Genetic mean components as selection criteria for identifying promising yellow melon hybrids for inbred line extraction

Componentes de média genética como critérios de seleção para identificação de híbridos de melão amarelo promissores para extração de linhagens endogâmicas

Abstract

Estimates of the mean components m + a′ and d represent a promising strategy for guiding the selection of populations and lines in plant breeding programs. This study estimated m + a′ and d to identify commercial yellow melon hybrids suitable for inbred line extraction to develop new hybrids. The experiments evaluated seven commercial single-cross (F₁) yellow melon hybrids and their self-pollinated progenies (F₂), for a total of 14 populations. Experiments were conducted in a randomized complete block design with three replications in a split-plot arrangement, with generation assigned to main plots and hybrids to subplots. Evaluated traits included fruit number per plot, mean fruit mass, yield, flesh thickness, flesh firmness, and total soluble solids. The Mulamba and Mock selection index (MMI) provided joint selection across all traits. Individual and combined analyses of variance applied a split-plot model and treated generation and hybrid effects as fixed. In the F₁ generation, homozygous gene effects (m + a′) predominated and contributed substantially to all traits and the MMI. Estimates of m + a′ and d classified hybrids by success probability in inbred line extraction and showed that inbreeding depression was not significant for all melon traits. Hybrids AF-13395 and Goldmine showed the highest suitability for selecting superior inbred lines.

Keywords:
Cucumis melo L.; plant breeding; homozygosity; genetic mean components; inbreeding depression

Resumo

As estimativas dos componentes de média m + a′ e d fornecem um método para seleção de populações e linhagens em programas de melhoramento genético vegetal. Este estudo estimou

m + a′ e d para identificar híbridos comerciais de melão amarelo adequados para extração de linhagens endogâmicas visando o desenvolvimento de novos híbridos. Os experimentos avaliaram sete híbridos comerciais de cruzamento simples (F₁) de melão amarelo e suas respectivas progênies autopolinizadas (F₂), totalizando 14 populações. Os experimentos utilizaram delineamento de blocos casualizados completos com três repetições em arranjo de parcelas subdivididas, com gerações alocadas nas parcelas principais e híbridos nas subparcelas. As características avaliadas incluíram número de frutos por parcela, massa média de frutos, produtividade, espessura de polpa, firmeza de polpa e sólidos solúveis totais. O índice de seleção de Mulamba e Mock (MMI) permitiu seleção conjunta de todas as características. Análises individuais e combinadas de variância aplicaram modelo de parcelas subdivididas, considerando efeitos de geração e híbrido como fixos. Na geração F₁, efeitos gênicos homozigotos (m + a′) predominaram e contribuíram substancialmente para todas as características e para o MMI. As estimativas de m + a′ e d classificaram os híbridos pela probabilidade de sucesso na extração de linhagens consanguíneas e demonstraram que a depressão por endogamia não foi significativa para todas as características avaliadas em melão. Os híbridos AF-13395 e Goldmine apresentaram maior adequação para seleção de linhagens endogâmicas superiores.

Palavras-chave:
Cucumis melo L.; melhoramento genético vegetal; homozigose; componentes de média genética; depressão por endogamia

1. Introduction

Melon (Cucumis melo L.) ranks among the fruit crops most widely cultivated and consumed worldwide (Cui et al., 2022) and constitutes a primary cucurbit crop in the Brazilian Semiarid region. In Brazil, melon production concentrates in the states of Rio Grande do Norte, Bahia, and Ceará, which account for more than 95% of national production (IBGE, 2024).

Sustaining and expanding melon production in the Northeast region of Brazil requires ongoing breeding to develop new hybrids that exceed current ones. These efforts depend on quantifying genetic components of hybrid performance. The mean value of any trait in a single-cross hybrid (HS) can be expressed as HS = m + a′ + d, where m + a′ represents the contribution of homozygous loci already fixed in the parental inbred lines (corresponding to the mean of the two parents), and d denotes the contribution of heterozygous loci, commonly referred to as heterosis or hybrid vigor (Bernardo, 2020; Ramalho et al., 2024). Superior hybrid development requires productive inbred lines and dominance effects from their crosses.

Extracting high-performing inbred lines requires identifying source populations with high success probability in breeding. Commercial hybrids, subjected to intensive selection for adaptation and performance in regional conditions, serve as genetic sources for extracting inbred lines. In melon breeding, researchers have used commercial hybrids from seed companies to identify populations suitable for inbred line extraction. This practice aligns with the Brazilian Plant Variety Protection Law (Law No. 9,456/1997), which allows inbred line extraction from protected commercial hybrids (Brasil, 1997).

Estimates of the mean components m + a′ and d from commercial hybrids identify those suitable for inbred line development. Brazilian melon hybrids consist mainly of single crosses from two parental lines, with segregating locus frequency of 0.5. Under these conditions, the m + a′ estimate represents the expected mean of inbred lines after complete inbreeding (F∞), without selection. This parameter quantifies segregating population potential from fixed loci in parental lines.

The d estimate quantifies contributions of heterozygous loci and expected genetic variability across inbreeding generations. Higher d values indicate greater heterozygosity in the hybrid and superior inbred lines via selection, as favorable alleles fix in homozygosity and join fixed alleles (Ramalho et al., 2024).

Researchers apply mean component estimates to select populations in breeding programs for self-pollinated crops, including common bean (Phaseolus vulgaris) (Oliveira et al., 1996; Abreu et al., 2002; Carneiro et al., 2002; Mendonça et al., 2002; Rocha et al., 2015) and cowpea (Vigna unguiculata) (Matos et al., 2020). In maize (Zea mays), this method applies to commercial hybrids (Lima et al., 2000; Souza Sobrinho, 2001; Viana et al., 2009). No studies report mean component estimates across multiple commercial melon hybrids.

Most studies on m + a′ and d estimates examine yield due to its economic value in major crops. In melon breeding, yield and fruit quality traits—soluble solids, flesh thickness, and flesh firmness—require joint consideration for market acceptance (Oliveira et al., 2019). Melon fruit, as a consumer product, necessitates the evaluation of yield and quality traits together when selecting populations for inbred line extraction.

This study estimated homozygous (m + a′) and heterozygous (d) locus contributions to agronomic traits in commercial melon (Cucumis melo L.) hybrids to identify those suitable for inbred line extraction and hybrids adapted to the Northeast region of Brazil.

2. Material and Methods

2.1. Experimental conditions

Two experiments were conducted: Experiment 1 from September to November 2017 in Baraúna (5°04′S, 37°37′W; 109 m altitude) and Experiment 2 from August to October 2018 in Alagoinha, Mossoró, Rio Grande do Norte, Brazil (5°03′S, 37°24′W; 70 m altitude). The regional climate is classified as BSh, hot semiarid, according to the Köppen-Geiger classification, with a dry season from June to January and a rainy season from February to May (Alvares et al., 2013). Crop management practices followed recommendations for Rio Grande do Norte, Brazil (Nunes et al., 2016).

2.2. Plant material and experimental design

Seven single-cross commercial yellow melon hybrids (F₁) were evaluated: Iracema, AF-13395, Goldex, Gladial, Goldmine, Mandacaru. and Solares. Self-pollinated progenies (F₂) yielded 14 populations total. Experiments were conducted in a randomized complete block design with three replications in a split-plot arrangement, with generation assigned to main plots and hybrids to subplots. Each plot had seven plants in a single row, spaced 0.4 m apart. Over two years, F₁ and F₂ populations of the seven hybrids formed 14 treatments. Harvest samples of five commercial fruits per plot were evaluated.

2.3. Traits

The agronomic traits evaluated were fruit number per plot, mean fruit mass, yield, flesh thickness, flesh firmness, and soluble solids. Fruit number per plot was determined by a total fruit count. Mean fruit mass (g) was calculated as the average mass of five fruits per plot. Yield (Mg ha⁻¹) was determined by summing the mass all fruits harvested in the plot; fruits were weighed individually on an electronic scale. Flesh thickness (mm) was measured as the mean pericarp thickness of five fruits per plot (two measurements per fruit), using a graduated scale. Flesh firmness (kgf) was measured by two penetration measurements per side of the equatorial flesh region after a transverse cut, using a Wagner penetrometer equipped with an 8.0 mm conical plunger (FT 011 Fruit Pressure Tester). Soluble solids content (°Brix) was determined as the mean from two equatorial flesh pieces (lower two-thirds) of five fruits per plot, pressed into a Palette 100 digital refractometer. The samples were manually pressed directly into a digital refractometer (Palette 100), and results were expressed as the mean of five fruits per plot.

2.4. Statistical analysis

Individual and joint analyses of variance applied a split-plot model to all traits, with generation and hybrid as fixed effects and year as random. The Mulamba and Mock (1978) selection index (MMI) enabled joint multi-trait selection. MMI weights (w) followed farmer input on trait importance: fruit number (w₁ = 0.15); mean fruit mass (w₂ = 0.15); yield (w₃ = 0.30); flesh thickness (w₄ = 0.10); flesh firmness (w₅ = 0.10); and soluble solids (w₆ = 0.20).

The contributions of homozygous and heterozygous loci were estimated using the generation mean components $F_1=m+a^{\prime }+d$ and $F_2=m+a^{\prime }+\frac{1}{2}d$, where m is the general mean, a′ is the additive effect of homozygous loci, and d is the dominance effect of heterozygous loci. From these expressions, the homozygous contribution (m + a′) was calculated as $2{\overline{F}}_2-{\overline{F}}_1$, and the heterozygous contribution (d) as $2\left({\overline{F}}_1-{\overline{F}}_2\right)$. The contributions of homozygous and heterozygous loci were expressed as percentages. The significance of these parameters was assessed via Student’s t-test. Analysis of variance (ANOVA) was performed using the SISVAR^® 5.6 software (Ferreira, 2019). Alls analyses used a significance level of α = 0.05.

3. Results

3.1. ANOVA and inbreeding estimates

Generations differed significantly for fruit number, yield, flesh thickness, and Mulamba-Mock index (MMI) (Table 1). Hybrids differed for all traits (Table 1). The effect of the hybrid × year interaction was significant for fruit number and soluble solids, indicating hybrid rank changes across years. The effect of the hybrid × generation × year interaction was significant for soluble solids (Table 1), with performance varying across generation-year combinations.

Table 1
Snedecor's F-values, overall means, mean square errors, and coefficients of variation from analysis of variance for six traits and Mulamba-Mock index (MMI) in F₁ and F₂ generations of yellow melon hybrids across two years in Baraúna and Mossoró, Rio Grande do Norte, Brazil, 2017–2018.

Source of variation	DF	Snedecor's F-values (Trait)
		FN	MFM	YLD	FT	FF	SS	MMI
Block (Year)	4	8.04*	0.71ns	1.86	5.08ns	0.99ns	1.93ns	0.04
Year (Y)	1	0.02ns	16.69*	69.86**	41.80**	94.99**	4.08ns	0.08
Gen. (G)	1	16.18*	3.25ns	13.28*	14.46*	0.27ns	3.57ns	12.2**
Y × G	1	0.13ns	0.10ns	0.02ns	0.55ns	0.14ns	0.04ns	0.07
Hybrid (H)	6	7.06**	12.01**	10.06**	11.68**	7.24**	7.02**	11.4**
H × Y	6	2.58*	0.51ns	2.23ns	0.87ns	1.29ns	7.15**	0.08
H × G	6	1.70ns	0.24ns	1.92ns	0.62ns	0.49ns	1.04ns	1.24
H × G × Y	6	1.35ns	1.75ns	1.09ns	1.48ns	0.24ns	2.74**	1.98
MS Error 1	4	2.23	0.14	27.09	0.08	0.64	1.98	3.49
MS Error 2	48	8.48	0.06	38.63	0.09	0.27	1.15	4.17
Mean		12.98	1.95	23.08	4.34	3.53	10.89	7.65
CV1 (%)		11.50	18.92	22.55	6.41	22.70	12.90	26.67
CV2 (%)		22.44	12.82	26.93	6.80	14.84	9.85	24.41
As		0.93	0.96	0.95	0.96	0.93	0.93	0.93

• FN = fruit number per plot; MFM = mean fruit mass (kg); YLD = yield (Mg ha⁻¹); FT = flesh thickness (cm); FF = flesh firmness (kgf); SS = soluble solids (°Brix); MMI = Mulamba-Mock index; DF = degrees of freedom; MS = mean square; CV = coefficient of variation; A_s = selective accuracy.
• ^**
  p < 0.01.
• ^*
  p < 0.05.
• ^ns
  Not significant (Snedecor's F-test).

Hybrids showed similar means for mean fruit mass, flesh thickness, flesh firmness, and soluble solids across F₁ and F₂ (Table 2). F₂ means for all hybrids fell within the export range for yellow melon fruit, indicating that both F₁ and F₂ generations supported commercialization. Yield means declined from F₁ to F₂ for most hybrids (Table 2). Some hybrids yielded below 25 Mg ha⁻¹, the farmer-defined minimum yield to ensure profitability. MMI showed low variation among hybrids, particularly in F₂. F₁ exceeded F₂ for fruit number, yield, and flesh thickness (Table 2).

Table 2
Mean of agronomic traits of the F₁ and F₂ generations of yellow melon hybrids evaluated in two consecutive years in Baraúna and Mossoró, Rio Grande do Norte, Brazil, 2017–2018.

Generation		Mean (Trait)
		FN	MFM	YLD	FT	FF	SS	MMI
F1	Iracema	11.33	1.61	15.38	3.88	3.83	10.93	5.28
	AF-13395	16.33	2.00	29.89	4.48	3.77	11.99	10.15
	Goldex	15.00	1.99	26.59	4.64	3.56	11.50	9.47
	Gladial	11.17	2.30	22.77	4.70	3.29	11.83	8.32
	Goldmine	14.33	2.30	31.50	4.61	3.31	10.75	9.63
	Mandacaru	14.50	1.99	26.47	4.52	3.91	11.23	9.10
	Solares	12.75	1.95	23.45	4.36	2.82	11.37	7.23
F 2	Iracema	7.83	1.41	8.49	3.57	3.65	10.90	3.71
	AF-13395	13.83	1.82	22.15	4.19	4.01	12.01	7.63
	Goldex	12.83	1.91	21.46	4.21	3.48	11.46	6.37
	Gladial	12.00	2.27	26.37	4.60	3.22	11.82	8.52
	Goldmine	13.00	2.08	24.42	4.31	3.42	10.72	7.52
	Mandacaru	17.00	1.85	28.84	4.37	3.87	11.19	8.67
	Solares	9.75	1.78	15.36	4.41	2.73	11.36	5.53
F1 mean		13.63a	2.02a	25.15a	4.46a	3.50a	11.37a	8.45a
F2 mean		12.32b	1.87a	21.01b	4.24b	3.48a	11.35a	6.85b

• FN = fruit number per plot; MFM = mean fruit mass (kg); YLD = yield (Mg ha⁻¹); FT = flesh thickness (cm); FF = flesh firmness (kgf); SS = soluble solids (°Brix); MMI = Mulamba-Mock index. Means followed by the same letter within columns do not differ (Snedecor’s F-test, p < 0.05).

Inbreeding depression estimates and Mulamba-Mock index (MMI) are shown in Table 3. Inbreeding depression magnitude varied among traits and MMI, with the highest estimates for yield and the lowest for soluble solids. Hybrids varied in inbreeding depression for all traits and MMI. Yield inbreeding depression ranged from 44.84% to -15.80% (60.64 percentage points total range) (Table 3). Soluble solids inbreeding depression ranged from -0.17% to 0.36% (Table 3). MMI showed the highest mean inbreeding depression.

Table 3
Inbreeding depression estimates for agronomic traits of yellow melon hybrids across two years in Baraúna and Mossoró, Rio Grande do Norte, Brazil, 2017–2018.

Hybrid	Inbreeding depression (%)
	FN	MFM	YLD	FT	FF	SS	MMI
Iracema	30.88	12.24	44.84	7.83	4.70	0.27	29.81
AF-13395	15.31	9.07	25.90	6.58	-6.37	-0.17	24.79
Goldex	14.44	3.77	19.30	9.20	2.25	0.33	32.75
Gladial	-7.46	1.16	-15.80	2.23	2.13	0.11	-2.40
Goldmine	9.30	9.51	22.45	6.37	-3.32	0.25	21.97
Mandacaru	-17.24	7.03	-8.96	3.32	1.02	0.36	4.76
Solares	23.53	9.05	34.51	-1.15	3.19	0.04	23.62
Mean	9.82	7.40	17.46	4.91	0.51	0.17	19.33

• FN = fruit number per plot; MFM = mean fruit mass (kg); YLD = yield (Mg ha⁻¹); FT = flesh thickness (cm); FF = flesh firmness (kgf); SS = soluble solids (°Brix); MMI = Mulamba-Mock index.

3.2. Estimates of m + a′ and d

Homozygous loci (m + a′) and heterozygous loci (d) contributions to the six traits are shown in Table 4. Student's t-test indicated significant m + a′ for all hybrids and traits. Hybrid Mandacaru exhibited the highest m + a′ estimates for fruit number and yield, whereas hybrid Iracema showed the lowest. The hybrid Gladial exhibited the highest m + a′ value for mean fruit mass and flesh thickness, whereas hybrid Iracema showed the lowest. Hybrids AF-13395 exhibited the highest m + a′ for flesh firmness, whereas hybrid Solares showed the lowest. The m + a′ estimates for soluble solids varied little among hybrids, with hybrid AF-13395 showing the highest value. The heterozygous loci (d) contribution was significant for yield (all hybrids) and fruit number (all hybrids except Gladial). Hybrid AF-13395 exhibited the highest yield d, whereas Gladial showed the lowest (Table 4).

Table 4
Homozygous locus (m + a′) and heterozygous locus (d) contributions to six traits and MMI of yellow melon hybrids across two years in Baraúna and Mossoró, Rio Grande do Norte, Brazil, 2017–2018.

Hybrid	Trait
	FN	MFM	YLD	FT	FF	SS	MMI
Contribution of homozygous loci (m + a′)
Iracema	4.33**	1.21**	1.59**	3.27**	3.47**	10.87**	2.13**
AF-13395	11.33**	1.64**	14.40**	3.89**	4.68**	12.03**	5.12**
Goldex	10.67**	1.84**	16.33**	3.78**	3.46**	11.42**	3.27**
Gladial	12.83**	2.25**	29.97**	4.49**	3.20**	11.81**	8.72**
Goldmine	11.67**	1.86**	17.35**	4.02**	3.89**	10.69**	5.40**
Mandacaru	19.50**	1.71**	31.21**	4.22**	3.83**	11.15**	8.23**
Solares	6.75**	1.60**	7.27**	4.46**	2.76**	11.36**	3.82**
Mean	11.01**	1.21**	16.87**	4.02**	3.61**	11.33**	5.24**
Contribution of heterozygous loci (d)
Iracema	7.00**	0.39ns	13.79**	0.61ns	0.36ns	0.06ns	3.15**
AF-13395	5.00**	0.36ns	15.49**	0.59ns	-0.48ns	-0.04ns	5.03**
Goldex	4.33**	0.15ns	10.26**	0.85ns	0.16ns	0.08ns	6.20**
Gladial	-1.67**	0.05ns	-7.20**	0.21ns	0.14ns	0.03ns	-0.40ns
Goldmine	2.67**	0.44ns	14.14**	0.59ns	-0.22ns	0.05ns	4.23**
Mandacaru	-5.00**	0.28ns	-4.74**	0.30ns	0.08ns	0.08ns	0.87ns
Solares	6.00**	0.35ns	16.19**	-0.10ns	0.18ns	0.01ns	3.42**
Mean	2.62**	0.29ns	8.28**	0.44ns	0.36ns	0.04ns	3.21**

• FN = fruit number per plot; MFM = mean fruit mass (kg); YLD = yield (Mg ha⁻¹); FT = flesh thickness (cm); FF = flesh firmness (kgf); SS = soluble solids (°Brix); MMI = Mulamba-Mock index.
• ^**
  p < 0.01.
• ^ns
  not significant (Student's t-test).

The mean m + a′ of the seven hybrids exceeded d means, indicating greater additive effects from parental lines than F₁ dominance effects. The hybrid m + a′ mean percentages exceeded d mean percentages in 80% for five traits, from 67.08% (mean fruit mass) to 99.65% (soluble solids) (Figure 1). These results suggest that additive genetic effects predominate over dominance effects for melon traits.

Figure 1
Percentage of homozygous loci (m + a′) and heterozygous loci (d) to six traits and MMI of yellow melon hybrids across two years in Baraúna and Mossoró, Rio Grande do Norte, Brazil, 2017–2018. FN = fruit number per plot; AFW = mean fruit mass (kg); YID = yield (Mg ha⁻¹); FT = flesh thickness (cm); FF = flesh firmness (kgf); SS = soluble solids (°Brix); MMI = Mulamba-Mock index.

4. Discussion

Genotype evaluation requires high experimental precision, as cultivar differences are small in most crops. High precision improves the detection of differences among treatments (Ramalho et al., 2000). Selective accuracy (A_s) quantifies experiment quality. This study recorded high selective accuracy (A_s > 0.90) for all traits, confirming high experimental quality (Resende and Alves, 2022). The coefficient of variation (CV) provides the standard experimental quality measure. According to the classification of Lima et al. (2004), CV estimates for melon yield and total soluble solids content were considered average (Table 1). The CV values are consistent with those observed in the Mossoró–Jaguaribe hub experiments (Oliveira et al., 2019; Cavalcante Neto et al., 2025; Lima et al., 2025).

Hybrid effect significance showed genotypic variation for all traits (Table 1). Generations differed for fruit number, yield, and flesh thickness. F₁ means exceeded F₂ means for these traits, confirming inbreeding depression (Table 3). Inbreeding depression reflects genotypic value reduction from increased homozygosity, decreased heterozygosity, and dominance interactions.

Inbreeding depression varied among traits (Table 3). Trait-specific heritability, numbers of genes (loci), allelic interactions, and gene interactions (epistasis) explained this variation. Quantitative trait inheritance amplified environmental influence on phenotypic expression. Hybrid variation in inbreeding depression resulted from parental allele complementarity (genetic divergence) and dominance or epistasis (non-additive effect) (Table 3). Limited parental allele complementarity or low dominance (non-additive effect) results in low inbreeding depression, whereas greater parental allele complementarity increases inbreeding depression and reduces heterosis (Charlesworth and Willis, 2009).

Classical models explain inbreeding depression and heterosis via dominance and overdominance, the literature also describes the effects additive × additive epistasis, which may result in negative heterosis (Lamkey and Edwards, 1999; Bernardo, 2020; Ramalho et al., 2024) or pseudo-dominance (Wallace et al., 2014). Models may operate simultaneously across allogamous and autogamous crops, with inbreeding depression and heterosis varying by genotype.

Cucurbitaceae species show low inbreeding depression despite cross-pollination favored by floral structure. Evolutionary processes of these species reduced inbreeding effects, as plants cover large areas with few individuals, promoting related mating (inbreeding), and natural selection eliminate deleterious alleles over time (Allard, 1999). Low inbreeding depression has been observed in pumpkin (Cardoso, 2004), watermelon (Wehner, 2008), and melon (Monforte et al., 2004).

Studies that guide selection of populations and lines increase breeding program efficiency because evaluating segregating populations is costly and time-consuming. Traditionally, breeders generate segregating populations with superior agronomic traits using diallel crosses, an effective method in several crops (Pouyesh et al., 2017; Cavalcante Neto et al., 2020; Napolitano et al., 2020; Shajari et al., 2021; Melgoza et al., 2022; Rad et al., 2023; Soltani et al., 2023). However, diallel crosses demand substantial time and labor, increasing breeding program cost. Therefore, alternative strategies are needed to manage segregating populations and identify superior lines in early generations.

Estimating m + a′ and d provides an alternative to identify superior populations early and efficiently. The m + a′ estimate identifies populations with higher means when all loci are homozygous, assuming no selection. The d estimate indicates dominance in genetic control of the trait. Higher d values imply more loci in heterozygosity and therefore greater genetic variability expressed in subsequent inbreeding generations. If selection is efficient, additional loci with favorable alleles increase mean inbred line performance. Thus, d functions as an indicator of potential population variability. Breeders should select populations with high m + a′ and high d to develop new inbred lines. Abreu et al. (2002) reported this pattern in common bean, where m + a′ estimates from F₁ and F₂ efficiently predicted seed yield and variance among F₇ populations.

Estimates of m + a′ varied among traits (Table 4), indicating fixed homozygous loci for these traits. Populations with higher m + a′ show greater frequency of fixed favorable alleles at allele frequency of 0.5 (Viana et al., 2009; Rocha et al., 2015; Ramalho et al., 2024). These m + a′ and d estimates indicate low heterosis in melon. However, yield showed heterosis in some hybrids. Yield and soluble solids represent primary traits in melon breeding (Oliveira et al., 2019). Single-cross hybrids remain viable in melon despite low heterosis and inbreeding depression, as Brazilian farmers have used them since the mid-1980s in major production regions. Fruit uniformity and resistance to pathogens and insect pests further support the use of simple hybrids.

Prior studies estimated these parameters in allogamous species (e.g., maize) and autogamous species (e.g., cowpea). Maize single-cross hybrids showed higher d than m + a′ for grain yield (Lima et al., 2000). Contrastingly, cowpea yield showed predominant m + a′ (Matos et al., 2020). Thus, genetic interactions vary by species, requiring crop-specific analysis.

Previous studies emphasized grain yield as the primary selection trait in major crops. However, melon breeding requires consideration of fruit yield and quality traits. Thus, the present study evaluated the primary traits identified by melon farms in the Mossoró–Jaguaribe hub, Brazil's primary production and export region. Hybrids AF-13395 and Goldmine showed the highest m + a′ and d for MMI, indicating suitability for line extraction, as they exhibited high yield and fruit number. These hybrids showed similar m + a′ and non-significant d for soluble solids content (Table 4). Hybrids Gladial and Mandacaru showed the highest m + a′ values for MMI, with non-significant d values.

Deriving m + a′ and d estimates from F₁ and F₂ means identifies superior hybrids suitable for inbred line development. This methodology identified AF-13395 and Goldmine as optimal hybrids and supported early identification of potential inbred lines as homozygosity increases. Early selection reduces population development costs and enables evaluation of more populations in less time than conventional traditional methods, providing an efficient and cost-effective alternative.

5. Conclusions

In the melon F₁ generation, homozygous loci (m + a′) contributed substantially to fruit number, mean fruit mass, yield, flesh thickness, flesh firmness, and soluble solids. Assessments of m + a′ and d quantified the suitability of single-cross hybrid for melon inbred line development. AF-13395 and Goldmine were the most suitable hybrids for superior line selection.

Acknowledgements

This study was partly funded by the Brazilian Coordination for the Improvement of Higher Education Personnel (CAPES, Finance Code 001).

Data Availability Statement

Data are available upon request.

References

• ALLARD, R.W., 1999. History of plant population genetics. Annual Review of Genetics, vol. 33, no. 1, pp. 1-27. https://doi.org/10.1146/annurev.genet.33.1.1 PMid:10690402.
  » https://doi.org/10.1146/annurev.genet.33.1.1
• ABREU, Â.D.F.B., RAMALHO, M.A.P. and SANTOS, J.B.D., 2002. Prediction of seed-yield potential of common bean populations. Genetics and Molecular Biology, vol. 25, no. 3, pp. 323-327. https://doi.org/10.1590/S1415-47572002000300013
  » https://doi.org/10.1590/S1415-47572002000300013
• ALVARES, C.A., STAPE, J.L., SENTELHAS, P.C., GONÇALVES, J.L.M. and SPAROVEK, G., 2013. Köppen’s climate classification map for Brazil. Meteorologische Zeitschrift, vol. 22, no. 6, pp. 711-728. https://doi.org/10.1127/0941-2948/2013/0507
  » https://doi.org/10.1127/0941-2948/2013/0507
• BERNARDO, R., 2020. Breeding for quantitative traits in plants 3rd ed. Woodbury, Minnesota: Stemma Press, 422 p.
• BRASIL, 1997. Lei nº 9.456, de 25 de abril de 1997. Institui a Lei de Proteção de Cultivares Diário Oficial da República Federativa do Brasil, Brasília, 28 abr. Seção 1.
• CARDOSO, A.I.I., 2004. Depression by inbreeding after four successive self-pollination squash generations. Scientia Agrícola, vol. 61, no. 2, pp. 224-227. https://doi.org/10.1590/S0103-90162004000200016
  » https://doi.org/10.1590/S0103-90162004000200016
• CARNEIRO, J.E.S., RAMALHO, M.A.P., ABREU, A.F.B. and GONÇALVES, F.M.A., 2002. Breeding potential of single, double and multiple crosses in common bean. Crop Breeding and Applied Biotechnology, vol. 2, no. 4, pp. 515-524. https://doi.org/10.12702/1984-7033.v02n04a04
  » https://doi.org/10.12702/1984-7033.v02n04a04
• CAVALCANTE NETO, J.G., FERREIRA, K.T.C., ARAGÃO, F.A.Z., ANTÔNIO, R.P. and NUNES, G.H.S., 2020. Potential of parents and hybrids experimental of the yellow melon. Ciência Rural, vol. 50, no. 2, pp. e20190452. https://doi.org/10.1590/0103-8478cr20190452
  » https://doi.org/10.1590/0103-8478cr20190452
• CAVALCANTE NETO, J.G., NUNES, E.W.L.P., SILVA, E.M., MARTINS, A.F., DIÓGENES, L.J.P., MELO, S.B., COSTA FILHO, J.H. and NUNES, G.H.S., 2025. Genotype × environment interaction in yellow melon hybrids in different locations and growing seasons. Bragantia, vol. 84, pp. e20240047. https://doi.org/10.1590/1678-4499.20240047
  » https://doi.org/10.1590/1678-4499.20240047
• CHARLESWORTH, D. and WILLIS, J.H., 2009. The genetics of inbreeding depression. Nature Reviews. Genetics, vol. 10, no. 11, pp. 783-796. https://doi.org/10.1038/nrg2664 PMid:19834483.
  » https://doi.org/10.1038/nrg2664
• CUI, H., DING, Z., ZHU, Z., LIU, S., WANG, X., LUAN, F. and GAO, P., 2022. Identification of major-effect QTL CmFpl3.1 controlling fruit pedicel length in melon (Cucumis meloL.). Scientia Horticulturae, vol. 293, pp. 110717. https://doi.org/10.1016/j.scienta.2021.110717
  » https://doi.org/10.1016/j.scienta.2021.110717
• FERREIRA, D.F., 2019. SISVAR: a computer analysis system to fixed effects split plot type designs. Revista Brasileira de Biometria, vol. 37, no. 4, pp. 529-535. https://doi.org/10.28951/rbb.v37i4.450
  » https://doi.org/10.28951/rbb.v37i4.450
• INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA – IBGE, 2024 [viewed 16 August 2024]. Produção Agrícola Municipal: Culturas temporárias e permanentes [online]. Available from: https://sidra.ibge.gov.br/Tabela/1612#resultado
  » https://sidra.ibge.gov.br/Tabela/1612#resultado
• LAMKEY, K.R. and EDWARDS, K.R., 1999. Quantitative genetics of heterosis In: J. G. COORS and S. PANDEY, eds. Genetics and exploitation of heterosis in crops Madison: American Society of Agronomy, pp. 523-524. https://doi.org/10.2134/1999.geneticsandexploitation.c4
  » https://doi.org/10.2134/1999.geneticsandexploitation.c4
• LIMA, L.L., NUNES, G.H.S. and BEZERRA-NETO, F., 2004. Coeficientes de variação de algumas características do meloeiro: uma proposta de classificação. Horticultura Brasileira, vol. 22, no. 1, pp. 14-17. https://doi.org/10.1590/S0102-05362004000100003
  » https://doi.org/10.1590/S0102-05362004000100003
• LIMA, M.W.P., SOUZA, E.A. and RAMALHO, M.A., 2000. Procedimento para escolha de populações de milho promissoras para extração de linhagens. Bragantia, vol. 59, no. 2, pp. 153-158. https://doi.org/10.1590/S0006-87052000000200005
  » https://doi.org/10.1590/S0006-87052000000200005
• LIMA, M.M.F., MARTINS, A.F., DA SILVA, A.S., ALVES, C.S., NUNES, E.W.L.P., SILVA, E.M. and NUNES, G.H.S., 2025. Adaptability and stability of melon genotypes through linear regression and segmented linear regression models: a bayesian approach. Euphytica, vol. 221, no. 6, pp. 80. https://doi.org/10.1007/s10681-025-03528-1
  » https://doi.org/10.1007/s10681-025-03528-1
• MATOS, R.F., BARROSO NETO, A.M., DOVALE, J.C., BERTINI, C.H.C.M. and FRITSCHE-NETO, R., 2020. A new proposal for the m + a methodology in segregating populations of cowpea. Bragantia, vol. 79, no. 2, pp. 242-249. https://doi.org/10.1590/1678-4499.20190271
  » https://doi.org/10.1590/1678-4499.20190271
• MELGOZA, F.A.G., ESCALANTE, F.B., DEL RIO, A.J.L., BENÍTEZ, A.L., MENDOZA, A.B., TORRES, N.A.R., MARTÍNEZ, R. H., ARANDA, C.P.A., 2022. Diallel analysis for yield and quality characteristics in melon. Journal of Experimental Agriculture International, vol. 44, pp. 29-36. https://doi.org/10.9734/jeai/2022/v44i122075
  » https://doi.org/10.9734/jeai/2022/v44i122075
• MENDONÇA, H.A., SANTOS, J.B. and RAMALHO, M.A.P., 2002. Selection of common bean segregating populations using genetic and phenotypic parameters and RAPD markers. Crop Breeding and Applied Biotechnology, vol. 2, no. 2, pp. 219-226. https://doi.org/10.12702/1984-7033.v02n02a08
  » https://doi.org/10.12702/1984-7033.v02n02a08
• MONFORTE, A.J., OLIVER, M., GONZALO, M.J., ALVAREZ, J.M., DOLCET-SANJUAN, R. and ARÚS, P., 2004. Identification of quantitative trait loci involved in fruit quality traits in melon (Cucumis melo L.). Theoretical and Applied Genetics, vol. 108, no. 4, pp. 750-758. https://doi.org/10.1007/s00122-003-1483-x PMid:14576985.
  » https://doi.org/10.1007/s00122-003-1483-x
• MULAMBA, N.N. and MOCK, J.J., 1978. Improvement of yield potential of the Eto Blanco maize (Zea mays L.) population by breeding for plant traits. Egyptian Journal of Genetics and Cytology, vol. 7, pp. 40-51.
• NAPOLITANO, M., TERZAROLI, N., KASHYAP, S., RUSSI, L., JONES-EVANS, E. and ALBERTINI, E., 2020. Exploring heterosis in melon (Cucumis melo L.). Plants, vol. 9, no. 2, pp. 1-19. https://doi.org/10.3390/plants9020282 PMid:32098173.
  » https://doi.org/10.3390/plants9020282
• NUNES, G.H.S., ARAGÃO, F.A.S., NUNES, E.W.L.P., COSTA, J.M. and RICARTE, A.O., 2016. Melhoramento de Melão. In: C. NICK and A. BORÉM, eds. Melhoramento de hortaliças Viçosa: Editora UFV, pp. 331-363.
• OLIVEIRA, L.A.A., CARDOSO, E.A., RICARTE, A.O., MARTINS, A.F., COSTA, J.M. and NUNES, G.H.S., 2019. Stability, adaptability and shelf life of Cantaloupe melon hybrids. Revista Brasileira de Fruticultura, vol. 41, no. 5, pp. e-418. https://doi.org/10.1590/0100-29452019418
  » https://doi.org/10.1590/0100-29452019418
• OLIVEIRA, L.B., RAMALHO, M.A.P., ABREU, A.F.B. and FERREIRA, D.F., 1996. Alternative procedures for parent choice in a breeding program for the common bean (Phaseolus vulgaris L.). Revista Brasileira de Genetica, vol. 19, no. 4, pp. 611-615. https://doi.org/10.1590/S0100-84551996000400014
  » https://doi.org/10.1590/S0100-84551996000400014
• POUYESH, A., LOTFI, M., RAMSHINI, H., KARAMI, E., SHAMSITABAR, A. and ARMIYOUN, E., 2017. Genetic analysis of yield and fruit traits in cantaloupe cultivars. Plant Breeding, vol. 136, no. 4, pp. 569-577. https://doi.org/10.1111/pbr.12486
  » https://doi.org/10.1111/pbr.12486
• RAD, M.R.N., BAKHSHI, B., MORADGHOLI, A. and RAFEZI, R., 2023. Heritability and combining ability in half diallel cross of melon (Cucumis melo L.). Journal of Horticulture Postharvest Research, vol. 6, pp. 181-192. https://doi.org/10.22077/jhpr.2023.5783.1294
  » https://doi.org/10.22077/jhpr.2023.5783.1294
• RAMALHO, M.A.P., FERREIRA, D.F. and OLIVEIRA, A.C., 2000. Experimentação em genética e melhoramento de plantas Lavras: Universidade Federal de Lavras.
• RAMALHO, M.A.P., NUNES, J.A.R., PÁDUA, J.M.V. and TOLEDO, F.H.B., 2024. Aplicações da Genética Quantitativa no melhoramento de plantas autógamas Lavras: Universidade Federal de Lavras.
• RESENDE, M.D.V. and ALVES, R.S., 2022. Statistical significance, selection accuracy, and experimental precision in plant breeding. Crop Breeding and Applied Biotechnology, vol. 22, no. 3, pp. e42712238. https://doi.org/10.1590/1984-70332022v22n3a31
  » https://doi.org/10.1590/1984-70332022v22n3a31
• ROCHA, G.S., CARNEIRO, J.E.S., CARNEIRO, P.C.S., POERSCH, N.L., LIMA, M.S. and SILVA, L.C., 2015. Evaluation of common bean segregant populations: prediction strategies and environmental effects. Revista Ceres, vol. 62, pp. 438-445. https://doi.org/10.1590/0034-737X201562050003
  » https://doi.org/10.1590/0034-737X201562050003
• SOUZA SOBRINHO, F., 2001. Divergência genética de híbridos simples e alternativas para a obtenção de híbridos duplos de milho Lavras: Universidade Federal de Lavras, 96 p. Tese de Doutorado em Genética e Melhoramento de Plantas.
• SOLTANI, F., NOURI, M., SHOKRPOUR, M., ALABBOUD, M., ERBAN, A., FEHRLE, I. and KOPKA, J., 2023. Breeding of Persian melon: purpose to increase volatile compounds in melon (Cucumis melo) intergroup F1 cross population. Plant Breeding, vol. 142, no. 3, pp. 372-387. https://doi.org/10.1111/pbr.13089
  » https://doi.org/10.1111/pbr.13089
• SHAJARI, M., SOLTANI, F., REZA BIHAMTA, M. and ALABBOUD, M., 2021. Genetic analysis and inheritance of floral and fruit traits in melon (Cucumis melo) in the full diallel cross. Plant Breeding, vol. 140, no. 3, pp. 486-496. https://doi.org/10.1111/pbr.12915
  » https://doi.org/10.1111/pbr.12915
• VIANA, L.F., SOUZA, J.C., MACHADO, J.C. and LIMA, J.L., 2009. Predição de medias de linhagens obtidas de híbridos simples de milho (Zea mays L.). Ciência e Agrotecnologia, vol. 33, no. spe, pp. 1999-2004. https://doi.org/10.1590/S1413-70542009000700051
  » https://doi.org/10.1590/S1413-70542009000700051
• WALLACE, J.G., LARSSON, S.J. and BUCKLER, E.S., 2014. Entering the second century of maize quantitative genetics. Heredity, vol. 112, no. 1, pp. 30-38. https://doi.org/10.1038/hdy.2013.6 PMid:23462502.
  » https://doi.org/10.1038/hdy.2013.6
• WEHNER, T.C., 2008. Watermelon. In: J. PROHENS and F. NUEZ, eds. Vegetables I. Handbook of plant breeding New York: Springer, pp. 381-418. https://doi.org/10.1007/978-0-387-30443-4_12
  » https://doi.org/10.1007/978-0-387-30443-4_12

Edited by

• Editor:
  Takako Matsumura Tundisi

Publication Dates

• Publication in this collection
  10 Apr 2026
• Date of issue
  2026

History

• Received
  16 Nov 2025
• Accepted
  02 Feb 2026