Genetic variability and conservation of progenies (Myracrodruon urundeuva allemão) in the South Mato Grosso, Brazil

Variabilidade genética e conservação de progênies (myracrodruon urundeuva allemão) no Mato Grosso do Sul, Brasil

Abstract

This study aimed to evaluate the genetic variability and conservation potential of Myracrodruon urundeuva in the face of climate change. The study was conducted at the UFGD Experimental Farm in Dourados, MS, with 77 open-pollinated progenies originating from Selvíria, MS. The methodology involved measuring survival, height, diameter, and branching in 2024, with data analyzed using the REML/BLUP method and Tocher clustering to estimate genetic parameters at the juvenile stage. The results indicated no significant effect of progeny, reflecting low genetic variability detectable at an early stage. Individual heritabilities were low (0.01 to 0.06), while the plot effect was significant, indicating a strong environmental influence. Survival stood out with the highest coefficient of genetic variation (11.56%). Despite low initial accuracy (0.16 to 0.28), the genetic correlations between height and diameter were positive and high (r_g= 0.78). Tocher’s clustering revealed the formation of four distinct groups, indicating the existence of genetic divergence among matrices. The ecological relevance of this plant lies in identifying sufficient genetic diversity to support conservation and pre-breeding strategies in the Cerrado. Maintaining this genetic base is crucial to ensuring the species’ adaptive resilience and guiding controlled crossbreeding aimed at the restoration and sustainable use of regional biodiversity.

Keywords:
aroeira; heritability; progeny test; genetic conservation; genotype-environment interaction

Resumo

Este estudo teve como objetivo avaliar a variabilidade genética e o potencial de conservação de Myracrodruon urundeuva frente as mudanças climaticas. O trabalho foi conduzido na Fazenda Experimental da UFGD, em Dourados-MS, com 77 progênies de polinização aberta oriundas de Selvíria-MS. A metodologia envolveu a mensuração de sobrevivência, altura, diâmetro e bifurcações em 2024, com dados analisados via método REML/BLUP e agrupamento de Tocher para estimar parâmetros genéticos em estádio juvenil. Os resultados indicaram ausência de significância para o efeito de progênie, refletindo baixa variabilidade genética detectável precocemente. As herdabilidades individuais foram baixas (0,01 a 0,06), enquanto o efeito de parcela foi significativo, evidenciando forte influência ambiental. A sobrevivência destacou-se com o maior coeficiente de variação genética (11,56%). Apesar da baixa acurácia inicial (0,16 a 0,28), as correlações genéticas entre altura e diâmetro foram positivas e elevadas (r_g = 0,78). O agrupamento de Tocher revelou a formação de quatro grupos distintos, indicando a existência de divergência genética entre matrizes. A relevância ecológica vegetal reside na identificação de diversidade genética suficiente para subsidiar estratégias de conservação e pré-melhoramento no Cerrado. A manutenção dessa base genética é crucial para garantir a resiliência adaptativa da espécie e orientar cruzamentos controlados que visem a restauração e o uso sustentável da biodiversidade regional.

Palavras-chave:
aroeira; herdabilidade; teste de progênies; conservação genética; interação genótipo x ambiente

1. Introduction

Myracrodruon urundeuva is a dioecious tree species with a straight, cylindrical trunk; its wood is dense and highly durable due to its high concentration of tannins (Queiroz et al., 2002). It is a species native to South America, widely distributed throughout Brazil, with the highest occurrence in the Caatinga, Cerrado, and Atlantic Forest biomes, as well as records in fragments of the Pantanal and the Pampa. Among these habitats, the Cerrado biome stands out, possessing a vast wealth of plant species and high endemism (Strassburg et al., 2017).

Thus, the distribution of species in this environment is directly related to the interaction between genotype and environment, especially under conditions marked by the occurrence of forest fires—whether natural or anthropogenic—and by illegal deforestation (Santos et al., 2020a).

From this perspective, plant growth, reproduction, and survival depend on morphological, physiological, and genetic characteristics that determine their performance and adaptation (Violle et al., 2017). Added to this are climatic factors, which become a central concern in the establishment of germplasm banks because they involve complex interactions of temperature and precipitation (Loarie et al., 2009).

Given this scenario, silvicultural practices play a fundamental role in the selection of genetic materials from tree species, as well as in the exploration of biodiversity and genetic resources. These practices enable the sustainable use of these resources, promoting human well-being and contributing to the ecological preservation of ecosystems (Clement, 2001; Clement et al., 2005).

Through these practices, trees with desirable phenotypic characteristics are selected over time and introduced into genetic improvement programs, with the aim of obtaining genotypes that are more productive and/or better adapted to the agroecosystem (Gartland et al, 2003). However, these programs face limitations related to phenotypic stability, reproductive maturity, and the difficulty of controlling interspecific crosses (Riva et al., 2020).

To overcome these limitations, progeny testing emerges as a fundamental tool (Freitas et al., 2007). It allows for the evaluation of the magnitude of adaptive genetic variation and quantitative indicators that, although influenced by the environment, are essential for breeding (Cruz, 2005). After all, obtaining variability in natural populations requires an understanding of regional genetic control (Kampa et al., 2020). Therefore, continuous monitoring of these tests ensures the maintenance of diversity and maximizes genetic gains (Avelar et al., 2021).

Canuto et al. (2017), when evaluating nine progenies of Myracrodruon urundeuva in Selvíria, MS, observed high levels of experimental precision, indicating reliability in genetic estimates and highlighting the potential for selection, as well as the viability of genetic conservation in seed banks. However, they also noted that genetic variability was higher in populations established in environments with less human interference, highlighting the impact that human disturbances have on the preservation of the species.

In light of the above, the objective of this study was to evaluate and characterize genetic parameters, as well as to promote the conservation of genetic variability in a progeny test of Myracrodruon urundeuva, in response to bioclimatic conditions.

2. Materials and Methods

The 77 progenies of Myracrodruon urundeuva were obtained by selecting mother trees from an open-pollinated seed orchard established in 1987, at the Teaching, Research, and Extension Farm (FEPE), located in Selvíria, MS, of the Júlio de Mesquita Filho São Paulo State University (UNESP) Ilha Solteira Campus, SP.

The progeny test was established in March 2023 at the Experimental Farm of Agricultural Sciences (FAECA) of the Federal University of Grande Dourados – UFGD in Dourados – MS (Figure 1), georeferenced by the geographic coordinates of latitude 22°13‟16” S, longitude 54°48‟2” W, and an altitude of 430 meters.

Figure 1
Location of the Myracrodruon urundeuva progeny test at the Experimental Farm of Agricultural Sciences (FAECA) of UFGD in Dourados, MS. Source: author.

The soil of the experimental area was classified as a Distrophic Red Latosol with a clay texture. The region’s climate is of the humid mesothermal type, with a rainy summer (Cwa), an annual average temperature of 23.4°C, and annual precipitation of 1,419 mm, according to Köppen. Data on annual precip, especially for species itation and maximum and minimum temperatures were obtained from the weather station located at FAECA via the Guia Clima website (Embrapa Agropecuária Oeste) (Figure 2).

Figure 2
Monthly precipitation (mm) and monthly average maximum and minimum temperatures (°C) from January 2023 to September 2025 at the Experimental Farm of Agricultural Sciences of the Federal University of Grande Dourados – FAECA – UFGD in Dourados, MS. Source: Guia Clima Embrapa CPAO (EMBRAPA, 2026).

Biometric data were collected in June 2024, and the following development indicators were evaluated: survival rate, total height, initial diameter, and number of branches.

Height was measured using a tape measure, diameter was measured using a digital caliper, survival was estimated through visual assessment, and the number of branches was determined by counting the number of branches per plant, with “1” assigned to living plants and “0” to dead plants.

2.1. Estimates of variance components and genetic parameters

Variance estimates and genetic parameter components were obtained using the REML/BLUP (Restricted Maximum Likelihood/Linear Unbiased Prediction) method from unbalanced data, using the genetic-statistical software SELEGEN-REML/BLUP, developed by Resende (2007). Different models were used to perform statistical analyses and estimate genetic parameters, as described below.

2.2. Individual analysis, estimation of genetic parameters, and genetic and phenotypic correlations

The estimation of genetic parameters was based on the mixed linear model (univariate additive model), applying the REML/BLUP (restricted maximum likelihood/best linear unbiased prediction) procedure of the genetic-statistical software SELEGEN-REML/BLUP developed by Resende (2007).

The adopted model considered the following effects: half-sibling progenies, completely randomized blocks, multiple plants per plot, a single location, and a single population, where y = Xr + Za + Wp + e, where y is the data vector, r is the vector of replication effects (fixed) added to the overall mean, a is the vector of individual additive genetic effects (random), p is the vector of plot effects (random), and e is the vector of errors or residuals (random); X, Z, and W are the incidence matrices of r, a, and p, respectively.

The significance test was performed using the likelihood ratio test (LRT), obtained by the difference between the deviations for models with and without the effect to be tested and using the chi-square test.

The estimated genetic parameters were: Individual heritability in the strict sense, where $h_a^2=$ ${\sigma }_a^2÷$ ${\sigma }_{a+{\sigma }_e^2}^2$, where ${\sigma }_a^2$ is the additive genetic variance and ${\sigma }_e^2$is the residual variance. Average heritability of progenies, where $h_{m=}^2\left[1+\left(n-1\right){\rho }_a\right]h_a^2÷1+\left(n-1\right)x\left({\rho }_a{h}_a^2\right)$, where 𝜌_𝑎 is the additive genetic intraclass correlation among individuals of the progeny type considered, 𝑛 is the number of trees per progeny, and$h_a^2$ is the individual heritability in the strict sense. The accuracy (ACprog), where $r_{âa}=$ $\sqrt{n÷n+\left(4-h_a^2÷h_a^2\right)}$, where 𝑛 are the number of trees per progeny and$h_a^2$is the individual heritability in the strict sense. Individual genetic variance coefficient, where $C{V}_{gi}=$ $100.\sqrt{{\sigma }_a^2}÷$ $\mu$, where${\sigma }_a^2$ is the additive genetic variance and 𝜇 is the overall mean of the trait. Coefficient of genetic variation between progenies, where $C{V}_{gp}=100$ x $\sqrt{0.25{\sigma }_a^2}$÷ $\mu$, where${\text{σ}}_a^2$ is the additive genetic variance and 𝜇 is the overall mean of the trait. Experimental coefficient of variation $C{V}_e=100$ x $\sqrt{SMres}$÷ $\mu ,$ where 𝑆𝑀𝑟𝑒𝑠 is the mean of the residual squares and 𝜇 is the overall mean of the trait.

Genetic and phenotypic correlations among indicators were estimated using statistical models from the genetic-statistical software SELEGEN-REML/BLUP.

2.3. Estimates of genetic divergence

Genetic diversity among the progenies was estimated using the Generalized Mahalanobis Distance (D²), and this methodology is suitable for analyzing quantitative data, considering the standard deviation and the invariant scale. Thus, D² is estimated by the expression (Cruz et al., 2004): $D\frac{2}{ii}=\delta {\Psi }^{-1}\delta$, where $D\frac{2}{ii}$ = Mahalanobis distance between genotypes i and i’; ẟ = [d1, d2, ..., dv], where𝑑_𝑗 =𝑌_𝑖𝑗 −𝑌_𝑖′𝑗; ᴪ = residual variance and covariance matrices and Y_𝑖𝑗 = mean of the i^th genotype with respect to the i^th variable.

In the analysis of quantitative data, this methodology is suitable for considering the variance matrices and residual covariances among measured traits.

Using the Mahalanobis distance matrix estimated by the genetic statistical software SELEGEN REML/BLUP (Resende, 2007), we can construct a hierarchical structure in the form of a dendrogram using the average of distances method (UPGMA—Unweighted Pair Grouping Method Using Arithmetic Means). To perform this analysis, we used the statistical software R.

After obtaining D², Tocher’s optimization method was applied to identify groups of homogeneous genotypes. This method uses criteria based on the mean values of D² (intracluster) and must be lower than the values of D² (intercluster) (Cruz et al., 2004).

3. Results and Discussion

The deviance analysis (DEV) revealed that the progeny effect was not significant for any of the indicators evaluated (Table 1), with LRT values ranging from 0.02 to 0.17. This lack of statistical significance, confirmed by the likelihood ratio test at a 5% probability level, indicates the absence of detectable additive genetic variability among M. urundeuva families for height, diameter, branching, and survival.

Table 1
Deviance analysis of Myracrodruon urundeuva progenies in Dourados, MS.

ANADEV
Effect	Height (m)		Diameter (cm)		Branches		Survival rate (%)
	DEV	LRT	DEV	LRT	DEV	LRT	DEV	LRT
Progeny	-530.26	0.06 ns	3506.03	0.02 ns	970.48	0.02 ns	-729.85	0.17 ns
Parcel	-495.09	35.23*	3,533.26	27.25*	984.53	14.07*	-695.28	34.74*
Full model	-530.32	_	3,506.01	_	970.46	_	-730.02	_

• Tables chi-square values: 3.84 and 6.63 for significance levels of 5% (*). ns: Not significant.
• Dev = Deviance; LRT = Likelihood Ratio Test.

On the other hand, the plot effect was significant for all indicators, with high LRT values (35.23 for height; 27.25 for diameter; 14.07 for branching; and 34.74 for survival) (Table 1). This highlights the influence of environmental variation between plots, possibly related to edaphic, microclimatic, or experimental management differences. Thus, environmental control within the experimental design is crucial to reduce the interference of variation between plots and increase the precision of genetic estimates (Avelar et al., 2021).

The full model presented low deviance values (without considering each effect in isolation), indicating that the inclusion of progeny and plot effects contributed to a better model fit. However, since only the plot effect was significant, it is evident that the variation observed in the evaluated indicators is predominantly explained by environmental factors, rather than by genetic differences among the progenies.

It is worth noting that the evaluations were conducted on plants in the juvenile stage, and phenotypic expression may not yet reflect the genetic potential of the individuals, since quantitative traits in forest species tend to manifest more consistently throughout development. Thus, the low genetic differentiation observed may be associated with the early age of the progenies, combined with the possible influence of environmental conditions during the initial establishment period.

Consequently, the progeny effect was not significant, indicating low genetic differentiation among the evaluated progenies. In contrast, the plot effect was significant, highlighting the predominance of environmental variation over the expression of the indicators. This result suggests that the development and survival of the progenies were strongly influenced by environmental conditions throughout the experimental period, especially by the climatic variations observed between 2023 and 2024. It is noteworthy that, in 2024, higher average temperatures and lower rainfall indices were also recorded compared to 2023, as shown in Figure 2, which may have directly affected the plants’ performance since they were still in the juvenile stage.

Therefore, there is a trend toward the migration of M. urundeuva progenies to regions with milder climates due to rising temperatures and rainfall in regions where the species occurs (Capo et al., 2022).

The low estimated heritability values (Table 2) were not statistically significant, which reinforces the small magnitude of these estimates at the individual level. This result highlights the predominant influence of environmental factors such as water availability, temperature, precipitation regime, soil fertility, and environmental heterogeneity in modulating the phenotypic expression of the evaluated indicators, indicating that the observed variability is more associated with environmental conditions than with the genetic control of these traits (Cruz et al., 2020).

Table 2
Estimates of genetic parameters for the indicators in progenies of Myracrodruon urundeuva in Dourados, MS.

Parameter estimates	Height (m)	Diameter (cm)	Branching	Survival (%)
$h_a^2$	0.04 (±0.04)	0.01 (±0.03)	0.01 (± 0.02)	0.06 (±0.04)
$h_{aj}^2$	0.05	0.02	0.01	0.06
$h_m^2$	0.06	0.02	0.02	0.08
$h_{ad}^2$	0.04	0.01	0.01	0.04
ACprog	0.23	0.16	0.16	0.28
CVe (%)	28.55	29.89	14.69	34.21
CVgi (%)	7.97	5.44	2.67	11.56
CVgp (%)	3.98	2.72	1.34	5.78
LRT	0.06 ns	0.02 ns	0.02 ns	0.17 ns
Average (µ)	0.99	18.23	5.17	76

• h²a: Individual heritability in the strict sense; h²aj: Individual heritability adjusted for plot effects; h²m: Average progeny heritability; h²ad: Additive heritability within plot; ACprog: Progeny selection accuracy; CVgi (%): Individual genetic coefficient of variation; CVgp (%): Genetic coefficient of variation between progenies; CVe (%): Experimental coefficient of variation; µ: LRT: General mean; likelihood ratio test; Tabulated chi-square: 3.84 and 6.63 for significance levels of 5% and 1%, respectively. ns: Not significant.

In this context, the genotype × environment (G×E) interaction becomes a central component in the interpretation of genetic parameters, since different genotypes may respond differently to environmental conditions, and is also important for defining efficient strategies in genetic improvement programs, allowing the identification of genotypes better adapted to specific environments (Souza et al., 2020).

Individual heritability in the strict sense ($h_a^2$) showed low values for diameter (0.01), branching (0.01), height (0.04), and survival (0.06) (Table 2). This demonstrates that the fraction of phenotypic variance attributed to additive genetic effects is low, indicating a predominance of environmental influence on the evaluated indicators. Despite the low values, survival revealed a relevant genetic contribution for selection in conservation programs, considering that the Myracrodruon urundeuva seedlings, still in the juvenile stage, had been introduced into the environment one year prior.

The conservation of genetic variability requires maintaining, within the experiment, individuals with better adaptation, ensuring the preservation of rare alleles (Cornacini et al., 2017). However, the$h_a^2$ is important for selecting individuals based on additive genetic variance, since it determines the variance that will be transmitted to offspring (Henriques et al., 2017).

Heritability estimates based on progeny means ($h_m^2$) were higher than those based on individual means, with diameter and branching showing a value of 0.02, height 0.06, and survival 0.08 (Table 2). This behavior was expected, as progeny means tend to reduce environmental variance, thereby increasing selection accuracy. This reinforces the notion that selection based on families may be more efficient than individual selection.

The experimental coefficients of variation (CV_e) were relatively high, ranging from 14.69% (branching) to 34.21% (survival) (Table 2). These values indicate a significant influence of uncontrolled environmental factors (Henriques et al., 2017), as these effects contribute to errors that can reduce experimental precision (Pupin et al., 2017), particularly for survival, where environmental variability was more pronounced.

Thus, according to Pupin et al. (2017), high CV_e values may be related to environmental effects that occurred on M. urundeuva progenies, ranging from competition with wild plants to environmental variation.

The coefficients of individual genetic variation (CV_gi) and between progenies (CV_gp) showed low values for all indicators, especially for branching (2.67% and 1.34%, respectively), confirming the limited genetic variability detected. For survival, however, these values were relatively higher (11.56% and 5.78%), suggesting greater potential for response to selection for this trait compared to the others (Table 2). Coefficients greater than 7% are considered high and can be used as estimates of genetic variation, allowing identification of whether the greater magnitude of this variation occurs between individuals (CV_gi) or within progenies (CV_gp) (Cruz et al., 2020).

Although the individual genetic variation coefficients (CV_gi) and between progenies (CV_gp) were low and the progenies were influenced by environmental conditions, these factors were not sufficient to completely eliminate the existing genetic variability, possibly due to the broad genetic base of the evaluated individuals (Pupin et al., 2017). Thus, significant genetic variability is still observed, indicating potential for the application of selection strategies, even with Myracrodruon urundeuva plants in the juvenile stage of development.

Accuracy (AC_prog) showed low values, ranging from 0.16 (diameter and branching) to 0.28 (survival) (Table 2). Accuracy is a fundamental parameter for measuring the degree of reliability of estimates obtained in progeny test experiments (Resende and Duarte, 2007). According to Resende (2002), accuracy values between 0 and 25% are considered low, between 25 and 75% are classified as good, and above 75% are considered excellent. These results indicate that the reliability of progeny-based selection is limited but consistent for survival. Moraes et al. (2014) observed accuracy values between 0.59 and 0.75, which fall within the desirable parameters (0.50) for estimates of genetic and phenotypic variation, indicating good precision. In the present study, however, accuracy fell below the threshold considered adequate, implying reduced precision of the estimates.

The phenotypic means (µ) revealed values close to those expected for the initial growth stage: height of 0.99 m, mean diameter of 18.23 cm, branching of 5.17, and survival of 76% (Table 2). These results are consistent with tree species in the juvenile stage, in which phenotypic variation is conditioned by environmental factors. This behavior corroborates Santos (2023), who identified the influence of invasive vegetation, soil chemical and physical properties, and abiotic conditions on plant survival. In his study, evaluating 2,016 initially established individuals, the recorded survival rate was only 50.94%.

The results indicate low heritability and reduced genetic variability for height, diameter, and branching, suggesting that these traits are strongly influenced by the environment and associated with the juvenile stage. Survival, although with relatively low genetic potential, stood out compared to the other traits, emerging as a strategic parameter for M. urundeuva conservation programs.

The genetic correlation coefficients (Table 3) showed positive and significant values (p<0.01), indicating that the indicators have a favorable genetic association.

Table 3
. Estimates of genetic (rg = above the diagonal) and phenotypic (rf = below the diagonal) correlations between the indicators and form in Myracrodruon urundeuva progenies in Dourados, MS.

Indicators	Indicators
	Height	Diameter	Bifurcation
Height	1	0.78**	0.24**
Diameter	0.65**	1	0.29**
Bifurcation	0.20**	0.22**	1

• ^**
  (p<0.01) by t-test.

The highest correlation was observed between height and diameter (rg = 0.78); correlations between height and branching (rg = 0.24) and between diameter and branching (rg = 0.29), although of lesser magnitude, were also positive. Genetically taller individuals tend to have a larger diameter, indicating that these indicators may respond jointly to selection. Height and diameter may be associated with a higher incidence of forks, a factor considered important since forking can compromise stem quality. For indicators with a positive correlation and high magnitude, superior results are expected, which will ensure an advantage in progeny selection (Sant’Ana et al., 2013).

For the phenotypic correlation coefficients (Table 3), the associations followed the same trend as the genetic correlations; however, they exhibited lower magnitudes, indicating a lesser influence of environmental factors on the phenotypic expression of these indicators.

The highest correlation was between height and diameter (rf = 0.65) and between height and branching (rf = 0.20) and between diameter and branching (rf = 0.22), reinforcing that more vigorous individuals, in terms of vertical growth, also exhibit greater stem thickness; however, from a phenotypic perspective, branching is little influenced by the environment, maintaining a low association with growth, although still significant.

Moraes et al. (2014), working with Eucalyptus clones, found similar results regarding phenotypic correlations at 24 months of age, demonstrating values of 70% for selection based on DBH and other evaluated indicators, making it a good alternative for selecting young progeny.

Additionally, Alcantara (2019) identified, in Astronium fraxinifolium, the presence of genetic variability for both height and diameter, with genotypic and phenotypic variance coefficients indicating potential for selection gains. In this regard, the existence of genetic variability associated with growth indicators, combined with the observed phenotypic correlations, suggests that selection strategies based on indicators such as height and diameter can be effective, even under environmental influence and in early stages of development.

Based on estimates of Mahalanobis distances between the progenies, greater genetic divergence was observed between M10 and FC4 (D² = 29.45), while the smallest distance was observed between FC18 and AC18 (D² = 0.034) (Table 4). Individuals from progenies with greater divergence among themselves can be selected for controlled crossbreeding.

Table 4
Mahalanobis genotypic (D²) in Myracrodruon urundeuva in Dourados, MS.

D2 Mahalanobis - Distances
Prog.	Largest		Prog.		Min		Prog.		Prog.	Larger	Prog.	Minor	Pro.
PS19	21.458		FC10		1.283		AC6		M5	11.271	AC13	0.861	M9
AC21	21.368		FC12		0.530		M17		M10	10.006	AC10	0.265	M9
AC2	18.696		FC5		0.267		AC7		M10	18.203	AC12	0.150	M1
PS5	23.436		FC6		0.269		M2		AC2	13.671	AC5	0.206	AC7
PS5	13.413		FC13		0.145		PS8		AC2	18.076	AC15	0.109	M16
M10	14.872		FC14		0.141		AC17		M10	15.906	AC19	0.231	M3
PS19	14.755		FC1		0.149		PS16		M10	25.450	AC20	0.346	M20
PS7	21.418		FC17		0.167		M6		M10	25.066	AC16	1.403	M20
M10	29.458		FC4		1.241		M8		M20	17.294	AC6	0.873	M13
PS5	21.165		FC18		0.034		AC18		AC14	13.873	AC11	1.539	AC7
PS5	19.893		FC15		0.115		M1		M5	12.699	M17	0.912	M19
PS5	8.563		PS9		0.275		AC10		M5	18.562	AC14	0.558	M18
PS19	14.279		PS16		0.254		PS1		AC2	15.874	AC7	0.448	M16
M10	26.557		PS10		1.000		PS21		AC2	8.974	AC9	0.181	M11
PS5	16.389		PS14		0.216		PS2		M5	19.326	M7	3.757	M2
M7	28.467		PS5		2.514		AC4		M10	13.033	M4	0.156	M1
M10	21.644		PS11		0.588		AC16		M5	16.116	M18	0.291	FC7
AC21	14.101		PS17		0.646		AC4		M10	21.541	M8	0.172	M3
PS19	17.376		PS2		0.288		M12		M20	12.140	M13	2.686	M10
M10	13.330		PS20		0.027		AC17		M10	13.317	M19	0.817	M9
PS7	17.334		PS22		0.490		M4		M10	11.889	M2	0.436	M1
PS7	11.838		PS8		0.066		AC9		AC2	21.301	M15	0.782	M3
M10	19.282		PS15		0.150		AC12		AC2	16.368	M16	0.159	M6
PS19	24.445		PS7		2.202		M17		AC2	26.268	M5	1.766	M14
M10	29.272		PS21		0.734		M20		M10	19.903	M3	0.702	M1
AC16	12.555		PS3		0.434		AC1		M10	13.350	M21	0.379	AC17
AC2	14.888		PS13		0.064		M22		M10	11.915	M12	0.399	AC18
M10	13.850		PS12		0.042		M21		M10	15.099	M1	0.165	AC18
AC2	29.289		PS19		0.612		M5		M10	12.669	M9	0.503	AC17
M10	15.275		PS1		0.463		PS6		M20	24.594	M10	3.241	AC2
M10	16.288		PS18		0.729		PS6		AC2	13.691	M22	0.340	M6
M10	15.645		PS6		0.170		M9		AC2	18.856	M20	2.934	AC17
AC16	11.292		AC1		0.235		AC9		AC2	21.564	M14	1.635	AC17
AC14	16.951		AC8		0.264		AC11		AC2	17.785	M6	1.836	AC17
AC2	19.687		AC3		0.151		M6		AC2	6.211	FC7	1.242	AC18
M7	16.366		AC4		0.421		M19		AC2	11.512	AC18	2.206	AC17
AC2	27.982		AC21		0.609		M15		AC17	12.295	AC2
Maximum:		29.45		(100%)		M10		FC4
Minimum:		0.034		(0.12%)		FC18		AC18

Clustering using the Tocher method, based on Mahalanobis genetic distances, resulted in the formation of four similarity groups (Figure 3). The first group consisted of 70 progenies, the second of 4, the third of 2, and the fourth of only 1 progeny, considering the heritability parameters and the individual genetic coefficient of variation.

Figure 3
Clustering of Myracrodruon urundeuva progenies using Tocher’s method, based on Mahalanobis genetic distances (D²) in Dourados, MS.

The clustering of individuals or progenies, performed using the Mahalanobis distance method combined with Tocher’s agglomeration method, aims to maximize homogeneity within groups and heterogeneity between them, thereby using predicted genetic values in place of phenotypic ones. This allows for the incorporation of genetic variance and covariance matrices into the analyses, increasing the precision in assessing divergence (Resende, 2007).

The clustering results highlight the importance of Myracrodruon urundeuva genotypes in the conservation of progenies and in supporting genetic improvement programs, especially for species with the potential to adapt to different ecosystems (Alves et al., 2024). Therefore, understanding the genetic variability of the species is essential for characterizing and identifying new genes of interest in plant breeding (Santos et al., 2020b).

4. Conclusion

The low genetic variability observed may be associated with both environmental interference and the physiological immaturity of the progenies at the time of evaluation.

Survival showed genetic significance, establishing itself as a strategic indicator for genotype conservation and selection programs, suggesting future evaluations of progeny performance.

The results demonstrated the predominance of environmental variation over genetic variation among progenies, while genetic and phenotypic correlations were positive, indicating potential for selection.

Clustering using the Tocher method revealed sufficient genetic diversity to form distinct groups, highlighting the importance of conserving these genotypes and maintaining this genetic base for species conservation strate

Acknowledgements

The authors acknowledge the research incentive received from the Federal University of Grande Dourados (UFGD) and the financial support from The Coordination for the Improvement of Higher Education Personnel (CAPES).

Data Availability Statement

The data generated from this study are not publicly available in any repository. However, all relevant information is fully presented within the study and may be obtained from the corresponding author upon reasonable request.

References

• ALCANTARA, M. A. M., 2019. Genotypic variability and population structure of Astronium fraxinifolium Schott (Anacardiaceae) in a degraded Cerrado area Botucatu – SP: Universidade Estadual Paulista (Unesp).
• ALVES, E.O., FREITAS, C.B.S., OLIVEIRA, M.G. and SILVA, M.S., 2024. Genetic parameters and dissimilarity between progenies of the ‘cupuaçu’ tree (Theobroma grandiflorum, Malvaceae) from free pollination in southeastern Pará, Brazil. Revista Verde de Agroecologia e Desenvolvimento Sustentável, vol. 19, no. 1, pp. 30-36. https://doi.org/10.18378/rvads.v19i1.9752
  » https://doi.org/10.18378/rvads.v19i1.9752
• AVELAR, M.L.M., DA SILVA JÚNIOR, V.P., ROSADO, L.R., GONÇALVES, F.M.A., PAVAN, A.J. and MELO, L.A., 2021. Genetic characterization in the initial growth of progenies of Eremanthus erythropappus (DC.) MacLeish in Aiuruoca, Minas Gerais State, Brazil. Hoehnea, vol. 48, pp. e142020. https://doi.org/10.1590/2236-8906-14/2020
  » https://doi.org/10.1590/2236-8906-14/2020
• CANUTO, D.S.O., DA SILVA, A.M., FREITAS, M.L.M., SEB BENN, A.M. and DE MORAES, M.L.T., 2017. Genetic Variability in Myracrodruon urundeuva (Allemão) Engl. Progeny Tests. Open Journal of Forestry, vol. 7, no. 1, pp. 1-10. https://doi.org/10.4236/ojf.2017.71001
  » https://doi.org/10.4236/ojf.2017.71001
• CAPO, L.F.M., MORAES, M.L.T.D., ZULIAN, D.F., WREGE, M.S., PORTELA, R.M., CAMBUIM, J., DA SILVA, A.M., SOARES, M.T.S., SOUZA, V.A. and AGUIAR, A.V., 2022. Natural distribution of Myracrodruon urundeuva Fr. All. in Brazil under current and future climate scenarios due to global climate change. Revista Árvore, vol. 46, pp. e4609. https://doi.org/10.1590/1806-908820220000009
  » https://doi.org/10.1590/1806-908820220000009
• CLEMENT, C.R., 2001. Breeding of native species. In: L.L. NASS, A.C.C. VALOIS, I.S. MELO and M.C. VALADARES INGLIS. Genetic resources & breeding: plants Rondonópolis: Fundação MT, pp. 423-441.
• CLEMENT, C.R., LLERAS PÉREZ, E. and VAN LEEUWEN, J., 2005. The potential of tropical palms in Brazil: successes and failures of the last decades. Agrociencia, vol. 9, no. 1-2, pp. 67-71.
• CORNACINI, M.R., SILVA, J.R., LUZ, K.C., CAMBUIM, J., SANTOS, W., MORAES, M.L.T. and AGUIAR, A.V., 2017. Selective thinning in a test of provenances and progenies of Astronium fraxinifolium Schott based on genetic variability. Scientia Forestalis, vol. 45, no. 115, pp. 581-591. https://doi.org/10.18671/scifor.v45n115.15
  » https://doi.org/10.18671/scifor.v45n115.15
• CRUZ, C.D. 2005. Principles of quantitative genetics Viçosa: UFV, 394 p.
• CRUZ, C.D., REGAZZI, A.J. and CARNEIRO, P.C.S., 2004. Biometric models applied to genetic improvement 3rd ed. Viçosa: UFV.
• CRUZ, L.S., PEDROZO, C.Â., OLIVEIRA, V.X.A., SILVA, A.M., RESENDE, M.D.V. and GONÇALVES, D.A., 2020. Genetic parameters and initial selection of breeding lines and progenies of white-tailed deer (Tachigali vulgaris) in Roraima. Ciência Florestal, vol. 30, no. 1, pp. 258-269. https://doi.org/10.5902/1980509831631
  » https://doi.org/10.5902/1980509831631
• EMPRESA BRASILEIRA DE PESQUISA AGROPECUÁRIA – EMBRAPA, 2026 [viewed 28 April 2026]. Guia clima [online]. Available from: https://www.cpao.embrapa.br/gc/guia-clima?cod_estacao=5
  » https://www.cpao.embrapa.br/gc/guia-clima?cod_estacao=5
• FREITAS, M.L.M., SEBBENN, A.M., ZANATTO, A.C.S. and MORAIS, E., 2007. Seed orchard for seedlings from selection within progeny trials of Myracrodruon urundeuva Fr. All. Journal of the Forestry Institute, vol. 19, no. 2, pp. 65-72.
• GARTLAND, K.M.A., CROW, R.M., FENNING, T.M. and GARTLAND, J.S., 2003. Genetically modified trees: production, properties, and potential. Arboriculture & Urban Forestry, vol. 29, no. 5, pp. 259-266. https://doi.org/10.48044/jauf.2003.030
  » https://doi.org/10.48044/jauf.2003.030
• HENRIQUES, E.P., MORAES, C.B., SEBBENN, A.M., TOMAZELLO FILHO, M., MORAES, M.L.T. and MORI, E.S., 2017. Genetic parameters estimates for silvicultural traits and wood density in Eucalyptus urophylla progeny test. Scientia Forestalis, vol. 45, no. 113, pp. 119-128. https://doi.org/10.18671/scifor.v45n113.11
  » https://doi.org/10.18671/scifor.v45n113.11
• KAMPA, M.B., HOMCZINSKI, I., ROQUE, R.H., FIGUEIREDO FILHO, A., PERES, F.S.B. and TAMBARUSSI, E.V., 2020. Genetic variability in nusery-grown progenies of Campomanesia xanthocarpa Mart. ex O. Berg. Scientia Forestalis, vol. 48, no. 125, pp. e2935. https://doi.org/10.18671/scifor.v48n125.10
  » https://doi.org/10.18671/scifor.v48n125.10
• LOARIE, S.R., DUFFY, P.B., HAMILTON, H., ASNER, G.P., FIELD, C.B. and ACKERLY, D.D., 2009. The velocity of climate change. Nature, vol. 462, no. 7276, pp. 1052-1055. https://doi.org/10.1038/nature08649 PMid:20033047.
  » https://doi.org/10.1038/nature08649
• MORAES, C.D., FREITAS, T.D., PIERONI, G.B., RESENDE, M.D., ZIMBACK, L. and MORI, E.S., 2014. Estimativas de parâmetros genéticos para seleção precoce de clones de Eucalyptus para região com ocorrência de geadas. Scientia Forestalis, vol. 42, no. 102, pp. 219-227.
• PUPIN, S., FREITAS, M.L.M., CANUTO, D.S.O., SILVA, A.M., ALZATE-MARIN, A.L. and MORAES, M.L.T., 2017. Genetic variability and selection gains in progenies of Myracrodruon urundeuva Fr. All. Nativa (Sinop), vol. 5, pp. 59-65. https://doi.org/10.31413/nativa.v5i1.3819
  » https://doi.org/10.31413/nativa.v5i1.3819
• QUEIROZ, C.R.A.A., MORAIS, S.A.L.D. and NASCIMENTO, E.A.D., 2002. Caracterização dos taninos da aroeira-preta (Myracrodruon urundeuva). Revista Árvore, vol. 26, no. 4, pp. 485-492.
• RESENDE, M.D.V. and DUARTE, J.B., 2007. Precision and experimental quality control in cultivar evaluation experiments. Pesquisa Agropecuária Tropical, vol. 37, no. 3, pp. 182-194.
• RESENDE, M.D.V., 2002. Biometric genetics and statistics in the breeding of perennial plants Brasília: EMBRAPA Technological Information, 975 p.
• RESENDE, M.D.V., 2007. SELEGEN-REML/BLUP: statistical system and computerized genetic selection via mixed linear models Colombo: Embrapa Florestas, 359 p.
• RIVA, L.C., MORAES, M.A.D., CAMBUIM, J., ZULIAN, D.F., SATO, L.M., CALDEIRA, F.A. and MORAES, M.L.T.D., 2020. Genetic control of wood quality of Myracrodruon urundeuva populations under anthropogenic disturbance. Crop Breeding and Applied Biotechnology, vol. 20, no. 4, pp. e320920411. https://doi.org/10.1590/1984-70332020v20n4a64
  » https://doi.org/10.1590/1984-70332020v20n4a64
• SANT’ANA, V.Z., FREITAS, M.L.M., MORAES, M.L.T., ZANATA, M., ZANATTO, A.C.S., MORAES, M.A. and SEBBENN, A.M., 2013. Genetic parameters in open-pollinated progenies of Enterolobium contortisiliquum (Vell.) Morong in Luiz Antonio, SP, Brazil. Hoehnea, vol. 40, no. 3, pp. 515-520. https://doi.org/10.1590/S2236-89062013000300011
  » https://doi.org/10.1590/S2236-89062013000300011
• SANTOS, L.A.C., DE MIRANDA, S.D.C., DE MELO, C. and NETO, S., 2020a. Cerrado phytophysiognomies: definitions and trends. Élisée-UEG. The Geographical Journal, vol. 9, no. 2, pp. e922022.
• SANTOS, S.C., CARVALHO, R.S. and DAVIDE, L.M.C., 2020b. Genetic polymorphism among natural populations of Anacardium humile A. ST-HIL. Revista Brasileira de Fruticultura, vol. 42, no. 1, pp. e-476. https://doi.org/10.1590/0100-29452020476
  » https://doi.org/10.1590/0100-29452020476
• SANTOS, T.C.S., 2023. Selection of Pinus caribaea var. bahamensis progenies for timber and resin production and quality Botucatu: Faculty of Agronomic Sciences, São Paulo State University, Thesis Master’s in Forest Product Technology.
• SOUZA, B.M., FREITAS, M.L.M., SEBBENN, A.M., GEZAN, S.A., ZANATTO, B., ZULIAN, D.F., LOPES, M.T.G., LONGUI, E.L., GUERRINI, I.A. and DE AGUIAR, A.V., 2020. Genotype-by-environment interaction in Corymbia citriodora (Hook.) KD Hill, & LAS Johnson progeny test in Luiz Antonio, Brazil. Forest Ecology and Management, vol. 460, pp. 117855. https://doi.org/10.1016/j.foreco.2019.117855
  » https://doi.org/10.1016/j.foreco.2019.117855
• STRASSBURG, B. B.N., BROOKS, T., FELTRAN-BARBIERI, R., IRIBARREM, A., CROUZEILLES, R., LOYOLA, R., BALMFORD, A., 2017. The moment of truth for the Cerrado hotspot. Nature Ecology & Evolution, vol. 1, pp. 0099.
• VIOLLE, C., THUILLER, W., MOUQUET, N., MUNOZ, F., KRAFT, N.J., CADOTTE, M.W. and MOUILLOT, D., 2017. Functional rarity: the ecology of outliers. Trends in Ecology & Evolution, vol. 32, no. 5, pp. 356-367. https://doi.org/10.1016/j.tree.2017.02.002 PMid:28389103.
  » https://doi.org/10.1016/j.tree.2017.02.002

Edited by

• Editor:
  Takako Matsumura Tundisi

Publication Dates

• Publication in this collection
  21 Sept 2026
• Date of issue
  2026

History

• Received
  28 Apr 2026
• Accepted
  19 Aug 2026