strength, fiber fineness, and lint yield were released. Three cultivars combining superior fiber and spinning properties, along with lint yields equivalent to southeastern commercial cultivars, were also released. A simple breeding procedure consisting of pedigree selection in populations derived from single crosses among Pee Dee lines and between Pee Dee germplasm and commercial cultivars was utilized to develop superior genotypes. However, such a simple breeding procedure was deemed ineffective in improving fiber traits in single-cross populations from Pee Dee x Chinese cultivar crosses. Selections produced adequate seedcotton yields but fibers were generally short, coarse, and weak. The relationship between two fiber strength measurements, stelometer and HVI strength, as well as their association with yarn strength were assessed in three populations at Florence, SC. Neither fiber strength measurement was found to be consistently correlated with yarn strength. Additionally, the genetic correlation between stelometer and HVI fiber strength was low (r=0.10). The genetic correlation between stelometer strength and yarn strength was 0.70, while that between HVI strength and yarn strength was less than 0.01. This discrepancy in genetic correlation between fiber strength measurements and yarn strength was evidence that the two methods of assessing fiber strength are not the same genetic properties. Thus, it is not known whether or not equivalent progress in improving yarn strength could be made by selecting for stelometer or HVI fiber strength alone. Additionally, higher yarn strength may require selection for fiber properties in addition to fiber strength. Morphological trait effects on nonlint trash in cotton fiber were studied in Louisiana. Twelve near-isolines involving four leaf shapes (normal, semi-okra, okra, and super-okra), two bract types (normal and frego), and two leaf-pubescence levels (hairy and semismooth) were evaluated for nonlint trash content before and after ginning followed by zero, one, and two lint cleanings. Semismooth and super-okra leaf traits reduced motes, small-leaf trash before ginning, and resulted in grades similar to the check with one less lint cleaner. Frego-bract isolines had less leaf and bract trash in lint than the normal bract check at any level of lint cleaning. Progress in improving fiber traits of Pima cottons was also made in this time period. The cultivar Pima S-7, released as a replacement for 'Pima S-6,' has longer and stronger fiber with higher yarn strength. Fiber properties were measured on 234 doubled haploids from 48 American Pima germplasm sources. The doubled haploids were uniform within but varied extensively among genotypes for fiber traits. Doubled haploids of American Pima cotton, however, were found with few exceptions to be inferior to standards in at least one fiber property. An investigation of the effect of G. barbadense parental genotype upon interspecific hybrid fiber length and micronaire revealed a significant parental contribution. Simple regressions of hybrid on parental 2.5 percent span length and micronaire resulted in R2 of 0.95 and 0.52, respectively. Heterotic effects in hybrids far exceeded the observed incremental effects due to G. barbadense parental genotype. In this investigation, heterosis opposed the direction of selection practiced in the G. barbadense parents. Assuming that extrapolation beyond the range of data were valid, a G. barbadense parent fiber length of 17.3 mm and micronaire of 6.38 would be required to produce an interspecific hybrid with an Acala-type fiber with a length of 30.5 mm and micronaire of 4.20. Ovule abortion and mote production result in fiber immaturity, low fiber strength, poor dyeing quality, and reduced yield potential. In an investigation of ovule abortion rates in G. hirsutum, G. barbadense, and interspecific hybrids, lowest mote numbers were observed in G. hirsutum, with increased numbers in G. barbadense, and highest numbers in hybrids. These results were in agreement with previous studies. Ovule abortion rates, as a response to environmental stress, were also found to vary between the two species and their hybrids. Increase in percent motes between a high elevation, moderate environment (daily maximum and minimum = 36.8 and 19.8 °C, respectively) and a low elevation, extreme environment (daily maximum and minimum = 40.6 and 23.7 °C, respectively) in Arizona were -1.5 percent in G. hirsutum, 3 percent in G. barbadense, and 8.8 percent in hybrids. The rate of mote production decrease under a moderating environment also varied between species and their hybrid. In a situation in which mean maximum temperature 21 days pre-anthesis moderated from 41.1 to 27.6 °C and in which mean minimum temperature moderated from 25.1 to 23.3 °C, the percentage of motes decreased 20 percent in G. hirsutum, 31 percent in G. barbadense, and 33 percent in hybrids. Previous studies have suggested that a genetic incompatability is responsible for increased mote production in interspecific hybrids. This study reveals that hybrids also possess an increased environmental sensitivity, leading to greater mote production under stress environments.   TABLE 2. LIST OF OLD AND NEW SYMBOLS AND PHENOTYPES FOR GENES REGULATING TRICHOMES. Symbol Phenotype New Old   T1 H2 Densely pubescent leaves, stems, and fruits T1to H2 Densely pubescent leaves and stems, glabrous fruits T1an H2 Densely pubescent leaves and stems, glabrous fruits T1h H1 Densely pubescent leaves and stems, glabrous fruits, (tomentum longer and less dense) T1sm Sm2 Trichomes confined to margins of leaves t1 h1,sm2 "normally pubescent" T2 Sm1 Glabrous stems and pubescent leaves T2arm Sm,Sm1sl Glabrous stems and nearly glabrous leaves T2b Sm1sl Removes trichomes only from stems in G. barbadense T2to   Glabrous stem and semi-glabrous leaves in G. hirsutum T2rai H6 Densely pubescent leaves and stems t2 H3sm1 "normally pubescent" T3 Sm3 Trichomes confined to leaf veins and margin and stems t3h Sm3h "normally pubescent" t3 sm3 phenotype of T3 when homozygous T4 H4 Trichomes on adaxial surface with T1h t4   "normal" T5 H5 Increases length of trichomes t5   "normal" Annotated Bibliography Visit: DAFVM || USDA Search our Site || Need more information about this subject? Last Modified: Friday, 11-Oct-02 11:06:08 URL: http://msucares.com/pubs/southerncoop/scsb386.htm Mississippi State University is an equal opportunity institution. Recommendations on this web site do not endorse any commercial products or trade names.