Incomplete sexual isolation in sympatry between subspecies of the butterfly Danaus chrysippus (L.) and the creation of a hybrid zone

Subspecies chrysippus, dorippus and alcippus of the butterfly Danaus chrysippus differ at three biallelic colour gene loci. They have partially vicariant distributions, but their ranges overlap over a substantial part of central and East Africa, where hybridism is commonplace. We now report that the West African subspecies alcippus differs from other subspecies, not only in nuclear genotype but also in mitochondrial haplotype in both allopatry and sympatry. The maintenance of concordant nuclear and cytoplasmic genetic differences in sympatry, and in the face of hybridisation, is prima facie evidence for sexual isolation. Other evidence that suggests alcippus may be isolated from chrysippus and dorippus include differences in sex ratio (SR), heterozygote deficiency at one site and deduced differences in patterns of migration. We suggest that, within the hybrid zone, differential infection of subspecies by a male-killing Spiroplasma bacterium causes SR differences that restrict female choice, triggering rounds of heterotypic mating and consequent heterozygote excess that is largely confined to females. The absence of these phenomena from hybrid populations that test negative for Spiroplasma supports the hypothesis. The incomplete sexual isolation and partial vicariance of alcippus suggests that it is a nascent species.     

Larval mortality and population regulation in the butterfly Danaus chrysippus in Ghana

An ecological study was made of a population of caterpillars of the African queen butterfly (Danaus chrysippus) feeding on three species of asclepiad plants at Nungua, Ghana, over a nine month period in 1972-73. The principal sources of mortality of caterpillars are probably the parasitic hymenopterans Apanteles chrysippi and Charops sp. There is an inverse correlation between the population of caterpillars and incidence of parasitization. This, and other evidence, suggests that the two parasites are important in limiting the population of Danaus chrysippus below the level imposed by the available food supply. Early instar caterpillars are probably palatable to birds but later instars may be edible or emetic depending on the toxicity of their food plant. It is suggested that the early instar caterpillars are cryptic whilst late instars are cryptic from a distance but conspicuous from nearby, and these may be aposematic or mimetic according to the nature of the food plant. Caterpillars parasitized by Apanteles are paler and hence more cryptic than normal caterpillars. There is also a green, cryptic morph present in the population at low frequency. It is suggested that there is a correlation between colour of caterpillars and the principal source of mortality: i.e., cryptic caterpillars are palatable to birds, suffer heavy predation but a low incidence of parasitization, whilst conspicuous caterpillars are unpalatable to many birds, suffer little predation but have a high incidence of parasitization.     

Population structure in African Drosophila melanogaster revealed by microsatellite analysis

Tropical sub-Saharan regions are considered to be the geographical origin of Drosophila melanogaster. Starting from there, the species colonized the rest of the world after the last glaciation about 10 000 years ago. Consistent with this demographic scenario, African populations have been shown to harbour higher levels of microsatellite and sequence variation than cosmopolitan populations. Nevertheless, limited information is available on the genetic structure of African populations. We used X chromosomal microsatellite variation to study the population structure of D. melanogaster populations using 13 sampling sites in North, West and East Africa. These populations were compared to six European and one North American population. Significant population structure was found among African D. melanogaster populations. Using a Bayesian method for inferring population structure we detected two distinct groups of populations among African D. melanogaster. Interestingly, the comparison to cosmopolitan D. melanogaster populations indicated that one of the divergent African groups is closely related to cosmopolitan flies. Low, but significant levels of differentiation were observed for sub-Saharan D. melanogaster populations from West and East Africa.     

Some adaptations between Danaus plexippus and its food plant, with notes on Danaus chrysippus and Euploea core (Insecta: Lepidoptera)

Behaviour of the egg-laying Monarch in captivity suggests that the concentration and quality of cardiac glycosides in the food plant are not important oviposition cues. The presence of eggs (as previously noted by Urquhart, 1960) and larvae feeding on the food plant, act as mild deterrents.
The butterfly's emetic potency (see Table XIII(a)) can sometimes surpass that of the leaves of the host plant itself. Unidentified factors, providing the internal plant environment, are more important as cardiac glycoside storage stimulants than either the quantity or quality of the cardenolides present. In the laboratory D. plexippus oviposited preferentially on a plant with relatively low cardiac glycoside content, but which produced the most powerfully protected (emetic) adult.
Metabolic changes during the pharate pupal stage, but also, in the case of Euploea core , in the larval fifth instar, rather than larval sequestration, may account for the major increase or decrease in butterfly toxicity compared with that of the food plant.
Temperature does not affect the storage of cardenolides except indirectly by altering metabolic rate. There is no evidence to support the concept that current "physiological cost" of cardenolide storage is high. Like the toad, this butterfly can be assumed to have evolved an enzvmatic system well adjusted to the presence of cardenolides in its bodv tissues.     

Population genetics of Eldana saccharina Walker (Lepidoptera: Pyralidae) and the implications for management using biocontrol

The African sugarcane stalk borer, Eldona saccharina Walker (Lepidoptera: Pyralidae), is widely distributed throughout sub-Saharan Africa and is an important insect pest of maize and sugarcane. The insect shows significant variation in behaviour, host plant and natural enemy guild in different regions. Several attempts to redistribute the natural enemies of E. saccharina from West Africa to South Africa were unsuccessful. The significant behavioural, host plant and natural enemy variations as well as failures of biocontrol attempts evoked a hypothesis of genetic diversification. To evaluate this hypothesis a molecular analysis was conducted on geographically isolated populations of E. saccharina from East, North, South and West Africa, using the cytochrome c oxidase subunit I (COI) region of the mitochondrial genome. The results revealed that E. saccharina populations are separated into four major units corresponding to the West Africa, Rift Valley, South/East Africa and southern African populations. Mitochondrial DNA divergence among the four populations ranged from 1% to 4.98%. To examine the impact of the observed genetic variation on the fertility of inter-population crosses, a mating experiment was conducted between the Rift valley and South African population to produce an F1 generation, and these were backcrossed with the South African parent population. Fertility of eggs produced by the F1/parent population cross was significantly reduced when compared to fertility of the "true" South African line, and the F1/F1 cross. The contributions of the observed genetic differences and inter population incompatibility for the failure of previous biocontrol attempts are discussed and recommendations on future biocontrol practices are given.     

Origin and taxonomic status of the Palearctic population of the stem borer Sesamia nonagrioides (Lefèbvre) (Lepidoptera: Noctuidae)

The major pest of maize in Mediterranean Europe, the stem borer Sesamia nonagrioides (Lefèbvre) (Lepidoptera: Noctuidae), has a fragmented distribution, north and south of the Sahara. The present study aimed: (1) to clarify the uncertain taxonomic status of the Palearctic and sub‐Saharan populations which were first considered as different species and later on as subspecies (Sesamia nonagrioides nonagrioides and Sesamia nonagrioides botanephaga) and (2) to investigate the origin of the Palearctic population which extends from Spain to Iran, outside what is considered typical for this mainly tropical genus. We reconstructed the evolutionary history of both populations using one nuclear and two mitochondrial genes. The sub‐Saharan taxon was fragmented in two isolated populations (West and East) whose mitochondrial genes were distant by 2.3%. The Palearctic population was included in the East African clade and its genes were close or identical to those of a population from Central Ethiopia, where the species was discovered for the first time. Similarly, in Africa, the alleles of the nuclear gene were distributed mainly in two West and East clades, whereas some Palearctic alleles belonged to the West clade. The Palearctic population originated therefore from East and West Africa and is the progeny of the cross between these two African populations. The main species concepts were in agreement, leading to the conclusion that the three populations are still conspecific. In the surveyed regions, the species therefore does not include two subspecies but three isolated populations. The Palearctic population suffered from severe bottlenecks that resulted in the fixation of one East African mitochondrial genome and the large reduction in its genetic diversity compared to the African populations. The data suggest that natural colonization of the Palearctic region was more plausible than human introduction. The allelic distribution of the Palearctic population was similar to that of species that survived the last glaciation. It is concluded that the African populations expanded during the last interglacial, crossed the Sahara and mixed in North Africa where fixation of the East mitochondrial genome occurred. The species then colonized Europe westward through only one eastern entrance. The coalescent‐based estimate of the time to the ancestor of the Palearctic population was 108 000 years, which is consistent with this scenario. © 2011 The Linnean Society of London, Biological Journal of the Linnean Society, 2011, 103 , 904–922.     

Genetic diversity,evolutionary history and implications for conservation of the lion (Panthera leo) in West and Central Africa

Aim In recent decades there has been a marked decline in the numbers of African lions (Panthera leo), especially in West Africa where the species is regionally endangered. Based on the climatological history of western Africa, we hypothesize that West and Central African lions have a unique evolutionary history, which is reflected by their genetic makeup. Location Sub‐Saharan Africa and India, with special focus on West and Central Africa. Method In this study 126 samples, throughout the lion’s complete geographic range, were subjected to phylogenetic analyses. DNA sequences of a mitochondrial region, containing cytochrome b, tRNAPro, tRNAThr and the left part of the control region, were analysed. Results Bayesian, maximum likelihood and maximum parsimony analyses consistently showed a distinction between lions from West and Central Africa and lions from southern and East Africa. West and Central African lions are more closely related to Asiatic lions than to the southern and East African lions. This can be explained by a Pleistocene extinction and subsequent recolonization of West Africa from refugia in the Middle East. This is further supported by the fact that the West and Central African clade shows relatively little genetic diversity and is therefore thought to be an evolutionarily young clade. Main conclusions The taxonomic division between an African and an Asian subspecies does not fully reflect the overall genetic diversity within lions. In order to conserve genetic diversity within the species, genetically distinct lineages should be prioritized. Understanding the geographic pattern of genetic diversity is key to developing conservation strategies, both for in situ management and for breeding of captive stocks.     

The lesser wanderer butterfly, Danaus petilia (Stoll 1790) stat. rev. (Lepidoptera: Danainae), reinstated as a species

Abstract  The lesser wanderer butterfly, Danaus ( Anosia ) chrysippus petilia (Stoll) (Lepidoptera: Danainae), has been treated as a subspecies for the last 100 years. New mitochondrial DNA sequence data for D. petilia , in conjunction with allozyme, structural, morphometric and pattern characters, constitute a compelling case for its specific rank. The holotype of D. petilia has never been located and, as the type location is uncertain, a neotype is designated. Fresh material was collected and examined for this project. Danaus petilia and D. chrysippus have been separated at Lydekker's Line for an estimated 1.1 million years, and they remain interfertile. However, as an allopatric taxon, diagnosable from D. chrysippus , D. petilia merits specific status under the phylogenetic species concept. The following pairs of D. chrysippus subspecies are considered to be synonyms, the first member of each pair having priority: chrysippus  +  aegyptius (subspecies), klugii  +  infumata (hybrid phenotypes) and orientis  + liboria (subspecies).     

The making of the African mtDNA landscape

Africa presents the most complex genetic picture of any continent, with a time depth for mitochondrial DNA (mtDNA) lineages >100,000 years. The most recent widespread demographic shift within the continent was most probably the Bantu dispersals, which archaeological and linguistic evidence suggest originated in West Africa 3,000-4,000 years ago, spreading both east and south. Here, we have carried out a thorough phylogeographic analysis of mtDNA variation in a total of 2,847 samples from throughout the continent, including 307 new sequences from southeast African Bantu speakers. The results suggest that the southeast Bantu speakers have a composite origin on the maternal line of descent, with approximately 44% of lineages deriving from West Africa, approximately 21% from either West or Central Africa, approximately 30% from East Africa, and approximately 5% from southern African Khoisan-speaking groups. The ages of the major founder types of both West and East African origin are consistent with the likely timing of Bantu dispersals, with those from the west somewhat predating those from the east. Despite this composite picture, the southeastern African Bantu groups are indistinguishable from each other with respect to their mtDNA, suggesting that they either had a common origin at the point of entry into southeastern Africa or have undergone very extensive gene flow since.     

Assessing introgression of Sahelian zebu genes into native Bos taurus breeds in Burkina Faso

A total of 350 samples were analyzed to estimate zebu gene proportions into two different taurine cattle breeds of Burkina Faso (Lobi and N’Dama) using 38 microsatellites and various statistical methodologies. West African and East African zebu samples were sequentially used as reference parental populations. Furthermore, N’Dama cattle from Congo, the composite South African Bonsmara cattle breed and a pool of European cattle were used successively as second parental populations. Independently of the methodology applied: (a) the use of West African zebu samples gave higher admixture coefficients than the East African zebu; (b) the higher zebu proportions were estimated when the European cattle was used as parental population 2; and (c) the use of the N’Dama population from Congo as parental population 2 gave the more consistent zebu proportion estimates for both the Lobi and the N’Dama breeds. In any case, the zebu admixture proportions estimated were not negligible and were always higher in the N’Dama cattle than in the Lobi cattle of Burkina Faso. This suggested that the introgression of Sahelian zebu genes into the taurine cattle of Southern West Africa can follow a complex pattern that can depend on local agro-ecological features. The current research pointed out that the estimation of admixture coefficients is highly dependent on both the assumptions underlying the methodologies applied and the selection of parental populations. Our analyses suggest that either too high or nil genetic identity between the parental and the expectedly derived populations must be avoided.     

East African pigs have a complex Indian,Far Eastern and Western ancestry

In this study, we have characterized the mitochondrial diversity of 81 swine from Uganda. Median‐joining network analysis of D‐loop sequences from these individuals and others characterized in previous studies allowed us to determine that Ugandan pigs cluster with populations from the West (Europe/North Africa), Far East and India. In addition, partial sequencing of the Y‐chromosome UTY locus in 18 Ugandan domestic pigs revealed the segregation of a single HY1 lineage that has a cosmopolitan distribution. A Western and Far Eastern ancestry for East African pigs had been already reported, but this is the first study demonstrating an additional contribution from the Indian porcine gene pool. This result is consistent with the high frequency of zebuine alleles in cattle from East Africa. The geographic coordinates of East Africa, at the crossroads of many trading routes that, through the ages, linked Europe, Africa and Asia, might explain the rich and complex genetic heritage of livestock native to this area.     

Mitochondrial DNA variability in Giraffa camelopardalis: consequences for taxonomy, phylogeography and conservation of giraffes in West and central Africa

The giraffe (Giraffa camelopardalis) still survives in four countries of West and central Africa. The populations of Niger and Cameroon are generally assigned to the subspecies peralta, but those of Chad and the Central African Republic are taxonomically problematic, as they are referred to as either peralta, or antiquorum, or congoensis. In this study, a mitochondrial fragment of 1765 nucleotide sites, covering the complete cytochrome b gene, three transfer RNAs and a large part of the control region, was sequenced to assess the relationships between several populations of giraffe. The phylogenetic analyses performed on the 12 identified haplotypes indicate that northern giraffes constitute a natural group, distinct from that of southern giraffes. Surprisingly, the giraffes of Niger are found to be more closely related to the giraffes of East Africa (subspecies rothschildi and reticulata) than to those of central Africa. We conclude therefore that the subspecies peralta contains only the Niger giraffes, whereas the subspecies antiquorum includes all populations living in Cameroon, Chad, the Central African Republic, and southwestern Sudan. We suggest that the ancestor of the Nigerian giraffe dispersed from East to North Africa during the Quaternary period and thereafter migrated to its current Sahelian distribution in West Africa, in response to the development of the Sahara desert. This hypothesis implies that Lake Mega-Chad acted as a strong geographical barrier during the Holocene, preventing any contact between the subspecies peralta and antiquorum. Our study has direct implications for conservation management, as we show that no subspecies peralta is represented in any European zoos, only in Niger, with a small population of less than 200 individuals.     

Regional differences in the distribution of the sub-Saharan, West Eurasian, and South Asian mtDNA lineages in Yemen

Despite its key location for population movements out of and back into Africa, Yemen has not yet been sampled on a regional level for an investigation of sub-Saharan, West Eurasian, and South Asian genetic contributions. In this study, we present mitochondrial DNA (mtDNA) data for regionally distinct Yemeni populations that reveal different distributions of mtDNA lineages. An extensive database of mtDNA sequences from North and East African, Middle Eastern and Indian populations was analyzed to provide a context for the regional Yemeni mtDNA datasets. The groups of western Yemen appear to be most closely related to Middle Eastern and North African populations, while the eastern Yemeni population from Hadramawt is most closely related to East Africa. Furthermore, haplotype matches with Africa are almost exclusively confined to West Eurasian R0a haplogroup in southwestern Yemen, although more sub-Saharan L-type matches appear in more northern Yemeni populations. In fact, Yemeni populations have the highest frequency of R0a haplotypes detected to date, thus Yemen or southern Arabia may be the site of the initial expansion of this haplogroup. Whereas two variants of the sub-Saharan haplogroup M1 were detected only in southwestern Yemen close to the Bab el-Mandeb Strait, different non-African M haplotypes were detected at low frequencies (approximately 2%) in western parts of the country and at a higher frequency (7.5%) in the Hadramawt. We conclude that the Yemeni gene pool is highly stratified both regionally and temporally and that it has received West Eurasian, Northeast African, and South Asian gene flow.     

Historical biogeography of Drosophila simulans based on Y-chromosomal sequences

Y-chromosomal sequences have been used for phylogeographic studies in humans and other mammals, but so far have been ignored as a source of historical information in Drosophila and other insects with X/Y sex determination. Here, we present the first phylogeographic study of Drosophila simulans based on the Y chromosome. Geographic distribution of Y-chromosomal haplotypes suggests a high degree of population subdivision within Africa, as well as between the African and cosmopolitan groups of populations. Consistent with earlier studies based on autosomal and X-linked loci, our results suggest that D. simulans originated in Madagascar or East Africa, and that the South and West African populations of this species are derived.     

The butterfly Danaus chrysippus (L.) in East Africa comprises polyphyletic, sympatric lineages that are, despite behavioural isolation, driven to hybridization by female-biased sex ratios

Mitochondrial and nuclear DNA information was analysed among four subspecies of the African Queen butterfly, Danaus (Anosia) chrysippus sensu lato (s.l.), along with four other Danaus species drawn from all three subgenera (D. (Danaus) plexippus, D. (Salatura) genutia, D. (A.) gilippus, D. (A.) eresimus) and two outgroup species from the same tribe, Tirumala septentrionis and Amauris niavius. A mitochondrial phylogeny derived from the 12S rRNA (347 bp) and COI (537 bp) loci indicates two very distinct haplotypes for subspecies D. (A.) c. dorippus, dorippus‐1 and dorippus‐2. Interestingly, dorippus‐1, on the one hand, and all other D. (A.) chrysippus haplotypes, on the other, are the most distantly related clades within the genus and have different most recent ancestors from different subgenera, though sharing the common ancestor of the monophyletic genus. A phylogeny based on the EF1‐α nuclear locus (400 bp) shows that the two well‐separated mitochondrial lineages of dorippus are identical for this gene and reciprocally monophyletic to the other D. (A.) chrysippus lineages. Thus, nuclear and cytoplasmic phylogenies are not only discordant, but also suggest that both D. (A.) chrysippus s.l. and subspecies dorippus are polyphyletic. Paradoxically, four African subspecies, chrysippus‐orange, chrysippus‐brown, alcippus and dorippus, though substantially vicariant, hybridize extensively in East Africa wherever the ranges of two or more of them overlap. Linkage disequilibrium, and hence sexual isolation, in sympatry between colour (nuclear) genes and unlinked mitochondrial (cytoplasmic) loci is consistent across populations and therefore indicates the operation of positive natural selection. Together with data from previous experimental and field work, our results suggest that extensive hybridization occurs among once allopatric or parapatric lineages, that are now nascent species. We deduce that hybridism among lineages in sympatry is currently enforced, in the face of assortative mate choice, by a bacterial symbiont, Spiroplasma, a male‐killer that forces females in female‐biased populations to pair with heterotypic males. In discussion we emphasize that neither D. (A.) chrysippus s.l. as presently circumscribed, nor its component clades, conform to any established concept of species. © 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 86 , 117–131.     

Allozyme variation in 16 natural populations of Faidherbia albida (Del.) A. Chev

The objectives of this study were (1) to investigate the genetic variability; and (2) to determine the phylogenetic relationship in the leguminous tree Faidherbia albida through out its range of distribution in East, West and Southern Africa. A total of 16 populations were subjected to enzyme electrophoresis and 6 enzyme systems (Adh, Mdh, G6pdh, Idh, 6Pgdh, and alpha-Est) encoded by 12 loci were scored. A total of fifty one different alleles were detected, with an average of 2.5 alleles per locus. Forty three percent of the loci were polymorphic at a 95% criterion. The average expected heterozygosity (gene diversity index H(e)) was 0.141. All provenances showed significant deviation from Hardy-Weinberg expectation. The UPGMA cluster analysis, based on Modified Rogers distance revealed close similarities between Eastern and Southern African provenances, except Debre zeit from Ethiopia, which was closest to West African populations than to the East African populations. Also, Bignona from West Africa was peculiarly closer to East African provenances. Differentiation of West African populations from Eastern and Southern African provenances was quite evident, emphasising the clinical pattern of variation in these two major geographical races of F. albida in Africa. Wright's F-statistics showed an overall significant deficit of heterozygotes, a common feature in mixed mating, entomophilous, widespread species such as F. albida. The dendrogram analysis showed wide separation among the three Ethiopian provenances indicating a high level of genetic differentiation and diversity among them.     

Phylogeography and population genetics of the maize stalk borer Busseola fusca (Lepidoptera, Noctuidae) in sub-Saharan Africa

The population genetics and phylogeography of African phytophagous insects have received little attention. Some, such as the maize stalk borer Busseola fusca, display significant geographic differences in ecological preferences that may be congruent with patterns of molecular variation. To test this, we collected 307 individuals of this species from maize and cultivated sorghum at 52 localities in West, Central and East Africa during the growing season. For all collected individuals, we sequenced a fragment of the mitochondrial cytochrome b. We tested hypotheses concerning the history and demographic structure of this species. Phylogenetic analyses and nested clade phylogeographic analyses (NCPA) separated the populations into three mitochondrial clades, one from West Africa, and two--Kenya I and Kenya II--from East and Central Africa. The similar nucleotide divergence between clades and nucleotide diversity within clades suggest that they became isolated at about the same time in three different refuges in sub-Saharan Africa and have similar demographic histories. The results of mismatch distribution analyses were consistent with the demographic expansion of these clades. Analysis of molecular variance (amova) indicated a high level of geographic differentiation at different hierarchical levels. NCPA suggested that the observed distribution of haplotypes at several hierarchical levels within the three major clades is best accounted for by restricted gene flow with isolation by distance. The domestication of sorghum and the introduction of maize in Africa had no visible effect on the geographic patterns observed in the B. fusca mitochondrial genome.     

Population biology of tropical African butterflies. Sex ratio and genetic variation in Acraea encedon

In Uganda the butterfly, Acraea encedon (L.), occurs in well-defined populations that are largely genetically and ecologically isolated from each other. During the course of a large-scale capture-recapture programme about 18,000 different individuals were marked and released, and from these and many recaptures aspects of the population structure were investigated. Some populations are composed of almost all females while in other populations the sex ratio is normal. In the predominantly female populations many females remain unmated and produce infertile eggs, laying them on plants that are not foodplants and on each other. In such populations there is intense aggregating behaviour. Males live longer than females and each male may mate with several different females. There is evidence for the existence of a self-regulating mechanism in the predominantly female populations in which population size and male frequency are reciprocally dependent. Acraea encedon exists in a large variety of sympatric colour forms, the frequency of which varies from population to population. Some colour forms survive better than others. The hindwing of Acraea encedon has 20 black spots each of which can be absent in any particular butterfly. The relative frequency of missing spots varies from population to population. The ecological genetics of Acraea encedon is discussed in relation to theories of mimicry and polymorphism.     

Plio-Pleistocene history and phylogeography of Acacia senegal in dry woodlands and savannahs of sub-Saharan tropical Africa: evidence of early colonisation and recent range expansion

Drylands are extensive across sub-Saharan Africa, socio-economically and ecologically important yet highly sensitive to environmental changes. Evolutionary history, as revealed by contemporary intraspecific genetic variation, can provide valuable insight into how species have responded to past environmental and population changes and guide strategies to promote resilience to future changes. The gum arabic tree (Acacia senegal) is an arid-adapted, morphologically diverse species native to the sub-Saharan drylands. We used variation in nuclear sequences (internal transcribed spacer (ITS)) and two types of chloroplast DNA (cpDNA) markers (PCR-RFLP, cpSSR) to study the phylogeography of the species with 293 individuals from 66 populations sampled across its natural range. cpDNA data showed high regional and rangewide haplotypic diversity (h_T(cpSSR)=0.903–0.948) and population differentiation (G_ST(RFLP)=0.700–0.782) with a phylogeographic pattern that indicated extensive historical gene flow via seed dispersal. Haplotypes were not restricted to any of the four varieties, but showed significant geographic structure (G_ST(cpSSR)=0.392; R_ST=0.673; R_ST>R_ST (permuted)), with the major division separating East and Southern Africa populations from those in West and Central Africa. Phylogenetic analysis of ITS data indicated a more recent origin for the clade including West and Central African haplotypes, suggesting range expansion in this region, possibly during the Holocene humid period. In conjunction with paleobotanical evidence, our data suggest dispersal to West Africa, and across to the Arabian Peninsula and Indian subcontinent, from source populations located in the East African region during climate oscillations of the Plio-Pleistocene.