Agronomical Characters and Rapd Markers Associated with the Resistant Allele to the Erysiphe Polygoni in Common Bean

Aiming at verifying whether the common bean allele of resistance to Erysiphe polygoni is associated with undesirable agronomic characters and identifying RAPD markers linked to the resistance allele of the line ESAL 686, the reaction of 64 F1:2 families of the backcross (Jalo x ESAL 686) x Jalo to the pathogen was evaluated. The F1:3RC1 families were evaluated, in the absence of oidium, by grain yield (g/ plot), weight of 100 seeds (g) number of days to flowering and grain aspect. The association noticed of the reaction gene with grain aspect should not prevent the selection of resistant lines with grains similar to Jalo. DNA of F1RC1 plants, resistant and susceptible, were pooled making up two bulks and amplified by RAPD. The four markers identified, amplified by the primers OPAC15, OPC14, OPC18 and OPL16, are closely linked among them and relatively far from the resistance allele, having little usefulness for indirect selection of resistant plants.


INTRODUCTION
In Brazil, common bean (Phaseolus vulgaris L.) of the Jalo type, with yellow and large grains, have a great commercial acceptance in some regions and in general possess a higher market price.However, those beans bear some problems which make their cultivation difficult such as susceptibility to pests and some pathogens specially Erysiphe polygoni, causal agent of oidium.
Oidium is a disease of importance in the common bean when it is cultivated at the times of "drought" and fall-winter, when environmental conditions favor the occurrence of that pathogen.The losses of grain yield ascribed to oidium may reach up to 69% depending on the cultivar and environmental conditions (Shwartz et al., 1981).In a work performed in the South of Minas Gerais State there were reductions in yield of 50% in the cultivar Eriparza and 40% in Rio Vermelho (Arriel et al., 1991).
Although there are efficient alternatives to disease control with chemical products, the use of resistant cultivars is the control measure most advised for not increasing production costs or causing environmental unbalance and pollution (Rava and Sartorato, 1993).A large number of cultivars with high resistance to the pathogen are known, however, most of the resistant cultivars belong to the mesoamerican groups, with medium to small sized grains and indeterminate growth habit (Schwartz et al., 1981;Sartorato et al., 1993).ESAL 686 is one of the few lines with medium grains, determinate habit and resistant (Rezende et al., 1996).
As the Jalo cultivar presents crossing incompatibility with most cultivars of the Mesoamerican group (Singh and Gutierrez, 1984;Vieira et al., 1989) the introduction of the allele of resistance to E. polygoni in that cultivar is difficult.With the identification of the pathogen resistance in the line ESAL 686, which is compatible to large grained cultivars, the possibility of the introduction of the resistance allele into the cultivar Jalo appeared.
Crop Breeding and Applied Biotechnology, v. 1, n. 1, p. 11-21, 2001 In the development of a breeding program, obtaining information on the genetic control and association among the characters help breeders in decision making.Another aspect to be taken into account is that the pathogen is obligatory and its occurrence is not frequent through the year which may delay the selection program.One alternative is the selection of resistant genotypes by indirect way through markers linked to the resistance allele (Ferreira and Grattapaglia, 1995;Kelly, 1994).RAPD (Random Amplified Polymorphic DNA) is one of the most utilized markers.In this context, the objectives of the present work were to verify whether the oidium resistance allele is associated with undesirable agronomic characters and identify RAPD markers linked to the resistance allele of the line ESAL 686 to Erysiphe polygoni.

MATERIAL AND METHODS
The experiments were carried out at the Universidade Federal de Lavras, located in South of Minas Gerais State (21º58' S, 45º22' W and 910 m of altitude), with dark red latosol and CWb climate according to Koppen climate classification.

Parents and crosses
One of the parents utilized was the Jalo cultivar, with large yellow grains, type III indeterminate growth habit and high oidium susceptibility but with a grain type well accepted by consumers.The other parent was the line ESAL 686, with medium sized grains and dark yellow in color, type I determinate growth habit and resistant to oidium.Each plant from the first backcrossing generation F 1 RC 1 with the susceptible parent [(Jalo x ESAL 686) x Jalo] was utilized for DNA extraction, obtaining F 1:2 RC 1 families.

Evaluation of oidium reaction in the F 1:2 RC 1 families
Sixty four F 1:2 RC 1 families were evaluated, aiming at identifying those segregating ones in regards to oidium reaction and also those which were completely susceptible.An 8 x 8 simple lattice design was utilized, the plot being made up of a one meter row with 15 plants.Twenty days before the experiment was set up in the winter of 1998, two rows were sowed with Jalo around the experiment, which served as a border and inoculum source.Twenty days after the sowing of the experiment, an inoculation with the pathogen was done through the contact of pieces of infected leaves with healthy leaves in addition to the natural inoculation done by the wind.
The first evaluation of oidium incidence was performed two months after sowing by using a score scale of 1 to 10 ( Rezende et al., 1999).After six days, another evaluation was performed.During the evaluation of each plot, whether the plants were evenly susceptible or whether segregation occurred was recorded.An average score from two evaluators at two scoring times was utilized for variance analysis by decomposing the source of variation treatments into susceptible families (group 1), segregating families (group 2) and between groups.
The heritability (h 2 ) among means of families, in the wider sense, was estimated, utilizing the methodology presented by Vencovsky and Barriga (1992).Also, the h 2 confidence intervals estimation was obtained, by the expressions presented by Knapp et al. (1985) with 95% confidence (1 -α = 0.95).The relationship between the genetic and the experimental coefficient of variation was estimated (Vencovsky and Barriga 1992) as well.The number of segregating families and susceptible families observed was compared with the expected one, monogenic segregation being considered by the χ 2 test.

Evaluation of the F 1:3 RC 1 families in the absence of oidium
The same 64 families were evaluated in the F 1:3 RC 1 generation, in the rainy season of 1998/1999 in a 8 x 8 triple lattice design.Each plot was made up of two rows 2m long, 0.5 meters apart.For that generation, a weekly control of oidium was done by utilizing a fungicide based on sulphur.The objective of that control was to verify the effect of the resistance allele, in the absence of disease, on the following characters: number of days to flowering, grain yield (g/plot), weight of 100 seeds (g) and grain aspect.This last character was evaluated by utilizing a score scale of 1 to 5 in which: 1 -grains of uniform aspect and yellow, similar to Jalo; 2 -grains similar to Jalo in shape and differing in color; 3 -smaller-sized grains, rather round and dark; 4 -round and dark grains and, 5 -grains completely different from Jalo.
The analyses of variance were conducted for each character decomposing the treatment source of variation treatments into families within groups, and between groups of families with and without the resistance allele identified in the F 1:2 RC 1 (Fujimaki, 1978).Two heritability coefficients were estimated, one including the total genetic variation (h 2 ) and other omitting the variation among groups (H).Their respective intervals of confidence at 95% (1 -α = 0.95) was estimated as well.Genetic correlation was estimated utilizing the means of F 1:2 RC 1 and F 1:3 RC 1 .

DNA extraction
DNA extraction was made from F 1 RC 1 plants by means of a procedure similar to that employed by Nienhuis et al. (1995).DNA concentration was determined by using a fluorometer (Hoeffer Scientific, San Francisco, Ca, U.S.A).

DNA bulks and RAPD analysis
Two bulks were built from DNA samples extracted from leaves of F 1 RC 1 plants previously identified as to their reaction to Erysiphe polygoni.One of the bulks contained equimolar amounts of DNA extracted from seventeen resistant F 1 RC1 plants and the other bulk seventeen susceptible F 1 RC 1 plants.DNA of the bulks was amplified with 700 decamer primers (Operon Technologies Inc. Alameda, Ca, USA) to identify polymorphisms between the bulks.
The RAPD reactions were prepared in volumes of 10 µl in a way similar to the procedure used by Nienhuis et al. (1995).The reactions were performed in glass capillary tubes in an airrefrigerated thermal-cycler (Idaho Technology, Idaho Falls, ID, USA).The thermal cycler was programmed for 40 cycles under the following conditions: 1) DNA denaturation at 91 0 C, for 60s, in the first 2 cycles, and 1s in the next 38 cycles; 2) Elongation at 72 0 C, for 70s; 3) Primer annealing at 42 0 C, for 7s.After amplification, the reaction products were separated by eletrophoresis in 1% agarose gel, stained with ethidium bromide, viewed in ultraviolet light transiluminator and photographed.

Analysis of genetic distances
The potential RAPD markers detected in segregating bulks were firstly tested in the constituent individuals of each bulk to check the co-segregation between those markers and the resistance/susceptibility phenotype determined in 64 F 1 RC 1 plants.Chi-square tests were utilized to compare the observed and expected phenotypic ratios under the supposition of independent assortment.
To determine the positions of the identified markers relative to the resistance allele of the ESAL 686 line the MAPMAKER program (Lander et al., 1987) version 3.0 with minimum lod score of 0.83 was used.The map-funtion of Haldane (1919) was employed to calculate the genetic distances in centimorgans (cM).The recombination frequency (r) was estimated by using an iterative process through the non-linear least square procedure (Mendonça et al., 1998), by utilizing the PROC NLIN (SAS®, 1990) procedure.The model adjustment was evaluated by the determination coefficient.

Genetic control of oidium resistance
The summary of the analysis of variance of the oidium incidence scores is presented in Table 1.Initially it is worthwhile to point out that practically no lattice efficiency was found which suggests that the distribution of the inoculum was uniform.Along with that, the experimental precision, calculated by the CV (%), was superior to those reported with other common bean characters (Abreu et al., 1994) as well as to those reported by Marques Crop Breeding and Applied Biotechnology, v. 1, n. 1, p. 11-21, 2001

2001, Brazilian Society of Plant Breeding
Júnior (1997) by utilizing scores in the evaluation of diseases.
It was found that the families evaluated differed in relation to pathogen incidence (P<0.01).When proceeding the decomposition of the families into two groups (segregating and susceptible), no significant differences were observed among the susceptible ones.The opposite took place in the case of the segregating families.As it was expected, a significant difference was detected (P<0.01) among groups, highlighting the highest incidence of disease occurred among the susceptible families.The non-significance of the susceptible families is due to the fact that those families come from the self-fertilization of homozygous plants with the susceptibility allele, while the segregating families come from the self-fertilization of heterozygous plants to the gene that control the reaction to oidium having, therefore, genetic variation possible to be explored for this character.
The 35 segregating families for the 29 susceptibles are very close to the 1:1 ratio, as expected in the backcrossing considering the involvement of a single gene (χ 2 =0.5625).
In a previous work involving the same crossing, but with other segregating generations, the hypothesis which two genes with double recessive epistasis interaction was tested (Rezende et al., 1999).However, the segregation observed in this research shows that as a matter of fact, only one gene is involved.Results of the literature leave doubt about the gene number involved and the sort of gene action (Dundas, 1936;Dundas, 1942;Bett and Michaels, 1995).It is probable that this difference occurs in terms of the criterion of scores utilized in each case and further, the score from which a plant or a family must be considered as resistant or susceptible.In addition, in the evaluation of families by Rezende et al. (1999), an average score was given and it was not taken into consideration whether there was or not segregation within the families as was the case of this work.
The high estimate of heritability in the wide sense as well as the relationship between the coefficient of genetic and experimental variation are a clue that the genetic control of oidium reaction is due to one or few genes (Rezende et al., 1999).Those values show the existence of genetic variability among families and a reduced environmental variation which allows to foresee the success with the selection of resistant plants and/or families in segregating populations.

Association of oidium resistance allele with agronomical characters
The summary of the analysis of variance of agronomical characters is presented in Table 2.It is found that the lattice design was of little efficiency Table 1 -Analysis of variance of the oidium incidence scores in F 1:2 families of the cross (Jalo x ESAL 686) x Jalo of bean in the winter of 1998.
Crop Breeding and Applied Biotechnology, v. 1, n. 1, p. 11-21, 2001  relative to the randomized blocks, specially for 100 seeds weight and days to flowering.The coefficient of variation (CV%) was less for the number of days to flowering (3.43%) and larger for grain aspect (18.85%).Even this last value is of magnitude similar or inferior to that normally reported in experiment with common bean crop (Marques Júnior, 1997;Abreu et al., 1994), standing out that the families were evaluated with precision.
Genetic differences were found among families for all the characters evaluated (P<0,01).It was still observed that the plot of the total genetic variation due to groups was significant only for the weight of 100 seeds (P<0.05) and grain aspect (P<0.01),indicating that those characters may be associated with the oidium resistance allele.The genetic variation of the families within groups was significant (P<0.01) for all the characters.
The heritability estimates including total genetic variation (h 2 ) and omitting the variation between group (H), were distinct only for grain aspect, showing a more marked effect of the allele resistance for the formation of undesirable grains.Nevertheless, there is an overlapping between the two estimates when taking the intervals of confidence of each of them into consideration, which suggests that the unfavorable effect in grain aspect will not be able to prevent the selection of resistant lines and with grains similar to Jalo.In addition, such similarity took place probably due to the recombination of the resistance allele relative to one or more genes which affect grain aspect, by the fact of eight families with grain aspect close to the ideal and carriers of the resistance allele having been detected.Therefore, the association of the oidium resistance allele with those which condition the grain undesirable aspect may be due to genic linkage (Fujimaki and Comstock, 1977;Fujimaki, 1978).
On the other hand, the differences among the average weights of 100 seeds of the two groups cannot be interpreted as due to gene linkage or pleiotropy (Table 6) because the mean of the families with the resistance allele (2.855g) was slightly superior to the mean of the susceptible families (2.532g), a result opposite to the average weight of 100 seeds of the two parents.The resistant line ESAL 686 possess seeds smaller than Jalo, the susceptible parent.Therefore, no difficulty in the selection of resistant families and of large seeds as Jalo is expected.
The estimates of the correlations between the score ascribed to the incidence of the pathogen with the other characters evaluated were of little magnitude, nevertheless, when involving grain aspect it was rather superior (Table 3).As the occurrence score of the pathogen was given in an environmental condition different from that in which the other characters were evaluated, the estimated covariance is genetic, the same occurring with correlation which is a measure of genetic association among the characters and may be due to linkage of genes or pleiotropy (Falconer, 1987).Thus, it may be inferred that there is an association of the oidium reaction gene with those responsible for grain aspect.That result confirms what was previously reported with analysis of variance, showing that Fujimaki's methodology (1978) is effective for detecting the association of the characters.Similar results were obtained in wheat (Cox et al., 1997) and rice (Fujimaki, 1978).
In addition, in the present results, the high estimates of heritability of all characters evaluated allowed to foresee the success with the selection of plants/ families resistant to oidium, early, high yielding, with large grains and similar to Jalo.

RAPD markers linked to the oidium resistance allele
Four potential RAPD markers at the coupling phase were identified.Figure 1 shows the amplification patterns of the parents and of the DNA bulks with the four primers which enabled the detection of the RAPD markers.It was found that the four markers (arrows) exhibited higher intensity in the parents which are homozygous relative to the resistant bulks because the F 1 RC 1 plants are heterozygous (Table 7).

2001, Brazilian Society of Plant Breeding
The segregation analysis of the markers of 1/2 presence of bands: 1/2 absence of bands in the 64 individuals of the F 1 RC 1 population was consistent with the dominant monogenic inheritance expected for the four markers which characterize them as genetic markers (Table 4).
To confirm whether the markers were really linked to the oidium resistance allele, the co-segregation analysis between each marker and the resistance/ susceptibility allele was done.A co-segregation analysis between the markers two by two was also conducted (Table 5).It is found that the four markers are closely linked and all of them are relatively distant from the resistance allele.
The genetic distances, LOD scores (log of oddsratio) as well as the recombination frequency, standard error, and determination coefficient (R 2 ) are in  Table 3 -Genetic correlations (r G ) involving the oidium incidence scores in the F 1:2 families, the number of days to flowering, grain yield (g/plot), weight of 100 seeds (g) and grain aspect of the F 1:3 families of the crossing (Jalo x ESAL 686) x Jalo.
Crop Breeding and Applied Biotechnology, v. 1, n. 1, p. 11-21, 2001 Pairs considered that all the markers are linked to the oidium resistance allele with 95% confidence.The high values of R 2 shows that the estimated   According to the results on Table 5 and 6, the four markers are closely linked among each other and weakly linked to the resistance allele.Taking into account the marker closest to the resistance allele, amplified by the primer OPAC15, the recombination frequency estimated is of 32.51%, which means that if the indirect selection of plants is performed with the marker in a F 2 population, 18% of the selected plants will be susceptible (Table 7).This frequency of erroneously selected plants by means of the marker shows that it has little utility in aiding selection.
Crop Breeding and Applied Biotechnology, v. 1, n. 1, p. 11-21, 2001  The magnitude of the error expected in the selection reduces proportionally as the marker is closer to the allele of interest.This is the case, for example, of the Co.5 allele, which confers resistance to some races of Colletotrichum lindemuthianum in common bean, where only 7.2% of the plants selected indirectly in F 2 by means of the marker amplified by the primer OPF10 (r=0.115), will be susceptible (Castanheira et al., 1999).
Table 5 -Segregation analysis of the four marker loci and oidium reaction gene, taken two by two and considering independent assortment in the F 1 generation of the crossing (Jalo x ESAL 686) x Jalo.
1/ AB (presence of the resistance allele and marker), Ab (presence of the resistance allele and absence of the marker).AB (presence of the susceptibility allele and marker), ab (presence of the susceptibility allele and absence of the marker).
2/ AB (presence of the band in the two markers), Ab (presence of the band in the first marker and absence in the second marker), ab (absence of the band in the two markers).
Table 6 -Distance and frequency of recombination of RAPD markers and the oidium resistance allele in the F 1 generation of the crossing (Jalo x ESAL 686) x Jalo.
Crop Breeding and Applied Biotechnology, v. 1, n. 1, p. 11-21, 2001 1/ Total distance=58.20 cM (Haldane); 2/ Recombination frequency.It is noticed, in Table 7, that the other markers are slightly less efficient for indirect selection purposes than OPAC15, due to they are farther from the resistance allele.Considering the distance of the markers identified relative to the resistance allele, efforts must be done to identify others closer, which will be useful for indirect selection specially to fasten the selection program in times of the year when the disease generally does not develop.

CONCLUSIONS
Association of the oidium resistance allele only with grains aspect was observed however, that association must not prevent the selection of resistant lines and grains similar to Jalo.
Four RAPD markers amplified by the primers OPAC15, OPC14, OPC18 and OPL16 which are closed linked and relatively distant from the resistance allele were identified, therefore they have little usefulness to the indirect selection of plants with resistance allele.

INTRODUCTION
Aluminum is one of the most abundant earth metals and is found in large amounts in tropical soils.The Brazilian tropical and subtropical areas are formed by acid soils, where the toxicity of this metal is one of the most important environmental stresses affecting cereal development (Delhaize et al., 1993;Camargo, et al., 1992).Associated to this fact, the action of the lime application is limited to the zone of incorporation, and it certainly will not penetrate deeper than the plowing levels.Its application in sub superficial levels represents great cost to the farmer.Thus, the existence of cultivars with tolerance to toxic aluminum is of great importance for oat breeding programs.
Roots affected by Al +++ present a peculiar development at their tips.The meristematic region of the main and lateral roots, in the presence of Al +++ , gets darker, smaller and thicker, with fewer ramifications (Foy, 1974), resulting in low nutritional efficiency and water supply for the plant (Carver et al., 1988;Foy and Fleming, 1978).Such phenomenon allows the identification of genotypes that are sensitive to this cation.
According to Camargo (1984), Dornelles (1994) and Sànchez-Chacòn (1998), tolerance to Al +++ is easy to detect in tests under controlled conditions (in greenhouse or laboratories) and nutritive solution.This is, therefore, an easy method to identify plants that are tolerant to aluminum, saving space and time.
Several studies have investigated aluminum toxicity tolerance in cereals.However, few were carried out with oats.In an seminal study, Sànchez-Chacòn (1998) characterized the germoplasm from the UFRGS oat breeding program and identified one dominant gene for aluminum tolerance in nine populations from crosses between tolerant and sensitive genotypes.However, the presence of different sources of genes for aluminum tolerance among oat genotypes from the UFRGS that can be combined in a breeding program is unknown.The present work has been developed to study the inheritance of Al +++ tolerance in other source genotypes and to investigate the allelic relations between genes present in different oat genotypes.

MATERIALS AND METHODS
The assessment of the tolerance to aluminum toxicity was done with the method described by Camargo and Oliveira (1981) and adapted to oats by Sánchez-Chacón (1998).Seedlings from P 1 , P 2 and F 2 generations derived from the cross between the tolerant (UFRGS 15, UFRGS 17 and UFRGS 93605) and sensitive parents (UFRGS 911715 and UFRGS 93598) (Table 1) were evaluated for their response to Al +++ toxicity through the regrowth of the primary root.The F 2 segregating populations with their respective parents were submitted to nutritional solution in pots with 20 ppm of cation arranged in a completely randomized experimental design, considering that each pot constituted a population.
The seedlings assessed were divided in sensitive and tolerant in relation to their response to toxicity to Al +++ , having the class separation limit at 0.8 mm for all populations.The same class limit separation was used in the work done with oats by Sànchez-Chacòn (1998).A genetic hypothesis in relation to the number of segregating genes was proposed for each population, based on the frequency distribution obtained in the F 2 generation, and tested using the chi-square analysis (Steel and Torrie, 1960).

RESULTS AND DISCUSSION
We began testing the genetic hypothesis of just one gene segregating in crosses from populations between tolerant x sensitive genotypes since, according to several studies, one to two genes for tolerance explain most of the variation for this character in cereal crops (Kerridge et al., 1971;Camargo, 1984;Lagos et al., 1991;Camargo et al., 1992;Riede and Anderson, 1996;Johnson et al., 1997;Sànchez-Chacòn, 1998).This genetic hypothesis was confirmed for the populations from the crosses UFRGS 17 x UFRGS 911715, UFRGS 17 x UFRGS 93598 and UFRGS 93605 x UFRGS 911715 (Table 2) and is in agreement with that reported by Sànchez-Chacòn (1998).
A segregation ratio of 9 tolerant to 7 sensitive genotypes was observed for the F 2 populations derived from crosses involving the tolerant genotype UFRGS 15.This indicates that there are Crop Breeding and Applied Biotechnology, v. 1, n. 1, p. 22-26, 2001 Genotypes UFRGS two genes segregating in those populations and their interaction is shown through epistasis (Table 2).
Crosses between the tolerant UFRGS 17 x UFRGS 93605 and UFRGS 17 x UFRGS 15 showed continuous segregation and since the parental genotypes were both tolerant and overlapped for their response to Al +++ , it was not possible to classify the progeny accurately in two distinct phenotypic classes only (Wagner, 1999).
The fact that crosses of sensitive genotypes with UFRGS 15 segregate as two instead of one gene, indicates that this source of tolerance to aluminum toxicity differs from those of UFRGS 17 and UFRGS 93605.It is possible, however, that one of these loci is common to all tolerant sources studied here.Unfortunately, due to the variation in the measurement of primary root regrowth from environmental effects and to the size of the F 2 populations examined, this hypothesis could not be accurately tested in this work.Thus, further studies using larger populations and genetic designs to control environmental effects may help to determine these allelic relationships.
Crosses between the sensitive genotypes UFRGS 911715 and UFRGS 93598 did not show discontinuous segregation and most F 2 progeny had from 0.3 to 0.8 mm of primary root regrowth, indicating that these sources have similar genetic constitution.Nevertheless, UFRGS 93598 is the most sensitive genotype that has been identified in our studies and is more affected by Al +++ than UFRGS 911715 as can be seen from the means and standard deviations of primary root regrowth of both genotypes in Table 1.
Variation within the fixed genotypes P 1 and P 2 for primary root regrowth and some overlapping of tolerant and sensitive classes were observed, indicating the influence of environmental effects on the expression of this trait.This result shows that, although the methodology used in this work was widely utilized in tolerance assessments in different species (Camargo, 1984;Ferreira et al., 1997;Sànchez-Chacòn, 1998), it still needs improvement in assessing oats.On the other hand, it is important to point out that the variation observed in this work was not bigger than that observed by Sànchez-Chacòn (1998).It was possible to distinguish genotypes that were tolerant and sensitive to aluminum.The parents from this study behaved like those described by Sànchez-Chacòn (1998).
Crop Breeding and Applied Biotechnology, v. 1, n. 1, p. 22-26, 2001 Table 2 -Segregation of the F 2 generation from crosses between three tolerant and two sensitive oat genotypes to aluminum toxicity and the Chi-square test for each genetic hypothesis.Agosto, 1998.

Number of
Seedlings

Figure 1 -
Figure 1 -RAPD markers (arrows) obtained by DNA amplification with the primers OPC14, OPC18, OPAC15 and OPL16 in which: P1 is the resistant parent ESAL 686, P2 is the susceptible parent Jalo, B1 is the resistant bulk and B2 is the susceptible bulk.The column M corresponds to the lambda DNA digested with the restriction enzyme Hae III.
frequencies fit those observed in RC 1 , showing the good adjustment of the model.1/presence of band: absence of band, considering monogenic inheritance.

Table 6
. The least value of LOD score presented in Table 6 (0.88) exceeds the minimum LOD score of 0.83.Therefore, it may be

Table 2 -
Analyses of variance of the number of days to flowering, grain yield (g/plot), weight of 100 seeds (g) and grain aspect of the F 1:3 families of the crossing (Jalo x ESAL 686) x Jalo, in the rainy season of 1998/1999.
* and ** -significant at the level of 5% and 1% by the F test; ns -non-significant;

Table 4 -
Segregation analysis of the markers OPC14, OPC18, OPAC15 and OPL16 in the F 1 generation of the crossing (Jalo x ESAL 686) x Jalo.

Table 7 -
Estimate of the frequency of susceptible plants (EFSP) selected as resistant in F 2 by the four markers, the respective number of pair of bases (pb) of each marker, each one amplified by a primer of known sequence.

Table 1 -
Genealogy, type of response to aluminum toxicity and mean± standard deviation (sd) of primary root regrowth (cm) of the oat genotypes included in this study.Janeiro, 1998.
1/ Mean±sd of each genotype grown in all pots of the experiment.