Comparison of two indices for evaluating regeneration ability in rice (Oryza sativa L.) through a diallel analysis

 A full diallel analysis was performed among seven rice cultivars, all of which showed different abilities of regeneration from seed-derived calli. Number of regenerated shoots and regeneration frequency were used as indices of regeneration ability. In both cases, additive effects were significant at the 0.1% level, and dominant genes had a positive effect, that is, they increased regeneration ability. Non-allelic interaction (epistasis) and maternal effects were not detected. Dominance effects were significant at the 1% and the 0.1% level when the number of regenerated shoots and regeneration frequency were used as indices, respectively. Average degree of dominance was 0.531 for the shoot regeneration index and 0.990 for the regeneration frequency index. Since broad-sense heritability was 0.919 for number of regenerated shoots and 0.736 for regeneration frequency, the former was considered to be a better index of regeneration ability than the latter. Received: 10 June 1996 / Accepted: 23 August 1996     

PCR-based molecular markers for the fragrance gene in rice (Oryza sativa. L.)

The genomic DNA clone RG28, linked to the major fragrance gene of rice (fgr), was assessed for polymorphism in order to produce a PCR-based marker for fragrance. A small mono-nucleotide repeat, that was polymorphic between a pair of fragrant and non-fragrant cultivars, was identified and developed into a co-dominant PCR-based marker. The polymorphism-information-content determinations for three microsatellite markers, that have been genetically mapped near RG28, are also presented. These PCR-based markers will be highly useful in distinguishing fragrance-producing alleles from non-fragrance-producing alleles at the fgr locus. Received: 19 October 1999 / Accepted: 16 December 1999     

A novel plant nuclear gene encoding chloroplast ribosomal protein S9 has a transit peptide related to that of rice chloroplast ribosomal protein L12

We have cloned a novel nuclear gene for a ribosomal protein of rice and Arabidopsis that is like the bacterial ribosomal protein S9. To determine the subcellular localization of the gene product, we fused the N-terminal region and green fluorescent protein and expressed it transiently in rice seedlings. Localized fluorescence was detectable only in chloroplasts, indicating that this nuclear gene encodes chloroplast ribosomal protein S9. The N-terminal region of rice ribosomal protein S9 was found to have a high sequence similarity to the transit peptide region of the rice chloroplast ribosomal protein L12, suggesting that these transit peptides have a common lineage.     

Mapping of the Rf-3 nuclear fertility-restoring gene for WA cytoplasmic male sterility in rice using RAPD and RFLP markers

The cytoplasmic male sterility (CMS) of wild-abortive (WA) cytoplasm has been widely used for breeding hybrid rice. Two restorer genes for the CMS have been found by traditional genetic analysis. To tag the restorer genes we used a set of near-isogenic lines (NILs) of Zhenshan 97 carrying different genotypes for fertility restoration from IR24, to perform RAPD analysis. From the survey of 720 random primers, six RAPD markers were identified to be associated with Rf-3. Three of these OPK05-800, OPU10-1100 and OPW01-350, were mapped on chromosome 1. Two populations from the crosses between Zhenshan 97 A and a near-isogenic restorer line ZSR21 and between Zhenshan 97 A and IR24 were used for mapping Rf-3. The three RAPD markers and three RFLP markers, RG532, RG140 and RG458, were found to be closely linked to Rf-3 in the two populations. The same location of Rf-3 was also found in a population from the cross of IR58025 A//IR36/IR58025 B. At the RG532 locus, different alleles were found between two CMS lines, Zhenshan 97 A and IR58025 A, and between two restorer lines, IR24 and IR36. The use of these molecular markers closely linked to Rf-3 in facilitating the development of hybrid rice is discussed. Received: 3 January 1996 / Accepted: 17 May 1996     

Mapping QTLs for phosphorus deficiency tolerance in rice (Oryza sativa L.)

 The amplified fragment length polymorphism (AFLP) technique combined with selective genotyping was used to map quantitative trait loci (QTLs) associated with tolerance for phosphorus (P) deficiency in rice. P deficiency tolerant cultivar IR20 was crossed to IR55178-3B-9-3 (sensitive to P-deficiency) and 285 recombinant inbred lines (RILs) were produced by single-seed descent. The RILs were phenotyped for the trait by growing them in P-sufficient (10.0 mg/l) and P-deficient (0.5 mg/l) nutrient solution and determining their relative tillering ability at 28 days after seeding, and relative shoot dry weight and relative root dry weight at 42 days after seeding. Forty two of each of the extreme RILs (sensitive and tolerant) and the parents were subjected to AFLP analysis. A map consisting of 217 AFLP markers was constructed. Its length was 1371.8 cM with an average interval size of 7.62 cM. To assign linkage groups to chromosomes, 30 AFLP and 26 RFLP markers distributed over the 12 chromosomes were employed as anchor markers. Based on the constructed map, a major QTL for P-deficiency tolerance, designated PHO, was located on chromosome 12 and confirmed by RFLP markers RG9 and RG241 on the same chromosome. Several minor QTLs were mapped on chromosomes 1, 6, and 9. Received: 21 April 1998 / Accepted: 9 June 1998     

RFLP mapping of the genes controlling hybrid breakdown in rice (Oryza sativa L.)

 Complementary recessive genes hwd1 and hwd2 controlling hybrid breakdown (weakness of F₂ and later generations) were mapped in rice using RFLP markers. These genes produce a plant that is shorter and has fewer tillers than normal plants when the two loci have only one or no dominant allele at both loci. A cultivar with two dominant alleles at the hwd1 locus and a cultivar with two dominant alleles at the hwd2 locus were crossed with a double recessive tester line. Linkage analysis was carried out for each gene independently in two F₂ populations derived from these crosses. hwd1 was mapped on the distal region of rice genetic linkage map for chromosome 10, flanked by RFLP markers C701 and R2309 at a distance of 0.9 centiMorgans (cM) and 0.6 cM, respectively. hwd2 was mapped in the central region of rice genetic linkage map for chromosome 7, tightly linked with 4 RFLP markers without detectable recombination. The usefulness of RFLP mapping and map information for the genes controlling reproductive barriers are discussed in the context of breeding using diverse rice germplasm, especially gene introduction by marker-aided selection.     

Field assessments of gene flow from transgenic to cultivated rice (Oryza sativa L.) using a herbicide resistance gene as tracer marker

Development of plant genetic engineering has led to the deployment of transgenic crops and, simultaneously, to the need for a thorough assessment of the risks associated with their environmental release. This study investigated the occurrence of gene flow from transgenic rice to non-transgenic rice plants under agronomic conditions using a herbicide resistance gene as a tracer marker. Two field experiments were established in the paddy fields of two main Mediterranean rice-growing areas of Spain and Italy. In both locations analyses of phenotypic, molecular and segregation data showed that pollination of recipient plants with pollen of the transgenic source occurred at a significant frequency. A gene flow slightly lower than 0.1% was detected in a normal side-by-side plot design. Similar results were found in a circular plot when the plants were placed at 1-m distance from the transgenic central nucleus. A strong asymmetric distribution of the gene flow was detected among this circle and highest values (0.53%) were recorded following the direction of the dominant wind. A significant lowest value (0.01%) was found in the other circle (5 m from the transgenic plants) as was expected according to the characteristics of rice pollen. Such circular-field trial designs could also prove to be very useful in studying the gene flow to other commercial cultivars of rice with the aim of establishing strategies to prevent pollen dispersal from commercial transgenic fields to the neighbouring conventional fields. Received: 23 February 2001 / Accepted: 31 March 2001     

cDNA cloning and sequencing of tobacco chloroplast ribosomal protein L12

Tobacco chloroplast ribosomal protein L12 was isolated as a ssDNA-cellulose-binding protein from a chloroplast soluble protein fraction. Based on the N-terminal amino acid sequence of chloroplast L12, a cDNA clone was isolated and characterized. The precursor protein deduced from the DNA sequence consists of a transit peptide of 53 amino acid residues and a mature L12 protein of 133 amino acid residues. The chloroplast L12 protein was synthesized with a reticulocyte lysate and subjected to nucleic acid-binding assays. L12 synthesized in vitro does not bind to ssDNA, dsDNA nor ribonucleotide homopolymers, but it binds to cellulose matrix.     

QTL analysis and mapping of pi21, a recessive gene for field resistance to rice blast in Japanese upland rice

Field resistance is defined as the resistance that allows effective control of a parasite under natural field conditions and is durable when exposed to new races of that parasite. To identify the genes for field resistance to rice blast, quantitative trait loci (QTLs) conferring field resistance to rice blast in Japanese upland rice were detected and mapped using RFLP and SSR markers. QTL analysis was carried out in F₄ progeny lines from the cross between Nipponbare (moderately susceptible, lowland) and Owarihatamochi (resistant, upland). Two QTLs were detected on chromosome 4 and one QTL was detected on each of chromosomes 9 and 12. The phenotypic variation explained by each QTL ranged from 7.9 to 45.7% and the four QTLs explained 66.3% of the total phenotypic variation. Backcrossed progeny lines were developed to transfer the QTL with largest effect using the susceptible cultivar Aichiasahi as a recurrent parent. Among 82 F₃ lines derived from the backcross, resistance segregated in the expected ratio of resistant 1 : heterozygous 2 : susceptible 1. The average score for blast resistance measured in the field was 4.2 ± 0.67, 7.5 ± 0.51and 8.2 ± 0.66, for resistant, heterozygous and susceptible groups, respectively. The resistance gene, designated pi21, was mapped on chromosome 4 as a single recessive gene between RFLP marker loci G271 and G317 at a distance of 5.0 cM and 8.5 cM, respectively. The relationship to previously reported major genes and QTLs conferring resistance to blasts, and the significance of marker-assisted selection to improve field resistance, are discussed. Received: 8 June 2000 / Accepted: 24 November 2000     

QTLs for cell-membrane stability mapped in rice (Oryza sativa L.) under drought stress

Cell-membrane stability (CMS) is considered to be one of the major selection indices of drought tolerance in cereals. In order to determine which genomic region is responsible for CMS, 104 rice (Oryza sativa L.) doubled haploid (DH) lines derived from a cross between CT9993–5-10–1-M and IR62266-42–6-2 were studied in the greenhouse in a slowly developed drought stress environment. Drought stress was induced on 50-day-old plants by withholding water. The intensity of stress was assessed daily by visual scoring of leaf wilting and by measuring leaf relative water content (RWC). The leaf samples were collected from both control (well-watered) and stressed plants (at 60–65% of RWC), and the standard test for CMS was carried out in the laboratory. There was no significant difference (P>0.05) in RWC between the two parental lines as well as among the 104 lines, indicating that all the plants were sampled at a uniform stress level. However, a significant difference (P<0.05) in CMS was observed between the two parental lines and among the population. No significant correlation was found between CMS and RWC, indicating that the variation in CMS was genotypic in nature. The continuous distribution of CMS and its broad-sense heritability (34%) indicates that CMS should be polygenic in nature. A linkage map of this population comprising of 145 RFLPs, 153 AFLPs and 17 microsatellite markers was used for QTL analysis. Composite interval mapping identified nine putative QTLs for CMS located on chromosomes 1, 3, 7, 8, 9, 11 and 12. The amount of phenotypic variation that was explained by individual QTLs ranged from 13.4% to 42.1%. Four significant (P<0.05) pairs of digenic interactions between the detected QTLs for CMS were observed. The identification of QTLs for this important trait will be useful in breeding for the improvement of drought tolerance in rice. This is the first report of mapping QTLs associated with CMS under a natural water stress condition in any crop plants. Received: 8 September 1999 / Accepted: 13 October 1999     

A combined RFLP and AFLP linkage map of upland rice (Oryza sativa L.) used to identify QTLs for root-penetration ability

Acombined RFLP and AFLP linkage map of an F₆ recombinant inbred population, which was derived from a previously mapped F₂ of a cross between the two drought resistant upland rice varieties Bala and Azucena, is presented. The map contains 101 RFLP and 34 AFLP markers on 17 linkage groups covering 1680 cM. Also presented is the approximate mapping position of a further four RFLP and 75 AFLP markers, which either could not be given a unique place on the map or for which the available data is not sufficient to allow confident positioning, and the result of quantitative trait locus (QTL) mapping of traits related to root-penetration ability. Root penetration was assessed by counting the number of root axes that penetrated a 3 mm-thick layer consisting of 80% wax and 20% white soft paraffin. Good root penetration would be expected to increase drought resistance where soil strength is high. Single-marker analysis revealed seven QTLs for the number of roots which penetrate the wax layer. In identical locations were seven QTLs for the ratio of penetrated to the total number of roots. Transgressive inheritance of positive alleles from Bala explained four of these QTLs. Comparison of the QTLs identified here with previous reports of QTLs for root morphology suggest that alleles which improve root penetration ability may also either make the roots longer or thicker. Received: 3 February 1999 / Accepted: 30 April 1999     

Inter-simple sequence repeat (ISSR) amplification for analysis of microsatellite motif frequency and fingerprinting in rice (Oryza sativa L.)

 Inter-simple sequence repeat (ISSR) amplification was used to analyze microsatellite motif frequency in the rice genome and to evaluate genetic diversity among rice cultivars. A total of 32 primers, containing different simple sequence repeat (SSR) motifs, were tested for amplification on a panel of 59 varieties, representative of the diversity of cultivated rice (Oryza sativa L.). The ISSR analysis provided insights into the organization, frequency and levels of polymorphism of different simple sequence repeats in rice. The more common dinucleotide motifs were more amenable to ISSR analysis than the more infrequent tri-, tetra- and penta-nucleotide motifs. The ISSR results suggested that within the dinucleotide class, the poly(GA) motif was more common than the poly(GT) motif and that the frequency and clustering of specific tri- and tetra-nucleotide simple sequence repeats was variable and motif-specific. Furthermore, trinucleotide ISSR markers were found to be less polymorphic than either dinucleotide or certain tetranucleotide ISSR markers, suggesting which motifs would be better targets for microsatellite marker development. The ISSR amplification pattern was used to group the rice genotypes by cluster analysis. These results were compared to surveys of the same varieties for amplified fragment length polymorphism (AFLP), restriction fragment length polymorphism (RFLP) and isozyme markers. The ISSR fingerprint could be used to differentiate the genotypes belonging to either Japonica or Indica sub species of cultivated rice and to dissect finer levels of diversity within each subspecies. A higher percentage of polymorphic bands was produced with the ISSR technique than the AFLP method, based on a similar PCR reaction. Therefore, ISSR amplification proved to be a valuable method for determining genetic variability among rice varieties and for rapidly identifying cultivars. This efficient genetic fingerprinting technique would be useful for characterizing the large numbers of rice accessions held in national and international germplasm centers. Received: 25 May 1998 / Accepted: 17 September 1998     

Localization of the rice stripe disease resistance gene, Stv-bi, by graphical genotyping and linkage analyses with molecular markers

 We used graphical genotyping and linkage analyses with molecular markers to determine the chromosomal location of the rice stripe disease resistance gene, Stv-b ⁱ . The stripe resistance gene from the indica rice (Oryza sativa) cv ‘Modan’ was introgressed into several Japanese rice varieties. We found 4 RFLP markers in ‘Modan’, five susceptible parental rice varieties (‘Norin No. 8’, ‘Sachihikari’, ‘Kanto No. 98’, ‘Hokuriku No.103’ and ‘Koganebare’) and four resistant progeny varieties (‘St. No. 1’, ‘Aichi No. 6’, ‘Aoisora’ and ‘Asanohikari’). Graphical genotyping of the resistant progeny revealed a chromosomal segment ascribable to ‘Modan’ and associated with stripe resistance. The chromosomal segment from ‘Modan’ was located at 35.85 cM on chromosome 11. Linkage analysis using 120 F₂ individuals from a cross between ‘Koshihikari’ (susceptible) and ‘Asanohikari’ (resistant) revealed another 8 RFLP markers in the same chromosome. We performed a bioassay for rice stripe resistance in F₃ lines of the F₂ individuals using infective small brown planthoppers and identified an 1.8-cM segment harboring the rice stripe disease resistance gene, Stv-b ⁱ , between XNpb220 and XNpb257/ XNpb254. Furthermore, Stv-b ⁱ was linked by 0.0 cM to a RFLP marker, ST10, which was developed on the basis of the results of RAPD analysis. These DNA markers near the Stv-b ⁱ locus may be useful in marker-assisted selection and map-based cloning of the Stv-b ⁱ gene. Received: 26 September 1997 / Accepted: 4 November 1997     

Molecular mapping of a resistance-specific PCR-based marker linked to a gall midge resistance gene (Gm4t) in rice

 A PCR-based marker (E20₅₇₀) linked to the gene Gm4t, which confers resistance to a dipteran pest gall midge (Orseolia oryzae), has been mapped using the restriction fragment length polymorphism (RFLP) technique in rice. Gm4t is a dominant resistance gene. We initially failed to detect useful polymorphism for this marker in a F₃ mapping population derived from a cross between two indica parents, ‘Abhaya’בShyamala’, with as many as 35 restriction enzymes. ‘Abhaya’ carries the resistance gene Gm4t and ‘Shyamala’ is susceptible to gall midge. Subsequently, E20₅₇₀ was mapped using another mapping population represented by a F₂ progeny from a cross between ‘Nipponbare’, a japonica variety, and ‘Kasalath’, an indica variety, in which the gene Gm4t was not known to be present. Gm4t mapped onto chromosome 8 between markers R1813 and S1633B. Our method, thus, presents an alternative way of mapping genes which otherwise would be difficult to map because of a lack of polymorphism between closely related parents differing in desired agronomic traits. Received: 1 April 1997 / Accepted: 13 May 1997     

Identification and mapping of two brown planthopper resistance genes in rice

The brown planthopper (BPH) is one of the most serious insect pests of rice. In this study, we conducted a molecular marker-based genetic analysis of the BPH resistance of ’B5’, a highly resistant line that derived its resistant genes from the wild rice Oryza officinalis. Insect resistance was evaluated using 250 F₃ families from a cross between ’B5’ and ’Minghui 63’, based on which the resistance of each F₂ plant was inferred. Two bulks were made by mixing, respectively, DNA samples from highly resistant plants and highly susceptible plants selected from the F₂ population. The bulks were surveyed for restriction fragment length polymorphism using probes representing all 12 chromosomes at regular intervals. The survey revealed two genomic regions on chromosome 3 and chromosome 4 respectively that contained genes for BPH resistance. The existence of the two loci were further assessed by QTL (quantitative trait locus) analysis, which resolved these two loci to a 14.3-cM interval on chromosome 3 and a 0.4-cM interval on chromosome 4. Comparison of the chromosomal locations and reactions to BPH biotypes indicated that these two genes are different from at least nine of the ten previously identified BPH resistance genes. Both of the genes had large effects on BPH resistance and the two loci acted essentially independent of each other in determining t he resistance. These two genes may be a useful BPH resistance resource for rice breeding programs. Received: 6 March 2000 / Accepted: 28 July 2000     

Sequence and expression analysis of the thioredoxin protein gene family in rice

Thioredoxin (Trx) proteins play important biological functions in cells by changing redox via thioldisulfied exchange. This system is especially widespread in plants. Through database search, we identified 30 potential Trx protein-encoding genes (OsTrx) in rice (Oryza sativa L.). An analysis of the complete set of OsTrx proteins is presented here, including chromosomal location, conserved motifs, domain duplication, and phylogenetic relationships. Our findings suggest that the expansion of the Trx gene family in rice, in large part, occurred due to gene duplication. A comprehensive expression profile of Trx genes family was investigated by analyzing the signal data of this family extracted from the whole genome microarray analysis of Minghui 63 and Zhenshan 97, two indica parents, and their hybrid Shanyou 63, using 27 different tissues representing the entire life cycle of rice. Results revealed specific expression of some members at germination transition as well as the 3-leaf stage during the vegetative growth phase of rice. OsTrx genes were also found to be differentially up- or down-regulated in rice seedlings subjected to treatments of phytohormones and light/dark conditions. The expression levels of the OsTrx genes in the different tissues and under different treatments were also checked by RT-PCR analysis. The identification of OsTrx genes showing differential expression in specific tissues among different genotypes or in response to different environmental cues could provide a new avenue for functional analyses in rice.     

Quantitative trait locus analysis for rice panicle and grain characteristics

 The development of molecular genetic maps has accelerated the identification and mapping of genomic regions controlling quantitative characters, referred to as quantitative trait loci or QTLs. A molecular map derived from an F₂ population of a tropical japonica×indica cross (Labelle/Black Gora) consisted of 116 restriction fragment length polymorphism (RFLP) markers. Composite interval mapping was used to identify the QTLs controlling six panicle and grain characteristics. Two QTLs were identified for panicle size at LOD>3.0, with one on chromosome 3 accounting for 16% of the phenotypic variation. Four loci controlling spikelet fertility accounted for 23% of the phenotypic variation. Seven, four, three and two QTLs were detected for grain length, breadth, shape and weight, respectively, with the most prominent QTLs being on chromosomes 3, 4, and 7. Grain shape, measured as the ratio of length to breadth, was mostly controlled by loci on chromosomes 3 and 7 that coincided with the most important QTLs identified for length and breadth, respectively. A model including three loci accounted for 45% of the phenotypic variation for this trait. The identification of economically important QTLs will be useful in breeding for improved grain characteristics. Received: 18 July 1997 / Accepted: 9 December 1997     

A low-temperature method for maintaining plant regeneration activity in embryogenic callus of rice (Oryza sativa L.)

 A method was developed to maintain plant regeneration activity of rice cells (Oryza sativa L.) using embryogenic callus. Calluses were cultured in suspension, then on solid medium, to form compact globular callus resistant to low-temperature stress and with high plant regeneration activity. Callus preserved at 5  °C for 5 months regenerated plants from protoplasts at a frequency higher than from non-preserved callus from cv. Nipponbare, and cv. Koshihikari, but at lower rates from cv. Akitakomachi. Similar results were obtained from protoplasts of the three cultivars. Callus preserved at 5  °C for 8 months incurred cell damage, yet some surviving cells divided in suspension culture and eventually regenerated whole plants. Preserved and non-preserved regenerated plants showed similar levels of somaclonal variation. Received: 7 January 1999 / Revision received: 28 April 1999 / Accepted: 26 May 1999     

The expression of two engrailed-related genes in an apterygote insect and a phylogenetic analysis of insect engrailed-related genes

 Homologues of the Drosophila segment polarity gene engrailed have been cloned from many insect species, as well as other arthropods and non-arthropods. We have cloned partial cDNAs of two engrailed homologues, which we call engrailed-related genes, from the phylogenetically basal insect, Thermobia domestica (Order Thysanura) and possibly as many as four engrailed-related genes from the phylogenetically intermediate insect, Oncopeltus fasciatus (Order Hemiptera). Previous to our findings, only single engrailed-related homologues had been found in phylogenetically intermediate insect species (Tribolium and Schistocerca) and in the crustacean Artemia, while two engrailed-related homologues have been found in more derived orders (Hymenoptera and the engrailed and invected genes of lepidopterans and dipterans). Consequently, we performed a phylogenetic analysis of insect engrailed-related genes to determine whether insects ancestrally had one or two engrailed-related genes. We have found evidence of concerted evolution among engrailed-related paralogues, however, that masks the true phylogenetic history of these genes; the phylogeny may only be decipherable, therefore, by examining the presence or absence of engrailed-specific and invected-specific motifs, which will require cloning the full length cDNAs from more species. In addition, we examined the embryonic expression pattern of the two Thermobia engrailed-related genes; like Drosophila engrailed and invected, they are expressed in very similar patterns, but show one temporal difference in pregnathal segments that correlates with the tentative phylogenetic placement of the genes. Thermobia engrailed-related expression also confirms that the dorsal ridge is an ancient structure in insects. Received: 4 May 1998 / Accepted: 2 August 1998     

QTLs and epistasis for aluminum tolerance in rice (Oryza sativa L.) at different seedling stages

To investigate the genetic background for aluminum (Al) tolerance in rice, a recombinant inbred (RI) population, derived from a cross between an Al-sensitive lowland indica rice variety IR1552 and an Al-tolerant upland japonica rice variety Azucena, was used in culture solution. A molecular linkage map, together with 104 amplified fragment length polymorphism (AFLP) markers and 103 restriction fragment length polymorphism (RFLP) markers, was constructed to map quantitative trait loci (QTLs) and epistatic loci for Al tolerance based on the segregation for relative root length (RRL) in the population. RRL was measured after stress for 2 and 4 weeks at a concentration of 1mM of Al³⁺ and a control with a pH 4.0, respectively. Two QTLs were detected at both the 2nd and the 4th weeks on chromosomes 1 and 12 from unconditional mapping, while the QTL on chromosome 1 was only detected at the 2nd stress week from conditional mapping. The effect of the QTL on chromosome 12 was increased with an increase of the stress period from 2 to 4 weeks. The QTL on chromosome 1 was expressed only at the earlier stress, but its contribution to tolerance was prolonged during growth. At least one different QTL was detected at the different stress periods. Mean comparisons between marker genotypic classes indicated that the positive alleles at the QTLs were from the Al-tolerant upland rice Azucena. An important heterozygous non-allelic interaction on Al tolerance was found. The results indicated that tolerance in the younger seedlings was predominantly controlled by an additive effect, while an epistatic effect was more important to the tolerance in older seedlings; additionally the detected QTLs may be multiple allelic loci for Al tolerance and phosphorus-uptake efficiency, or for Al and Fe²⁺ tolerance. Received: 29 July 1999 / Accepted: 13 October 1999