HTML

ePub

PDF

Introduction

The Jaboticabeiras [Plinia cauliflora (Mart.) Kausel] occur generally in clusters that possibly were formed by related individuals, as already studied in the Southwest region of Paraná (DANNER, 2009; DANNER et al., 2010). However, due to the expansion of agricultural frontiers and forest fragmentation, occurrence sites of many species may have been lost, as they were previously found inside forests, with probable dispersal by animals (DANNER, 2009) and by Indians living in the region.

Forest fragmentation reduces the size of the reproductive population and the population density of plant species. Thus, the main consequences of population fragmentation and reduction are genetic drift, increased inbreeding and decreased gene flow (KAGEYAMA; GANDARA, 1998), and in the case of jaboticabeira, these conditions have been aggravated by the presence of polyembryony resulting from apomixis.

This phenomenon tends to decrease intrapopulation genetic diversity and increase interpopulation genetic diversity, compared to that of the original conditions. However, one should also consider the reproductive system of the species, which is not yet fully understood (DANNER, 2009; DANNER et al., 2011a), the pollen dispersal distance, which is dependent on the area of the reproductive neighborhood, the density and behavior of pollinators, and seed dispersal. These factors influence gene flow and are determinants for genetic diversity and structuring (SEBBENN, 2006).

As genetic diversity represents the potential of a population to produce different genotypes, it is critical for species survival, adaptation and evolution, especially under environmental changes and for selection in genetic breeding (ISAGI et al., 2007; CARVALHO, 2004), being primordial for the species in question.

It is important to emphasize that through the characterization of genetic diversity and its manner of distribution, it is possible to make rational and sustainable use of genetic resources, and contribute to prebreeding studies, in order to understand the available germplasm before introducing it in the breeding program (DANTAS et al., 2012).

In genetic breeding programs, information about genetic divergence within a species is fundamental to the rational use of genetic resources (LOARCE et al., 1996), as well as for the biological conservation of species and for studies of evolution and ecology of populations (DURAN et al., 2009).

Genetic divergence refers to the level of heterogeneity of a population or individuals of a particular species and can be assessed by means of predictive processes or multivariate techniques. The multivariate techniques are especially useful because they allow obtaining estimates of the general combining ability, in order to provide information on the concentration of predominantly additive genes in their effects, thereby aiding in the indication of parents used in intrapopulation breeding programs (CRUZ et al., 2012; MIRANDA et al., 2003). Among the multivariate statistical techniques, principal component analysis and clustering methods (CRUZ; REGAZZI, 1997) will be described.

The principal component analysis consists of retaining, in order of estimation, the maximum information in terms of total variation contained in the original data. The cluster analysis aims to use classification criteria to gather parental groupings in such a way that there is homogeneity within the group and heterogeneity between groups (CRUZ, 1990).

The study of genetic divergence also contributes to the evolution of the species with the identification of divergent genotypes, enabling inference about the specific capacity of combinations and heterosis (OLIVEIRA et al., 2003).

Studies involving native jaboticabeiras provide parameters for the identification of favorable parents to obtain segregating populations, for hybridization programs, in the selection of superior genotypes and, as a consequence, the result of obtaining genetically improved populations.

Given this information, it is known that the characterization of existing genetic resources serves as a strategy to mitigate damages already caused and to increase the exploitation of jaboticabeira in commercial crops (CITADIN et al., 2010), in agroforestry systems, and for use in permanent preservation areas and legal reserves (ZERBIELLI et al., 2016).

The genetic divergence between native jaboticabeiras plants, based on physical and chemical characteristics, was characterized in order to identify parents for use in future hybridizations, thereby seeking to broaden the genetic basis for these characteristics and making available information pertinent to the breeding program of this species.

Material and Methods

The study was carried out on a population of jaboticabeiras [Plinia cauliflora; (Mart.) Kausel] found in fragmented mixed ombrophilous forests or forests containing araucaria (DANNER et al., 2010). (26°26'17" S; 52°19'20" W; 963 m in altitude), located in the municipality of Clevelândia, in the Southwest region of Paraná. The study area has 12.3 hectares and 930 adult jaboticabeiras.

In 2013, 60 (sixty) fruits per plant were collected, and 80 were collected in 2014 (31/Oct/2013 and 29/Oct/2014), with 70 genotypes characterized in 2013 and 56 in 2014, of adult reproductive individuals, within the sampled plot.

Of the genotypes collected in 2013 (designated 54, 106, 169, and 212), the identifications were removed the next year; in numerically equivalent terms, in the 2014 survey, these genotypes received new codes: J7-01, J7-02, 345, and 347, respectively. Thus, it is possible that these renamed genotypes had also been evaluated in 2013, though there is no way to distinguish them.

The samples of ripe fruits were taken to the Laboratory of Vegetal Physiology, Universidade Tecnológica Federal do Paraná - Câmpus Dois Vizinhos, where the following assessments were performed.

The physicochemical characteristics were determined: polar diameter (PD, in mm) and equatorial fruit (ED, in mm); mass of the total fresh matter (g, gram); mass of the fresh matter of the bark (g); mass of the fresh seed matter (g); percentage of pulp (%, percentage); and content of total soluble solids (SST, ° Brix). The pulps were stored in a freezer (-18°C) for 15 days, and biochemical analyzes were then performed, including protein content [method of Bradford (1976)], total sugars [method of Dubois (1956)], anthocyanins and flavonoids [method of Lees; Francis (1972)], total phenols [method of Singleton et al. (1999)] and total titratable acidity (ATT) (INSTITUTO ADOLFO LUTZ, 2008). The ratio SST / ATT was estimated by the simple division of the means of the repetitions of SST by ATT.

To calculate the frequency, genotypes were ranked in terms of each of the 15 analyzed variables, from the 1^st to the 70^th or 56^th genotype evaluated (in 2013 or 2014, respectively). Of these, 13 variables were classified by means of the variables in decreasing order, except for the variables "shell mass" and "total ATT", which were ranked in ascending order. The rank of each genotype in each of the variables was added, and applying the total value of the sum, the genotypes were then classified in ascending order. Genotypes ranked among the top 20% were selected (up to 14^th in 2013 and up to 11^th in 2014) (WAGNER JÚNIOR, 2007).

For the analysis of the genetic divergence, 33 genotypes were selected, with repeatability based on the characteristics of the main components. The cluster analysis was performed by the Tocher and nearest neighbor methods, using as a measure of dissimilarity the distance of Mahalanobis (CRUZ et al., 2012).

Results and Discussion

For each evaluation year, the genotypes ranked among the top 20% were selected (Table 1) (up to 14^th in 2013 and up to 11^th in 2014).

Genotypes 16 and 194 were highlighted in both selection indications, which allows them to be used as future cultivars and / or parents in order to obtain quality fruits, as based on physicochemical and nutraceutical variables.

Genotypes 101, 103, 104 and 105, J7-01 and J7-02, and 153 and 154 have proximity within the sampled area (Figure 1). It was assumed that these genotypes could be clones from apomixis which, according to Danner et al. (2011b), can occur with jaboticabeira, as they are polyembryonic. In this case, there is a zygote embryo and one or more asexual embryos formed by apomixis, in which genome clones were generated by plants, with the intention of forming uniform orchards.

As for the assessment of divergence by the principal components method carried out in 2013, it was demonstrated that six variables [equatorial (CP1) and polar (CP2) diameters and mass of fresh fruit (CP3), bark (CP4), seed (CP5) and pulp (CP6)] explain approximately 85% of the variation obtained by the 33 accessions studied.

The equatorial diameter explains 38.48% of the total variation, considering the component of greater relevance. The variables that contributed minimally to the study of genetic divergence among jaboticabeira genotypes were SST, ATT, proteins, total sugars, flavonoids, anthocyanins and total phenols, with a variation of less than 4%. In 2014, the same six main components were used, with these explaining 82.5% of the variation between the 33 accessions (Table 2). The importance of a variable being evaluated was related to the percentage of total variance that it explained (CRUZ et al., 2012).

The maximum heterotrophic effect in controlled crosses between the most divergent genotypes can be observed in Figures 2 A, B, C, D and E, with the formation of seven, six, five, five and four groups, respectively. In all the main components, it was always observed that jaboticabeiras 148 and 16 were each in isolated groups due to the divergence of these with the others in the characteristics analyzed. It is known that the use of parental genotypes with greater divergence increases the heterosis manifested in the hybrids, with probability of occurrence of superior segregants in advanced generations, and must be used to recommend crosses, avoiding those genotypes within the same group (DESTRO, 1991).

It was verified that the largest group comprised jaboticabeiras 204, 120, 118, 100, 107, 11, 104, 105, 108, 79, 117, 49, 88, 119, 47, 10, 42, 194, 163 and 80 (Figure 2-A). The second largest group occurred with genotypes 35, 101, 98 and 65, along with 57, 177, 96 and 162, both groups being followed by the formation of 166 with 102. Genotype 41, as well as those already cited (16 and 148), appeared in clusters formed by an individual.

The jaboticabeiras were framed together in a single group including 10, 11, 41, 42, 47, 49, 57, 65, 79, 88, 96, 98, 100, 101, 102, 105, 107, 108, 117, 118, 119, 120, 162, 163, 166, 177, 194, and 204; and the rest were isolated, forming five groups with a single genotype in each, these being formed by 148, 35, 104, 80 and 16 (Figure 2-B).

A similar grouping can be observed in Figure 2C, differing only by the presence of five groups and replacing of genotype 104 by genotype 100. In this way, the groups were constituted by 148 (I), 35 (II), 100 (III), 16 (IV), and 10, 11, 41, 42, 47, 49, 57, 65, 79, 80, 88, 96, 98, 101, 102, 104, 105, 107, 108, 117, 118, 119, 120, 162, 163, 166, 177, 194, and 204 (V).

Analyzing Figure 2D, which involves the relationship of the components, equatorial diameter × mass of the fresh seed matter, jaboticabeiras 10, 11, 35, 42, 47, 49, 57, 65, 79, 88, 98, 100, 102, 104, 105, 107, 108, 117, 118, 119, 120, 148, 166, 194 and 204 compose one group, which is followed by the group constituted by genotypes 163, 80, 162, 96, 41 and 177 as the second major grouping, and finally, three isolates are represented by jaboticabeiras 148, 101 and 16.

In Figure 2-E, the number of groups was reduced in relation to that of the other main components present (Figure 2A, B, C, and D); however, the largest group involved 27 of the 33 jaboticabeira described, separating this from genotypes 35, 101, 98, and 102, which constituted a group, and from 148 and 16, which formed two other isolates.

Figure 3-A, B, C, D and E reveals the formation of seven, six, nine, nine and seven groups, respectively. The results show greater divergence between the main components in 2014 compared to that in 2013. As seen in Figure 3A, there were two groups with one genotype (35 and 105), others with three (102, 117, and 119; 16, 104, and 88), four (57, 47, 49, and 42), and seven (204, 162, 80, 194, 118, 108, and 101), and the largest with fourteen (107, 96, 100, 163, 41, 177, 11, 79, 65, 148, 120, 10, 98, and 166).

In Figure 3B, six groups were formed, but in five of these, there were few accesses [57 (group I), 88 (group II), 105 (group III), 47 and 49 (group IV), 108, 117 and 119 (group V), and in the largest group, 25 accesses, 10, 11, 16, 35, 41, 42, 65, 79, 80, 96, 98, 100, 101, 102, 104, 107, 118, 120, 148, 162, 163, 166, 177, 194 and 204 (group VI)].

In relation to Figures 3C and 3D, both formed nine genotype groups; however, the groups with the lowest number of genotypes did not have similarity, the only exception being 105, which appeared isolated in both figures. This genotype was also isolated in the other analyzed components (Figures 3A, 3B, and 3E), indicating such divergence in relation to the others.

In the analysis demonstrated by Figure 3, the other eight groups were formed by isolated genotypes 119, 108 and 16, along with two groups of two genotypes each (104 and 88; 42 and 100). In addition, two groups each consisted of six genotypes, with 57, 117, 107, 102, 49 and 47 in one group and 194, 98, 166, 101, 35, and 118 in the other. The group with the highest number of genotypes was formed by 10, 11, 41, 65, 79, 80, 96, 120, 163, 177 and 204.

When comparing Figures 3-C and 3D, a larger number of jaboticabeiras was observed in a single group, with both figures revealing that nine groups were formed: the groups formed by an individual involved genotypes 57, 49, 162, 194 and 105; one group of two genotypes with 42 and 107; one group of four with 102, 47, 117 and 119; one group of nine consisting of 11, 41, 79, 80, 96, 100, 163, 177 and 204; and a group with thirteen genotypes including 10, 148, 166, 101, 35, 65, 120, 98, 118, 108, 3, 104 and 88.

One group in figure 3-E is the largest, containing 22 of the 33 genotypes analyzed, these being 11, 35, 41, 42, 65, 79, 80, 96, 98, 100, 101, 107, 108, 118, 120, 148, 162, 163, 166, 177, 194 and 204. The other groups had three (119, 117 and 47; 88, 104 and 16), two (102 and 49), and one genotype each (57, 10 and 105).

In general, when comparing the results of the main components in both years of analysis (Figures 2-A, B, C; D, and E, and Figures 3-A, B, C, D, and E) it was verified that, even when using the same variables, most of the groups formed did not have the same formation with the exception of some joined accesses.

This fact was related to the variation in the values of the averages obtained from one year to the next in the analyzed variables; similar behavior was previously described by Citadin et al. (2005). This relationship also reflects the fact that such plants are found in nature, without any kind of management, which allows such behavior from one year to the next. It is important to perform such evaluations in additional cycles because in this way, we can apply repeatability and stability analyses based on such hypotheses described above.

For the formation of the dendrogram using the nearest neighbor method, regarding the analyses of 2013, the largest distance was considered to be the value between 84 and 40 (obtained by D2), which was set as 100%. The greatest divergence observed among populations was between accesses 96 and 148, belonging to groups I and IV, by the Tocher method (Table 3) and/or groups II and I by the nearest neighbor method (Figure 4), thus demonstrating the greater heterotic potential of these two combinations.

The smallest divergence occurred between jaboticabeiras 79 and 108, both belonging to the same group in both analyses (Table 3 and Figure 4). Thus, the use of these individuals in controlled hybridizations is not recommended, since endogamy can lead to inbreeding depression, causing loss of vigor by the homozygosis of deleterious genes (MIRANDA FILHO, 2001; FALCÃO et al., 2001).

These results emphasize divergence, as the physical location between accesses 79 and 108 (72 m) within the sampled area, although presenting a smaller divergence, was more distant than were accesses 96 and 148 (57 m), which were among the most distant in relation to the others (Figure 4). It was assumed that these factors might be related to some seed disperser that geographically spread the accesses of lower divergence (79 and 108).

The seed dispersers of this species are rodents, birds and monkeys (GRESSLER et al., 2006), which can be found in such fragmented forests. In addition, it is possible to suggest the presence of common kinship, a fact that remarkably illustrates how the local ecosystem originally involved the existence of a contiguous Araucaria Forest in this region before anthropic exploration, which facilitated gene flow between individuals.

In Figure 4, three groups were identified, two with a single individual (148 and 119) and the other with one. In the dendrogram formed based on the analyses of 2014, by the nearest neighbor method, it was considered to be greater distance than that characterized by the values of 230 and 31. This greater divergence was between jaboticabeiras 79 and 119, with less divergence between 96 and 98.

The most divergent genotypes were categorized in groups I and IV by the Tocher method (Table 3), and in groups I and V by the nearest neighbor method (Figure 5), verifying among groups a greater heterotic potential in these two combinations as well.

The distance of the most divergent genotypes (148 and 119) was 79 m within the area, with those of smaller value (96 and 98) being 6.6 m. In the latter, it may have been influenced by inbreeding or even by material of the same genetic constitution, since in jaboticabeira polyembryony is common, as results from apomixis.

By the Tocher method, it was observed in Table 4 that there were six groups, consisting of the following: one genotype in groups VI (119), V (118), IV (57), two in group III (16 and 42), five in group II (47, 204, 49, 162 and 194), and the other 21 jaboticabeiras in group I.

Genetic variability is of fundamental importance for species evolution. It is also in populations with genetic variability that the selection of genotypes yielding phenotypes with characteristics of agronomic interest, such as larger and tastier fruits and resistance to diseases and pests, can be accomplished.

It was verified in this study that there were some differences between the populations grouped by the Tocher method versus those obtained with the dendrogram using the closest neighbor grouping method (Table 4 and Figure 5, respectively).

However, what was visualized is that accesses 47, 204, 49 and 162 exhibited union within the same group in both analyses (Table 4 and Figure 5), both 119 and 57 were present as isolates, and the largest group was formed by 10 (1), 11 (2), 35 (4), 41 (5), 65 (10), 79 (11), 80 (12), 88 (13), 96 (14), 98 (15), 100 (16), 101 (17), 102 (18), 104 (19), 105 (20), 107 (21), 108 (22), 117 (23), 120 (26), 148 (27), 163 (29), 166 (30), and 177 (31).

According to Figure 5, five groups were formed: group I by jaboticabeira 119 (25), group II with 42 (6), group III having 162 (28), 49 (8), 204 (33), and 47 (7), group IV by 57 (9) and group V by the others [10 (1), 11 (2), 16 (3), 35 (4), 41 (5), 65 (10), 79 (11), 80 (12), 88 (13), 96 (14), 98 (15), 100 (16), 101 (17), 102 (18), 104 (19), 105 (20), 107 (21), 108 (22), 117 (23), 118 (24), 120 (26), 148 (27), 163 (29), 166 (30), 177 (31), and 194 (32)].

Carpentieri-Pípolo et al. (2000) recommend the use of parental genotypes; however, these have the greatest possible divergence to maximize heterosis in hybrids, increasing the probability of occurrence of higher segregants in advanced generations and broadening the genetic base. This factor must be taken into account in the choice of which jaboticabeiras will be used as mothers.

Conclusion

There are differences between the populations of jaboticabeira plants grouped by the Tocher method versus those obtained with the dendrogram by the nearest neighbor grouping method.

There is genetic variability between the jaboticabeira plants analyzed, and therefore, it is recommended to perform hybridization between genotypes 79 with 119 and 96 with 148.

Acknowledges

The CNPq by financial support and CAPES by doctoral scholarship from second, fourth, fifth and sixth author

References

BRADFORD, M. M. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Analitycal Biochemistry, Orlando, v. 72, n. 1-2, p. 248-254, 1976.

CARPENTIERI-PÍPOLO, V.; DESTRO, D.; PRETE, C. E. C.; GONZALES, M. G. N.; POPPER, I.; ZANATTA, S.; SILVA, F. A. Seleção de genótipos parentais de acerola com base na divergência genética multivariada. Pesquisa Agropecuária Brasileira, Brasília, v. 35, n. 8, p. 1613-1619, 2000.

CARVALHO, M. A. Germplasm characterization of Arachis Pintoi Krap. and Greg. (Leguminosae). 2004. Tesis (Doctor in Agronomy) - University of Florida, Flórida.

CITADIN, I.; DANNER, M. A.; SASSO, S. A. Z. Jabuticabeiras. Revista Brasileira de Fruticultura, Jaboticabal, v. 32, n. 2, p. 577-583, 2010.

CITADIN, I.; VICARI, I. J.; SILVA, T. T.; DANNER, M. A. Qualidade de frutos de jabuticabeira (Myrciaria cauliflora) sob influência de duas condições de cultivo: sombreamento natural e pleno sol. Nota técnica. Revista Brasileira de Agrociência, Pelotas, v. 11, n. 3, p. 373-375, 2005.

CRUZ, C. D. Aplicação de algumas técnicas multivariadas no melhoramento de plantas. 1990. Tese (Doutorado em Agronomia) - Escola Superior de Agricultura Luiz de Queiroz, Piracicaba.

CRUZ, C. D.; REGAZZI, A. J. Modelos biométricos aplicados ao melhoramento genético. Viçosa, MG: Imprensa Universitária, 1997. 390 p.

CRUZ, C. D.; REGAZZI, A. J.; CARNEIRO, P. C. S. Modelos biométricos aplicados ao melhoramento genético. 4. ed. Viçosa: UFV, 2012. 514 p. v. 1.

DANNER, M.A. Diagnóstico ecogeográfico e caracterização morfogenética de jabuticabeiras. 2009. Dissertação (Mestrado em Agronomia) - Universidade Tecnológica Federal do Paraná. Programa de Pós-Graduação em Agronomia, Pato Branco.

DANNER, M. A.; CITADIN, I.; SASSO, S. A. Z.; AMBROSIO, R.; WAGNER JUNIOR, A. Armazenamento a vácuo prolonga a viabilidade de sementes de jabuticabeira. Revista Brasileira de Fruticultura, Jaboticabal, v. 33, n. 1, p. 246-252, 2011b.

DANNER, M. A.; CITADIN, I.; SASSO, S. A. Z.; SCARIOT, S.; BENIN, G. Genetic dissimilarity among jabuticaba trees native to Southwestern Paraná, Brazil. Revista Brasileira de Fruticultura, Jaboticabal, v. 33, n. 2, p. 517-525, 2011a.

DANNER, M. A.; CITADIN, I.; SASSO, S. A. Z.; TOMAZONI, J. C. Diagnóstico ecogeográfico da ocorrência de jabuticabeiras nativas no Sudoeste do Paraná. Revista Brasileira de Fruticultura, Jaboticabal, v. 32, n. 3, p. 746-753, 2010.

DANTAS, A. C. A.; NUNES, G. H. S.; ARAÚJO, I. S.; ALBUQUERQUE, L. B. Caracterização molecular de acessos de melão coletados no nordeste brasileiro. Revista Brasileira de Fruticultura, Jaboticabal, v. 34, n. 1, p. 183-189, 2012.

DESTRO, D. Capacidade de combinação de genótipos de soja (Glycine max (L.) Merrill) apropriados para o consumo humano. 1991. Tese (Doutorado em Agronomia) - Escola Superior de Agricultura Luiz de Queiroz, Piracicaba.

DUBOIS, M.; GILLES, K. A.; HAMILTON, J. K.; REBERS, P. A.; SMITH, F. Colorimetric method for determination of sugars and related substances. Analitycal Biochemistry, Orlando, v. 28, n. 3, p. 350-356, 1956.

DURAN, C.; APPLEBY, N.; EDWARDS, D.; BATHEY, J. Molecular genetic markers: discovery, applications, data storage and visualisation. Current Bioinformatics, Roma, v. 4, n. 1, p. 16-27, 2009.

FALCÃO, A. J. S.; MARTINS-FILHO, R.; MAGNABOSCO, C. U.; BOZZI, R.; LIMA, F. A. M. Efeitos da endogamia sobre características de reprodução, crescimento e valores genéticos aditivos de bovinos da raça Pardo-Suíça. Revista Brasileira de Zootecnia, Piracicaba, v. 30, n. 1, p. 83-92, 2001.

GRESSLER, E.; PIZO, M. A.; MORELLATO, L. P. C. Polinização e dispersão de sementes em Myrtaceae do Brasil. Revista Brasileira de Botânica, São Paulo, v. 29, n. 4, p. 509-530, 2006.

INSTITUTO ADOLFO LUTZ - IAL. Métodos físico-químicos para análise de alimentos. Coordenadores Odair Zenebon, Neus Sadocco Pascuet e Paulo Tiglea. Versão eletrônica. 4. ed. digital. São Paulo: IAL, 2008. 1020 p.

ISAGI, Y.; TATENO, R.; MATSUKI, Y.; HIRAO, A.; WATANABE, S.; SHIBATA, M. Genetic and reproductive consequences of forest fragmentation for populations of Magnolia obovata. Ecological Research, Tokyo, v. 22, p. 382-389, 2007.

KAGEYAMA, P. Y.; GANDARA, F. B. Consequências genéticas da fragmentação sobre populações de espécies arbóreas. Série Técnica IPEF, Piracicaba, v. 12, n. 32, p. 65-70, 1998.

LEES, D. H.; FRANCIS, F. J. Standardization of pigment analyses in cranberries. HortScience, Alexandria, v. 7, n. 1, p. 83-84, 1972.

LOARCE, Y.; GALLEGO, R.; FERRER, E. A comparative analysis of the genetic relationship between rye cultivars using RFLP and RAPD markers. Euphytica, Wageningen, v. 88, n. 1 p. 107-115, 1996.

MIRANDA FILHO, J. B. Endogamia ou consaguinidade. In: NASS, L. L.; VALOIS, A. C. C.; MELO, I. S.; VALADARES-INGLIS, M. C. (Ed.). Recursos genéticos e melhoramento: plantas. Rondonópolis: Fundação MT, 2001. p. 629-648.

MIRANDA, G. V.; COIMBRA, R. R.; GODOY, C. L.; SOUZA, L. V.; GUIMARÃES, L. J. M.; MELO, A. V. de. Potencial de melhoramento e divergência genética de cultivares de milho-pipoca. Pesquisa Agropecuária Brasileira, Brasília, v. 38, n. 6, p. 681-688, 2003.

OLIVEIRA, F. J.; ANUNCIAÇÃO FILHO, C. J.; BASTOS, G. Q.; REIS, O. V. Divergência genética entre cultivares de caupi. Pesquisa Agropecuária Brasileira, Brasília, v. 38, n. 5, p. 605-611, 2003.

SEBBENN, A. M. Sistemas de reprodução em espécies tropicais e suas implicações para a seleção de árvores matrizes para reflorestamentos ambientais. In: HIGA, A. R.; SILVA, L. D. (Coord.). Pomar de sementes de espécies florestais nativas. Curitiba: FUPEF, 2006. p. 93-138.

SINGLETON, V. L.; ORTHOFER, R.; LAMUELA-RAVENTÓS, R. M. Analysis of total phenols and other oxidation substrates and antioxidants by means of Folin-Ciocalteu reagent. Methods in Enzymology, New Jersey, v. 299, p. 152-178, 1999.

WAGNER JÚNIOR, A. Seleção de pessegueiro adaptado ao clima subtropical. 2007. Tese (Doutorado em Fitotecnia) - Universidade Federal de Viçosa, Viçosa, MG.

ZERBIELLI, L.; NIENOW, A. A.; DALACORTE, L.; JACOBS, R.; DARONCH, T. Diversidade físico-química dos frutos de jabuticabeiras em um sítio de ocorrência natural. Revista Brasileira de Fruticultura, Jaboticabal, v. 38, n. 1, p. 107-116, 2016.

Author notes

*Author for correspondence

IR