Genetic structure and relationships within and between cultivated and wild sorghum (<Emphasis Type="Italic">Sorghum bicolor</Emphasis> (L.) Moench) in Kenya as revealed by microsatellite markers

Understanding the extent and partitioning of diversity within and among crop landraces and their wild/weedy relatives constitutes the first step in conserving and unlocking their genetic potential. This study aimed to characterize the genetic structure and relationships within and between cultivated and wild sorghum at country scale in Kenya, and to elucidate some of the underlying evolutionary mechanisms. We analyzed at total of 439 individuals comprising 329 cultivated and 110 wild sorghums using 24 microsatellite markers. We observed a total of 295 alleles across all loci and individuals, with 257 different alleles being detected in the cultivated sorghum gene pool and 238 alleles in the wild sorghum gene pool. We found that the wild sorghum gene pool harbored significantly more genetic diversity than its domesticated counterpart, a reflection that domestication of sorghum was accompanied by a genetic bottleneck. Overall, our study found close genetic proximity between cultivated sorghum and its wild progenitor, with the extent of crop-wild divergence varying among cultivation regions. The observed genetic proximity may have arisen primarily due to historical and/or contemporary gene flow between the two congeners, with differences in farmers’ practices explaining inter-regional gene flow differences. This suggests that deployment of transgenic sorghum in Kenya may lead to escape of transgenes into wild-weedy sorghum relatives. In both cultivated and wild sorghum, genetic diversity was found to be structured more along geographical level than agro-climatic level. This indicated that gene flow and genetic drift contributed to shaping the contemporary genetic structure in the two congeners. Spatial autocorrelation analysis revealed a strong spatial genetic structure in both cultivated and wild sorghums at the country scale, which could be explained by medium- to long-distance seed movement.     

SPECIAL PAPER: SYSTEMATICS AND EVOLUTION OF SORGHUM SECT. SORGHUM (GRAMINEAE)

The genus Sorghum Moench is subdivided into sections Chaeotosorghum, Heterosorghum, Parasorghum, Stiposorghum and Sorghum. Section Sorghum includes two rhizomatous species, S. halepense (L.) Pers. (2n = 40) and S. propinquum (Kunth) Hitchcock (2n = 20), as well as the annual S. bicolor (L.) Moench (2n = 20). Sorghum bicolor is divided into subspecies bicolor to include all domesticated grain sorghums, subspecies drummondii (Steud.) de Wet comb. nov. to include stabilized derivatives of hybridization among grain sorghums and their closest wild relatives and subspecies arundinaceum (Desv.) de Wet et Harlan to include the wild progenitors of grain sorghums. Four ecotypes of subspecies arundinaceum are recognized: race aethiopicum of the arid African Sahel. race virgatum of northeastern Africa, race arundinaceum of the African tropical forest, and race verticilliflorum of the African Savanna. The numerous, usually recognized grain sorghums are divided among five basic races, bicolor, caudatum, durra, guinea and kafir, and ten hybrid races that each combine characteristics of at least two of these basic races. Races of grain sorghum are morphologically distinct, and they maintain their unity of type through spacial and ethnological isolation.     

Patterns of allozyme variation in cultivated and wild Sorghum bicolor

Summary Patterns of allozyme variation were surveyed in collections of cultivated and wild sorghum from Africa, the Middle East, and Asia. Data for 30 isozyme loci from a total of 2067 plants representing 429 accessions were analyzed. Regional levels of genetic diversity in the cultivars are greater in northern and central Africa compared to southern Africa, the Middle East, or Asia. The spatial distribution of individual alleles at the most variable loci was studied by plotting allele frequencies on geographic maps covering the distribution of sorghum. Generally, many of the alleles with frequencies below 0.25 are localized in specific portions of the range and are commonly present in more than one race in that region. Several alleles occur in both wild and cultivated sorghum of one region and are absent from sorghum elsewhere, suggesting local introgression between the wild and cultivated forms. Although the same most common allele was found in the wild and cultivated gene pools at 29 of the 30 loci, phenetic analyses separated the majority of wild collections from the cultivars, indicating that the two gene pools are distinct. Wild sorghum from northeast and central Africa exhibits greater genetic similarities to the cultivars compared to wild sorghum of northwest or southern Africa. This is consistent with the theory that wild sorghum of northeast-central Africa is ancestral to domesticated sorghum. Wild sorghums of race arundinaceum of northwest Africa and race virgatum from Egypt are shown to be genetically distinct from both other forms of wild sorghum and from the cultivars. Suggestions for genetic conservation are presented in light of these data.     

Domestication patterns in common bean (Phaseolus vulgaris L.) and the origin of the Mesoamerican and Andean cultivated races

Chloroplast DNA polymorphisms were studied by PCR sequencing and PCR-restriction fragment length polymorphism in 165 accessions of domesticated landraces of common bean from Latin America and the USA, 23 accessions of weedy beans, and 134 accessions of wild beans covering the entire geographic range of wild Phaseolus vulgaris. Fourteen chloroplast haplotypes were identified in wild beans, only five of which occur also in domesticated beans. The chloroplast data agree with those obtained from analyses based on morphology and isozymes and with other DNA polymorphisms in supporting independent domestications of common bean in Mesoamerica and the Andean region and in demonstrating a founder effect associated with domestication in each region. Andean landraces have been classified into three different racial groups, but all share the same chloroplast haplotype. This suggests that common bean was domesticated once only in South America and that the races diverged post-domestication. The haplotype found in Andean domesticated beans is confined to the southern part of the range of wild beans, so Andean beans were probably domesticated somewhere within this area. Mesoamerican landraces have been classified into four racial groups. Our limited samples of Races Jalisco and Guatemala differ from the more widespread and commercially important Races Mesoamerica and Durango in types and/or frequencies of haplotypes. All four Mesoamerican races share their haplotypes with local wild beans in parts of their ranges. Independent domestications of at least some of the races in Mesoamerica and/or conversion of some locally adapted wild beans to cultigens by hybridization with introduced domesticated beans, followed by introgression of the domestication syndrome seem the most plausible explanations of the chloroplast and other molecular data.     

Molecular evolution of the phytochrome gene family in sorghum: changing rates of synonymous and replacement evolution

The photoreceptor phytochromes, encoded by a small gene family, are responsible for controlling the expression of a number of light-responsive genes and photomorphogenic events, including agronomically important phenotypes such as flowering time and shade-avoidance behavior. The understanding and control of flowering time are particularly important goals in sorghum cultivar development for diverse environments, and naturally occurring variation in the phytochrome genes might prove useful in breeding programs. Also of interest is whether variation observed at the phytochrome loci in domesticated sorghum, or in particular races, is a result of human selection. Population genetic studies can reveal evidence of such selection in patterns of polymorphism and divergence. In this study we report a population genetic analysis of the PHY gene family in Sorghum bicolor (L.) Moench in a diverse panel including both cultivated and wild accessions. We show that the level of nucleotide variation in all gene family members is about half the average for this species, consistent with purifying selection acting on these loci. However, the rate of amino acid substitution is accelerated at PHYC compared to the other two loci. In comparisons to a closely related sorghum species, PHYC shows a pattern of intermediate frequency amino acid changes that differ from the patterns observed in comparisons across longer evolutionary distances. There is also a departure from expected patterns of polymorphism and divergence at synonymous sites in PHYC, although the data do not fit a simple model of directional or diversifying selection. Cultivated sorghum has a level of variation similar to that of wild relatives (ssp. verticilliflorum), but many polymorphisms are subspecies-specific, including several amino acid variants.     

Assessment of genetic diversity among sorghum landraces and their wild/weedy relatives in western Kenya using simple sequence repeat (SSR) markers

Understanding the extent of gene exchange between cultivated sorghum and its wild/weedy relatives and the evolutionary processes (including farmers’ practices) that act to shape the structure of genetic diversity within and between them is an important aspect for germplasm conservation strategies, biosafety risk assessment, and crop improvement programs. In this study, molecular characterization and genetic diversity analyses were conducted on wild, weedy and cultivated sorghums collected at a local-scale in a traditional farming system in the Lambwe Valley of western Kenya. Nine simple sequence repeat (SSR) markers were used to genotype 294 cultivated sorghum and 200 wild sorghum individuals. The nine SSR markers were highly polymorphic with a number of alleles that varied from 2 to 19. Overall, wild sorghums had higher genetic diversity, observed heterozygosity, total number of alleles, polymorphic information content and more genotypes per locus than the cultivated types. A Mantel test demonstrated that there was significant isolation-by-distance for wilds and cultivated materials. STRUCTURE, cluster and principal coordinate analyses consistently assigned wild and cultivated individuals to different groups but failed to place hybrids/weedy types as a single separate group from wilds. Our results provide strong evidence of significant genetic diversity retained within wilds, larger divergence between wild and cultivated materials and reduced gene flow than those previously reported in Kenya. These results demonstrate the value of the Lambwe Valley region as a genetic reservoir and the importance to conduct genetic diversity studies at the local scale to design and execute appropriate in situ conservation programs and policies.     

Gene Flow and its Consequences in Sorghum spp.

Gene flow between crops and their weedy or wild relatives can be problematic in modern agricultural systems, especially if it endows novel adaptive genes that confer tolerance to abiotic and biotic stresses. Alternatively, gene flow from weedy relatives to domesticated crops may facilitate ferality through introgression of weedy characteristics in the progeny. Cultivated sorghum (Sorghum bicolor), is particularly vulnerable to the risks associated with gene flow to several weedy relatives, johnsongrass (S. halepense), shattercane (S. bicolor ssp. drummondii) and columbusgrass (S. almum). Johnsongrass and shattercane are common weeds in many sorghum production areas around the world. Sorghum varieties with adaptive traits developed through conventional breeding or novel transgenesis pose agronomic and ecological risks if transferred into weedy/wild relatives. Knowledge of the nature and characteristics of gene flow among different sorghum species is scarce, and existing knowledge is scattered. Here, we review current knowledge of gene flow between cultivated sorghum and its weedy and wild relatives. We further discuss potential avenues for addressing gene flow through genetic, molecular, and field level containment, mitigation and management strategies to facilitate successful deployment of novel traits in this economically important crop species.     

Comparative Analysis of Qtls Affecting Plant Height and Maturity across the Poaceae, in Reference to an Interspecific Sorghum Population

Correspondence among QTLs affecting height and/or flowering was investigated across the five races of sorghum, an interspecific sorghum F(2) population, and 32 previously published sorghum, maize, rice, wheat, and barley populations revealing 185 QTLs or discrete mutants. Among nine QTLs mapped in the interspecific sorghum population (six affecting height and three affecting flowering), at least seven (78%) are associated with ``conversion,'''' backcross-introgression of alleles imparting reduced height or earlier flowering from cultivated sorghums into one or more exotic Sorghum bicolor races. One chromosomal region was ``converted'''' in all S. bicolor races--in the interspecific F(2), this region explined 54.8% of height variation (putatively the Dw2 gene) and 85.7% of flowering time variation (putatively Ma1). Comparative data suggest that Ma1 and Dw2 orthologs influence height and flowering of other Poaceae taxa and support classical dogma that the sorghum phenotypes attributed to Ma1 and Dw2 (respectively) are due to different genetic loci. Other sorghum QTLs also showed correspondence with those in other Poaceae, more frequently than would be expected by chance. Possible homoeologous QTLs were found within both the maize and sorghum genomes. Comparative QTL mapping provides a means to unify, and thereby simplify, molecular analysis of complex phenotypes.     

Genetic diversity, structure, gene flow and evolutionary relationships within the Sorghum bicolor wild�Cweedy�Ccrop complex in a western African region

Gene flow between domesticated plants and their wild relatives is one of the major evolutionary processes acting to shape their structure of genetic diversity. Earlier literature, in the 1970s, reported on the interfertility and the sympatry of wild, weedy and cultivated sorghum belonging to the species Sorghum bicolor in most regions of sub-Saharan Africa. However, only a few recent surveys have addressed the geographical and ecological distribution of sorghum wild relatives and their genetic structure. These features are poorly documented, especially in western Africa, a centre of diversity for this crop. We report here on an exhaustive in situ collection of wild, weedy and cultivated sorghum assembled in Mali and in Guinea. The extent and pattern of genetic diversity were assessed with 15 SSRs within the cultivated pool (455 accessions), the wild pool (91 wild and weedy forms) and between them. F (ST) and R (ST) statistics, distance-based trees, Bayesian clustering methods, as well as isolation by distance models, were used to infer evolutionary relationships within the wild-weedy-crop complex. Firstly, our analyses highlighted a strong racial structure of genetic diversity within cultivated sorghum (F (ST) = 0.40). Secondly, clustering analyses highlighted the introgressed nature of most of the wild and weedy sorghum and grouped them into two eco-geographical groups. Such closeness between wild and crop sorghum could be the result of both sorghum's domestication history and preferential post-domestication crop-to-wild gene flow enhanced by farmers' practices. Finally, isolation by distance analyses showed strong spatial genetic structure within each pool, due to spatially limited dispersal, and suggested consequent gene flow between the wild and the crop pools, also supported by R (ST) analyses. Our findings thus revealed important features for the collection, conservation and biosafety of domesticated and wild sorghum in their centre of diversity.     

Genetic structure and diversity of wild sorghum populations (<Emphasis Type="Italic">Sorghum</Emphasis> spp.) from different eco-geographical regions of Kenya

Wild sorghums are extremely diverse phenotypically, genetically and geographically. However, there is an apparent lack of knowledge on the genetic structure and diversity of wild sorghum populations within and between various eco-geographical regions. This is a major obstacle to both their effective conservation and potential use in breeding programs. The objective of this study was to assess the genetic diversity and structure of wild sorghum populations across a range of eco-geographical conditions in Kenya. Sixty-two wild sorghum populations collected from the 4 main sorghum growing regions in Kenya were genotyped using 18 simple sequence repeat markers. The study showed that wild sorghum is highly variable with the Coast region displaying the highest diversity. Analysis of molecular variance showed a significant variance component within and among wild sorghum populations within regions. The genetic structure of wild sorghum populations indicated that gene flow is not restricted to populations within the same geographic region. A weak regional differentiation was found among populations, reflecting human intervention in shaping wild sorghum genetic structure through seed-mediated gene flow. The sympatric occurrence of wild and cultivated sorghums coupled with extensive seed-mediated gene flow, suggests a potential crop-to-wild gene flow and vice versa across the regions. Wild sorghum displayed a mixed mating system. The wide range of estimated outcrossing rates indicate that some environmental conditions may exist where self-fertilisation is favoured while others cross-pollination is more advantageous.     

Microsatellites uncover extraordinary diversity in native American land races and wild populations of cultivated sunflower

The contemporary oilseed sunflower (Helianthus annuus L.) gene pool is a product of multiple breeding and domestication bottlenecks. Despite substantial phenotypic diversity, modest differences in molecular genetic diversity have been uncovered in anciently and recently domesticated sunflowers. The paucity of molecular marker polymorphisms in early analyses led to the hypothesis of a single domestication origin. Phylogenetic analyses were performed on 47 domesticated and wild germplasm accessions using 122 microsatellite loci distributed throughout the sunflower genome. Extraordinary allelic diversity was found in the Native American land races and wild populations, and progressively less allelic diversity was found in germplasm produced by successive cycles of domestication and breeding. Of 1,341 microsatellite alleles, 489 were unique to land races, exotic domesticates and wild populations, whereas only 15 were unique to elite inbred lines. The number of taxon-specific alleles was 35-fold greater among wild populations (26.27) than elite inbred lines (0.75). Microsatellite genotyping uncovered the possibility of multiple domestication origins. Land races domesticated by Native Americans of the southwestern US (Hopi and Havasupai) formed a clade independent of land races domesticated by Native Americans of the Great Plains and eastern US (Arikara and Seneca). Predictably, domestication and breeding have ratcheted genetic diversity down in sunflower. The contemporary oilseed sunflower gene pool, while not imperiled, could profit from an infusion of novel alleles from the reservoir of latent genetic diversity present in wild populations and Native American land races.     

Structure of genetic diversity in the two major gene pools of common bean (Phaseolus vulgaris L., Fabaceae)

Domesticated materials with well-known wild relatives provide an experimental system to reveal how human selection during cultivation affects genetic composition and adaptation to novel environments. In this paper, our goal was to elucidate how two geographically distinct domestication events modified the structure and level of genetic diversity in common bean. Specifically, we analyzed the genome-wide genetic composition at 26, mostly unlinked microsatellite loci in 349 accessions of wild and domesticated common bean from the Andean and Mesoamerican gene pools. Using a model-based approach, implemented in the software STRUCTURE, we identified nine wild or domesticated populations in common bean, including four of Andean and four of Mesoamerican origins. The ninth population was the putative wild ancestor of the species, which was classified as a Mesoamerican population. A neighbor-joining analysis and a principal coordinate analysis confirmed genetic relationships among accessions and populations observed with the STRUCTURE analysis. Geographic and genetic distances in wild populations were congruent with the exception of a few putative hybrids identified in this study, suggesting a predominant effect of isolation by distance. Domesticated common bean populations possessed lower genetic diversity, higher F _ST, and generally higher linkage disequilibrium (LD) than wild populations in both gene pools; their geographic distributions were less correlated with genetic distance, probably reflecting seed-based gene flow after domestication. The LD was reduced when analyzed in separate Andean and Mesoamerican germplasm samples. The Andean domesticated race Nueva Granada had the highest F _ST value and widest geographic distribution compared to other domesticated races, suggesting a very recent origin or a selection event, presumably associated with a determinate growth habit, which predominates in this race. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Variation in physical and chemical characteristics of common bean (Phaseolus vulgaris L.) grain along a domestication gradient

The objective of this study was to compare six samples of Mexican wild common bean (Phaseolus vulgaris L.) against three landraces and three improved cultivars with respect to physical and chemical attributes, and the culinary quality potential of their grain. A completely randomized experimental design was used to characterize the twelve genotypes. Data were analyzed by analysis of variance and pair-wise comparison of the treatment means by the Tukey test. In addition, correlation and principal-component analysis (PCA) were carried out using twelve characteristics of raw and four of cooked wild and domesticated grains. The results show a larger variability of the physical and chemical characteristics in wild than in domesticated beans. The PCA confirmed that grain gigantism was the main physical characteristic resulting of domestication, whereas the protein and tryptophan contents tended to be higher in wild than domesticated genotypes. Some wild samples from Chihuahua and Durango, Mexico, showed to be a genetic resource to improve food quality, because of their richness in minerals, protein, lysine, tryptophan, and dietary fibers.     

Restriction fragment variation in the nuclear and chloroplast genomes of cultivated and wild Sorghum bicolor

Summary Fifty-six accessions of cultivated and wild sorghum were surveyed for genetic diversity using 50 low-copy-number nuclear DNA sequence probes to detect restriction fragment length polymorphisms (RFLPs). These probes revealed greater genetic diversity in wild sorghum than in cultivated sorghum, including a larger number of alleles per locus and a greater portion of polymorphic loci in wild sorghum. In comparison to previously published isozyme analyses of the same accessions, RFLP analysis reveals a greater number of alleles per locus. Furthermore, many RFLP alleles have frequencies between 0.25–0.75, while the vast majority of isozyme alleles are either rare (< 0.25) or near fixation (> 0.75). Correlations between genetic and geographic distances among the accessions were stronger when calculated with RFLP than with isozyme data. Systematic relationships revealed by nuclear and chloroplast restriction site analysis indicate that cultivated sorghum is derived from the wild ssp. arundinaceum. The portion of the wild gene pool most genetically similar to the cultivars is from central-northeastern Africa. Previous published data also suggested that this is most likely the principal area of domestication of sorghum. Introgression between wild and cultivated sorghum was inferred from disconcordant relationships shown by nuclear and chloroplast DNA markers. Introgression apparently occurs infrequently enough that the crop and its wild relatives maintain distinct genetic constitutions.     

New insight into the history of domesticated apple: secondary contribution of the European wild apple to the genome of cultivated varieties

The apple is the most common and culturally important fruit crop of temperate areas. The elucidation of its origin and domestication history is therefore of great interest. The wild Central Asian species Malus sieversii has previously been identified as the main contributor to the genome of the cultivated apple (Malus domestica), on the basis of morphological, molecular, and historical evidence. The possible contribution of other wild species present along the Silk Route running from Asia to Western Europe remains a matter of debate, particularly with respect to the contribution of the European wild apple. We used microsatellite markers and an unprecedented large sampling of five Malus species throughout Eurasia (839 accessions from China to Spain) to show that multiple species have contributed to the genetic makeup of domesticated apples. The wild European crabapple M. sylvestris, in particular, was a major secondary contributor. Bidirectional gene flow between the domesticated apple and the European crabapple resulted in the current M. domestica being genetically more closely related to this species than to its Central Asian progenitor, M. sieversii. We found no evidence of a domestication bottleneck or clonal population structure in apples, despite the use of vegetative propagation by grafting. We show that the evolution of domesticated apples occurred over a long time period and involved more than one wild species. Our results support the view that self-incompatibility, a long lifespan, and cultural practices such as selection from open-pollinated seeds have facilitated introgression from wild relatives and the maintenance of genetic variation during domestication. This combination of processes may account for the diversification of several long-lived perennial crops, yielding domestication patterns different from those observed for annual species.     

Variation in levels of lipid components and protein in ecogeographic races of Sorghum bicolor

Sorghum bicolor (L.) Moench is a widespread grass of the African savanna. It contains wild and cultivated subspecies which have morphologically distinct ecogeographic races. Measurements of the amount of oil and protein in sorghum grains revealed significant differences in protein levels between wild and cultivated subspecies. Measurements of the relative percentages of fatty acid components of oil also revealed significant differences in levels of oleic and linoleic acid among wild and cultivated races. We conclude that morphological differences acquired in the course of differentiation of ecogeographic races have been paralleled by differentiation in levels of lipid components and grain protein levels.     

Asymmetry of gene flow and differential geographical structure of molecular diversity in wild and domesticated common bean (<Emphasis Type="Italic">Phaseolus vulgaris</Emphasis> L.) from Mesoamerica

Using amplified fragment length polymorphisms (AFLPs), we analyzed the genetic structure of wild and domesticated common bean (Phaseolus vulgaris L.) from Mesoamerica at different geographical levels to test the hypothesis of asymmetric gene flow and investigate the origin of weedy populations. We showed both by phenetic and admixture population analyses that gene flow is about three- to four-fold higher from domesticated to wild populations than in the reverse direction. This result, combined with other work, points to a displacement of genetic diversity in wild populations due to gene flow from the domesticated populations. The weedy populations appear to be genetically intermediate between domesticated and wild populations, suggesting that they originated by hybridization between wild and domesticated types rather than by escape from cultivation. In addition, the domesticated bean races were genetically similar confirming a single domestication event for the Mesoamerican gene pool. Finally, the genetic diversity of the domesticated bean population showed a lower level of geographic structure in comparison to that of the wild populations.     

Host-plant preference and oviposition responses of the sorghum midge, Stenodiplosis sorghicola (Coquillett) (Dipt., Cecidomyiidae) towards wild relatives of sorghum

Sorghum midge, Stenodiplosis ( Contarinia ) sorghicola (Coquillett) is an important pest of grain sorghum world-wide. Considerable progress has been made in screening and breeding for resistance to sorghum midge. However, some of the sources of resistance have become susceptible to sorghum midge in Kenya, in eastern Africa. Therefore, the wild relatives of Sorghum bicolor were studied as a possible source of new genes conferring resistance to sorghum midge. Midge females did not lay eggs in the spikelets of Sorghum amplum , Sorghum bulbosum , and Sorghum angustum compared to 30% spikelets with eggs in Sorghum halepense when infested with five midge females per panicle under no-choice conditions. However, one egg was laid in S. amplum when infested with 50 midges per panicle. A larger number of midges were attracted to the odours from the panicles of S. halepense than to the panicles of Sorghum stipoideum , Sorghum brachypodum , S. angustum , Sorghum macrospermum , Sorghum nitidium , Sorghum laxiflorum , and S. amplum in dual-choice olfactometer tests. The differences in midge response to the odours from S. halepense and Sorghum intrans were not significant. Under multi-choice conditions, when the females were also allowed a contact with the host, more sorghum midge females were attracted to the panicles of S. bicolor compared with S. amplum , S. angustum , and S. halepense . In another test, numerically more midges responded to the panicles of IS 10712 compared with S. halepense , whereas the differences in midge response to the panicles of ICSV 197 ( S. bicolor ) and S. halepense were not apparent, indicating that S. halepense is as attractive to sorghum midge females as S. bicolor . The wild relatives of sorghum (except S. halepense ) were not preferred for oviposition, and they were also less attractive to the sorghum midge females. Thus, wild relatives of sorghum can prove to be an alternative source of genes for resistance to sorghum midge.     

RFLP-based assay of Sorghum bicolor (L.) Moench genetic diversity

Sixty-two single-copy sorghum DNA clones were used to compare restriction fragment patterns of 53 sorghum accessions from Africa, Asia and the United States. Included were accessions from five morphological races of the cultivated subspecies bicolor, and four races of the wild subspecies verticilliflorum. From two to twelve alleles were detected with each probe. There was greater nuclear diversity in the wild subspecies (255 alleles in ten accessions) than in the domestic accessions (236 alleles in 37 accessions). Overall, 204 of the 340 alleles (60%) that were detected occurred in both subspecies. Phylogenetic analysis using parsimony separated the subspecies into separate clusters, with one group of intermediate accessions. Though exceptions were common, especially for the race bicolor, accessions classified as the same morphological race tended to group together on the basis of RFLP similarities. Selection for traits such as forage quality may have led to accessions genetically more similar to other races being classified as bicolors, which have a loose, small-grained panicle similar to wild races. Population statistics, calculated using four nuclear and four cytoplasmic probes that detect two alleles each, revealed a low but significant amount of heterozygosity, and showed little differentiation in alleles in the wild and cultivated subspecies. Outcrossing with foreign pollen appears to have been more important than migration via seed dispersal as a mechanism for gene flow between the wild and domestic accessions included in this study.