Marker-assisted selection to introgress rice QTLs controlling root traits into an Indian upland rice variety

A marker-assisted back-crossing (MABC) breeding programme was conducted to improve the root morphological traits, and thereby drought tolerance, of the Indian upland rice variety, Kalinga III. This variety, the recurrent parent in the MABC, had not previously been used for quantitative trait locus (QTL) mapping. The donor parent was Azucena, an upland japonica variety from Philippines. Five segments on different chromosomes were targeted for introgression; four segments carried QTLs for improved root morphological traits (root length and thickness) and the fifth carried a recessive QTL for aroma. Some selection was made at non-target regions for recurrent parent alleles. We describe the selection made in three backcross (BC) generations and two further crosses between BC₃ lines to pyramid (stack) all five target segments. Pyramids with four root QTLs were obtained in eight generations, completed in 6 years using 3,000 marker assays in a total of 323 lines. Twenty-two near-isogenic lines (NILs) were evaluated for root traits in five field experiments in Bangalore, India. The target segment on chromosome 9 (RM242-RM201) significantly increased root length under both irrigated and drought stress treatments, confirming that this root length QTL from Azucena functions in a novel genetic background. No significant effects on root length were found at the other four targets. Azucena alleles at the locus RM248 (below the target root QTL on chromosome 7) delayed flowering. Selection for the recurrent parent allele at this locus produced early-flowering NILs that were suited for upland environments in eastern India.     

Mapping QTLs associated with drought avoidance in upland rice

The identification of molecular markers linked to genes controlling drought resistance factors in rice is a necessary step to improve breeding efficiency for this complex trait. QTLs controlling drought avoidance mechanisms were analyzed in a doubled-haploid population of rice. Three trials with different drought stress intensities were carried out in two sites. Leaf rolling, leaf drying, relative water content of leaves and relative growth rate under water stress were measured on 105 doubled haploid lines in two trials and on a sub-sample of 85 lines in the third one. Using composite interval mapping with a LOD threshold of 2.5, the total number of QTLs detected in all trials combined was 11 for leaf rolling, 10 for leaf drying, 11 for relative water content and 10 for relative growth rate under stress. Some of these QTLs were common across traits. Among the eleven possible QTLs for leaf rolling, three QTLs (on chromosomes 1, 5 and 9) were common across the three trials and four additional QTLs (on chromosomes 3, 4 and 9) were common across two trials. One QTL on chromosome 4 for leaf drying and one QTL on chromosome 1 for relative water content were common across two trials while no common QTL was identified for relative growth rate under stress. Some of the QTLs detected for leaf rolling, leaf drying and relative water content mapped in the same places as QTLs controlling root morphology, which were identified in a previous study involving the same population. Some QTL identified here were also located similarly with other QTLs for leaf rolling as reported from other populations. This study may help to chose the best segments for introgression into rice varieties and improvement of their drought resistance.     

Evaluation of near-isogenic lines of rice introgressed with QTLs for root depth through marker-aided selection

Drought is one of the main abiotic constraints in rice. A deep root system contributes efficiently to maintaining the water status of the crop through a stress period. After identifying QTLs affecting root parameters in a doubled-haploid (DH) population of rice derived from the cross IR64/Azucena, we started a marker-assisted backcross program to transfer the Azucena allele at four QTLs for deeper roots (on chromosomes 1, 2, 7 and 9) from selected DH lines into IR64. We selected the backcross progenies strictly on the basis of their genotypes at the marker loci in the target regions up to the BC₃F₂. We assessed the proportion of alleles remaining from Azucena in the non-target areas of the BC₃F₂ plants, which was in the range expected for the backcross stage reached. Twenty nine selected BC₃F₃ near-isogenic lines (NILs) were developed and compared to IR64 for the target root traits and three non-target traits in replicated experiments. Of the three tested NILs carrying target 1, one had significantly improved root traits over IR64. Three of the seven NILs carrying target 7 alone, as well as three of the eigth NILs carrying both targets 1 and 7, showed significantly improved root mass at depth. Four of the six NILs carrying target 9 had significantly improved maximum root length. Five NILs carrying target 2 were phenotyped, but none had a root phenotype significantly different from that of IR64. A re-analysis of the initial data with the composite interval mapping technique revealed two linked QTLs with opposite effects in this area. Some NILs were taller than IR64 and all had a decreased tiller number because of a likely co-introgression of linked QTLs. The usefulness of NILs, the efficiency of marker-aided selection for QTLs and the relationship between root traits are discussed. The NILs with an improved root system will permit testing the importance of root depth for water-limited environments. Received: 17 July 2000 / Accepted: 20 October 2000     

Accuracy of marker-assisted selection with auxiliary traits

Genetic information on molecular markers is increasingly being used in plant and animal improvement programmes particularly as indirect means to improve a metric trait by selection either on an individual basis or on the basis of an index incorporating such information. This paper examines the utility of an index of selection that not only combines phenotypic and molecular information on the trait under improvement but also combines similar information on one or more auxiliary traits. The accuracy of such a selection procedure has been theoretically studied for sufficiently large populations so that the effects of detected quantitative trait loci can be perfectly estimated. The theory is illustrated numerically by considering one auxiliary trait. It is shown that the use of an auxiliary trait improves the selection accuracy; and, hence, the relative efficiency of index selection compared to individual selection which is based on the same intensity of selection. This is particularly so for higher magnitudes of residual genetic correlation and environmental correlation having opposite signs, lower values of the proportion of genetic variation in the main trait associated with the markers, negligible proportion of genetic variation in the auxiliary trait associated with the markers, and lower values of the heritability of the main trait but higher values of the heritability of the auxiliary trait.     

普通野生稻中增产相关QTL的发掘

普通野生稻(Oryza Rufipogon)是重要的遗传资源,发掘其优良等位基因将对水稻遗传改良产生重要影响。文章从以珍汕97为轮回亲本,普通野生稻为供体的BC2F1群体中选择一个与珍汕97表型明显不同的单株BC2F1-15,经过连续自交获得回交重组自交系BC2F5群体。均匀分布于12条染色体的126个多态性SSR(Simplesequence repeats)标记基因型分析,发现BC2F1-15单株在30%的标记位点为杂合基因型;利用该群体共检测到4个抽穗期、3个株高、4个每穗颖花数、2个千粒重和1个单株产量QTL。在第7染色体RM481-RM2区间,检测到抽穗期、每穗颖花数和产量QTL,野生稻等位基因表现增效作用;其他3个每穗颖花数QTL位点,野生稻等位基因也均具有增效作用。结果表明野生稻携带有增产相关的等位基因,这些有利等位基因无疑是水稻遗传改良可资利用的新资源。     

Mapping QTLs of root morphological traits at different growth stages in rice

Roots are a vital organ for absorbing soil moisture and nutrients and influence drought resistance. The identification of quantitative trait loci (QTLs) with molecular markers may allow the estimation of parameters of genetic architecture and improve root traits by molecular marker-assisted selection (MAS). A mapping population of 120 recombinant inbred lines (RILs) derived from a cross between japonica upland rice 'IRAT109' and paddy rice 'Yuefu' was used for mapping QTLs of developmental root traits. All plant material was grown in PVC-pipe. Basal root thickness (BRT), root number (RN), maximum root length (MRL), root fresh weight (RFW), root dry weight (RDW) and root volume (RV) were phenotyped at the seedling (I), tillering (II), heading (III), grain filling (IV) and mature (V) stages, respectively. Phenotypic correlations showed that BRT was positively correlated to MRL at the majority of stages, but not correlated with RN. MRL was not correlated to RN except at the seedling stage. BRT, MRL and RN were positively correlated to RFW, RDW and RV at all growth stages. QTL analysis was performed using QTLMapper 1.6 to partition the genetic components into additive-effect QTLs, epistatic QTLs and QTL-by-year interactions (Q x E) effect. The results indicated that the additive effects played a major role for BRT, RN and MRL, while for RFW, RDW and RV the epistatic effects showed an important action and Q x E effect also played important roles in controlling root traits. A total of 84 additive-effect QTLs and 86 pairs of epistatic QTLs were detected for the six root traits at five stages. Only 12 additive QTLs were expressed in at least two stages. This indicated that the majority of QTLs were developmental stage specific. Two main effect QTLs, brt9a and brt9b, were detected at the heading stage and explained 19% and 10% of the total phenotypic variation in BRT without any influence from the environment. These QTLs can be used in breeding programs for improving root traits.     

A method for marker-assisted selection based on QTLs with epistatic effects

A method for marker-assisted selection (MAS) based on quantitative trait loci (QTLs) with epistatic effects is proposed. The efficiency of such method is investigated by simulations under a wide range of situations. In the presence of epistasis, MAS generally yields longer persistence response than that based exclusively on additive or additive and dominance. Neglecting epistasis could result in considerable loss in response, and more pronounced at later generations. In addition to population size and trait heritability, genetic variance configurations play an important role in determining both the short- and long-term efficiencies of MAS. MAS using breeding values not only achieves higher response, but also tends to have smaller standard error than other methods in most cases. Errors in QTL detection cause distinct reductions in responses to MAS in most cases. It is thus concluded that verifications of putative QTL and its magnitude of effect and accurate map chromosome location are imperative to realize the potentials of MAS.     

Mapping QTLs associated with drought avoidance in upland rice grown in the Philippines and West Africa

Localizing genes that contribute to drought avoidance in a quantitative way should enable the exploitation of these genes in breeding through marker-assisted selection, and may lead to the discovery of gene identity and function. Between 110 and 176 F₆ recombinant inbred lines from a mapping population derived from a cross of upland rice varieties Bala and Azucena have been evaluated for indicators of drought avoidance in sites in the Philippines and West Africa over two dry seasons. A molecular map with 102 RFLP, 34 AFLP and six microsatellite markers has been used to map (by composite interval mapping) quantitative trait loci (QTLs) for the visual scores of leaf rolling and leaf drying and leaf relative water content. QTLs were mapped for each site and across sites. A total of 17 regions were identified which contained QTLs with a LOD score greater than 3.2. For leaf rolling, Bala was the parent contributing the majority of positive alleles whilst for the other traits, Bala and Azucena contributed more evenly. Six of the 17 regions influenced more than one trait, explaining the phenotypic correlations between traits that were observed. Three QTLs appeared to be specific to the Philippines experiments. One QTL had opposing effects in the Philippines and West Africa. QTLs for relative water content were detected on chromosome 8, congruent with an osmotic adjustment QTL identified in another population. Only three of the QTLs identified here have not been reliably identified in the two other populations that have been screened for drought avoidance. By using several populations assessed for drought avoidance in different sites, the distribution and utility of QTLs for drought avoidance in rice is being elucidated.     

Locating QTLs controlling several adult root traits in an elite Chinese hybrid rice

This study aimed to elucidate the genetics of the adult root system in elite Chinese hybrid rice. Several adult root traits in a recombinant inbred line (RIL) population of Xieyou 9308 and two backcross F₁ (BCF₁) populations derived from the RILs were phenotyped under hydroponic culture at heading stage for quantitative trait locus (QTL) mapping and other statistical analysis. There a total of eight QTLs detected for the root traits. Among of them, a pleiotropic QTL was repeatedly flanked by RM180 and RM5436 on the short arm of chromosome 7 for multiple traits across RILs and its BCF₁ populations, accounting for 6.88% to 25.26% of the phenotypic variances. Only additive/dominant QTLs were detected for the root traits. These results can serve as a foundation for facilitating future cloning and molecular breeding.     

Simultaneous marker-assisted selection for multiple traits in autogamous crops

A method is presented for the selection of parents with the aim of obtaining improved genotypes in the progeny of a cross. The procedure is designed to select in several unrelated traits simultaneously and is based on the selection of molecular markers that are linked to QTLs. The method was compared with conventional phenotypic selection in simulation experiments for a number of genetic structures underlying the traits and several types of parental populations. Although the method in general provides good results, some of the underlying assumptions may be violated quite easily, thereby reducing the applicability of the procedure in practice. Received: 10 September 1999 / Accepted: 24 August 2000     

Molecular marker-assisted selection for malting quality traits in barley

Selection for malting quality in breeding programs by micromalting and micromashing is time-consuming, and resource-intensive. More efficient and feasible approaches for identifying genotypes with good malting quality would be highly desirable. With the advent of molecular markers, it is possible to map and tag the loci affecting malting quality. The objective of this study was to assess the effectiveness of molecular marker assisted selection for malting quality traits. Two major quantitative trait loci (QTL) regions in six-row barley for malt extract percentage, -amylase activity, diastatic power, and malt -glucan content on chromosomes 1 (QTL1) and 4 (QTL2) have been previously identified. The flanking markers, Brz and Amy2, and WG622 and BCD402B, for these two major QTL regions were used in marker-assisted selection. Four alternative selection strategies; phenotypic selection, genotypic selection, tandem genotypic and phenotypic selection, and combined phenotypic and genotypic selection, were compared for both single and multiple trait selection in a population consisting of 92 doubled haploid lines derived from Steptoe × Morex crosses. Marker assisted selection for QTL1 (tandem genotypic and phenotypic selection, and combined phenotypic and genotypic selection) was more effective than phenotypic selection, but for QTL2 was not as effective as phenotypic selection due to a lack of QTL2 effects in the selection population. The effectiveness of tandem genotypic and phenotypic selection makes marker assisted selection practical for traits which are extremely difficult or expensive to measure such as most malting quality traits. It can substantially eliminate undesirable genotypes by early genotyping and keeping only desirable genotypes for later phenotypic selection.     

QTLs associated with chlorimuron ethyl sensitivity in soybean: Effects on seed yield and related traits

 Soybean, Glycine max (L.) Merr., genotypes are known to differ in chlorimuron ethyl sensitivity (CS). Earlier we have reported two putatively independent marker loci linked to two quantitative trait loci (QTLs) controlling CS in a soybean population derived from a cross of PI97100 (sensitive to chlorimuron ethyl) and ‘Coker 237’ (tolerant to chlorimuron ethyl). The objective of the present study was to quantify the association of the two marker loci with seed yield and related traits in this soybean population following application of chlorimuron ethyl. Phenotypic data were collected for 111 F₂-derived lines of the cross grown in replicated plots at Athens, G.A., in 1994 and 1995, and at Blackville, S.C., in 1995. The two CS marker loci explained as much as 50% of the genetic variation in seed yield and seed number m^-2, but had no association with seed weight, plant height, lodging, seed protein, and seed oil. There were no epistatic interactions between the two marker loci for any of the traits. The marker locus (cr168-1 on USDA linkage group E) linked to the major CS QTL explained between 13 and 23% of the variation in seed yield. The Coker 237 allele at this locus was associated with decreased CS and increased seed yield. The marker locus (Blt015-2 on an unknown linkage group) linked to the minor CS QTL accounted for a maximum of 11% of the variation in seed yield. The Coker 237 allele at this locus was associated with an increase in CS and a decrease in seed yield. The association of the two marker loci with seed number m^-2 strongly resembled their association with seed yield. Seed yield had a strong positive correlation (r=0.74 – 0.94) with seed number m^-2, and the effect of chlorimuron ethyl on seed yield was due mainly to its effect on seed number m^-2 rather than seed weight. Received: 6 August 1996 / Accepted: 28 February 1997     

Molecular marker-assisted selection for enhanced yield in malting barley

Brewers are reluctant to change malting barley (Hordeum vulgare ssp. vulgare L.) cultivars due to concerns of altered flavor and brewing procedures. The U.S. Pacific Northwest is capable of producing high yielding, high quality malting barley but lacks adapted cultivars with desirable malting characteristics. Our goal was to develop high yielding near isogenic lines that maintain traditional malting quality characteristics by transferring quantitative trait loci (QTL) associated with yield, via molecular marker-assisted backcrossing, from the high yielding cv. Baronesse to the North American two-row malting barley industry standard cv. Harrington. For transfer, we targeted Baronesse chromosome 2HL and 3HL fragments presumed to contain QTL that affect yield. Analysis of genotype and yield data suggests that QTL reside at two regions, one on 2HL (ABG461C-MWG699) and one on 3HL (MWG571A-MWG961). Genotype and yield data indicate that additional Baronesse genome regions are probably involved, but need to be more precisely defined. Based on yield trials conducted over 22 environments and malting analyses from 6 environments, we selected one isogenic line (00-170) that has consistently produced yields equal to Baronesse while maintaining a Harrington-like malting quality profile. We conclude there is sufficient data to warrant experiments testing whether the 2HL and 3HL Baronesse QTL would be effective in increasing the yield of other low yielding barley cultivars.     

Mapping QTLs for root traits in a recombinant inbred population from two indica ecotypes in rice

Evaluation of root traits in rainfed lowland rice is very difficult. Molecular genetic markers could be used as an alternative strategy to phenotypic selection for the improvement of rice root traits. This research was undertaken to map QTLs associated with five root traits using RFLP and AFLP markers. Recombinant inbred lines (RILs) were developed from two indica parents, IR58821–23-B-1–2-1 and IR52561-UBN-1–1-2, that were adapted to rainfed lowland production systems. Using wax-petrolatum layers to simulate a hardpan in the soil, 166 RILs were evaluated for total root number (TRN), penetrated root number (PRN), root penetration index (RPI, the ratio of PRN to TRN), penetrated root thickness (PRT) and penetrated root length (PRL) under greenhouse conditions during the summer and the fall of 1997. A genetic linkage map of 2022 cM length was constructed comprising 303 AFLP and 96 RFLP markers with an average marker space of 5.0 cM. QTL analysis via interval mapping detected 28 QTLs for these five root traits, which were located on chromosomes 1, 2, 3, 4, 6, 7, 10 and 11. Individual QTLs accounted for between 6 and 27% of the phenotypic variation. Most of the favorable alleles were derived from the parent IR58821–23-B-1–2-1, which was phenotypically superior in root traits related to drought resistance. Three out of six QTLs for RPI were detected in both summer and fall experiments and they also were associated with PRN in both experiments. Out of eight QTLs for RPT, five were common in both seasons. Two genomic regions on chromosome 2 were associated with three root traits (PRN, PRT and RPI), whereas three genomic regions on chromosomes 2 and 3 were associated with two root traits (PRT and RPI). Two QTLs affecting RPI and two QTLs affecting PRT were also found in similar genomic regions in other rice populations. The consistent QTLs across genetic backgrounds and the common QTLs detected in both experiments should be good candidates for marker-assisted selection toward the incorporation of root traits in a drought resistance breeding program, especially for rainfed lowland rice. Received: 17 November 1999 / Accepted: 19 March 2000     

Mix of trinexapac-ethyl and nitrogen application to reduce upland rice plant height and increase yield

Plant Growth Regulation - Techniques to increase grain production such as nitrogen fertilization and plant growth regulators are essential to enable and enhance rice yield. Therefore, our goal was...     

Mapping of QTLs governing agronomic and yield traits in chickpea

Chickpea is one of the most important leguminous cool season food crops, cultivated prevalently in South Asia and Middle East. The main objective of this study was to identify quantitative trait loci (QTLs) associated with seven agronomic and yield traits in two recombinant inbred line populations of chickpea derived from the crosses JG62 × Vijay (JV population) and Vijay × ICC4958 (VI population) from at least three environments. Single locus QTL analysis involved composite interval mapping (CIM) for individual traits and multiple-trait composite interval mapping (MCIM) for correlated traits to detect pleiotropic QTLs. Two-locus analysis was conducted to identify the main effect QTLs (M-QTLs), epistatic QTLs (E-QTLs) and QTL × environment interactions. Through CIM analysis, a total of 106 significant QTLs (41 in JV and 65 in VI populations) were identified for the seven traits, of which one QTL each for plant height and days to maturity was common in both the populations. Six pleiotropic QTLs that were consistent over the environments were also identified. LG2 in JV and LG1a in VI contained at least one QTL for each trait. Hence, concentrating on these LGs in molecular breeding programs is most likely to bring simultaneous improvement in these traits.     

Identification of main effects, epistatic effects and their environmental interactions of QTLs for yield traits in rice

Quantitative trait loci (QTLs) controlling yield and yield components were identified by using a doubled haploid (DH) population of 120 lines from a sub-specific cross between ‘Samgang’ (Indica) and ‘Nagdong’ (Japonica). Main effects, epistatic effects, their environment interactions of QTLs were analyzed via mixed linear model approach across different environments. A total of 17 putative QTLs were identified on 8 chromosomes and five QTLs were detected over two years. 7 QTLs of main effects and 23 epistatic interactions were observed for five traits. Epistatic interactions played an important role in controlling the expression of yield related traits. The epistatic effects explained higher percentages of phenotype variation for panicles per plant, seed set percentage and yield. Significant QTL×environment (QE) interactions effects were identified for all traits, including 5 main effect QTLs. However, the present study failed to identify the significant interactions between epistatic loci containing main effect QTLs and the environment. The information provided in the present study could be used in the marker-assisted selection to enhance selection efficiency and to improve yield in rice.     

Genetic enhancement for semi-dwarf and bacterial blight resistance with enhanced grain quality characteristics in traditional Basmati rice through marker-assisted selection

Ranbir Basmati is one of the traditional Basmati varieties of India and of the most popular traditional Basmati variety grown in Jammu's region (State of Jammu & Kashmir). It is a tall and short-duration variety with strong aroma and excellent cooking quality. However, it is susceptible to bacterial blight (BB) disease caused by Xanthomonas oryzae pv oryzae (Xoo) and prone to lodging. In this study, semi-dwarf (sd1) and BB resistance genes (Xa21 and xa13) were introgressed into Ranbir Basmati using marker-assisted backcross breeding (MABB) scheme. A high-yielding PAU148 carrying Xa21, xa13 and sd1 genes was used as a donor parent. On each generation target, genes were selected, while polymorphic SSR markers were used to select plants having maximum recovery of the recurrent genome. The maximum genome recovery of Ranbir Basmati in BC₂F₂ was 86.9% in introgressed line SBTIL121. The genotypes carrying resistant genes exhibited very high levels of tolerance against BB disease along with good Basmati rice grain quality traits. The agronomic traits of introgressed lines evaluated in the field and the laboratory showed that most of the agro-morphological traits were similar or superior to Ranbir Basmati. The identified lines can be further evaluated and released as Improved Ranbir Basmati variety.     

Identification of QTLs associated with physiological nitrogen use efficiency in rice

Demand for low-input sustainable crop cultivation is increasing to meet the need for environment-friendly agriculture. Consequently, developing genotypes with high nutrient use efficiency is one of the major objectives of crop breeding programs. This study was conducted to identify QTLs for traits associated with physiological nitrogen use efficiency (PNUE). A recombinant inbred population (DT-RILs) between Dasanbyeo (a tongil type rice, derived from an indica x japonica cross and similar to indica in its genetic make-up) and TR22183 (a Chinese japonica variety) consisting of 166 F8 lines was developed and used for mapping. A frame map of 1,409 cM containing 113 SSR and 103 STS markers with an average interval of 6.5 cM between adjacent marker loci was constructed using the DT-RILs. The RILs were cultivated in ordinary-N (N-P2O5-K2O = 100-80-80 kg/ha) and low-N (N-P2O5-K2O= 50-80-80 kg/ha) (100 kg/ha) conditions. PNUE was positively correlated with the harvest index and grain yield in both conditions. Twenty single QTLs (S-QTLs) and 58 pairs of epistatic loci (E-QTLs) were identified for the nitrogen concentration of grain, nitrogen concentration of straw, nitrogen content of shoot, harvest index, grain yield, straw yield and PNUE in both conditions. The phenotypic variance explained by these S-QTLs and E-QTLs ranged from 11.1 to 44.3% and from 16.0% to 63.6% , respectively. The total phenotypic variance explained by all the QTLs for each trait ranged from 35.8% to 71.3%, showing that the expression of PNUE and related characters depends significantly upon genetic factors. Both S-QTLs and E-QTLs may be useful for marker-assisted selection (MAS) to develop higher PNUE genotypes.     

Comparative analysis of marker-assisted and phenotypic selection for yield components in cucumber

Theoretical studies suggest that marker-assisted selection (MAS) has case-specific advantages over phenotypic selection (PHE) for selection of quantitative traits. However, few studies have been conducted that empirically compare these selection methods in the context of a plant breeding program. For direct comparison of the effectiveness of MAS and PHE, four cucumber (Cucumis sativus L.; 2n = 2x = 14) inbred lines were intermated and then maternal bulks were used to create four base populations for recurrent mass selection. Each of these populations then underwent three cycles of PHE (open-field evaluations), MAS (genotyping at 18 marker loci), and random mating without selection. Both MAS and PHE were practiced for yield indirectly by selecting for four yield-component traits that are quantitatively inherited with 2–6 quantitative trait loci per trait. These traits were multiple lateral branching, gynoecious sex expression (gynoecy), earliness, and fruit length to diameter ratio. Both MAS and PHE were useful for multi-trait improvement, but their effectiveness depended upon the traits and populations under selection. Both MAS and PHE provided improvements in all traits under selection in at least one population, except for earliness, which did not respond to MAS. The populations with maternal parents that were inferior for a trait responded favorably to both MAS and PHE, while those with maternal parents of superior trait values either did not change or decreased during selection. Generally, PHE was most effective for gynoecy, earliness, and fruit length to diameter ratio, while MAS was most effective for multiple lateral branching and provided the only increase in yield (fruit per plant). Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.