Relationships between cold- and salt-tolerance during seed germination in tomato: Analysis of response and correlated response to selection

Seeds of F₂ progeny of a cross between a slow-germinating (UCT5) and a fast-germinating tomato line (PI120256) were evaluated for germination under non-stress (control), cold-stress and salt-stress conditions, and in each treatment the most rapidly (first 5%) germinating seeds were selected, grown to maturity and self-pollinated to produce F₃ progeny. The selected F₃ progeny from each experiment were evaluated for germination in each of the three treatments, and compared with germination rate of unselected F₃ progeny. Selection for rapid seed germination was effective under cold stress and salt stress, but not effective under non-stress conditions. Furthermore, selection in either cold-stress or salt-stress treatment significantly improved progeny germination rate under both cold-stress and salt-stress treatments, as well as the non-stress treatment. The results support the suggestion that the same genes contribute to rapid seed germination under cold-, salt- and non-stress conditions. In practice therefore, selection for rapid seed germination under a single-stress environment may result in progeny with improved seed germination under a wide range of environmental conditions. Furthermore, to improve germination rate under non-stress conditions, it may be more effective to make selections under a stress treatment.     

Comparison of QTLs for seed germination under non-stress, cold stress and salt stress in tomato

The purpose of this study was to examine whether rate of tomato seed germination under non-stress, cold-stress and salt-stress conditions was under similar genetic control by identifying and comparing quantitative trait loci (QTLs) which affect germination rate under these conditions. A fast-germinating accession (LA722) of the wild tomato species Lycopersicon pimpinellifolium Jusl. and a slow-germinating cultivar (NC84173, maternal and recurrent parent) of tomato (Lycopersicon esculentum Mill.) were hybridized and BC₁ and BC₁S₁ progeny produced. The BC₁ population was used to construct a linkage map with 151 restriction fragment length polymorphism (RFLP) markers. The BC₁S₁ population (consisting of 119 BC₁S₁ families) was evaluated for germination under non-stress (control), cold-stress and salt-stress conditions and the mean time to 50% germination (T50) in each treatment was determined. Germination analyses indicated the presence of significant (P < 0.01) phenotypic correlations between T50 under control and cold stress (r = 0.71), control and salt stress (r = 0.58) and cold stress and salt stress (r = 0.67). The QTL analysis indicated the presence of genetic relationships between germination under these three conditions: a few QTLs were identified which commonly affected germination under both stress- (cold-, salt- or both) and non-stress conditions, and thus were called stress-nonspecific QTLs. A few QTLs were also identified which affected germination only under cold or salt stress and thus were called stress-specific QTLs. However, the stress-nonspecific QTLs generally exhibited larger individual effects and together accounted for a greater portion of the total phenotypic variation under each condition than the stress-specific QTLs. Whether the effects of stress-nonspecific QTLs were due to pleiotropic effects of the same genes, physical linkage of different genes, or a combination of both could not be determined in this study. The results, however, indicate that the rate of tomato seed germination under different stress and nonstress conditions is partly under the same genetic control.     

Breeding tomato for pollen tolerance to low temperatures by gametophytic selection

Haploid selection for traits related to pollen cold tolerance in tomato was performed in segregating populations derived from a Lycopersicon esculentum × L. pennellii hybrid. BC₁ populations were obtained by combining normal and low temperature treatments on two stages of pollen development: pollen formation, and germination and pollen tube growth. F₁ hybrids were cultivated under low and normal temperatures and their pollen was used to pollinate L. esculentum plants at low and normal temperatures. The four BC₁ populations obtained were tested for the quality and quantity of pollen produced at low temperatures. The population obtained by cold treatment at both stages had a significantly improved pollen germination ability at low temperatures. The two other coldselected BC₁ populations showed no differences compared with the unselected BC population. A second cycle of pollen selection, corresponding to BC₂, was applied in order to test its persistence in the subsequent generations and the possibility to further improve the character. This second cycle showed no improvement although some plants retained the high pollen germination ability at low temperatures that was observed in the first cycle. Hence, gametophytic selection of some characters related with tomato pollen performance may be feasible, at least for the first cycle of selection.     

Pollen selection in breeding tomato (lycopersicon esculentum Mill.) for adaptation to low temperature

Summary F₂ and BC₁ progenies from crosses between tomato (Lycopersicon esculentum) varieties differing for growth capacity at low temperature were produced under controlled conditions by hand pollination under two temperature regimes (22°C D/15°C N, normal temperature (N), and 15°C D/8°C N, low temperature (L)) with pollen formed under both regimes, resulting in four pollination treatments: NN, NL, LN and LL. Vegetative growth of the offspring was compared under a rather low temperature regime (19°C D/10°C N). Populations from different treatments within the progeny of a particular F₁ often differed significantly for average dry weight of 7 weeks old plants, the average of the NN population always being higher. Variances for dry weight were sometimes larger for LN populations, but this never resulted from a larger number of vigorous plants. Differences between populations within each progeny seemed to result in part from differences in the conditions for embryo development. Pollen selection at low temperature did not appear to be efficient for sporophyte breeding in this experiment.     

Absence of a genetic relationship between salt tolerance during seed germination and vegetative growth in tomato

Two approaches were used to determine the relationship between salt tolerance during seed germination and vegetative growth in tomato. First, F₄ progeny families of a cross between a breeding line, ‘UCT5’ (salt sensitive at all developmental stages), and a primitive cultivar, ‘PI 174263’ (salt tolerant during germination and vegetative growth), were evaluated in separate experiments for salt tolerance during germination and vegetative growth. There were significant differences among the F₄ families in both the rate of seed germination and the plant growth (dry matter production) under salt stress. There was, however, no significant correlation between the ability of seeds to germinate rapidly and the ability of plants to grow under salt stress. In the second approach, selection was made for rapid germination under salt stress in an F₂ population of the same cross and the selected progeny was evaluated for salt tolerance during both germination and vegetative growth. The results indicated that selection for salt tolerance during germination significantly improved germination performance under salt stress; a realized heritability estimate of 0.73 was obtained. Selection for salt tolerance during germination, however, did not affect plant salt tolerance during vegetative growth; there was no significant difference between the selected and unselected progeny based on either absolute or relative growth under salt stress. Obviously, in these genetic materials, salt tolerance during germination and vegetative growth are controlled by different mechanisms. Thus, to develop tomato cultivars with improved salt tolerance, selection protocols that include all critical developmental stages would be desirable.     

Genetic analysis of low-temperature tolerance during germination in tomato, Lycopersicon esculentum Mill.

The genetic basis of low-temperature tolerance during germination of tomato seed was investigated using two approaches. First, a cold-tolerant (PI 120256) and a cold-sensitive tomato cultivar (UCT5) and their reciprocal F₂, F₃ and BC₁ progeny (total of 10 generations) were evaluated for germination at a low (11 ± 0.5°C) and a high (control) temperature 20 ±0.5° C) Weighted least-square regression analysis indicated that in the low-temperature treatment most of the variation resulted from additive genetic effects, and dominance and epistatic interactions were nonsignificant. Partitioning of the total genetic variance into those attributable to the effects of embryo, endosperm, testa and the cytoplasm indicated that additive effects of endosperm and embryo could individually account for 80% and 77% of the total variance, respectively. In the control treatment, greater than 60% of the variation could be explained by individual additive effects of endosperm or embryo and ? 27% of the variation could be explained by embryo dominance effects. Across generations, there was a positive correlation (r = 0.78, P < 0.01) between germination in the control and low-temperature treatments and there were no significant genotype × temperature interactions. The results indicate the presence of similar or identical genes with predominantly additive effects on germination under both low and high temperatures. In the second approach, the effectiveness of directional phenotypic selection to improve tomato cold tolerance during germination was evaluated by selecting (in an F₂ population of the same cross) the fastest germinating seeds under low temperature and comparing the germination of the selected F₃ progeny with germination of an unselected F₃ population. The results indicated that selection was highly effective and significantly improved germination performance of the progeny; a realized heritability of 0.74 was obtained for low-temperature tolerance during germination. It is concluded that in these tomato lines germination under low temperature is genetically controlled, with additivity being the major genetic component, and thus the trait can be improved by phenotypic selection.     

Generation means analysis of right-censored response-time traits: Low temperature seed germination in tomato

Summary In studies to determine the inheritance of response-time traits, such as time to seed germination, some viable individuals may fail to respond during an experiment. If these right-censored observations are ignored, sample means and variances will be underestimated. This is illustrated using data from time to seed germination at 9°C for Lycopersicon esculentum (Mill.) fast germinating PI 120256, slow-germinating T3 and their reciprocal F₁, F₂ and backcross progeny. This paper presents methods to detect and to accommodate right-censored data in generation means analysis. Genetic interpretations derived from corrected and uncorrected estimates of generation means and variances are compared. Correction for right-censoring increased estimates of environmental and phenotypic variances, and decreased heritability estimates.     

Heritability of early blight resistance in a Lycopersicon esculentum×Lycopersicon hirsutum cross estimated by correlation between parent and progeny

Sixteen‐hundred BC₁ plants of a cross between an early blight (EB) susceptible tomato (Lycopersicon esculentum Mill.) breeding line (‘NC84173’ maternal and recurrent parent) and a resistant accession (‘PI126445’) of the tomato wild species Lycopersicon hirsutum Humb. and Bonpl. were grown in a field in 1998. This population was segregating (among other traits) for growth habit, self‐incompatibility and earliness in maturity. To eliminate confounding effects of these factors on disease evaluation and h² estimation, plants that were self‐incompatible, indeterminate and/or late‐maturing were eliminated. The remaining plants (146), which were self‐compatible and determinate (sp./sp.) in growth habit, with early‐ to mid‐season maturity, were evaluated for EB resistance and self‐pollinated to produce BC₁S₁ seed. The 146 BC₁S₁ progeny families, consisting of 30 plants per family, were grown in a replicated field trial in 1999 and evaluated for EB resistance and plant maturity. For each of the 146 BC₁ plants and corresponding BC₁ families, the area under the disease progress curve (AUDPC) and final disease severity (final percentage defoliation) were determined and used to measure disease resistance. The distributions of the AUDPC and final percentage defoliation values in the BC₁ and BC₁S₁ generations indicated that resistance from ‘PI126445’ was quantitative in nature. Estimates of h² for EB resistance, computed by correlation between BC₁S₁ progeny family means and BC₁ individual plant values, ranged from 0.69 to 0.70, indicating that EB resistance of ‘P1126445’ was heritable. Across BC₁S₁ families, a small, but significant, negative correlation (r = ‐0.26, P < 0.01) was observed between disease resistance and earliness in maturity. However, several BC₁S₁ families were identified with considerable EB resistance and reasonably early maturity. These families should be useful for the development of commercially acceptable EB‐resistant tomato lines.     

Introgression of resistance to Columbia and Northern root-knot nematodes fromSolanum bulbocastanum into cultivated potato

Summary Resistance toMeliodogyne chitwoodi races 1 (MC1) and 2 (MC2) andM. hapla (MH) derived fromSolanum bulbocastanum was introduced into the cultivated potato gene pool through somatic fusion. The initial F₁ hybrids showed resistance to the three nematodes. Resistance to reproduction on roots by MC1 was accompanied by resistance to tuber damage in F₁ clones. Tuber damage sometimes occurred, however, in hybrids of BC₁ progeny resistant to reproduction on roots when MC2 and MH were the challenging nematodes. Resistance to reproduction was transferred into BC₁ individuals, but a greater proportion of BC₁ progeny was resistant to MC1 than to MC2 or MH. Resistance to MC1 appears to be dominant and discretely inherited. F₁ and BC₁ progeny were pollen sterile, but seed were produced from crosses using cultivated tetraploid pollen sources. Approximately 11 and 33 per cent of pollinations produced berries on F₁ and BC₁ pistillate parents, respectively. Seed yield increased fourfold overall in crosses with F₁ compared to BC₁ individuals.Abbreviations MC1 Meloidogyne chitwoodi race 1 - MC2 Meloidogyne chitwoodi race 2 - MH Meloidogyne hapla - Rf Reproductive factor     

Relationships between cold- and salt-tolerance during seed germination in tomato: Germplasm evaluation

Thirty tomato accessions representing six Lycopersicon species were evaluated for the rate of seed germination under no stress, cold-stress and salt-stress conditions. Most accessions responded similarly to both cold- and salt-stress conditions (i.e. they were equally sensitive or tolerant to both stresses), however, a few accessions exhibited more sensitivity (or tolerance) to one stress than the other. In addition, some accessions that germinated relatively rapidly under non-stress conditions exhibited great sensitivity to both cold stress and salt stress. Across accessions, significant (P < 0.01) positive phenotypic correlations were observed between germination rate under control and cold stress (r_P= 0.89), control and salt stress (r_P= 0.63) and cold stress and salt stress (r_P= 0.77). The results indicate that the rate of tomato seed germination under non-stress, cold- and salt-stress conditions may be controlled by the same genes (or physiological mechanisms), but additional components may be involved which affect germination rate under specific stress conditions.     

Introgression of Cajanus platycarpus genome into cultivated pigeonpea, C. cajan

Summary Cajanus platycarpus, an incompatible wild species from the tertiary gene pool of pigeonpea (C. cajan (L.) Millspaugh), has many desirable characteristics for the improvement of cultivated varieties. To necessitate such transfers, embryo rescue techniques were used to obtain F₁ hybrids. The F₁ hybrids were treated with colchicine to obtain tetraploid hybrids, that were selfed to obtain F₂, F₃ and F₄ progenies. All of the hybrids and subsequent progenies had an intermediate morphology between the two parents. Backcrossing of the tetraploid hybrids with cultivated pigeonpea was not possible given embryo abortion, with smaller aborted embryos than those obtained in the F₀ parental cross.As a route of introgression, diploid F₁ hybrids were backcrossed with cultivated pigeonpea and BC₁ progeny obtained by in vitro culture of aborting embryos. BC₂ plants were obtained by normal, mature seed germination. Although embryo rescue techniques had to be used to obtain F₁ and BC₁ plants, it was possible to produce BC₂ and subsequent generations through direct mature seed. Every backcross to cultivated pigeonpea increased pollen fertility and the formation of mature seeds.Special project assistant till December, 2003.     

Interspecific cross of Brassica oleracea var. alboglabra and B. napus : effects of growth condition and silique age on the efficiency of hybrid production, and inheritance of erucic acid in the self-pollinated backcross generation

Interspecific hybrids were produced from reciprocal crosses between Brassica napus (2n = 38, AACC) and B. oleracea var. alboglabra (2n = 18, CC) to introgress the zero-erucic acid alleles from B. napus into B. oleracea. The ovule culture embryo rescue technique was applied for production of F₁ plants. The effects of silique age, as measured by days after pollination (DAP), and growth condition (temperature) on the efficiency of this technique was investigated. The greatest numbers of hybrids per pollination were produced under 20°/15°C (day/night) at 16 DAP for B. oleracea (♀) × B. napus crosses, while under 15°/10°C at 14 DAP for B. napus (♀) × B. oleracea crosses. Application of the ovule culture technique also increased the efficiency of BC₁ (F₁ × B. oleracea) hybrid production by 10-fold over in vivo seed set. The segregation of erucic acid alleles in the self-pollinated backcross generation, i.e. in BC₁S₁ seeds, revealed that the gametes of the F₁ and BC₁ plants carrying a greater number of A-genome chromosomes were more viable. This resulted in a significantly greater number of intermediate and a smaller number of high-erucic acid BC₁S₁ seeds.     

Interspecific hybrids between helianthus PetiolarisNutt. and H. Annuus L.: Effect of backcrossing on meiosis

Summary Interspecific substitutions of the nucleus of Helianthus annuus (2n=34) into the cytoplasm of H. petiolaris (2n=34) were obtained by successive backcrossing using cultivated sunflower, H. annuus, as the recurrent pollen parent.Meiosis in the F₁ was characterized by multivalents, suggesting that 10 of the 34 chromosomes were heterozygous for chromosomal interchanges. An additional pair of chromosomes also contained a paracentric inversion. Continued backcrossing resulted in rapid elimination of the meiotic aberrations evident in the F₁. In the BC₁, 1 of 11 plants had normal meiosis and by the BC₂, only 13 of 54 plants had meiotic aberrations similar to those of the F₁. However, trisomic progeny (2n=35) were found in 3 of the 11 BC₁ plants and 20 of the 54 BC₂ plants. No meiotic aberrations were observed in BC₃ or BC₄ plants. Plants with indehiscent anthers, and considered to be male sterile (M.S.), first occurred in the BC₁ and, by the BC₂, 51 of 54 plants were M.S. All 19 BC₃ and 16 BC₄ plants were M.S. Preliminary investigations suggest that the pollen from such plants is sterile and that the sterility is cytoplasmic rather than genetic.Disc-flower measurements were a useful technique for selecting samples at the correct stage of microsporogenesis, but could not be used to distinguish between successive backcrosses.     

The genetic relations between length of time to germination and seed dormancy in lisianthus (Eustoma grandiflorum)

Summary The inheritance of speed of germination and its genetic relations with seed dormancy was investigated in lisianthus (Eustoma grandiflorum). The study was based on data from parental, F1, BC1F1 and F2 generations of a cross between a normally germinating genotype (P1) and a chill-requiring genotype (P2). The mean post-chilling germination speed of P2 was considerably lower than that of P1. Germination speed was found to be under nuclear embryonic control. Analysis of generation means for prechilled seeds revealed an additive gene action with complete dominance of the alleles conferring higher germination speed, since the means of the F1 and the BC1 (P1) were indentical to that of P1. Such dominance was not found for unchilled seeds, in which the mean germination speed of the F1 and the BC1 (P1) was lower than that of P1. It was hypothesized that slow germination speed was induced by pleiotropic effects of seed dormancy alleles. Seed prechilling seemed to eliminate these effects in progeny heterozygous for dormancy alleles, but not in progeny homozygous for dormancy alleles.Contribution from the Agricultural Research Organization, The Volcani Center, Bet Dagan, No 1292-E, 1993 series.     

Increasing biodiversity of irrigated rice in Africa by interspecific crossing of Oryza glaberrima (Steud.) × O. sativa indica (L.)

The African rice Oryza glaberrima, traditionally cultivated since more than 3.500 years, is of poor agronomic performance but resistant/tolerant to various stresses and diseases. The introduction of these characters into O. sativa cultivars is difficult since crossing barriers cause spikelet sterility in F1. Backcrossing can restore fertility and recently facilitated the development of fertile O. glaberrima × O. sativa ssp. japonica hybrid progenies for rain fed systems. With the objective to gain access to African rice germplasm for improvement of irrigated rice, crosses were performed with eighteen O. glaberrima and twenty O. sativa ssp. indica accessions. In total about one hundred F1-hybrid grains were obtained. The F1 plants were all completely sterile and backcrossing (BC) to O. sativa was performed in order to restore spikelet fertility. Monitoring of Tog5681 × IR64 hybrid progenies under field conditions revealed a broad genetic diversity within the BC₁ and BC₂ populations. Some BC₁ and BC₂ progeny plants outperformed the O. sativa parent, indicating that the heterozygocity level and complementary gene action after two backcrosses are still sufficient to positively influence plant vigor. Spikelet fertility of progenies was highly variable, but almost complete fertility was already observed within the BC₁F₂ population. High spikelet fertility was preserved in one out of two analyzed BC₁F₃ families and inmost of the BC₂F₃ families. The ability to restore spikelet fertility within few generations and the potential of the genetic diversity present in interspecific progenies facilitates the development of plant types specifically designed for the African irrigated and lowland environment. This revised version was published online in August 2006 with corrections to the Cover Date.     

Effect of selection for pollen grain size on various traits in common bean

Summary Selection among microgametophytes usually exploits variation in pollen grain germination. Studies of variation in pollen grain size in common bean (Phaseolus vulgaris L.) suggested that selection for size might lead to changes in sporophytic traits. To determine whether microgametophytic selection based on size would affect pollen grain size in subsequent generations or sporophytic traits that were correlated with pollen grain size, pollen grains from three crosses were separated into two size categories by sieving and then used to pollinate cv. Diacol Calima. Selection resulted in changes in pollen grain diameter for pollen from F₁, F₂ and F₃ plants for all crosses. In vitro germination indicated no differences between vigor of large and small grains, but extraction and sieving reduced germinability. F₁ seed from two of the crosses with size-selected pollen varied in weight according to pollen grain size, but in subsequent generations, the effect disappeared. Both size categories of selected pollen resulted in F₂ progeny with reduced numbers of seeds per pod as compared to controls, suggesting that the size selection process may have resulted in indirect selection for traits reducing seed set. The overall results suggested that genes determining pollen grain size in bean have little or no effect on sporophytic traits such as seed size and seed yield.     

Salt-tolerance in Brassica juncea L. I. In vitro selection,agronomic evaluation and genetic stability

Summary In vitro selection of salt tolerant plants of Brassica juncea L. (Indian mustard) cv. Prakash has been accomplished by screening highly morphogenic cotyledon explant cultures on high NaCl media. Out of a total of 2,620 cotyledons cultured on high salt medium, 3 survived, showed sustained growth and regenerated shoots. They were multiplied by axillary bud culture on NaCl free medium. The salt-selected shoots retained salt tolerance following 3 month of growth and multiplication on control medium. While two of these somaclones flowered and set seeds, third one grew slowly, had abnormal leaf morphology and was sterile. The seed of the two fertile plants were sown in the field to raise R₁ segregating generation. Data were recorded for field, other agronomic components and oil content. The somaclonal lines, both selected salt-tolerant and non-selected, showed tremendous amount of variation for all the characters studied. One of the two tolerant somaclones invariably showed reduced height, longer reproductive phase and higher 1000 seed weight. Based on the agronomic performance of R₁ plants of these somaclones, some plants were selected and their progeny were evaluated for agronomic performance under standard field conditions and salt-tolerance in the greenhouse using sand pot culture method. Most of the lines bred true for their specific characteristics. In the greenhouse, selected salt-tolerant somaclones (SR-2 and SR-3) performed better for plant growth, yield and other agronomic traits at higher salt treatments, indicating thereby that salt-tolerance trait selected in vitro was expressed in the whole plants and is genetically stable and transmitted onto the progeny. The two tolerant lines, however, differed in their salt-tolerance during vegetative and reproductive phases as indicated by their salt-tolerance and stress susceptibility indices. The mechanism of salt-tolerance is not clear and needs to be further investigated.     

Inheritance of black leaf mold resistance in tomato

Summary Inheritance of black leaf mold (BLM) (caused by Pseudocercospora fuligena) resistance was studied in four crosses involving two resistant Lycopersicon accessions (PI134417, L. hirsutum and PI254655, L. esculentum) and four susceptible Asian Vegetable Research and Development Center tomato lines (CLN657BC₁F₂-267-0-3-12-7, CL143-0-10-3-0-1-10, CLN698BC₁F₂-358-4-13 and CL5915-93D₄-1-0-3). For each cross, six generations, i.e. P₁, P₂, F₁, F₂, BC₁F₁ and BC₁F₂ were evaluated following inoculations with isolate Pf-2 of P. fuligena. Chi-square analyses of the data based on the ratio of resistant to susceptible plants in the F₂ in three of four crosses gave a good fit to a segregation ratio of 1 R : 15 S, and BC₁F₂ data in three of four crosses gave an acceptable fit to the segregation ratio of 1 R : 63 S. The results indicate that resistance to BLM may be conditioned by two recessive genes acting epistatically in both PI134417 and PI254655.     

A novel source of resistance to the two-spotted spider mite in Lycopersicon pimpinellifolium (Jusl.) Mill.: its genetics as affected by interplot interference

We have previously found an accession of Lycopersicon pimpinellifolium (Jusl.) Mill. (`TO-937') that appeared to resist attack by the two-spotted spider mite (Tetranychus urticae Koch). L. pimpinellifolium is a very close relative of the cultivated tomato (Lycopersicon esculentum Mill.) and thereby a potential source of desirable traits that could be introgressed to the crop species. The objective of this study was to investigate the genetics of the resistance present in `TO-937'. Resistance to infestation by the spider mite was quantified in 24-plant plots of L. pimpinellifolium accessions `TO-937' and `PE-10', L. pennellii accession `PE-45', L. esculentum cultivars `Moneymaker', `Roma' and `Kalohi' (reported to be partially resistant: Stoner & Stringfellow, 1967), and the interspecific F₁ cross, L. esculentum `Moneymaker' × L. pimpinellifolium `TO-937'. Only `TO-937', the F<sub>1</sub>, and`PE-45' were found to be resistant. Resistance of `TO-937' was complete when evaluated in two small greenhouses completely planted with `TO-937' so as to simulate the genotypic homogeneity usual in commercial crops. Generations (P₁, P₂, F₁, F₂, BC₁P₁, and BC₁P₂) of a P₁ (susceptible) × P₂ (resistant) cross (`Moneymaker' × `TO-937') were studied for resistance in a single-plant per plot design. Resistance of `TO-937' was inherited with complete dominance and appeared to be controlled by either two or four genes according to whether segregation in the F₂ or the BC₁P₁, respectively, were considered. However, calculation of the number of genes involved in the resistance was complicated by negative interplot interference due to the high frequency of resistant genotypes within most of the generations. This revised version was published online in July 2006 with corrections to the Cover Date.     

Monogenic resistance in tomato to Fusarium oxysporum f. sp. lycopersici race 3

Summary Resistance to fusarium wilt, incited by Fusarium oxysporum (Schlecht.) f. sp. lycopersici (Sacc.) Snyder & Hansen race 3 in tomato (Lycopersicon esculentum Mill.) was discovered in LA 716, a L. pennellii accession. A resistant BC₁F₃ breeding line, E427, was developed from LA 716. E427 was crossed with the susceptible cv. Suncoast and F₁, BCP₁, BCP₂ (to Fla 7155, a susceptible parent) F₂, F₃, and BCP₂S₁ seeds were obtained. Segregation for resistance following root dip inoculation over three experiments indicated a single dominant gene controlled resistance. Five of the 12 BCP₁S₁'s segregated more susceptible plants, whereas one of the 12 segregated more resistant plants than expected (P<0.05). Three of 23 F₃ lines segregated more susceptible plants than expected while 1 of the 23 had more resistant plants than expected (P<0.05). Segregation in all other lines fit expected ratios. Five of the 23 F₃'s were homozygous resistant which was an acceptable fit to expectations (P=0.1–0.5). The gene symbol I ₃ is proposed for resistance to race 3 of the wilt pathogen. Deviations from expected ratios in data reported here and for other breeding lines indicate an effect of modifier genes and/or incomplete penetrance. Plant age at inoculation and seed dormancy did not affect results.Florida Agricultural Experiment Station Journal Series No. 8101.