Search

Severe Verticillium dahliae attacks have occurred in artichoke crops in the Comunidad Valenciana region of eastern-central Spain since the late 1990s. Knowledge of genetic and virulence diversity in the pathogen population is a key factor for the management of the disease through disease risk assessment as well as development and use of resistant cultivars. V. dahliae isolates from artichoke (109 isolates) and cotton (three isolates) in that region were characterized by vegetative compatibility grouping (VCG), and specific polymerase chain reaction assays using three sets of primer pairs that differentiate the cotton-defoliating (D) and -nondefoliating (ND) V. dahliae pathotypes. In all, 35 and 39 V. dahliae isolates representative of the identified VCGs and geographic origins were tested for virulence to artichoke cvs. Nun 6374 and Nun 9444, and cotton cv. Acala S J-2, respectively. Four VCGs were identified among 107 artichoke isolates, and 2 isolates were heterokaryon self-incompatible: VCG1A (one isolate), VCG2A (31 isolates), VCG2B (72 isolates), and VCG4B (three isolates). The three cotton isolates were VCG1A. Isolates in VCG2B were distributed across the region and were the most prevalent isolates in the northern part. Conversely, 83.9% of isolates in VCG2A were recovered from the southern part of the region. Two subgroups of isolates were identified in VCG2B based on heterokaryon compatibility with either international or local tester isolates, which further showed diversity in the amplification of 334- and 824-bp DNA fragments which are markers of the D and ND pathotypes, respectively. Virulence of isolates to artichoke and cotton correlated with VCG but the pattern of correlation varied with the host. VCG1A isolates from artichoke and cotton induced defoliation in cotton but not in artichoke. Collectively, isolates of VCG2B and VCG4B were the most virulent and isolates of VCG1A or HSI were the least virulent to artichoke; but isolates of VCG1A were more virulent to cotton than those of any other VCG. Also, molecular subgrouping in VCG2B determined by amplification of the 334- and 824-bp markers correlated with virulence of isolates to the two hosts tested. [ABSTRACT FROM AUTHOR]

Fusarium wilt of chickpea caused by Fusarium oxysporum f. sp. ciceris can be managed by risk assessment and use of resistant cultivars. A reliable method for the detection and quantification of F. oxysporum f. sp. ciceris in soil and chickpea tissues would contribute much to implementation of those disease management strategies. In this study, we developed a real-time quantitative polymerase chain reaction (q-PCR) protocol that allows quantifying F. oxysporum f. sp. ciceris DNA down to 1 pg in soil, as well as in the plant root and stem. Use of the q-PCR protocol allowed quantifying as low as 45 colony forming units of F. oxyporum f. sp. ciceris per gram of dry soil from a field plot infested with several races of the pathogen. Moreover, the q-PCR protocol clearly differentiated susceptible from resistant chickpea reactions to the pathogen at 15 days after sowing in artificially infested soil, as well as the degree of virulence between two E oxysporum f. sp. ciceris races. Also, the protocol detected early asymptomatic root infections and distinguished significant differences in the level of resistance of 12 chickpea cultivars that grew in that same field plot infested with several races of the pathogen. Use of this protocol for fast, reliable, and cost-effective quantification of F. oxysporum f. sp. ciceris in asymptomatic chickpea tissues at early stages of the infection process can be of great value for chickpea breeders and for epidemiological studies in growth chambers, greenhouses and field-scale plots. [ABSTRACT FROM AUTHOR]

A sensitive nested-polymerase chain reaction (PCR) protocol was developed using either of two primer pairs that improves the in planta detection of Peronospora arborescens DNA. The new protocol represented an increase in sensitivity of 100- to 1,000-fold of detection of the oomycete in opium poppy tissue compared with the detection limit of single PCR using the same primer pairs. The new protocol allowed amplification of 5 to 0.5 fg of Peronospora arborescens DNA mixed with Papaver somniferum DNA. The protocol proved useful for amplifying Peronospora arborescens DNA from 96-year-old herbarium specimens of Papaver spp. and to demonstrate that asymptomatic, systemic infections by Peronospora arborescens can occur in wild Papaver spp. as well as in cultivated opium poppy. Also, the increase in sensitivity of the protocol made possible the detection of seedborne Peronospora arborescens in commercial opium poppy seed stocks in Spain with a high frequency, which poses a threat for pathogen spread. Direct sequencing of purified amplicons allowed alignment of a Peronospora arborescens internal transcribed spacer (ITS) ribosomal DNA (rDNA) sequence up to 730-bp long when combining the sequences obtained with the two primer sets. Maximum parsimony analysis of amplified Peronospora arborescens ITS rDNA sequences from specimens of Papaver dubiun, P. hybridum, P. rhoeas, and P. somniferum from different countries indicated for the first time that a degree of host specificity may exist within populations of Peronospora arborescens. The reported protocol will be useful for epidemiological and biogeographical studies of downy mildew diseases as well as to unravel misclassification of Peronospora arborescens and Peronospora cristata, the reported causal agents of the opium poppy downy mildew disease. [ABSTRACT FROM AUTHOR]

Root-knot nematodes (Meloidogyne spp.) are sedentary, obligate endoparasites in plants, where they induce specialized feeding sites. The feeding sites act as strong metabolic sinks to which photosynthates are mobilized. The histopathological modifications in the nematode-induced feeding sites of artificially inoculated chickpea cv. UC 27 were qualitatively and quantitatively compared using five isolates of M. artiellia and one isolate each of M. arenaria, M. incognita, and M. javanica. All Meloidogyne isolates infected chickpea plants, but root gall thickening was significantly less for M. artiellia isolates than for the other Meloidogyne species. Nevertheless, neither the number of giant cells in the feeding site (averaging four to six) nor the area of individual giant cells was influenced by nematode species or isolate. However, the number of nuclei per giant cell was significantly smaller, and the maximum diameters of nuclei and nucleoli were significantly greater, in giant cells induced by M. artiellia isolates than in those induced by M. arenaria, M. incognita, or M. javanica. In a second experiment, M. artiellia-induced giant cells in faba bean and rapeseed also contained a small number of large nuclei. [ABSTRACT FROM AUTHOR]

Plant pathogens often exhibit variation in virulence, the ability to cause disease on host plants with specific resistance, evident from the diversity of races observed within pathogen species. The evolution of races in asexual fungal pathogens has been hypothesized to occur in a stepwise fashion, in which mutations to virulence accumulate sequentially in clonal lineages, resulting in races capable of overcoming multiple host plant resistance genes or multiple resistant cultivars. In this study, we demonstrate a simple stepwise pattern of race evolution in Fusarium oxysporum f. sp. ciceris, the fungus that causes Fusarium wilt of chickpeas. The inferred intraspecific phylogeny of races in this fungus, based on DNA fingerprinting with repetitive sequences, shows that each of the eight races forms a monophyletic lineage. By mapping virulence to each differential cultivar (used for defining races) onto the inferred phylogeny, we show that virulence has been acquired in a simple stepwise pattern, with few parallel gains or losses. Such a clear pattern of stepwise evolution of races, to our knowledge, has not been demonstrated previously for other pathogens based on analyses of field populations. We speculate that in other systems the stepwise pattern is obscured by parallel gains or losses of virulence caused by higher mutation rates and selection by widespread deployment of resistant cultivars. Although chickpea cultivars resistant to Fusarium wilt are available, their deployment has not been extensive and the stepwise acquisition of virulence is still clearly evident. [ABSTRACT FROM AUTHOR]

We used a population genomics approach to test the hypothesis of clonal expansion of a highly fit genotype in populations of Verticillium dahliae. This fungal pathogen has a broad host range and can be dispersed in contaminated seed or other plant material. It has a highly clonal population structure, with several lineages having nearly worldwide distributions in agricultural crops. Isolates in lineage 1A are highly virulent and cause defoliation in cotton, okra, and olive (denoted 1A/D), whereas those in other lineages cause wilting but not defoliation (ND). We tested whether the highly virulent lineage 1A/D could have spread from the southwestern United States to the Mediterranean basin, as predicted from historical records. We found 187 single-nucleotide polymorphisms (SNPs), determined by genotyping by sequencing, among 91 isolates of lineage 1A/D and 5 isolates in the closely related lineage 1B/ND. Neighbor-joining and maximum-likelihood analyses on the 187 SNPs showed a clear divergence between 1A/D and 1B/ND haplotypes. Data for only 77 SNPs were obtained for all 96 isolates (no missing data); lineages 1A/D and 1B/ND differed by 27 of these 77 SNPs, confirming a clear divergence between the two lineages. No evidence of recombination was detected within or between these two lineages. Phylogenetic and genealogical analyses resulted in five distinct subclades of 1A/D isolates that correlated closely with geographic origins in the Mediterranean basin, consistent with the hypothesis that the D pathotype was introduced at least five times in independent founder events into this region from a relatively diverse source population. The inferred ancestral haplotype was found in two isolates sampled before 1983 from the southwestern United States, which is consistent with historical records that 1A/D originated in North America. The five subclades coalesce with the ancestral haplotype at the same time, consistent with a hypothesis of rapid population expansion in the source population during the emergence of 1A/D as a severe pathogen of cotton in the United States. [ABSTRACT FROM AUTHOR]

Verticillium wilts caused by the soilborne fungus Verticillium dahliae are among the most challenging diseases to control. Populations of this pathogen have been traditionally studied by means of vegetative compatibility groups (VCGs) under the assumption that VCGs comprise genetically related isolates that correlate with clonal lineages. We aimed to resolve the phylogenetic relationships among VCGs and their subgroups based on sequences of the intergenic spacer region (IGS) of the ribosomal DNA and six anonymous polymorphic sequences containing singlenucleotide polymorphisms (VdSNPs). A collection of 68 V dahliae isolates representing the main VCGs and subgroups (VCGs 1A, 1B, 2A, 2B, 3, 4A, 4B, and 6) from different geographic origins and hosts was analyzed using the seven DNA regions. Maximum parsimony (MP) phylogenies inferred from IGS and VdSNP sequences showed five and six distinct clades, respectively. Phylogenetic analyses of individual and combined data sets indicated that certain VCG subgroups (e.g., VCGs 1A and 1B) are closely related and share a common ancestor; however, other subgroups (e.g., VCG 4B) are more closely related to members of a different VCG (e.g., VCG 2A) than to subgroups of the same VCG (VCG 4B). Furthermore, MP analyses indicated that VCG 2B is polyphyletic, with isolates placed in at least three distinct phylogenetic lineages based on IGS sequences and two lineages based on VdSNP sequences. Results from our study suggest the existence of main VCG lineages that contain VCGs 1A and 1B; VCGs 2A and 4B; and VCG 4A, for which both phytogenies agree; and the existence of other VCGs or VCG subgroups that seem to be genetically heterogeneous or show discrepancies in their phylogenetic placement: VCG 2B, VCG 3, and VCG 6. These results raise important caveats regarding the interpretation of VCG analyses: genetic homogeneity and close evolutionary relationship between members of a VCG should not be assumed. [ABSTRACT FROM AUTHOR]

Severe downy mildew diseases of opium poppy (Popover somniferum) can be caused by Peronospora arborescens and P cristata, hut differentiating between the two pathogens is difficult because they share morphological features and a similar host range. In Spain, where severe epidemics of downy mildew of opium poppy have occurred recently, the pathogen was identified as P. arborescens on the basis of morphological traits. In this current study, sequence homology and phylogenetic analyses of the internal transcribed spacer regions (ITS) of the ribosomal DNA (rDNA) were carried Out with DNA from P. arborescens and P. cristata from diverse geographic origins, which suggested that only P. arborescens occurs in cultivated Papaver somniferum in Spain. Moreover, analyses of the rDNA ITS region from 27 samples of downy-mildew-affected tissues from all opium-poppy-growing regions in Spain showed that genetic diversity exists within P. arborescens populations in Spain and that these are phylogenetically distinct from P. cristata. P. cristata instead shares a more recent, common ancestor with a range of Peronospora species that includes those found on host plants that are not members of the Papaveraceae. Species-specific primers and a PCR assay protocol were developed that differentiated P. arborescens and P. cristata and proved useful for the detection of P. arborescens in symptomatic and asymptomatic opium poppy plant parts. Use of these primers demonstrated that P. arborescens can be transmitted using seeds and that commercial seed stocks collected from crops with high incidence of the disease were frequently infected. Field experiments conducted in microplots free from P. arborescens using seed stocks harvested from infected capsules further demonstrated that transmission from seedborne P. arborescens to opium poppy plants can occur. Therefore, the specific-PCR detection protocol developed in this study can be of use for epidemiological studies and diagnosing the pathogen in commercial seed stocks; thus facilitating the sanitary control of the disease and avoidance of the pathogen distribution in seeds. [ABSTRACT FROM AUTHOR]

A degree of genetic diversity may exist among Verticillium dahliae isolates within vegetative compatibility groups (VCGs) that bears phytopathological significance and is worth investigating using molecular tools of a higher resolution than VCG characterization. The molecular variability within and among V. dahliae VCGs was studied using 53 artichoke isolates from eastern-central Spain, 96 isolates from cotton, 7 from cotton soil, and 45 from olive trees in countries of the Mediterranean Basin. Isolates were selected to represent the widest available diversity in cotton- and olive-defoliating (D) and -nondefoliating (ND) pathotypes, as well as for VCG. The VCG of 96 cotton and olive isolates was determined in this present study. Molecular variability among V. dahliae isolates was assessed by fluorescent amplified fragment length polymorphism (AFLP) analysis and by polymerase chain reaction (PCR) assays for DNA fragments associated with the D (462 bp) and ND (824 bp) pathotypes, as well as a 334-bp amplicon associated with D pathotype isolates but also present in some VCG2B isolates. Isolates from cotton were in VCGIA, VCGIB, VCG2A, VCG2B, and VCG4B and those from olive trees were in VCG1A, VCG2A, and VC.G4B. Artichoke isolates included representatives of VCGIA, VCG2A, VCG2B (including a newly identified VCG2Ba), and VCG4B. AFLP data were used to generate matrixes of genetic distance among isolates for cluster analysis using the neighbor-joining method and for analysis of molecular variance. Results demonstrated that V. dahliae isolates within a VCG subgroup are molecularly similar, to the extent that clustering of isolates correlated with VCG subgroups regardless of the host source and geographic origin. VCGs differed in molecular variability, with the variability being highest in VCG2B and VCG2A. For some AFLP/VCG subgroup clusterings, V. dahliae isolates from artichoke grouped in subclusters clearly distinct from those comprising isolates from cotton and olive trees. In addition, VCG2B isolates from artichoke formed two distinct clusters that correlated with PCR markers of 334 bp (VCG2B334) or 824 bp (VCG2B824). Artichoke isolates in the VCG2B334/2β334 cluster were molecularly similar to isolates of VCG1A. The molecular difference found among artichoke isolates in VCG2B correlates with virulence of isolates to artichoke and cotton cultivars demonstrated in a previous study. [ABSTRACT FROM AUTHOR]

In the Mediterranean Basin, Fusarium oxysporum f. sp. ciceris and the root-knot nematode Meloidogyne artiellia coinfect chickpea. The influence of root infection (after inoculation with 20 nematode eggs and second-stage juveniles per gram of soil) by two M. artiellia populations, from Italy and Syria, on the reaction of chickpea lines and cultivars with partial resistance to Fusarium wilt (CA 252.10.1.OM, CA 255.2.5.0, CPS 1, and PV 61) and with complete resistance to F. oxysporum f. sp. ciceris race 5 (CA 334.20.4, CA 336.14.3.0, ICC 14216 K, and UC 27) was investigated under controlled conditions. In genotypes with partial resistance, infection by M. artiellia significantly increased the severity of Fusarium wilt, irrespective of the fungal inoculum density (3,000 or 30,000 chlamydospores per gram of soil), except in cultivar CPS 1 at the lower fungal inoculum density. In genotypes with complete resistance to Fusarium wilt, infection by M. artiellia overcame the resistance to F. oxysporum f. sp. ciceris race 5 in CA 334.20.4 and CA 336.14.3.0 but not in ICC 14216 K, irrespective of the fungal inoculum density, and overcame the resistance in UC 27 only at the higher inoculum density. Infection by the nematode significantly increased the number of propagules of F. oxysporum f. sp. ciceris race 5 in root tissues of genotypes with complete resistance to Fusarium wilt, compared with roots that were not inoculated with the nematode, irrespective of the fungal inoculum density, except in ICC 14216 K, in which this effect occurred only at the higher inoculum density. Reproduction of an M. artiellia population from Syria in the absence of F. oxysporum f. sp. ciceris race 5 was significantly higher than that of a population from Italy in all tested chickpea genotypes except ICC 14216 K. However, there was no significant difference between the reproduction rates of the two nematode populations in plants infected with E oxysporum f. sp. ciceris race 5, irrespective of the fungal inoculum density and the reaction of the genotypes to the fungus. [ABSTRACT FROM AUTHOR]

Verticillium wilts caused by the soilborne fungus Verticillium dahliae are among the most challenging diseases to control. Populations of this pathogen have been traditionally studied by means of vegetative compatibility groups (VCGs) under the assumption that VCGs comprise genetically related isolates that correlate with clonal lineages. We aimed to resolve the phylogenetic relationships among VCGs and their subgroups based on sequences of the intergenic spacer region (IGS) of the ribosomal DNA and six anonymous polymorphic sequences containing single-nucleotide polymorphisms (VdSNPs). A collection of 68 V. dahliae isolates representing the main VCGs and subgroups (VCGs 1A, 1B, 2A, 2B, 3, 4A, 4B, and 6) from different geographic origins and hosts was analyzed using the seven DNA regions. Maximum parsimony (MP) phylogenies inferred from IGS and VdSNP sequences showed five and six distinct clades, respectively. Phylogenetic analyses of individual and combined data sets indicated that certain VCG subgroups (e.g., VCGs 1A and 1B) are closely related and share a common ancestor; however, other subgroups (e.g., VCG 4B) are more closely related to members of a different VCG (e.g., VCG 2A) than to subgroups of the same VCG (VCG 4B). Furthermore, MP analyses indicated that VCG 2B is polyphyletic, with isolates placed in at least three distinct phylogenetic lineages based on IGS sequences and two lineages based on VdSNP sequences. Results from our study suggest the existence of main VCG lineages that contain VCGs 1A and 1B; VCGs 2A and 4B; and VCG 4A, for which both phylogenies agree; and the existence of other VCGs or VCG subgroups that seem to be genetically heterogeneous or show discrepancies in their phylogenetic placement: VCG 2B, VCG 3, and VCG 6. These results raise important caveats regarding the interpretation of VCG analyses: genetic homogeneity and close evolutionary relationship between members of a VCG should not be assumed.

Severity of Verticillium wilt in olive trees in Andalusia, southern Spain is associated with the spread of a highly virulent, defoliating (D) Verticillium dahliae pathotype of vegetative compatibility group 1A (VCG1A) but the extent of this spread and the diversity of the pathogen population have never been documented. VCG typing of 637 V. dahliae isolates from 433 trees in 65 orchards from five olive-growing provinces in Andalusia indicated that 78.1% were of VCG1A, 19.8% of VCG2A, 0.6% of VCG2B, 1.4% of VCG4B, and one isolate was heterokaryon self-incompatible. A single VCG prevailed among isolates within most orchards but two and three VCGs were identified in 12 and 3 orchards, respectively, with VCG1A+VCG2A occurring in 10 orchards. VCG1A was the predominant VCG in the three most important olive-growing provinces, and was almost as prevalent as VCG2A in another one. Molecular pathotyping of the 637 isolates using specific polymerase chain reaction assays indicated that VCG1A isolates were of the D pathotype whereas isolates of VCG2A, -2B, and -4B were of the less virulent nondefoliating (ND) pathotype. The pathotype of isolates correlated with the disease syndrome affecting sampled trees. Only three (seq1, seq2, and seq4) of the seven known sequences of the V. dahliae-specific 539- or 523-bp amplicon were identified among the 637 isolates. Distribution and prevalence of VCGs and seq sequences among orchards indicated that genetic diversity within olive V. dahliae in Andalusia is higher in provinces where VCG1A is not prevalent. Log-linear analysis revealed that irrigation management, source of irrigation water, source of planting stock, and cropping history of soil were significantly associated with the prevalence of VCG1A compared with that of VCG2A. Multivariate analyses using a selected set of agricultural factors as variables allowed development of a discriminant model for predicting the occurrence of D and ND pathotypes in the area of the study. Blind tests using this model correctly indentified the V. dahliae pathotype occurring in an orchard. The widespread occurrence and high prevalence of VCG1A/D pathotype in Andalusia have strong implications for the management of the disease.

ABSTRACT Races 0 (Foc-0) and 5 (Foc-5) of Fusarium oxysporum f. sp. ciceris differ in virulence and induce yellowing or wilting syndrome, respectively, in chickpea. We modeled the combined effects of soil temperature and inoculum density of Foc-0 and Foc-5 on disease developed in chickpea cvs. P-2245 and PV-61 differing in susceptibility to those races, using quantitative nonlinear models. Disease development over time in the temperature range of 10 to 30 degrees C and inoculum densities between 6 and 8,000 chlamydospores g(1) of soil was described by the Weibull function. Four response variables (the reciprocal incubation period, the final disease intensity, the standardized area under the disease progress curve, and the intrinsic rate of disease development) characterized the disease development. Response surface models that expressed the combined effect of inoculum density and temperature were developed by substituting the intrinsic rate of disease development in the Weibull or exponential functions with a beta function describing the relationship of response variables to temperature. The models estimated 22 to 26 degrees C as the most favorable soil temperature for infection of cvs. P-2245 and PV-61 by Foc-5, and 24 to 28 degrees C for infection of cv. P-2245 by Foc-0. At 10 degrees C, no disease developed except in cv. P-2245 inoculated with Foc-5. At optimum soil temperature, maximum disease intensity developed with Foc-5 and Foc-0 at 6 and 50 chlamydospores g(1) of soil respectively, in cv. P-2245, and with Foc-5 at 1,000 chlamydospores g(1) of soil in cv. PV-61. The models were used to construct risk threshold charts that can be used to estimate the potential risk of Fusarium wilt epidemics in a geographical area based on soil temperature, the race and inoculum density in soil, and the level of susceptibility of the chickpea cultivar.

ABSTRACT A 3-year experiment was conducted in field microplots infested with Fusarium oxysporum f. sp. ciceris race 5 at Córdoba, Spain, in order to assess efficacy of an integrated management strategy for Fusarium wilt of chickpea that combined the choice of sowing date, use of partially resistant chickpea genotypes, and seed and soil treatments with biocontrol agents Bacillus megaterium RGAF 51, B. subtilis GB03, nonpathogenic F. oxysporum Fo 90105, and Pseudomonas fluorescens RG 26. Advancing the sowing date from early spring to winter significantly delayed disease onset, reduced the final disease intensity (amount of disease in a microplot that combines disease incidence and severity, expressed as a percentage of the maximum possible amount of disease in that microplot), and increased chickpea seed yield. A significant linear relationship was found between disease development over time and weather variables at the experimental site, with epidemics developing earlier and faster as mean temperature increased and accumulated rainfall decreased. Under conditions highly conducive for Fusarium wilt development, the degree of disease control depended primarily on choice of sowing date, and to a lesser extent on level of resistance of chickpea genotypes to F. oxysporum f. sp. ciceris race 5, and the biocontrol treatments. The main effects of sowing date, partially resistant genotypes, and biocontrol agents were a reduction in the rate of epidemic development over time, a reduction of disease intensity, and an increase in chickpea seedling emergence, respectively. Chickpea seed yield was influenced by all three factors in the study. The increase in chickpea seed yield was the most consistent effect of the biocontrol agents. However, that effect was primarily influenced by sowing date, which also determined disease development. Effectiveness of biocontrol treatments in disease management was lowest in January sowings, which were least favorable for Fusarium wilt. Sowing in February, which was moderately favorable for wilt development, resulted in the greatest increase in seed yield by the biocontrol agents. In March sowings, which were most conducive for the disease, the biocontrol agents delayed disease onset and increased seedling emergence. B. subtilis GB03 and P. fluorescens RG 26, applied either alone or each in combination with nonpathogenic F. oxysporum Fo 90105, were the most effective treatments at suppressing Fusarium wilt, or delaying disease onset and increasing seed yield, respectively. The importance of integrating existing control practices, partially effective by themselves, with other control measures to achieve appropriate management of Fusarium wilt and increase of seed yield in chickpea in Mediterranean-type environments is demonstrated by the results of this study.

ABSTRACT Specific primers and polymerase chain reaction (PCR) assays that identify Fusarium oxysporum f. sp. ciceris and each of the F. oxysporum f. sp. ciceris pathogenic races 0, 1A, 5, and 6 were developed. F. oxysporum f. sp. ciceris- and race-specific random amplified polymorphic DNA (RAPD) markers identified in a previous study were cloned and sequenced, and sequence characterized amplified region (SCAR) primers for specific PCR were developed. Each cloned RAPD marker was characterized by Southern hybridization analysis of Eco RI-digested genomic DNA of a subset of F. oxysporum f. sp. ciceris and nonpathogenic F. oxysporum isolates. All except two cloned RAPD markers consisted of DNA sequences that were found highly repetitive in the genome of all F. oxysporum f. sp. ciceris races. F. oxysporum f. sp. ciceris isolates representing eight reported races from a wide geographic range, nonpathogenic F. oxysporum isolates, isolates of F. oxysporum f. spp. lycopersici, melonis, niveum, phaseoli, and pisi, and isolates of 47 different Fusarium spp. were tested using the SCAR markers developed. The specific primer pairs amplified a single 1,503-bp product from all F. oxysporum f. sp. ciceris isolates; and single 900- and 1,000-bp products were selectively amplified from race 0 and race 6 isolates, respectively. The specificity of these amplifications was confirmed by hybridization analysis of the PCR products. A race 5-specific identification assay was developed using a touchdown-PCR procedure. A joint use of race 0- and race 6-specific SCAR primers in a single-PCR reaction together with a PCR assay using the race 6-specific primer pair correctly identified race 1A isolates for which no RAPD marker had been found previously. All the PCR assays described herein detected up to 0.1 ng of fungal genomic DNA. The specific SCAR primers and PCR assays developed in this study clearly identify and differentiate isolates of F. oxysporum f. sp. ciceris and of each of its pathogenic races 0, 1A, 5, and 6.