Mapping of a gene determining linolenic acid concentration in rapeseed with DNA-based markers

Rapeseed ranks third in world oil production. An important breeding objective to improve oil quality in this crop is to lower linolenic acid concentration in the seeds. Previous reports indicate that the concentration of this acid in Brassica napus is determined by two or three nuclear genes. Using DNA-based markers, we have successfully mapped a gene determining linolenic acid concentration in an F₂ population derived from crossing the cultivar Duplo and alow linolenic acid line, 3637-1. Linolenic acid concentration in this population ranged from 2.1% to 10.5% with-amean of 6.2%. A RAPD marker, K01-1100, displayed significantly different frequencies between two subpopulations consisting of either high or low linolenic acid concentration individuals sampled from the two extremes of the F₂ distribution. Marker K01-1100 segregated in a codominant fashion when used as an RFLP probe on DNA from individuals of this F₂ population. The linolenic acid concentration means for the three resulting RFLP genotypes in the F₂ population were 4.8% (homozygous 3637-1 allele), 6.4% (heterozygous), and 7.5% (homozygous Duplo allele), respectively. It is estimated that this marker accounts for 26.5% of the genetic variation of linolenic acid concentration in this population.     

Inheritance of low linolenic acid content of the seed oil of a mutant in Glycine max

Summary Linolenic acid content of the oil from F₁, F₂, and F₃ seeds was compared with the parental values from a cross between a soybean cultivar with high (7.0%) and a mutant line with low (3.4%) linolenate (183). Linolenic acid content of F₁ seeds was intermediate to that of selfed seeds from the two parents and values from reciprocal crosses were essentially the same. This demonstrated that in this cross, linolenic acid content of the oil was controlled by the embryo rather than by the maternal parent. The distribution of linolenic acid in F₂ seeds from F₁ plants was trimodal and extended across the range of parental values. High and low linolenate F₂ plants bred true for 183 content and the F₃ distribution of seeds from F₂ plants with intermediate levels of 183 was similar to the F₂ distribution. The data were consistent with a model for two alleles with additive effects at a single locus controlling percent linolenic acid in these progenies. The simply-inherited alleles for low linolenate could be readily transferred to agronomically superior soybean cultivars, which would improve the fatty acid composition of the oil.Cooperative Investigations of the Northern Regional Research Center, ARS, USDA, Peoria, Illinois and the Purdue Univ. Agric. Exp. Stn. Journal Paper No. 10,020 of the Purdue Univ. Agric. Exp. Stn.     

Inheritance of seed dormancy in Cummis sativus var. hardwickii (Royle) Alef.

Summary Reciprocal matings were made between two Cucumis sativus var. sativus L. inbred lines (WI 1606 and WI 2808) and two var. hardwickii (Royle) Alef. accessions (PI 215589 and PI 183967). Each case produced a series of reciprocal F₁, F₂, and BC₁ and BC₂ progenies which were used to evaluate seed dormancy in var. hardwickii. Under controlled conditions (25°±1°C and 85%±5% RH; 12 h fluorescent light, 30 mol s^–1 m^–2), no seed dormancy was observed in the var. sativus inbred lines 36 days following seed extraction from fruit. With rare exception, var. hardwickii accessions were dormant for at least 60 days. Seed dormancy in the F₁ was absent 36 days post extraction, indicating that dormancy in var. hardwickii is conditioned by recessive genes present. Seed of some F₁ progeny germinated between 36 and 50 days post-extraction, indicating the presence of transient dormancy or the more variable expression of the dormancy of var. hardwickii. No significant reciprocal differences in either germination rate or percentage were detected in either of the F₁ and F₂ progeny sets, suggesting lack of cytoplasmic or maternal control over these traits. It was estimated that three to seven factors or loci are involved with the expression of this trait depending on the method of calculation, and that a complex interaction between embryonic and non-embryonic tissue exists. Least square estimates indicate that both additive and dominance effects were important in the expression of dormancy. Comparison of theoretical geometric and arithmetic F₂ means to observed F₂ means also suggests that non-additive gene action contributes substantially to the observed variation. Broad-sense heritability ranged between 78 and 95%.     

The QTL analysis on maternal and endosperm genome and their environmental interactions for characters of cooking quality in rice (<Emphasis Type="Italic">Oryza sativa</Emphasis> L<Emphasis Type="Italic">.</Emphasis>)

Investigations to identify quantitative trait loci (QTLs) governing cooking quality traits including amylose content, gel consistency and gelatinization temperature (expressed by the alkali spread value) were conducted using a set of 241 RIL populations derived from an elite hybrid cross of “Zhenshan 97” × “Minghui 63” and their reciprocal backcrosses BC₁F₁ and BC₂F₁ populations in two environments. QTLs and QTL × environment interactions were analyzed by using the genetic model with endosperm and maternal effects and environmental interaction effects on quantitative traits of seed in cereal crops. The results suggested that a total of seven QTLs were associated with cooking quality of rice, which were subsequently mapped to chromosomes 1, 4 and 6. Six of these QTLs were also found to have environmental interaction effects.     

Mapping loci controlling the concentrations of erucic and linolenic acids in seed oil of Brassica napus L.

The quality of plant oil is determined by its component fatty acids. Relatively high levels of linolenic acid reduce the oxidative stability of the oil, and high levels of erucic acid in the diet have been associated with health problems. Thus, oilseed Brassica napus cultivars with low linolenic and low erucic acid contents are highly desirable for edible oil production. In order to identify genes controlling the levels of erucic and linolenic acids, we analyzed the oil composition of 99 F₁-derived doubled haploid lines from a cross between cv Major (high levels of erucic and linolenic acids) and cv Stellar (low levels of both fatty acids). A molecular marker linkage map of 199 loci for this population was used to identify quantitative trait loci (QTL) controlling oil composition. We identified two regions that accounted for nearly all of the phenotypic variation in erucic acid concentration and one region that accounted for 47% of the variation in linolenic acid concentration. The QTL associated with linolenic acid concentration mapped near a RFLP locus detected by a cDNA clone encoding an omega-3 desaturase, suggesting that the low linolenic acid content of Stellar may be due to a mutation in this gene.     

Development of monosomic alien addition lines and introgression of genes from Oryza australiensis Domin. to cultivated rice O. sativa L.

Oryza australiensis, a diploid wild relative of cultivated rice, is an important source of resistance to brown planthopper (BPH) and bacterial blight (BB). Interspecific hybrids between three breeding lines of O. sativa (2n=24, AA) and four accessions of O. australiensis (2n=24, EE) were obtained through embryo rescue. The crossability ranged from 0.25% to 0.90%. The mean frequency of bivalents at diakinesis/metaphase I in F₁ hybrids (AE) was 2.29 to 4.85 with a range of 0–8 bivalents. F₁ hybrids were completely male sterile. We did not obtain any BC₁ progenies even after pollinating 20,234 spikelets of AE hybrids with O. sativa pollen. We crossed the artificially induced autotetraploid of an elite breeding line (IR31917-45-3-2) with O. australiensis (Acc. 100882) and, following embryo rescue, produced six F₁ hybrid plants (AAE). These triploid hybrids were backcrossed to O. sativa. The chromosome number of 16 BC₁ plants varied from 28 to 31, and all were male sterile. BC₂ plants had 24–28 chromosomes. Eight monosomic alien addition lines (MAALs) having a 2n chromosome complement of O. sativa and one chromosome of O. australiensis were selected from the BC₂ F₂ progenies. The MAALs resembled the primary trisomies of O. sativa in morphology, and on the basis of this morphological similarity the MAALs were designated as MAAL-1, -4, -5, -7, -9, -10, -11, and -12. The identity of the alien chromosome was verified at the pachytene stage of meiosis. The alien chromosomes paired with the homoeologous pairs to form trivalents at a frequency of 13.2% to 24.0% at diakinesis and 7.5% to 18.5% at metaphase I. The female transmission rates of alien chromosomes varied from 4.2% to 37.2%, whereas three of the eight MAALs transmitted the alien chromosome through the male gametes. BC₂ progenies consisting of disomic and aneuploid plants were examined for the presence of O. australiensis traits. Alien introgression was detected for morphological traits, such as long awns, earliness, and Amp-3 and Est-2 allozymes. Of the 600 BC₂ F₄ progenies 4 were resistant to BPH and 1 to race 6 of BB. F₃ segregation data suggest that earliness is a recessive trait and that BPH resistance is monogenic recessive in two of the four lines but controlled by a dominant gene in the other two lines.     

Transfer of bacterial blight and blast resistance from the tetraploid wild rice Oryza minuta to cultivated rice,Oryza sativa

Summary Oryza minuta J. S. Presl ex C. B. Presl is a tetraploid wild rice with resistance to several insects and diseases, including blast (caused by Pyricularia grisea) and bacterial blight (caused by Xanthomonas oryzae pv. oryzae). To transfer resistance from the wild species into the genome of cultivated rice (Oryza sativa L.), backcross progeny (BC₁, BC₂, and BC₃) were produced from interspecific hybrids of O. sativa cv IR31917-45-3-2 (2n=24, AA genome) and O. minuta Acc. 101141 (2n=48, BBCC genomes) by backcrossing to the O. sativa parent followed by embryo rescue. The chromosome numbers ranged from 44 to 47 in the BC₁ progeny and from 24 to 37 in the BC₂ progeny. All F₁ hybrids were resistant to both blast and bacterial blight. One BC₁ plant was moderately susceptible to blast while the rest were resistant. Thirteen of the 16 BC₂ progeny tested were resistant to blast; 1 blast-resistant BC₂, plant 75-1, had 24 chromosomes. A 3 resistant: 1 susceptible segregation ratio, consistent with the action of a major, dominant gene, was observed in the BC₂F₂ and BC₂F₃ generations. Five of the BC₁ plants tested were resistant to bacterial blight. Ten of the 21 BC₂ progeny tested were resistant to Philippine races 2, 3, and 6 of the bacterial blight pathogen. One resistant BC₂, plant 78-1, had 24 chromosomes. The segregation of reactions of the BC₂F₂, BC₂F₃, and BC₂F₄ progenies of plant 78-1 suggested that the same or closely linked gene(s) conferred resistance to races 2, 3, 5, and 6 of the bacterial blight pathogen from the Philippines.     

Inheritance of reduced linolenic acid content in soybean seed oil

 Linolenic acid is the unstable component of soybean [Glycine max (L.) Merr.] oil that is responsible for the undesirable odors and flavors commonly associated with poor oil quality. Two mutants, M-5 and KL-8, have been identified that have lower linolenic acid levels in the seed oil than the ‘Bay’ cultivar. Our objective was to determine the relationships between the genetic systems controlling linolenic acid in these mutants. Reciprocal crosses were made between the mutants and ‘Bay’, and between the two mutants. No maternal effect for linolenic acid content was observed from the analysis of F₁ seeds in any of the crosses. The data for linolenic acid content in F₂ seeds of M-5בBay’ and KL-8בBay’ crosses satisfactorily fit a 1 : 2 : 1 and 3 : 1 ratio, respectively. For the M-5×KL-8 cross, segregation observed from the analysis of F₂ seeds for linolenic acid content satisfactorily fit a ratio of 3 more than either mutant: 12 within the range of the two mutants: 1 less than either mutant. The segregation ratio of F₂ seeds and the segregation of F₃ seeds from F₂ plants indicated that M-5 and KL-8 have alleles at different loci that control linolenic acid content. The allele in KL-8 has been designated as fanx (KL-8) to distinguish it from fan (M-5). The low linolenic acid segregates with the genotype fanfanfanxfanx provide additional germplasm to reduce the linolenic acid content from the seed oil of soybean. Received: 18 December 1995 / Accepted: 12 July 1996     

Molecular Analysis of Transgene (psy) Inheritance in a Golden Rice Line Developed in the Genetic Background of IR64

Stability in transgene inheritance and expression is an important consideration in utilizing transgenic germplasm for the development of commercially viable transgenic crops. In the present study, inheritance of the carotenogenic transgene encoding enzyme phytoene synthase (psy) was studied in rice through PCR analysis in F₂, BC₁F₁ (Swarna X F₁), its reciprocal cross (F₁ X Swarna) and BC₂F₂ populations. Segregation distortion was observed for the transgene in F₂ and BC₁F₁ (Swarna X F₁) populations. However, in the case of reciprocal BC₁F₁ (F₁ X Swarna) cross, the transgene showed normal Mendelian segregation ratio 1:1. This trend was also observed in the BC₂F₂ where the transgene followed an expected seregation ratio of 3:1. Segregation distortion of the transgene has been explained on the basis of the position effect of gametophytic selection locus ga2. The results obtained here are significant for the molecular marker assisted transfer of the golden rice transgene to widely grown Indian rice varieties.     

Mapping of genes affecting linolenic acid content in Brassica rapa ssp. oleifera

F₂ progeny segregating for linolenic acid content were used to identify genes and develop markers for linolenic acid in spring turnip rape (Brassica rapa ssp. oleifera). A candidate gene approach applying the rapeseed fad3 gene and bulked segregant analysis with RAPD markers was used. A total of 27 markers were distributed in three linkage groups which each exhibited a QTL for linolenic acid. Jointly the three QTLs accounted for 73.5% of the variation in linolenic acid level in this population. The fad3 gene was mapped near one QTL controlling 23.5% of the variation. Allele-specific markers were developed for fad3 and can be used for marker-assisted selection in future spring turnip rape breeding programmes.     

In vitro culture of Cucumis sativus L.

Summary The ability to regenerate plants from leaf explants has been tested for three highly inbred cucumber lines (B, G, S), their reciprocal hybrids, F₂ and BC₁ generations. The lines differed from each other in their regenerating ability, which was expressed by the percentage of explants regenerating embryoidal callus and mean number of plantlets per plant. Thus, the lines could be classified as frequently (B), intermediately (G) or occasionally regenerating ones (S). There were no reciprocal cross differences in the regeneration. It was found that the intermediately and intensively regenerating lines contain two pairs of dominant genes responsible for plant regeneration, characterized by complementary and probably additive interaction. The frequently regenerating line differed from the intermediately regenerating in the effect of one gene. It is supposed that the above-mentioned genes belong to three different loci. The ability to regenerate plants from leaf expiants had high heritability.     

Expression of Thinopyrum distichum NORs in wheat×Thinopyrum amphiploids and their backcross generations

Summary The C-banding pattern of the satellited chromosomes in Thinopyrum distichum and Triticum durum was established. Both T. durum and Th. distichum contained two pairs of satellited chromosomes, which could be distinguished from one another. In the amphiploids [T. durum×Th. distichum (2x=56)] and in the backcross T. durum/(T. durum×Th. distichum)², BC₁F₃, and BC₁F₅ (2n = 42) the satellite was visible on only 1B and 6B of T. durum. The vector pTa 71 containing the rRNA gene from wheat hybridized to two pairs of chromosomes (four hybridization sites) in T. durum and Th. distichum, to eight sites in the amphiploid hybrid (2n=56), and to six sites in the backcross populations BC₁F₁. BC₁F₃, and BC₁F₅ (2n=42). The two satellite pairs in Th. distichum could be distinguished by the chromosomal location of the rRNA site (median or subterminal) and by the centromere position. One copy of each pair was present in the BC₁F₁, but in the BC₁F₃ and BC₁F₅ populations the pair with the subterminal location of rRNA genes was absent. Silver nitrate staining indicated that the rRNA genes of T. durum did not completely suppress those of Th. distichum. The octoploid amphiploid (2n = 56) contained a maximum of four large and four small nucleoli and the hexaploid BC₁s (2n=42), four large and two small nucleoli.     

Fine mapping of S32(t), a new gene causing hybrid embryo sac sterility in a Chinese landrace rice (Oryza sativa L.)

Ketan Nangka, the donor of wide compatibility genes, showed sterility when crossed to Tuanguzao, a landrace rice from Yunnan province, China. Genetic and cytological analyses revealed that the semi-sterility was primarily caused by partial abortion of the embryo sac. Genome-wide analysis of the linkage map constructed from the backcross population of Tuanguzao/Ketan Nangka//Ketan Nangka identified two independent loci responsible for the hybrid sterility located on chromosomes 2 and 5, which explained 18.6 and 20.1% of phenotypic variance, respectively. The gene on chromosome 5 mapped to the previously reported sterility gene S31(t), while the gene on chromosome 2, a new hybrid sterility gene, was tentatively designated as S32(t). The BC₁F₂ was developed for further confirmation and fine mapping of S32(t). The gene S32(t) was precisely mapped to the same region as that detected in the BC₁F₁ but its position was narrowed down to an interval of about 1.9 cM between markers RM236 and RM12475. By assaying the recombinant events in the BC₁F₂, S32(t) was further narrowed down to a 64 kb region on the same PAC clone. Sequence analysis of this fragment revealed seven predicted open reading frames, four of which encoded known proteins and three encoded putative proteins. Further analyses showed that wide-compatibility variety Dular had neutral alleles at loci S31(t) and S32(t) that can overcome the sterilities caused by these two genes. These results are useful for map-based cloning of S32(t) and for marker-assisted transferring of the neutral allele in hybrid rice breeding.     

Inheritance of somatic embryogenesis in red clover (Trifolium pratense L.)

 Red clover genotypes capable of regenerating plantlets in vitro from non-meristematic tissue-derived callus are rare. Selection for genotypes capable of somatic embryogenesis identified a clone comprised of a group of plantlets regenerated from a hypocotyl-derived callus culture on L2-based media and another group of plantlets originating from crown divisions of the epicotyl-derived plant. The callus-derived plants of this clone were highly regenerative when reintroduced to callus culture, but the epicotyl-derived plants produced nonregenerative callus cultures. F₁, F₂ and BC₁ populations were evaluated to determine the mode of inheritance of the regeneration trait. Reciprocal crosses did not differ, indicating a lack of maternal effects. Results were compatible with genetic control of regeneration by two complementary genes. We propose the genotype Rn1-Rn2- for regenerative plants. Three petiole segment explants were sufficient to evaluate regenerative ability in seedlings. Regenerative ability was often associated with abnormal leaf morphology in a few to several leaves. Received: 17 February 1998 / Accepted: 7 April 1998     

Embryo development in reciprocal crosses of Phaseolus vulgaris L. and P. coccineus Lam.

Summary Embryo development was examined in reciprocal crosses of Phaseolus vulgaris cv. Great Northern and P. coccineus cv. Scarlet Runner. The formation of abnormal (shrunken and underdeveloped) embryos constituted the primary crossing barrier between the two species when P. coccineus was the female parent. Plants of P. coccineus X P. vulgaris were obtained by embryo culture. Although the P. vulgaris X P. coccineus cross resulted in normal seed development, the fertility of the resulting hybrids was much lower (27%) than that of the reciprocal hybrids (81%). Three classes of F₂ embryos, normal, shrunken, and underdeveloped were formed on reciprocal F₁s and the frequencies did not differ between reciprocal populations. Thus, the interactions between embryo and endosperm and/or maternal parent rather than cytoplasmic-nuclear effects seem to be important in the determination of the extent of embryo growth. The examination of pollen fertility of F₂ plants and the development of F₂ and F₃ embryos suggests that the formation of abnormal embryos and reduced male fertility are independent events. The P. vulgaris — P. coccineus crosses may be useful in studying the possible involvement of interspecific differences in hormonal metabolism in the development of hybrid embryos.     

Specific molecular marker of the genes controlling linolenic acid content in rapeseed

In rapeseed, which is an agronomically important oilseed, variation in the linolenic acid content of the oil has been obtained through chemical mutagenesis treatment. Conventional breeding of this quantitative trait, however requires specific molecular markers. By means of biochemical experiments, we have established that the induced variation in linolenic acid content is associated with the fad3 gene encoding the microsomal ¹⁵ desaturase. Using a pair of primers specific to this gene and a doubled haploid progeny derived from a low linolenic x high linolenic acid F₁hybrid, we have identified a polymorphism of the fad3 alleles between the low- and the high-linolenic acid genotypes. The structure exon/intron of the fad3 DNA sequence seems to be very similar to that of the Arabidopsis fad3 gene. The choice of the primer pair allows specific amplification of one of the two rapeseed fad3 genes. The value and contribution of specific markers to conventional plant breeding is discussed.     

Inheritance of the photoperiodically induced cold acclimation response in Cornus sericea L., red-osier dogwood

Summary Cold acclimation responses of latitudinal ecotypes of Cornus sericea L. (C. stolonifera Michx.) and F₁, F₂ and BC₁ hybrid progenies were measured under natural photoperiod conditions in St. Paul, MN and artificially shortened photoperiods in the glass-house. The 65 °N and 62 °N ecotypes (Alaska and Northwest Territories, respectively) were characterized by a short night length for hardiness induction, the 42 °N ecotype (Utah A and B) by a long night length for hardiness induction, while the F₁ was intermediate to the parents. Results from reciprocal crosses indicated there was no significant unilateral maternal influence on cold acclimation. Acclimation responses of the F₂ were highly variable but generally ranged between the parental extremes. However, three individuals from the 42 ° × 62 °N crosses exhibited greater cold resistance than the northern parent on two successive freezing test dates. F₂ plants were also found with less freezing resistance than the southern parent. Backcrosses to the southern parent produced progeny with acclimation patterns resembling that of the southern parent and were significantly less hardy than the F₂ in early freezing tests.Scientific Journal Series Paper No. 12,075 of the Minnesota Agricultural Experiment Station     

Genetic analysis of Karnal bunt (Neovossia indica) resistance in wheat

Embryos excised from seeds of six generations (P₁, P₂, F₁, BC₁, BC₂ and F₂) of a cross WH 283 x WH 533 were cultured on modified MS medium already inoculated with secondary sporidia ofNeovossia indica. Significant variations for callusing response (CR) (54.55–75.55%) were observed among generations but the presence or absence ofN. indicia did not affect callusing response. A clear inhibition zone (IZ) was formed around each embryo showing callusing. The diameter of IZ varied significantly among generations and was maximum in the resistant genotype, WH 283 (3.60 cm). Fresh weight and dry weight of calli, initiated from embryo cultured and inoculated withN. indica, varied significantly among generations. Coefficient of infection as well as percentage of infection reflected the overdominance of susceptibility. Generation mean analysis showed that the three parameter model was adequate for diameter of IZ only. Six-parameter model showed that additive (in presence ofN. indica), additive and additive x dominance (in absence ofN. indica) effects were also significant. Complementary type of epistasis for fresh weight of calli and dominance, and dominance x dominance effects for dry weight of calli were observed in the presence ofN. indica. Magnitude of additive effects was higher for diameter of IZ in three parameter model. Therefore, selection might assist in improving this trait and thus indirectly help in attaining the resistance towardsN. indica.     

Additive transgene expression and genetic introgression in multiple green-fluorescent protein transgenic crop × weed hybrid generations

The level of transgene expression in crop × weed hybrids and the degree to which crop-specific genes are integrated into hybrid populations are important factors in assessing the potential ecological and agricultural risks of gene flow associated with genetic engineering. The average transgene zygosity and genetic structure of transgenic hybrid populations change with the progression of generations, and the green fluorescent protein (GFP) transgene is an ideal marker to quantify transgene expression in advancing populations. The homozygous T₁ single-locus insert GFP/Bacillus thuringiensis (Bt) transgenic canola (Brassica napus, cv Westar) with two copies of the transgene fluoresced twice as much as hemizygous individuals with only one copy of the transgene. These data indicate that the expression of the GFP gene was additive, and fluorescence could be used to determine zygosity status. Several hybrid generations (BC₁F₁, BC₂F₁) were produced by backcrossing various GFP/Bt transgenic canola (B. napus, cv Westar) and birdseed rape (Brassica rapa) hybrid generations onto B. rapa. Intercrossed generations (BC₂F₂ Bulk) were generated by crossing BC₂F₁ individuals in the presence of a pollinating insect (Musca domestica L.). The ploidy of plants in the BC₂F₂ Bulk hybrid generation was identical to the weedy parental species, B. rapa. AFLP analysis was used to quantify the degree of B. napus introgression into multiple backcross hybrid generations with B. rapa. The F₁ hybrid generations contained 95–97% of the B. napus-specific AFLP markers, and each successive backcross generation demonstrated a reduction of markers resulting in the 15–29% presence in the BC₂F₂ Bulk population. Average fluorescence of each successive hybrid generation was analyzed, and homozygous canola lines and hybrid populations that contained individuals homozygous for GFP (BC₂F₂ Bulk) demonstrated significantly higher fluorescence than hemizygous hybrid generations (F₁, BC₁F₁ and BC₂F₁). These data demonstrate that the formation of homozygous individuals within hybrid populations increases the average level of transgene expression as generations progress. This phenomenon must be considered in the development of risk-management strategies.Communicated by J. Dvorak     

A quantitative nature of somatic embryogenesis in soybean

Summary The objectives of this research were to investigate the genetic parameters associated with the in vitro formation of somatic embryos in soybean and to determine the effect of light intensity on the embryogenic capability of F₁, F₂, and backcross (RC₁P₁ and RC₁P₂) progenies derived from crosses between embryogenic (IAS-5 and Embrapa-1) and non-embryogenic (Paraná) cultivars. Immature cotyledons (4–6 mm in length) derived from the parental lines, F₁, F₂, RC₁P₁, and RC₁P₂ were grown for 90 d on the inductive N10 medium, after which the number of somatic embryos was recorded. Chi-square tests for goodness of fit showed that the genetic component of the somatic embryogenesis trait is controlled in a quantitative manner by approximately 10 genes. A normal distribution for somatic embryo formation in the F₂ generations was observed reinforcing the quantitative nature of the trait. Variation in light intensity (8–12 and 27–33 μmol m⁻²s⁻¹) had no effect on somatic embryo formation in the parental material tested.