Molecular identification of <Emphasis Type="Italic">Schoenoplectiella</Emphasis> species (Cyperaceae) by use of microsatellite markers

Over the past few decades, use of molecular markers for species delimitation has drastically increased. Schoenoplectiella Lye has been recognized as a taxonomically difficult genus because of its morphological simplicity and frequent interspecific hybridization. The main objective of this study was to clarify the taxonomic identities of eight Schoenoplectiella species by use of molecular markers. We also evaluated the genetic relationships among S. × trapezoidea, known as a natural hybrid, and its close relatives. We used six microsatellite markers for 44 individuals from 31 natural populations of eight Schoenoplectiella species in South Korea. Six microsatellite marker combinations generated 59 amplification-detectable bands, of which 33 were specifically detected in one or more individuals of each species. Cluster analysis revealed that the grouping was consistent with the taxonomically recognized species. Our results do not support the hybrid origin of S. × trapezoidea. Rather, they suggest that this species is more closely related to S. hotarui. The informative microsatellite markers enabled us to clarify the distinctions among Schoenoplectiella species from South Korea and to identify the genetic relationships among these species. The molecular signatures found suitable for accurate identification of Schoenoplectiella species can be reliably used for studies of the phylogeny and evolution of this genus.     

Using progesterone 5β-reductase, a gene encoding a key enzyme in the cardenolide biosynthesis, to infer the phylogeny of the genus Digitalis

Summary The progesterone 5β-reductase (5β-POR), a key enzyme in the cardenolide biosynthesis, was sequenced for 21 species of Digitalis and Isoplexis to infer phylogenetic and biogeographic relationships. This new secondary metabolism molecular marker was compared to the previously applied nuclear ITS and plastid trnL-F sequences. The results from separate analyses show high congruence within the genus Digitalis and support the conclusion that all species of Isoplexis have a common origin and are embedded in Digitalis. The genus Isoplexis therefore should be reduced to sectional rank within the genus Digitalis. The sequence analyses give further evidence that additional sequence data increase support for relationships. It demonstrates that poorly supported relationships in smaller data sets may lead to erroneous conclusions about the evolution of the investigated taxa.     

Genetic relationships among Mexican white pines (Pinus, Pinaceae) based on RAPD markers

Genetic relationships among Mexican white pines have not been completely resolved by DNA sequencing analyses. The use of random amplified polymorphic DNA (RAPD) markers for the study of interspecific relationships has been questioned because of the possible lack of homology of co-migrating bands between species. However, several RAPD based studies on pines have provided sufficient information to discriminate between closely related taxa. Genetic relationships among four species of Mexican white pines (Pinus ayacahuite, Pinus strobiformis, Pinus lambertiana and Pinus chiapensis) were estimated based on RAPD markers. Sixty-nine primers generated 247 bands in pooled DNA samples from ten populations. In addition, four selected primers generated 27 bands in 176 individual DNA samples. Unweighted Pair Group Method with Arithmetic Average (UPGMA) dendrograms based on Jaccard similarity indices were constructed. The results suggest that the closest pine species analyzed were P. ayacahuite and P. strobiformis, followed by P. lambertiana. The most genetically distant species was P. chiapensis. Cluster analyses did not support P. strobiformis as a distinct species from P. ayacahuite.     

The Genus Syringa: Molecular Markers of Species and Cultivars

RAPD analysis was carried out with 22 accessions of the genus Syringa, including six species, one interspecific hybrid, and 15 cultivars. In total, 512 polymorphic fragments were detected; species-specific and cultivar-specific markers were identified. For the first time, genetic polymorphism and genome similarity coefficients were estimated and phylogenetic relationships were established for the genus Syringa.     

The Use of RAPD Markers to Assess Catfish Hybridization

Taiwan's endemic catfish Clarias fuscus is gradually disappearing from its native habitat, and has been proposed for genebank preservation. Environmental pressures, including exotic species interference and habitat destruction, as well as possible competitive advantages of the hybrids over this species. In order to quickly and effectively provide a reliable DNA fingerprint for the pure strain of C. fuscus we used RAPD markers to assess C. fuscus, C. mossambicus, and C.batrachus. Of the 200 primers screened to prime PCR amplification of DNA from wild-caught C. fuscus, 16 yielded reproducible DNA bands. Unique RAPD markers generated from 3 PCR primers (#211, #245 and #287) are shown to be alleles present in the genomes of C. mossambicus but absent in the genome of C. fuscus. Hybrids of C. fuscus and C. mossambicus, therefore, could possibly be distinguished by the use of these specific molecular markers. Catfish caught from the Mingder Dam were then cautiously removed from the preserved stock because of the appearance of hybrid markers in their genomes.     

Natural hybridization between <Emphasis Type="Italic">Phlomis lycia</Emphasis> D. Don × <Emphasis Type="Italic">P. bourgaei</Emphasis> Boiss., (Lamiaceae) revealed by RAPD markers

Randomly Amplified Polymorphic DNA markers (RAPD) were used to assess the hybrid identity of individuals sampled as Phlomis × termessi Davis. Out of 95 primers screened, 11 primers produced reproducible amplification patterns used for discrimination of P. × termessi and their parents. Eleven primers produced 81 bands. Forty two percent of the RAPD bands existed in parents. Of the 54 bands found in P. lycia, 19 were found only in this species and 7 of these were monomorphic. Similarly, of 57 RAPD bands observed in P. bourgaei, 18 were found only in P. bourgaei and 6 of these were monomorphic. Among hybrid individuals, 35 of the 73 markers were monomorphic. Fifteen of these existed in individual parents showing that parents were homozygous for these markers. Of the 35 monomorphic bands observed among hybrid individuals, 5 were present in the samples of one of the parents and completely absent from the samples of the other; therefore, additive inheritance is indicated. Of the 5 additive bands, 1 was inherited from P. bourgaei and 4 were inherited from P. lycia. Among 38 polymorhic markers observed in hybrid individuals, 9 were new and hybrid-specific. Pollen fertility was also investigated. Mean pollen fertility for P. lycia and P. bourgaei was 93% and 97% respectively. However, mean pollen fertility for hybrids was 65% (±10.5).     

Molecular discrimination of six species of Bagrid catfishes from Indus river system using randomly amplified polymorphic DNA markers

Bagrid catfishes constitute a very important group of fishes having immense commercial importance in south-east countries. The phylogenetic relationships and genome specificity among six species of Bagrid catfishes (Mystus bleekeri, M. cavasius, M. vittatus, M. tengara, M. aor and M. seenghala) were investigated using RAPD markers as discriminating characters for the first time. 511 RAPD fragments were generated using ten decamer primers of arbitrary nucleotide sequences. Amplification reactions resulted in fragments ranging in length between 92 and 2,863 bp, which were assigned to 155 RAPD loci. Clearly resolved and repeatable bands were scored for their presence or absence in a binary matrix. Different RAPD profiles were observed for all the six Mystus species. In the present study three group diagnostic, eleven group exclusive and 18 species-specific markers were generated. Thus six Mystus species can be successfully differentiated on the basis of these 18 species-specific RAPD markers. UPGMA dendrogram constructed on the basis of genetic distance formed two distinct clusters, M. seenghala and M. aor form one separate cluster from other four species i.e., M. tengara, M. cavasius, M. bleekeri and M. vittatus. The inferences drawn from the above study clearly showed their genetic distinctness from the other four Mystus species and supported their inclusion into a separate genus, Sperata.     

Molecular markers shared by diverse apomictic Pennisetum species

Two molecular markers, a RAPD (randomly amplified polymorphic DNA) and a RFLP/STS (restriction fragment length polymorphism/sequence-tagged site), previously were found associated with apomictic reproductive behavior in a backcross population produced to transfer apomixis from Pennisetum squamulatum to pearl millet. The occurrence of these molecular markers in a range of 29 accessions of Pennisetum comprising 11 apomictic and 8 sexual species was investigated. Both markers were specific for apomictic species in Pennisetum. The RFLP/STS marker, UGT 197, was found to be associated with all taxa that displayed apomictic reproductive behavior except those in section Brevivalvula. Neither UGT197 nor the cloned RAPD fragment OPC-04₆₀₀ hybridized with any sexually reproducing representatives of the genus. The cloned C04₆₀₀ was associated with 3 of the 11 apomictic species, P. ciliare, P. massaicum, and P. squamulatum. UGT197 was more consistently associated with apomictic reproductive behavior than OPC04₆₀₀ or cloned C04₆₀₀, thus it could be inferred that UGT197 is more closely linked to the gene(s) for apomixis than the cloned C04₆₀₀. The successful use of these probes to survey other Pennisetum species indicates that apomixis is a trait that can be followed across species by using molecular means. This technique of surveying species within a genus will be useful in determining the relative importance of newly isolated markers and may facilitate the identification of the apomixis gene(s).     

Identification of larch species (Larix decidua, Larix kaempferi and Larix X eurolepis) and estimation of hybrid fraction in seed lots by RAPD fingerprints

Species-specific RAPD markers were used to identify the different larch species (Larix decidua and Larix kaempferi) and their interspecific hybrid (Larix X eurolepis). Although morphological differences between pure species and the hybrids exist, differentiation is not always possible, especially at an early stage (seed or plantlet). Eleven RAPD markers differentiated the two larch species, and 4 species-specific markers were sufficient to estimate the F₁ hybrid fraction in a seed lot. The species-specific markers were tested on individual trees of European and Japanese larches of diverse geographic origins and on several seed lots of different origins (F₁, F₂ hybrids and pure species). The 4 specific markers found for the European larch and the Japanese larch were monomorphic and present in all provenances and in all F₁ hybrid trees tested. Polymorphic SCAR fragments were obtained for 3 of the 11 fragments originally selected for the RAPD screening phase. For 2 of them, the sequence had some homology with the mitochondrial genome of other organisms and is thus mitochondrial. The two mitochondrial fragments and the OPF-13₁₀₀₀ fragment exhibited one polymorphic band, thereby maintaining its species-specific identity: OPF-13₁₀₀₀ is specific to the European larch. The 4 RAPD primers selected in this study offer a reliable, quick and cheap tool for the identification of different larch species (Larix decidua and Larix kaempferi) and their interspecific hybrid (Larix X eurolepis). Received: 28 February 1999 / Accepted: 29 April 1999     

Construction of synteny groups of Brassica alboglabra by RAPD markers and detection of chromosome aberrations and distorted transmission under the genetic background of B. campestris

Interspecific hybrids were produced by crosses between the inbred lines of B. campestris and B. alboglabra, and were backcrossed twice to B. campestris. Genetical constitutions of the BC₂ plants were analyzed by RAPD (random amplified polymorphic DNA), flow cytometry and cytological observations. By using 140 arbitrary primers, a total of 137 polymorphic bands were obtained and 125 were found to be specific to B. alboglabra. Based on the presence and absence of the specific RAPD markers of B. alboglabra, 13 synteny groups were constructed. The number of markers in each synteny group was found to be different and varied from 2 to 28. This reflects the difference in the degree of genetic variability among the B. alboglabra chromosomes from those of B. campestris. Losses or gains of RAPD markers were observed frequently in most of the synteny groups, which indicated the occurrence of chromosome translocations and/or deletions in the chromosomes of B. alboglabra. In a population of 40 BC₂ plants, chromosome transmission rates were analyzed by using the RAPD markers in each synteny group. Most of the chromosomes of the synteny groups were transmitted with rates of 0.37–0.68. An extremely high transmission rate, 0.98, was however observed in one of the synteny groups. Inheritance data of the synteny groups revealed relationships among themselves. The plants lacking the RAPD markers of two synteny groups tended to lose others belonging to the rest of the synteny groups, indicating the effects of these groups on the transmission of B. alboglabra chromosomes to the B. campestris background. Received: 26 February 1999 / Accepted: 30 December 1999     

Evaluation of genetic relationships in Dalbergia species using RAPD markers

Conservation of identified germplasm is an important component forefficient and effective management of plant genetic resources. Traditionally,species identification has relied on morphological characters like growth habit,floral morphology like flower colour, and agronomic characteristics of the plant.Dalbergia species are important wind-dispersed tropicaltimber trees which exhibit high intrafruit seed abortion because of intensesibling competition for maternal resources. Studies were undertaken foridentification and genetic relationships in five species ofDalbergia and to evaluate genetic diversity withinpopulations of Dalbergia sisso, D.latifolia, D. paniculata, D.assamica and D. spinosa by using randomamplified polymorphic DNAs (RAPD) markers. Analysis was started by using 30decamer primers that allowed to distinguish five species and to select a reducedset of primers. The selected primers were used for identification and forestablishing a profiling system to estimate genetic relationships and toevaluate the genetic variability among the individuals in a population ofDalbergia species. A total of 120 distinct DNA fragments(bands), ranging from 0.3 to 4.0 kb, were amplified byusing nine selected random decamer primers. The genetic similarity was evaluated onthe basis of presence or absence of bands, which revealed a wide range ofvariability within the species. The cluster analysis indicated that five speciesof Dalbergia formed two major clusters. The first clusterconsisted of D. spinosa, D. latifolia and D.sisso. The second cluster was represented by two species, i.e.D. paniculata and D. assamica.A maximum similarity of 60% was observed in D. paniculata andD. assamica and they formed a minor cluster.Dalbergia latifolia and D. sissoformed another minor cluster with more than 50% similarity. Dalbergiaspinosa shared up to 40% similarity with D.latifolia and D. sisso. All the species sharemore than 20% similarity among themselves. The closest genetic distance existedwithin populations of different Dalbergia species. Thus,these RAPD markers have the potential for conservation of identified clones andcharacterization of genetic relatedness among the species. This is also helpful intree breeding programs and provides an important input into conservation biology.     

AFLP analysis of genetic relationships in the tribe Datureae (Solanaceae)

The AFLP technique was evaluated as a tool for assessing species relationships within the tribe Datureae and genetic distances were estimated for 47 accessions of over 12 species. The phenetic trees from various analyses of the AFLP data gave very high co-phenetic correlation values, and were found to be consistent with previous trees based on the analysis of different data types, in particular ITS-1 sequences, isozymes and morphology, carried out on the same accessions. These results indicated that the AFLP technique is both an efficient and effective tool for determining genetic relationships among taxa in the Solanaceae. A new classification is proposed for the tribe Datureae, which maintains the arborescent species as a separate genus, Brugmansia, and recognises three sections within the genus Datura; Stramonium, Dutra and Ceratocaulis. D. discolor, previously placed in section Dutra, was found to be intermediate between sections Dutra and Stramonium. Received: 11 January 1999 / Accepted: 1 February 1999     

Molecular analysis of interspecific and intergeneric relationships of Banksia using RAPDs and non-coding chloroplast DNA sequences

 Random amplified polymorphic DNA (RAPD) and chloroplast DNA (cpDNA) markers were used to study interspecific and intergeneric relationships of Banksia (Proteaceae) to aid breeding of the genus for cut flower production. The accepted morphological phylogeny of Banksia, with two subgenera, two sections and 13 series, is unclear regarding the relationships of the commercial cut flower species B. coccinea. Fifteen RAPD primers and a non-coding cpDNA sequence between the trnL (UAA) and trnF (GAA) gene were applied to species of Banksia, the related genus Dryandra, and to Musgravea heterophylla as the outgroup, with cluster analysis applied to the results. The two methods were in broad agreement with each other, and with the accepted taxonomy, with closely related species pairs and groups clustering together, but RAPDs were not informative between distantly related species or species pairs. Banksia coccinea clustered with Dryandra and formed a polytomy with 2 Dryandra species and the two sections of subgenus Banksia. Subgenus Isostylis formed a polytomy with D. formosa, basal to subgenus Banksia, but with B. cuneata and B. illicifolia (both in subgenus Isostylis) polyphyletic. Dryandra did not separate as a clade and fell within Banksia, raising questions about the currently accepted view of the two as sister genera with parallel morphological development. The results indicate that interspecific and intergeneric hybridisation with genus Dryandra and subgenus Banksia may be possible routes for improvement of the commercial species B. coccinea. Received: 25 September 1996 / Accepted: 8 November 1996     

Phylogenetic relationships in <Emphasis Type="Italic">Betula</Emphasis> (Betulaceae) based on AFLP markers

The genus Betula comprises various species in boreal and temperate climate zones of the Northern Hemisphere. The taxonomy of Betula is controversial and complicated by parallel evolution of morphological traits, polyploidization events, and extensive hybridization and introgression among species. Multilocus molecular data from AFLPs were used to provide phylogenetic information. A large number of polymorphic markers (321 variable bands) were produced in 107 Betula accessions from 23 species and 11 hybrids. The AFLP results were largely congruent with the results from previously examined nuclear DNA markers. Four distinct subgenera were identified within the genus Betula. These subgenera were partly in disagreement with the traditional (but disputed) division of the genus. In addition, the results indicated several groups of conspecific taxa. The majority of the species fell within subgenus Betula and shared a high degree of similarity with B. pendula. All hybrids were associated with this group, and the AFLP data contained signals on putative parents for some of the interspecific hybrids. Subgenus Chamaebetula and part of the Neurobetula species should be merged with Betula. The subgenera Betulenta, Betulaster, and the remaining part of Neurobetula are distinct and well supported. Although our results indicate that four major taxonomic groups can be recognized within the genus Betula, the relationship between them remains unclear. This may be due to the occurrence of hybridization and introgression, which would have a homogenizing effect on the relationships between species. Naturally occurring Betula species of hybrid origin may explain the low bootstrap values within the Betula clade. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Clonal propagation and production of cichoric acid in three species of Echinaceae

In vitro tissue culture protocols were tested for propagation of Echinacea purpurea, Echinacea pallida and Echinacea angustifolia in order to obtain biomass for the production of cichoric acid, which is the major active compound in the Echinacea extracts. The in vitro culture process was initiated by seed germination on half-strength Murashige and Skoog (MS) medium. Multiplication was achieved on MS medium supplemented with naphthaleneacetic acid (NAA), indole-3-butyric acid (IBA), 2-iso-pentenyladenine (2iP), and N⁶-benzyladenine (BA) in different concentrations. Shoot explants produced the highest number of shootlets on MS medium, which was supplemented with 0.1 mg/l 2iP and 0.1 mg/l IBA. RAPD markers revealed genetic polymorphism in some instances between in vitro generated plantlets such as for E. purpurea plantlets analyzed with the OPO-8 primer. RAPD markers generated with the primer 4A-29 revealed low levels of genetic variation between in vitro plantlets for all three species of Echinacea, while remaining RAPD markers revealed no variation. Content of cichoric acid in leaves, shoots, and callus was analyzed by high-performance liquid chromatography/MS and was identified in all studied samples, independent of species or tissue type. Highest levels (0.39–0.73 mg/g dw) were observed in shoots and leaves.     

Genetic diversity and phylogenetic relationship as revealed by inter simple sequence repeat (ISSR) polymorphism in the genus Oryza

Inter simple sequence repeat (ISSR) polymorphism was used to determine genetic diversity and phylogenetic relationships in Oryza. Forty two genotypes including 17 wild species, representing AA,BB,CC,EE,FF,GG,BBCC,CCDD, and HHJJgenomes, two cultivated species, Oryza sativa (AA) and Oryza glaberrima (AA), and three related genera, Porteresia coarctata, Leersia and Rhynchoryza subulata, were used in ISSR analysis. A total of 30 ISSR primers were screened representing di-, tri-, tetra- and penta-nucleotide repeats, of which 11 polymorphic and informative patterns were selected to determine the genetic diversity. The consensus tree constructed using binary data from banding patterns generated by ISSR-PCR clustered 42 genotypes according to their respective genomes. ISSR analysis suggests that the genus Oryza may have evolved following a polyphyletic pathway; Oryza brachyantha (FF genome) is the most divergent species in Oryza and Oryza australiensis (EE genome) does not fall under the Officinalis complex. DNA profiles based on ISSR markers have revealed potential diagnostic fingerprints for various species and genomes, and also for individual accessions/cultivars. Additionally ISSR revealed 87 putative genome/species-specific molecular markers for eight of the nine genomes of Oryza. The ISSR markers are thus useful in the fingerprinting of cultivated and wild species germplasm, and in understanding the evolutionary relationships of Oryza. Received: 23 August 1999 / Accepted: 10 November 1999     

Life history variation between species of the relictual genus Borderea (Dioscoreaceae): phylogeography, genetic diversity, and population genetic structure assessed by RAPD markers

The genus Borderea consists of two species, B. pyrenaica and B. chouardii, taxa which have been previously considered as conspecific due to their overall close morphology. These two sole species of the rare genus of Dioscoreaceae are endemic to the Pyrenees (Spain, France). This mountain range likely operated as a refugium for these plants during the last glaciations. B. chouardii is only known from a single population in the Spanish Prepyrenees and has been classified as at risk of extinction in the Red List of Endangered Species (IUCN); B. pyrenaica shows a narrow distribution range in the central Pyrenees and Prepyrenees. We analysed genetic variation, population structure and differentiation in these two taxa using RAPD markers. Our study was conducted on the same seven populations for which very low levels of genetic differentiation were detected previously through allozyme analysis. By contrast, high levels of genetic variability were detected through the RAPD hypervariable markers. Twelve RAPD primers produced 112 distinct bands in the 397 surveyed individuals, totalling 395 different RAPD phenotypes. Only four bands were monomorphic across all samples of Borderea, whereas 21 of the polymorphic bands were species‐specific (20 for B. chouardii, and one for B. pyrenaica). The largest genetic distances were those between the B. chouardii and the B. pyrenaica phenotypes. An analysis of molecular variance showed greater variance between groups (B. chouardii vs. B. pyrenaica, 76.08%) than within groups (3.60%). RAPD band specificity, phenotypic distances, and the partitioning of variance all support the taxonomic separation of the two species. Statistical evaluation of within‐ and among‐population RAPD genetic variability in B. pyrenaica showed that genetic variability was higher within populations (>80%) than among them. No clear pattern of RAPD differentiation could be observed among the six studied populations of this taxon though slight differences in genetic diversity could be observed in the more isolated Prepyrenean populations compared with the more widespread Pyrenean ones. These results suggest a recent postglacial origin of the present B. pyrenaica populations. © 2003 The Linnean Society of London, Biological Journal of the Linnean Society, 2003, 80 , 483–498.     

Comparative study of interspecific genetic divergence and phylogenic analysis of genus <Emphasis Type="Italic">Jatropha</Emphasis> by RAPD and AFLP

Genus Jatropha with 172 species having significant economic importance belongs to the family Euphorbiaceae. There are no reports on molecular characterization and phylogenetic relationship among the species of Jatropha. Hence, the present study was undertaken to assess the extent of genetic variability that exist and also to establish phylogenetic relationship among Jatropha curcas, J. glandulifera, J. gossypifolia, J. integerrima, J. multifida, J. podagrica and J. tanjorensis using RAPD and AFLP. The percentage of loci that are polymorphic among the species studied was found to be 97.74% by RAPD and 97.25% by AFLP. The mean percentage of polymorphism (PP) was found to be 68.48 by RAPD and 71.33 by AFLP. The phylogram generated with RAPD and AFLP data showed maximum similarity. With the generated data maximum relatedness was found between J. curcas and J. integerrima this may be the reason for the success of inter hybrid crosses between these two species. Neither RAPD nor AFLP data generated in this study supports the view of J. tanjorensis, a natural interspecific hybrid between J. curcas and J. gossypifolia. The present study concludes that both RAPD and AFLP techniques are comparable in divergence studies of Jatropha species. The markers generated by RAPD and AFLP can be employed efficiently for interspecific hybrids identification, marker assisted selection and genetic resource management.     

Assessment of random amplified polymorphic DNA (RAPD) for determining the origin ofYoungia koidzumiana Kitamura (compositae)

We determined the parental species ofYoungia koidzumiana (a natural interspecific hybrid) using PCR and arbitrary 10-mer primers to generate random amplified polymorphic DNA (RAPD) markers. These markers, generated by three primers, were sufficient to distinguishYoungia sonchifolia, Youngia denticulata, Youngia chelidoniifolia, andY. koidzumiana. The electrophoresis profiles of the amplified products from each of the four species were then compared. Three primers produced a total of 42 scorable markers; nine were specific markers forY. denticulata andY. chelidoni-ifolia. The length of the amplified DNA fragments ranged from 370 to 2500 b p. The three primers revealed polymorphic bands, which were indicators of the parental species ofY. koidzumiana. These bands showed a combination of specific profiles forY. denticulata andY. chelidoniifolia. Our results also were comparable to the data obtained for flowering times, floret numbers, and chromosome numbers of the four species. Therefore, we suggest thatY. koidzumiana is a hybrid betweenY. denticulata andY. chelidoniifolia}, and that RAPD markers are well suited for assessing the origins of plant species.     

Comparison of RFLP and RAPD markers to estimating genetic relationships within and among cruciferous species

Restriction fragment length polymorphism (RFLP) and random amplified polymorphic DNA (RAPD) markers are being used widely for evaluating genetic relationships of crop germplasm. Differences in the properties of these two markers could result in different estimates of genetic relationships among some accessions. Nuclear RFLP markers detected by genomic DNA and cDNA clones and RAPD markers were compared for evaluating genetic relationships among 18 accessions from six cultivated Brassica species and one accession from Raphanus sativus. Based on comparisons of genetic-similarity matrices and cophenetic values, RAPD markers were very similar to RFLP markers for estimating intraspecific genetic relationships; however, the two marker types gave different results for interspecific genetic relationships. The presence of amplified mitochondrial and chloroplast DNA fragments in the RAPD data set did not appear to account for differences in RAPD- and RFLP-based dendrograms. However, hybridization tests of RAPD fragments with similar molecular weights demonstrated that some fragments, scored as identical, were not homologous. In all these cases, the differences occurred at the interspecific level. Our results suggest that RAPD data may be less reliable than RFLP data when estimating genetic relationships of accessions from more than one species.