Haplotype divergence and multiple candidate genes at Rphq2, a partial resistance QTL of barley to Puccinia hordei.

KEY MESSAGE: Rphq2, a minor gene for partial resistance to Puccinia hordei , was physically mapped in a 188 kbp introgression with suppressed recombination between haplotypes of rphq2 and Rphq2 barley cultivars. ABSTRACT: Partial and non-host resistances to rust fungi in barley (Hordeum vulgare) may be based on pathogen-associated molecular pattern (PAMP)-triggered immunity. Understanding partial resistance may help to understand non-host resistance, and vice versa. We constructed two non-gridded BAC libraries from cultivar Vada and line SusPtrit. Vada is immune to non-adapted Puccinia rust fungi, and partially resistant to P. hordei. SusPtrit is susceptible to several non-adapted rust fungi, and has been used for mapping QTLs for non-host and partial resistance. The BAC libraries help to identify genes determining the natural variation for partial and non-host resistances of barley to rust fungi. A major-effect QTL, Rphq2, for partial resistance to P. hordei was mapped in a complete Vada and an incomplete SusPtrit contig. The physical distance between the markers flanking Rphq2 was 195 Kbp in Vada and at least 226 Kbp in SusPtrit. This marker interval was predicted to contain 12 genes in either accession, of which only five genes were in common. The haplotypes represented by Vada and SusPtrit were found in 57 and 43%, respectively, of a 194 barley accessions panel. The lack of homology between the two haplotypes probably explains the suppression of recombination in the Rphq2 area and limit further genetic resolution in fine mapping. The possible candidate genes for Rphq2 encode peroxidases, kinases and a member of seven-in-absentia protein family. This result suggests that Rphq2 does not belong to the NB-LRR gene family and does not resemble any of the partial resistance genes cloned previously.

Isolate Specificity and Polygenic Inheritance of Resistance in Barley to the Heterologous Rust Pathogen Puccinia graminis f. sp. avenae.

Barley is a near-nonhost to numerous heterologous (nonadapted) rust pathogens because a small proportion of genotypes are somewhat susceptible. We assessed 66 barley accessions and three mapping populations (Vada × SusPtrit, Cebada Capa × SusPtrit, and SusPtrit × Golden Promise) for response to three Swedish oat stem rust (Puccinia graminis f. sp. avenae) fungal isolates and determined that barley is a near-nonhost to P. graminis f. sp. avenae and that resistance was polygenically inherited. The parental genotypes Vada and Golden Promise were immune to all three isolates, whereas Cebada Capa was immune to two isolates and moderately resistant to the third. Phenotypic data from the Vada × SusPtrit mapping population and the barley accessions tested also demonstrated isolate-specific resistance. In particular, the SusPtrit parent and several other accessions allowed sporulation by isolate Ingeberga but were resistant to isolate Evertsholm. Nine chromosomal regions carried quantitative trait loci (QTL) (Rpgaq1 to Rpgaq9) of varying effect, most of which colocated to previously identified QTL for resistance to other heterologous rust pathogens. Rpgaq1 on chromosome 1H (Vada and Golden Promise) was effective toward all isolates tested. Microscopic examination indicated that resistance was prehaustorial in Vada whereas, in SusPtrit, both pre- and posthaustorial mechanisms play a role.

Golden SusPtrit: a genetically well transformable barley line for studies on the resistance to rust fungi.

KEY MESSAGE: We developed 'Golden SusPtrit', i.e., a barley line combining SusPtrit's high susceptibility to non-adapted rust fungi with the high amenability of Golden Promise for transformation. Nonhost and partial resistance to Puccinia rust fungi in barley are polygenically inherited. These types of resistance are principally prehaustorial, show high diversity between accessions of the plant species and are genetically associated. To study nonhost and partial resistance, as well as their association, candidate gene(s) for resistance must be cloned and tested in susceptible material where SusPtrit would be the line of choice. Unfortunately, SusPtrit is not amenable to Agrobacterium-mediated transformation. Therefore, a doubled haploid (DH) mapping population (n = 122) was created by crossing SusPtrit with Golden Promise to develop a 'Golden SusPtrit', i.e., a barley line combining SusPtrit's high susceptibility to non-adapted rust fungi with the high amenability of Golden Promise for transformation. We identified nine genomic regions occupied by resistance quantitative trait loci (QTLs) against four non-adapted rust fungi and P. hordei isolate 1.2.1 (Ph.1.2.1). Four DHs were selected for an Agrobacterium-mediated transformation efficiency test. They were among the 12 DH lines most susceptible to the tested non-adapted rust fungi. The most efficiently transformed DH line was SG062N (11-17 transformants per 100 immature embryos). The level of non-adapted rust infection on SG062N is either similar to or higher than the level of infection on SusPtrit. Against Ph.1.2.1, the latency period conferred by SG062N is as short as that conferred by SusPtrit. SG062N, designated 'Golden SusPtrit', will be a valuable experimental line that could replace SusPtrit in nonhost and partial resistance studies, especially for stable transformation using candidate genes that may be involved in rust-resistance mechanisms.

First Report of Powdery Mildew (Oidium sp.) on Pincushion Flower (Scabiosa columbaria) in New York.

Scabiosa columbaria (Dipsacaceae) is a popular perennial ornamental in the United States. It is native to Europe and was introduced to North America by nursery trade only recently. In the spring of 2006, symptoms of powdery mildew infection were observed on overwintered plants of S. columbaria cv. Butterfly Blue in a nursery in Cutchogue, NY. White powdery mildew mycelia with abundant sporulation were observed on upper and lower leaf surfaces. The portions of leaves with powdery mildew colonies often showed purplish discoloration. Conidia were cylindric to doliiform, measured 20 to 33 × 10 to 15 µm, and were produced singly on 60 to 130 µm long conidiophores consisting of a foot-cell measuring 20 to 50 × 6 to 10 µm, followed by one to three, 12 to 40 µm long cells. Hyphal appressoria were lobed or multilobed. The teleomorph stage was not found. On the basis of these characteristics, the pathogen was identified as an Oidium sp. belonging to the subgenus Pseudoidium. Recently, an anamorphic powdery mildew fungus with similar morphological characteristics, identified as Erysiphe knautiae, was reported on S. columbaria cv. Butterfly Blue in Washington (2). E. knautiae is a common powdery mildew species of dipsacaceous plants such as Scabiosa spp. and Knautia spp. in Europe and Asia (1). To determine whether the fungus reported here was E. knautiae, DNA was extracted from its mycelium, and the internal transcribed spacer (ITS) region of the ribosomal DNA was amplified and sequenced as described earlier (4). No ITS sequences are available in public DNA databases for E. knautiae, thus, we determined this sequence in a specimen of E. knautiae collected from Knautia arvensis in The Netherlands. Herbarium specimens of the Oidium sp. infecting S. columbaria in New York and E. knautiae from the Netherlands were deposited at the U.S. National Fungus Collections under accession numbers BPI 878259 and BPI 878258, respectively. The ITS sequence from Oidium sp. infecting S. columbaria in New York (GenBank Accession No. EU377474) differed in two nucleotides from that of E. knautiae infecting K. arvensis in the Netherlands (GenBank Accession No. EU377475). These two ITS sequences were also more than 99% similar to those of some newly emerged anamorphic powdery mildew fungi: Oidium neolycopersici and other Oidium spp. infecting Chelidonium majus, Passiflora caerulea, and some crassulaceous plants (3,4). Thus, it is unclear whether the fungus reported here was E. knautiae known from Eurasia or an Oidium sp. that has acquired pathogenicity to S. columbaria. To our knowledge, this is the first report of powdery mildew on S. columbaria in New York. References: (1) U. Braun. Beih. Nova Hedwigia 89:1, 1987. (2) D. A. Glawe and G. G. Grove. Online publication. doi:10.1094/PHP-2005-1024-01-BR. Plant Health Progress, 2005. (3) B. Henricot. Plant Pathol. 57:779, 2008. (4) T. Jankovics et al. Phytopathology 98:529, 2008.

Oidium neolycopersici: intraspecific variability inferred from amplified fragment length polymorphism analysis and relationship with closely related powdery mildew fungi infecting various plant species.

Previous works indicated a considerable variation in the pathogenicity, virulence, and host range of Oidium neolycopersici isolates causing tomato powdery mildew epidemics in many parts of the world. In this study, rDNA internal transcribed spacer (ITS) sequences, and amplified fragment length polymorphism (AFLP) patterns were analyzed in 17 O. neolycopersici samples collected in Europe, North America, and Japan, including those which overcame some of the tomato major resistance genes. The ITS sequences were identical in all 10 samples tested and were also identical to ITS sequences of eight previously studied O. neolycopersici specimens. The AFLP analysis revealed a high genetic diversity in O. neolycopersici and indicated that all 17 samples represented different genotypes. This might suggest the existence of either a yet unrevealed sexual reproduction or other genetic mechanisms that maintain a high genetic variability in O. neolycopersici. No clear correlation was found between the virulence and the AFLP patterns of the O. neolycopersici isolates studied. The relationship between O. neolycopersici and powdery mildew anamorphs infecting Aquilegia vulgaris, Chelidonium majus, Passiflora caerulea, and Sedum alboroseum was also investigated. These anamorphs are morphologically indistinguishable from and phylogenetically closely related to O. neolycopersici. The cross-inoculation tests and the analyses of ITS sequences and AFLP patterns jointly indicated that the powdery mildew anamorphs collected from the above mentioned plant species all represent distinct, but closely related species according to the phylogenetic species recognition. All these species were pathogenic only to their original host plant species, except O. neolycopersici which infected S. alboroseum, tobacco, petunia, and Arabidopsis thaliana, in addition to tomato, in cross-inoculation tests. This is the first genome-wide study that investigates the relationships among powdery mildews that are closely related based on ITS sequences and morphology. The results indicate that morphologically indistinguishable powdery mildews that differed in only one to five single nucleotide positions in their ITS region are to be considered as different taxa with distinct host ranges.

High-resolution mapping of genes involved in plant stage-specific partial resistance of barley to leaf rust.

Partial resistance quantitative trait loci (QTLs) Rphq11 and rphq16 against Puccinia hordei isolate 1.2.1 were previously mapped in seedlings of the mapping populations Steptoe/Morex and Oregon Wolfe Barleys, respectively. In this study, QTL mapping was performed at adult plant stage for the two mapping populations challenged with the same rust isolate. The results suggest that Rphq11 and rphq16 are effective only at seedling stage, and not at adult plant stage. The cloning of several genes responsible for partial resistance of barley to P. hordei will allow elucidation of the molecular basis of this type of plant defence. A map-based cloning approach requires to fine-map the QTL in a narrow genetic window. In this study, Rphq11 and rphq16 were fine-mapped using an approach aiming at speeding up the development of plant material and simplifying its evaluation. The plant materials for fine-mapping were identified from early plant materials developed to produce QTL-NILs. The material was first selected to carry the targeted QTL in heterozygous condition and susceptibility alleles at other resistance QTLs in homozygous condition. This strategy took four to five generations to obtain fixed QTL recombinants (i.e., homozygous resistant at the Rphq11 or rphq16 QTL alleles, homozygous susceptible at the non-targeted QTL alleles). In less than 2 years, Rphq11 was fine-mapped into a 0.2-cM genetic interval and a 1.4-cM genetic interval for rphq16. The strongest candidate gene for Rphq11 is a phospholipid hydroperoxide glutathione peroxidase. Thus far, no candidate gene was identified for rphq16.

Infection Structures of Host-Specialized Isolates of Uromyces viciae-fabae and of Other Species of Uromyces Infecting Leguminous Crops.

A study was made of the morphology of urediniospores and primary infection structures of 12 isolates of six legume-infecting species of Uromyces. Infection structures were sufficient to distinguish among species. Isolates of Uromyces viciae-fabae proved to be specialized with respect to host, because each isolate infected only cultivars of the species from which it was collected. Host-specialized isolates of U. viciae-fabae also were morphologically distinct, differing in both spore dimensions and infection structure morphology. In particular, the shape and dimensions of the substomatal vesicle were distinctive. These results support the view that U. viciae-fabae sensu lato is a species complex.

Specificity of Prehaustorial Resistance to Puccinia hordei and to Two Inappropriate Rust Fungi in Barley.

ABSTRACT To elucidate the specificity of prehaustorial resistance to inappropriate rust fungi, we studied two populations of recombinant inbred lines of barley that segregated for partial resistance (PR) to Puccinia hordei and for the resistance to the inappropriate rust species P. recondita f. sp. tritici and P. hordei-murini. PR to P. hordei is prehaustorial and nonhypersensitive, and its level can be assessed accurately by measuring the latent period of the fungus. The resistance to the inappropriate rust species is a combination of prehaustorial (nonhypersensitive) and posthaustorial (hypersensitive) mechanisms. The amount of nonhypersensitive, early abortion of P. recondita f. sp. tritici and P. hordei-murini sporelings reflects the degree of prehaustorial defense to the two inappropriate rust species. All lines showing a long latent period of P. hordei also had a relatively high level of early abortion of the growth of P. recondita f. sp. tritici and P. hordei-murini. This indicates that genes for PR to P. hordei are also effective against these two inappropriate rust species. The reverse was not necessarily true; some lines showing a high level of early abortion of P. recondita f. sp. tritici and P. hordei-murini had a low level of PR to P. hordei. Moreover, lines with a similar level of prehaustorial resistance to P. recondita f. sp. tritici could differ considerably in their prehaustorial resistance to P. hordei-murini. This indicates that genes for prehaustorial resistance may exhibit rust species specificity.

A high-density consensus map of barley to compare the distribution of QTLs for partial resistance to Puccinia hordei and of defence gene homologues.

A consensus map of barley was constructed based on three reference doubled haploid (DH) populations and three recombinant inbred line (RIL) populations. Several sets of microsatellites were used as bridge markers in the integration of those populations previously genotyped with RFLP or with AFLP markers. Another set of 61 genic microsatellites was mapped for the first time using a newly developed fluorescent labelling strategy, referred to as A/T labelling. The final map contains 3,258 markers spanning 1,081 centiMorgans (cM) with an average distance between two adjacent loci of 0.33 cM. This is the highest density of markers reported for a barley genetic map to date. The consensus map was divided into 210 BINs of about 5 cM each in which were placed 19 quantitative trait loci (QTL) contributing to the partial resistance to barley leaf rust (Puccinia hordei Otth) in five of the integrated populations. Each parental barley combination segregated for different sets of QTLs, with only few QTLs shared by any pair of cultivars. Defence gene homologues (DGH) were identified by tBlastx homology to known genes involved in the defence of plants against microbial pathogens. Sixty-three DGHs were located into the 210 BINs in order to identify candidate genes responsible for the QTL effects. Eight BINs were co-occupied by a QTL and DGH(s). The positional candidates identified are receptor-like kinase, WIR1 homologues and several defence response genes like peroxidases, superoxide dismutase and thaumatin.

A high density barley microsatellite consensus map with 775 SSR loci.

A microsatellite or simple sequence repeat (SSR) consensus map of barley was constructed by joining six independent genetic maps based on the mapping populations 'Igri x Franka', 'Steptoe x Morex', 'OWB(Rec) x OWB(Dom)', 'Lina x Canada Park', 'L94 x Vada' and 'SusPtrit x Vada'. Segregation data for microsatellite markers from different research groups including SCRI (Bmac, Bmag, EBmac, EBmag, HVGeneName, scsssr), IPK (GBM, GBMS), WUR (GBM), Virginia Polytechnic Institute (HVM), and MPI for Plant Breeding (HVGeneName), generated in above mapping populations, were used in the computer program RECORD to order the markers of the individual linkage data sets. Subsequently, a framework map was constructed for each chromosome by integrating the 496 "bridge markers" common to two or more individual maps with the help of the computer programme JoinMap 3.0. The final map was calculated by following a "neighbours" map approach. The integrated map contained 775 unique microsatellite loci, from 688 primer pairs, ranging from 93 (6H) to 132 (2H) and with an average of 111 markers per linkage group. The genomic DNA-derived SSR marker loci had a higher polymorphism information content value (average 0.61) as compared to the EST/gene-derived SSR loci (average 0.48). The consensus map spans 1,068 cM providing an average density of one SSR marker every 1.38 cM. Such a high-density consensus SSR map provides barley molecular breeding programmes with a better choice regarding the quality of markers and a higher probability of polymorphic markers in an important chromosomal interval. This map also offers the possibilities of thorough alignment for the (future) physical map and implementation in haplotype diversity studies of barley.

A novel major gene on chromosome 6H for resistance of barley against the barley yellow dwarf virus.

In a mapping population derived from the Ethiopian barley line L94 x Vada, natural infection by barley yellow dwarf virus (BYDV) occurred. While line L94 hardly showed symptoms, Vada was severely affected. The 103 recombinant inbred lines segregated bimodally. The major gene responsible for this resistance mapped to chromosome 6H. We propose to name the locus Ryd3. A subset of recombinant inbred lines, L94, and Vada were planted in a subsequent field test which confirmed the previous field observations. Double antibody sandwich enzyme-linked immunosorbent assays (DAS-ELISA) indicated that the epidemic was due to a combination of the serotypes BYDV-PAV and BYDV-MAV. In the accessions with the least BYDV symptoms no virus was detected, justifying the consideration of the gene as conferring true resistance rather than tolerance to these viruses. In a laboratory/gauze house trial a near-isogenic line carrying the Vada chromosome 6H fragment in an L94 background was affected as much as Vada. The effect of Ryd3 was quantified, and compared with that of the only other known major gene for resistance to BYDV, Ryd2, which is also of Ethiopian origin and is located on chromosome 3H. Both genes seemed to reduce the chance of the viral isolate used in this study to establish infection. In plants in which it became established, the virus concentration reached a similar level as in susceptible accessions, but with less dramatic symptom development. Inoculated plants in which the virus failed to multiply tended to show an increase in the number of ears per plant, resulting in higher grain yield per plant. Ryd3 co-segregates with several PCR-based molecular markers that may serve for marker assisted selection.

Partial resistance to leaf rust in a collection of ancient Spanish barleys.

A collection of 569 Spanish barley accessions was screened for resistance to leaf rust (Puccinia hordei Otth) in the field at Córdoba during the 2000-2001 season. The level of resistance ranged from very low to very high. In 14% of the accessions the relative AUDPC (L94 = 100%) was lower than 10%. Selected accessions that were most resistant in the field, were tested in the seedling stage under controlled conditions. Macroscopic components of resistance indicated that six lines had a high level of partial resistance close to check cv. Vada and one line a similar level of partial resistance. Histological studies indicated that the resistance was based on a high percentage of early aborted colonies and reduction in colony size without plant cell necrosis. Three of the selected lines showed high percentage of plant cell necrosis associated with established colonies, which indicates a combination of prehaustorial resistance with late acting incomplete posthaustorial resistance. Although the new barley varieties already incorporate some partial resistance, new sources of partial resistance like these are needed to improve durability of the resistance.

Genomic regions determining resistance to leaf stripe (Pyrenophora graminea) in barley.

Leaf stripe is a seed-borne disease of barley (Hordeum vulgare) caused by Pyrenophora graminea. Little is known about the genetics of resistance to this pathogen. In the present work, QTL analysis was applied on two recombinant inbred line (RIL) populations derived from two- and six-rowed barley genotypes with different levels of partial resistance to barley leaf stripe. Quantitative trait loci for partial resistance were identified using the composite interval mapping (CIM) method of PLABQTL software, using the putative QTL markers as cofactors. In the L94 x 'Vada' mapping population, one QTL for resistance was detected on chromosome 2H; the same location as the leaf-stripe resistance gene Rdg1 mapped earlier in 'Alf', where it confers complete resistance to the pathogen. An additional minor-effect QTL was identified by further analyses in this segregating population on chromosome 7H. In L94 x C123, two QTLs for resistance were mapped, one each on chromosomes 7H and 2H.

Characterization of Lr46, a gene conferring partial resistance to wheat leaf rust.

Components of resistance conferred by the Lr46 gene, reported as causing "slow rusting" resistance to leaf rust in wheat, were studied and compared with the effects of Lr34 and genes for quantitative resistance in cv. Akabozu. Lr34 is a gene that confers non-hypersensitive type of resistance. The effect of Lr46 resembles that of Lr34 and other wheats reported with partial resistance. At macroscopic level, Lr46 produced a longer latency period than observed on the susceptible recurrent parent Lalbahadur, and a reduction of the infection frequency not associated with hypersensitivity. Microscopically, Lr46 increased the percentage of early aborted infection units not associated with host cell necrosis and decreased the colony size. The effect of Lr46 is comparable to that of Lr34 in adult plant stage, but in seedling stage its effect is weaker than that of Lr34.

Prospects for exploitation of disease resistance from Hordeum chilense in cultivated cereals.

Hordeum chilense is a South American wild barley with high potential for cereal breeding given its high crossability with other members of the Triticeae. In the present paper we consider the resistance of H. chilense to several fungal diseases and the prospects for its transference to cultivated cereals. All H. chilense accessions studied are resistant to the barley, wheat and rye brown rusts, the powdery mildews of wheat, barley, rye and oat, to Septoria leaf blotch, common bunt and to loose smuts, which suggests that H. chilense is a non-host of these diseases. There are also lines resistant to wheat and barley yellow rust, stem rust and to Agropyron leaf rust, as well as lines giving moderate levels of resistance to Septoria glume blotch, tan spot and Fusarium head blight. Some H. chilense lines display pre-appressorial avoidance to brown rust. Lines differ in the degree of haustorium formation by rust and mildew fungi they permit, and in the degree to which a hypersensitive response occurs after haustoria are formed. Unfortunately, resistance of H. chilense to rust fungi is not expressed in tritordeum hybrids, nor in chromosome addition lines in wheat. In tritordeum, H. chilense contributes quantitative resistance to wheat powdery mildew, tan spot and loose smut. The resistance to mildew, expressed as a reduced disease severity, is not associated with macroscopically visible necrosis. Hexaploid tritordeums are immune to Septoria leaf blotch and to common bunt although resistance to both is slightly diluted in octoploid tritordeums. Studies with addition lines in wheat indicate that the resistance of H. chilense to powdery mildew, Septoria leaf blotch and common bunt is of broad genetic basis, conferred by genes present on various chromosomes.

Search for partial resistance to leaf rust in a collection of ancient Spanish wheats.

A collection of 917 accessions of Spanish durum and bread wheat was screened for resistance to leaf rust (Puccinia triticina) under field conditions at three locations. Resistance levels ranged from very low to very high, high susceptibility being most frequent. Relative disease severity (referred to the most susceptible accession = 100%) was lower than 20% in about 6% of the accessions in each location. In the collection most of the lines (84%) displayed a susceptible infection type. A final selection of seven accessions (one of them durum) displaying low severity level in the field and high infection type in a growth chamber was chosen for further studies. High levels of partial resistant with longer latency period and high percentage of early aborted colonies without necrosis were found. They might be used in breeding programmes.

Abnormal germling development by brown rust and powdery mildew on cer barley mutants.

The barley leaf rust fungus forms appressoria over host leaf stomata and penetrates via the stomatal pore. High levels of avoidance to leaf rust fungi have been described in some wild accessions of Hordeum species where a prominent wax layer on the stomata inhibits triggering of fungal appressorium differentiation. Leaf rust avoidance has not yet been found in H. vulgare. Since cuticular leaf waxes are implicated in the avoidance trait, we screened 27 eceriferum (cer) mutant lines of H. vulgare for avoidance to barley leaf rust. These mutations affect leaf waxes. Reduction in numbers of germ tubes forming appressoria over stomata was found in some lines, but the greatest reduction (ca 30%) was less than previously found in wild barley spp. or in an accession of H. chilense used here as a check. In one line (cer-zh654), avoidance was due to a combination of factors. Firstly, fewer germ tubes oriented towards stomata and so failed to contact them. Secondly, some germ tubes that encountered stomata did not form appressoria but over-grew them. In this line, therefore, the fungus tended to fail both to locate and to respond to stomata. The appressoria of barley powdery mildew form on leaf epidermal cells that they penetrate directly. On certain cer lines, a proportion of germlings of the barley powdery mildew fungus developed abnormally, suggesting that germlings failed to recognise and/or respond to the leaf surface waxes on these mutants.

QTL mapping provides evidence for lack of association of the avoidance of leaf rust in Hordeum chilense with stomata density.

In cereals, rust fungi are among the most harmful pathogens. Breeders usually rely on short-lived hypersensitivity resistance. As an alternative, "avoidance" may be a more durable defence mechanism to protect plants to rust fungi. In Hordeum chilense avoidance is based on extensive wax covering of stomata, which interferes with the induction of appressorium formation by the rust fungi. High avoidance levels are associated with a higher stoma density on the abaxial leaf epidermis. The avoidance level was assessed as the percentage of germ tube/stoma encounters that did not result in appressorium differentiation by Puccinia hordei, the barley leaf rust fungus. One hundred F(2) individuals from the cross between two H. chilense accessions with contrasting levels of avoidance showed a continuous distribution for avoidance of the rust fungus and for stoma density, indicating quantitative inheritance of the traits. No significant correlation was found between avoidance and stoma density in the segregating F(2) population. In order to map quantitative trait loci (QTLs) for both traits, an improved molecular marker linkage map was constructed, based on the F(2) population. The resulting linkage map spanned 620 cM and featured a total of 437 AFLP markers, thirteen RFLPs, four SCARs, nine SSRs, one STS and two seed storage protein markers. It consisted of seven long and two shorter linkage groups, and was estimated to cover 81% of the H. chilense genome. Restricted multiple interval mapping identified two QTLs for avoidance and three QTLs for stoma density in the abaxial leaf surface. The QTLs for avoidance were mapped on chromosome 3 and 5; those for stoma density on chromosomes 1, 3 and 7. Only the two QTLs regions located on chromosome 3 (one for avoidance and the other for stoma density) overlapped. The wild barley H. chilense has a high crossability with other members of the Triticeae tribe. The knowledge on the location of the QTLs responsible for the avoidance trait is a prerequisite to transfer this favourable agronomic trait from H. chilense to cultivated cereal genomes.