SciELO - Scientific Electronic Library Online

vol.35 número2Desempenho de sementes de feijão infectadas pelo fungo Sclerotinia sclerotiorumAspectos anatômicos e histoquímicos da semente de pupunha (Bactris gasipaes Kunth) índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados


Journal of Seed Science

versão impressa ISSN 2317-1537

J. Seed Sci. vol.35 no.2 Londrina  2013 

Coefficient of variation of normal seedlings obtained from the validation of methods for the seed germination testing of 20 species belonging to the family Fabaceae


Coeficientes de variação de plântulas normais obtidas na validação de métodos para teste de germinação de sementes de 20 espécies florestais da família Fabaceae



Vanderley José Pereira; Denise Garcia Santana*

Universidade Federal de Uberlândia, Instituto de Ciências Agrárias e Florestais, Caixa Postal 593, 38400-902 - Uberlândia, MG, Brasil




The standardization of inter-laboratory results of germination test of forest species seeds requires that the methods be robust. Therefore, the objective was to compare and discuss, through the coefficient of variation for normal seedlings, the variabilities present in the process of validation methods obtained in the germination test for seeds of 20 species of the family Fabaceae. Coefficients of variation for the experiment by lot and by laboratory were calculated for normal seedlings from the statistical analysis of method validations. For normal seedlings of 20 Brazilian forest species, the coefficients of variation are low (up to 9.84%), to average (up to 17.66%), contrary to expectations due to high genetic variability in these barely improved species. The increase of the coefficient is not related to treatment for breaking dormancy, but it grows as the lot quality decreases. The high coefficients by laboratory, overestimated by the lot effect, are uniform indicating that the methods are repeatable. The coefficient is not an indicator capable of predicting the heterogeneity of model variance. As normal distribution models random events, randomness is present in the validation process of the 20 forest species of the Fabaceae family.

Index terms: dormancy, statistical assumptions, forest seeds, data transformation, experimental variability.


A uniformização dos resultados inter laboratoriais de testes de germinação de sementes de espécies florestais exige que os métodos sejam robustos. Assim, o objetivo foi comparar e discutir, por meio do coeficiente de variação obtido para plântulas normais, as variabilidades presentes no processo de validação de métodos obtidas no teste de germinação de sementes de 20 espécies da família Fabaceae. Coeficientes de variação para o experimento, por lote e por laboratório para a variável plântulas normais foram calculados. Os coeficientes de variação obtidos para plântulas normais de 20 espécies florestais nativas, são de baixos (até 9,84%) a médios (até 17,66%), contrariando o esperado pela grande variabilidade genética dessas espécies pouco melhoradas. O aumento do coeficiente de variação não está relacionado ao tratamento utilizado para superação de dormência, porém cresce à medida que a qualidade do lote decresce. Os altos coeficientes obtidos por laboratório, superestimados pelo efeito de lotes, são uniformes indicando que os métodos são reproduzíveis. O coeficiente de variação não é um indício capaz de predizer a heterogeneidade das variâncias do modelo. Como a distribuição normal modela eventos aleatórios, a aleatoriedade está presente no processo de validação de métodos das 20 espécies florestais da família Fabaceae.

Termos para indexação: dormência, pressuposições estatísticas, sementes florestais, transformação de dados, variabilidade experimental.




Native species seed trading is still new compared to cultivated seed trading, due to the lack of supervision and the informality of the sector, thus, producers end up using seeds that they have collected themselves (Gonzáles and Torres, 2003). From 2001, the production of native seedlings boosted from the mandatory restoration of areas on farms (Lorza, 2009).

With a new perspective of plant resources, particularly in regard to forestry, Law No. 10.711/2003 (Brasil, 2004) has established a new phase and outlook for the sector of production and marketing of forest seeds and seedlings (Iede, 2005). Among the prospects, germination test standardization stands out.

Germination involves a sequence of biochemical, morphological and physiological events (Nonagaki et al., 2010), simplified in initial processes, such as seed imbibition, activation of metabolism (biochemical method), and ending in root protrusion (Labouriau, 1983 ; Nonagaki et al., 2010). The quantification and characterization of germination by seed technologists are determined by the development of embryo structures (Brasil, 2009; Nonagaki et al., 2010; ISTA, 2011).

There are several contributions to the standardization of laboratory procedures for analysis of forest seeds (Wielewicki et al., 2006; Brüning et al., 2011), however, with no assessment of replicability. The interest in standardization arises from the need to accurately determine the quality of seeds (Wielewicki et al., 2006). However, the fragmentation of information and the inconsistent reporting of the germination process, sometimes only measured by the number of germinated seeds, hinder the inclusion of a method into official rules (Santana et al., 2012), added to the difficulty of meeting principles of metrology (Kataoka et al., 2011).

It is noteworthy that in the Rules for Seed Testing, in its latest issue (Brasil, 2009), there are methods proposed for germination tests in 1365 species. Of these, just over 276 are forest and shrub species (Ferraz and Calvi, 2011), mostly not native to Brazil. In order for a method for germination test to be validated and included in the official rules, procedures proposed by ISTA (2007) must be followed. It is a collaborative process between the laboratories, to ensure that the procedure provides reliable and reproducible results. To be validated, the method has to have accuracy, robustness, reproducibility and repeatability (Kataoka et al., 2011). The manual of ISTA brings statistical techniques that are widely used and robust, but that do not restrict innovation. As the process involves a factorial design, consisting of laboratories and plots, one of the measures of statistical precision that can be used in the evaluation of the process is the coefficient of variation (CV). This measure allows the comparison of accuracy among experiments without the need to have equal units (Amaral et al., 1997), for characterizing the relative dispersion of data in relation to the mean value (Santana and Ranal, 2000).

To study the coefficient of variation of the same character in different experiments, the researcher must have experience with the variable (Steel and Torrie, 1980), although bands for classification have been proposed (Pimentel-Gomes, 2000). However, this classification is too broad and does not consider the peculiarities of the species studied, and mainly makes no distinction as to the nature of the trait (Garcia, 1989; Costa et al., 2002).

Several articles have demonstrated the need to establish the limits for classification of CV for agricultural crops of interest (Amaral et al., 1997; Costa et al., 2002; Carvalho et al., 2003), for forage species (Ambrosano and Schammas , 1994; Clemente and Muniz, 2002) and for the silvicultural characters of forest species of the genus Pinus and Eucalyptus (Garcia, 1989). For forest seeds, there is no rating scale; the measure is only used to ascertain the reliability of the experiment. The aim was to compare and discuss, through the coefficient of variation for normal seedlings, the variability present in the process of validation of methods obtained in the germination test of seeds of 20 species belonging to the family Fabaceae.


Material and Methods

The study used data from the process that resulted in the validation of methods for testing the germination of seeds of 20 tree species of the family Fabaceae. The family Fabaceae was chosen on the basis of available literature on dormancy and germination of seeds, and the availability for purchase, donation and seed collection. The details of the method with information on pre-germination treatments, scores, substrate, temperature and photoperiod are in Table 1. Before the formation of lots, number of samples in seven to 16 (Table 1) were formed from seeds of at least five sources, with at least four matrices. The plots, at least three, were formed by mixing the samples with seed quality physiological to generate similar variability, as well as similarity in size and shape. In lots of randomization sent to each lab (minimum six per species) were provided for four eight replications of 50 seeds, totaling 200 and 400 seeds, respectively. Moreover, even the laboratories have evaluated several features of germination of normal seedlings was examined only for purposes of validation.

In a preliminary analysis the test of Box-plot was applied, eliminating the outliers, known as "outliers" according to the Manual of Validation (ISTA, 2007). To test the assumptions of the variance analysis model, the Shapiro-Wilk test was used for the normality of model residuals, and Levene for the homogeneity of variances, both at 0.01 significance level. When both assumptions were met, the F test of Snedecor was applied, also at 0.01 of significance for the main effects (plot and lab) and for interaction. When at least one of the assumptions was not met, the percentage of normal seedlings were transformed by arcsine and assumptions were tested again. From the model of analysis, the experimental coefficient of variation was determined by the equation: ; where QMR: is the mean square of the residual, and :is the general mean of the experiment. The coefficient of variation per plot was found using the equation: , where Si..: is the standard deviation of normal seedling percentage obtained from the i-th plot, and : is the normal seedling percentage mean obtained from the i-th plot. The coefficient of variation per lab was determined using the equation , where S.j. : is the standard deviation of normal seedling percentage obtained from the j-th lab, and j.j: is the normal seedling percentage mean obtained from the j-th lab.


Results and Discussion

The synthesis of the statistical analysis of the seed germination test method validation of 20 native species of the family Fabaceae showed no significant laboratory effect, as well as interaction, and significant plots for the percentage of normal seedlings for all species (Table 2). The biggest impact of this result was the finding that even when seeds were subjected to specific pre-germinal treatments, such as clipping (lopping), and scarification, among others, the results were similar in different laboratories. According to the Manual of Validation of the International Seed Testing Association (ISTA, 2007), the goal of validation is to obtain comparable results among laboratories. The F statistics for the main effects and interaction were applied without neglecting the residual normality assumptions of the model and the homogeneity of variances. The statistical analysis was entirely based on the Manual of Validation of the International Seed Testing Association (ISTA, 2007).


Table 2. Summary of the analysis of variance for normal seedling percentages of the seed germination test method validation process of 20 Fabaceae species.

**; ns significant and non-significant, respectively at 0.01 of significance by ANOVA F-test.


The experimental coefficients of variation (CV) (model with main effects and interaction) for normal seedlings were low (up to 9.84%) to medium (up to 17.66%), with no records of high CV (20 <CV < 30%) nor very high (CV> 30%) (Table 3). The low and medium values of CV sensu Pimentel-Gomes (2000) showed accuracy in the variation of repetitions of same laboratory and plot combination, thus, it can be inferred that the proposed methods for testing the germination of the seeds of these species had repetitiveness. Residual normality assumptions and homoscedasticity were treated to the percentage of normal seedlings for all species, some with data transformation of the angular type. Species with low coefficients of variation for normal seedlings, such as Acacia polyphylla, Enterolobium contortisiliquum and Mimosa caesalpiniaefolia, showed heterogeneous variances and therefore required transformation (Table 4). However, species with the highest coefficients of variation were Mimosa scabrella, Dalbergia miscolobium and Ormosia arborea with 15.32 and 17.40 and 17.66%, respectively; variances were homogeneous with the original data and did not require data transformation. This revealed that the CV was no evidence capable of predicting the heterogeneity of variances and, consequently, there is a need to transform data.


Table 3. Coefficients of variation for the percentages of normal seedlings of the germination test methodology validation process of 20 species of the family Fabaceae, including the assumptions of the model and precision of the experiment, with adjectives.

1adjectives sensu Pimentel-Gomes (2000); 2W and F: statistics from the Shapiro-Wilk and Levene tests, values in bold indicate normality of residuals and homogeneity of variances, respectively.



Table 4. Coefficient of variation for normal seedlings per plot formed for the germination test methodology validation process of 20 species of the family Fabaceae.

1Data transformed by arcsine , where x is the percentage of normal seedlings, CV: coefficient of variation.


Data transformations reduce the scale of values and thus are more appropriate where the problem is heterogeneity of variances and not normality, since the latter reflects the behavior of the species (Santana and Ranal, 2000). Thus, meeting the assumption of normality is a desirable goal, but not the main one, because of the robustness of the linear models to non-normality, however sensitive to heteroscedasticity (Faraway, 2006). In the literature, several transformations have been suggested for data expressed as proportions or percentages (Piepho, 2003) and aim to break the dependency between mean and variance, which is characteristic of this type of variable (Santana and Ranal, 2000). When used arbitrarily, it results in undesirable effects, since there is no guarantee of the non-violation of the assumptions of the analysis of variance model (Sileshi, 2012). In addition, a transformation that corrects the violation of an assumption could result in the violation of another (Sileshi, 2007).

Of the 20 native species, only Peltogyne confertiflora presented residuals with non-normal distribution for the percentages of normal seedlings. Once transformed, the characteristic met the assumption (Table 3). However, some authors have reported that non-normality is prevalent in ecological data (Warton and Hui, 2011) and will likely be the rule and not the exception in germination (Sileshi, 2012). Unlike most major crops, native forest species have great genetic diversity (Sarmento and Villela, 2010). Forest trees mostly reproduce by crosses; they occupy large geographical areas and different "habitats", they are then expected to have high levels of genetic variation (Hamrick and Loveless, 1986). In fact, studies on the reproductive systems of tropical tree species, by isozymes, have shown that the vast majority is alogamous or have mixed system, predominantly alogamous (Murawski, 1995). Their wide geographical distribution, type of reproduction and interactions between genetic and environmental factors lead to great diversity, also noted in the physiological characteristics of seeds and in their germination process (Rego et al., 2005).

Normal distribution essentially shapes random events, which individual occurrence does not follow rules or standards. Because they do not have genetic improvement and are susceptible to several environmental factors, randomness was present in the germination test method validation process of 20 species, justifying the normal distribution of residuals. In summary, research has proved theory. Santana et al. (2012), when working on validation of methods for testing the germination of seeds of 10 species native to Brazil, have found that for all species the model residuals for normal seedlings had normal distribution and variances were homogeneous. The F statistic showed no differences in the percentage of normal seedlings among laboratories, nor interaction between laboratory and plot, only differences in the percentages of plots were noted (Table 2). In fact, plots were formed to provide different percentages of normal seedlings, which was confirmed by analysis of variance. It is consistent that seeds from plots with different qualities tend to have different responses to the method (Oliveira et al., 2005; Kataoka et al., 2011). Martins and Nakagawa (2008) found that plots may respond differently to pre-germination treatments. However, the methods have been extensively tested and validated after proved to be robust, ie efficiency ought to be maintained regardless of plot quality. The ideal method should withstand adverse conditions of application and be also precise and accurate for detecting subtle differences in quality (Waeny, 1980). Accordingly, one should prioritize treatments for breaking efficient cutaneous dormancy, which are independent of plot origin (Smiderle and Souza, 2003; Borges et al., 2004).

Except for Acacia polyphylla, Anadenanthera macrocarpa, Dalbergia miscolobium, Parapiptdenia rigida and Peltogyne confertiflora, which have not had dormant seeds, the seeds of other species underwent treatments to overcome dormancy. The methods of seed dormancy overcoming did not cause great variations among repetitions, considering that coefficients of variation were medium sensu (Pimentel-Gomes, 2000), reaching a maximum of 17.66% (Table 3). No difficulty of standardizing expected, due to differences in the intensity of scarification and lopping, control of heat treatment temperature, among other factors, was found.

The coefficients of variation per plot were below 10% for high quality plots of twelve species, and over 20% for low-quality plots of nine species (Table 4). This result indicated that as plots quality decreased, the classification of normal seedlings became doubtful and coefficients of variation increased. Kataoka et al. (2011) noted that plot coefficient of variation for seeds in advanced stages of deterioration, ie, lower quality, showed the highest values. Although, normal seedlings of Mimosa caesalpiniaefolia and Peltogyne confertiflora, with dormant and non-dormant seeds, respectively, had plot coefficients of variation below 10% regardless of quality.

Still on the coefficient of variation per plot (Table 4), no significant relationship was found between the coefficient of variation and treatments for breaking dormancy, because even species with non-dormant seeds, such as Anadenanthera macrocarpa and Dalbergia miscolobium, showed high coefficients of variation (from 20 to 30%) for low quality plots. Similarly, Schizolobium parahyba var. amazonicum, a species which seeds have high degree of tegumentar dormancy, had the lowest coefficient of variation per plot, 1.66% for normal seedlings from the plot of higher quality.

The coefficients of variation estimated by laboratory (Table 5) were higher than by plot (Table 4), for covering the spectrum of germination conferred by the three plots with different qualities. However, this variation was uniform for all laboratories, indicating that the proposed method for testing seed germination of species of the family Fabaceae was reproducible. The coefficient of variation is a comparative measure of the existing variability between two experiments regarding factors that are not controlled, but never on the quality of the assay (Ranal and Santana, 2000). The differences in the magnitudes of lab coefficient of variation among species occurred primarily due to differences among plots. Lab coefficients of variation for normal seedlings of Parkia pendula were the highest among the species, but such results cannot be associated with the species. Variations in seed quality, from 19.3% (low quality) to 88.9% (high quality), have overestimated the coefficient of variation. In contrast, Schizolobium parahyba var. amazonicum low lab coefficients of variation were due in part to smaller difference in the quality of plots (from 97.7 to 65.7%).



For normal seedlings of 20 Brazilian forest species, the coefficients of variation vary from low (up to 9.84%) to medium (up to 17.66%), contrary to what is expected from the great genetic diversity of these species.

The increased coefficient of variation is not related to the treatment of seed dormancy overcoming, however, it increases as plot quality decreases.

The high coefficients of variation estimated by laboratory, and overestimated by the effect of plots, are uniform, which indicates that the methods are reproducible.

The coefficient of variation is no evidence capable of predicting the heterogeneity of variances and, consequently, there is a need to transform data.

Randomness is present in the process of method validation of the 20 forest species of the family Fabaceae, due to the normal distribution of residuals.



We thank the National Council for Scientific and Technological Development (CNPq) for granting the scholarship and FAPEMIG for their financial support.



ALMEIDA, M.J.B.; FERRAZ, I.D.K.; BASSINI, F. Estudos sobre a permeabilidade do tegumento e a germinação de sementes de Hymenaea courbaril L. (Caesalpinoideae), uma espécie de uso múltiplo. Revista da Universidade da Amazônia, v.8, n.1-2, p.63-71. 1999.         [ Links ]

AMARAL, A.M.; MUNIZ, J.A.; SOUZA, M. Avaliação do coeficiente de variação como medida da precisão na experimentação com citrus. Pesquisa Agropecuária Brasileira, v.32, n.12, p.1221-1225, 1997.         [ Links ]

AMBROSANO, G.M.B.; SCHAMMAS, E.A. Avaliação dos coeficientes de variação em experimentos com forrageiras. Boletim da Indústria Animal, v.51, n.1, p.13-20, 1994.         [ Links ]

BARAZETTI, V.M.; SCCOTI, M.S.V. Quebra de dormência e tipos de substrato para avaliação da qualidade fisiológica de um lote de sementes de bracatinga (Mimosa scabrella Bentham). Unoesc & Ciência, v.1, n.1, p.69-76, 2010.        [ Links ]

BORGES, E.E.L.; RIBEIRO JÚNIOR, J.I.R.; RESENDE, S.T.; PEREZ, S.C.J.G.A. Alterações fisiológicas em sementes de Tachigalia multijuga (Benth.) relacionadas aos métodos para a superação da dormência. Revista Árvore, v.28, n.3, p.317-325, 2004.        [ Links ]

BRASIL. Ministério da Agricultura, Pecuária e Abastecimento. Regras para análise de sementes. Ministério da Agricultura, Pecuária e Abastecimento. Secretaria de Defesa Agropecuária. Brasília. MAPA/ACS. 2009. 395p.        [ Links ]

BRASIL. Decreto nº 5.153, de 23 de julho de 2004. Regulamentação da Lei nº 10.711, de 5 de agosto de 2003, que dispõe sobre o Sistema Nacional de Sementes e Mudas - SNSM, e dá outras providências. Diário oficial da República Federativa do Brasil, Poder executivo, Brasília, DF, 23 jul. 2004.        [ Links ]

BRAZ, V.S.; KANEGAE, M.F.; FRANCO, A.C. Estabelecimento e desenvolvimento de Dalbergia miscolobium Benth. em duas fitofisionomias típicas dos cerrados do Brasil Central. Acta Botanica Brasilica, v.14, n.1, p.27-35, 2000.        [ Links ]

BRÜNING, F.O.; LÚCIO, A.D.; MUNIZ, M.F.B. Padrões para germinação, pureza, umidade e peso de mil sementes em análises de sementes de espécies florestais nativas do Rio Grande do Sul. Ciência Florestal, v.21, n.2, p.193-202, 2011.        [ Links ]

CARVALHO, P.E.R. Espécies arbóreas brasileiras. EMBRAPA Informação Tecnológica, v.1. Colombo, Embrapa Florestas, 2003. 627p.         [ Links ]

CLEMENTE, A.L.; MUNIZ, J.A. Avaliação do coeficiente de variação em experimentos com gramíneas forrageiras. Ciência e Agrotecnologia, v.26, n.1, p.197-203, 2002.         [ Links ]

COSTA, N.H.A.D.; SERAPHIN, J.C; ZIMMERMANN, F.J.A. Novo método de classificação de coeficiente de variação para a cultura do arroz de terras altas. Pesquisa Agropecuária Brasileira, v.37, n.3, p.243-249, 2002.        [ Links ]

FARAWAY, J.J. Extending the linear model with R: generalized linear, mixed effects and nonparametric regression models. Boca Raton: Chapman and Hall, 2006. 312p.         [ Links ]

FERRAZ, I.D.K.; CALVI, G.P. Teste de germinação. In: LIMA JÚNIOR, M.J.V. (Ed.). Manual de procedimentos de análise de sementes florestais. Londrina: ABRATES, 2011, 83p.         [ Links ]

GARCIA, C.H. Tabelas para classificação do coeficiente de variação. Piracicaba: IPEF, 1989. 12p. (Circular técnica, 171).        [ Links ]

GONZALES, J.L.S.; VALERI, S.V.; PAULA, R.C. Germinación de semillas de Albizia hassleri a diferentes temperaturas, en condiciones de laboratório. Bosque (Valdivia), v.31, n.1, p.39-44, 2010.        [ Links ]

GONZÁLES, S.; TORRES, R.A.A. Coleta de sementes e produção de mudas. In: SALOMÃO, A.N., et al. Germinação de sementes e produção de mudas de plantas do Cerrado. Brasília: Rede de Sementes do Cerrado, 2003. p.11-22.         [ Links ]

HAMRICK, J.L.; LOVELESS, M.D. The influence of seed dispersal mechanisms on the genetic structure of plant populations. In: ESTRADA, A.; FLEMING, T.H. (Eds). Fugivores and seed dispersal. Dordrecht: Dr. Junk Publications, 1986. 392p.         [ Links ]

HENICKA, G.S.; BRAGA, L.F.; SOUSA, M.P.; CARVALHO, M.A.C. Germinação de sementes de Apuleia leiocarpa (Vogel.) J. F. Macbr.: temperatura, fotoblastismo e estresse salino. Revista de Ciências Agro-Ambientais, v.4, n.1, p.37-46, 2006.        [ Links ]

IEDE, E.T. Importância das pragas quarentenárias florestais no comércio internacional: estratégias e alternativas para o Brasil. Colombo: Embrapa Florestas. 2005. 28p.         [ Links ]

ISTA. International Seed Testing Association. International Rules for Seed Testing. Bassersdorf: International Seed Testing Association. 2011.         [ Links ]

ISTA. International Seed Testing Association. Method validation for seed testing. Bassersdorf: International Seed Testing Association. 2007.         [ Links ]

KATAOKA, V.Y.; CARVALHO, M.L.M.; OLIVEIRA, M.S.; CALDEIRA, C.M. Validação de metodologia para o teste de germinação em sementes de nabo forrageiro (Raphanus sativus L. var. oleiferus). Revista Brasileira de Sementes, v.33, n.1, p.69-79, 2011.        [ Links ]

LABOURIAU, L.G. A germinação das sementes. Washington: Secretaria Geral da organização dos Estados Americanos, Programa Regional de Desenvolvimento Científico e Tecnológico. 1983. 174p.         [ Links ]

LACERDA, D.R.; LEMOS FILHO, J.P.; GOULART, M.F.; RIBEIRO, R.A.; LOVATO, M.B. Seed-dormancy variation in natural populations of two tropical leguminous tree species: Senna multijuga (Caesalpinoideae) and Plathymenia reticulata (Mimosoideae). Seed Science Research, v.14, p.127 - 135, 2004.        [ Links ]

LEMOS FILHO, J.P.; GUERRA, S.T.M.; LOVATO, M.B.; SCOTTI, M.R.M.M. Germinação de sementes de Senna macranthera, Senna multijuga e Stryphnodendron polyphyllum. Pesquisa Agropecuária Brasileira, v.32, n.4, p.357-361, 1997.        [ Links ]

LORZA, R.F. Políticas públicas para produção e comercialização de sementes florestais no Brasil. Informativo ABRATES, v.19, n.2, p.108-109, 2009.         [ Links ]

MARQUES, M.A.; RODRIGUES, T.J.D.; PAULA, R.C. Germinação de sementes de Ormosia arborea (Vell.) Harms submetidas a diferentes tratamentos pré-germinativos. Científica, v.32, n.2, p.141-146, 2004.        [ Links ]

MARTINS, C. C.; NAKAGAWA, J. Germinação de sementes de Stryphnodendron adstringens (Mart.) Coville de diferentes origens submetidas a tratamentos para superação de dormência. Revista Árvore, v.32, n.6, p.1059-1067, 2008.        [ Links ]

MURAWSKI, D.A. Reproductive biology and genetics of tropical trees from a canopy perspective. In: LOWMAN, M.D.; NADKARMI, N.M. (Eds.). Forest canopies. New York: Academic Press, p.457-493, 1995.         [ Links ]

NASSIF, S.M.L.; PEREZ, S.C.J.G.A. Efeito da temperatura na germinação de sementes de amendoim-do-campo (Pterogyne nitens Tul.). Revista Brasileira de Sementes, v.22, n.1, p.1-6, 2000.        [ Links ]

NONAGAKI, H.; BASSEL, G.W.; BEWLEY, J.D. Germination still a mystery. Plant Science, v.179, p.574-581, 2010.         [ Links ]

NOVEMBRE, A.D.L.C.; FARIA, T.C.; PINTO, D.H.V.; CHAMMA, H.M.C.P. Teste de germinação de sementes de sansão-do-campo (Mimosa caesalpiniaefolia Benth. - Fabaceae-Mimosoideae). Revista Brasileira de Sementes, v.29, n.3, p.42-45, 2007.        [ Links ]

OLIVEIRA, L.M.; CARVALHO, M.L.M.; DAVIDE, A.C. Teste de tetrazólio para avaliação da qualidade de sementes de Peltophorum dubium (Sprengel) Taubert - Leguminosae- Caesalpinioideae. Revista Cerne, v.11, n.2, p.159-166, 2005.        [ Links ]

OLIVEIRA, L.M.; DAVIDE, A.C.; CARVALHO, M.L.M. Teste de germinação de sementes de Peltophorum dubium (Sprengel) Taubert - Fabaceae. Floresta, v.38, n.3, p.545-551, 2008.         [ Links ]

PIEPHO, H.P. The folded exponential transformation for proportions. Journal of the Royal Statistical Society (Series D), v.52, p.575-589, 2003.         [ Links ]

PIMENTEL-GOMES, F. Curso de estatística experimental. 14.ed. Piracicaba: Editora da Universidade de São Paulo, 2000. 477p.         [ Links ]

RAMOS, M.B.P.; VARELA, V.P.; MELO, M.F.F. Influência da temperatura e da água sobre a germinação de sementes de paricá (Schizolobium amazonicum Huber ex Ducke - Leguminosae-Caesalpinioideae). Revista Brasileira de Sementes, v.28, n.1, p.163-168, 2006.        [ Links ]

REGO, F.L.H.; COSTA, R.B.; CONTINI, A.Z.; MORENO, R.G.S.; RONDELLI, K.G.S.; KUMIMOTO, H.H. Variabilidade genética e estimativas de herdabilidade para o caráter germinação em matrizes de Albizia lebbeck. Ciência Rural, v.35, n.5, p.1209-1212, 2005.        [ Links ]

ROSSETO, J.; ALBUQUERQUE, M.C.F.; RONDON NETO, R.M.; SILVA, I.C.O. Germinação de sementes de Parkia pendula (Willd.) Benth. ex Walp. (Fabaceae) em diferentes temperaturas. Revista Árvore, v.33, n.1, p.47-55, 2009.        [ Links ]

SANTANA, D.G.; RANAL, M.A. Análise estatística na germinação. Revista Brasileira de Fisiologia Vegetal, v.12 (Edição Especial), p.205-237, 2000.        [ Links ]

SANTANA, D.G.; WIELEWICKI, A.P.; SALOMÃO, A.N. Validation of quality tests for forest seed species. Seed Science Research, v.22, p.74-79, 2012.         [ Links ]

SARMENTO, M.B.; VILLELA, F.A. Sementes de espécies florestais nativas do Sul do Brasil. Informativo ABRATES, v.20, n.1,2, p.39-44, 2010.        [ Links ]

SILESHI, G.W. A critique of current trends in the statistical analysis of seed germination and viability data. Seed Science Research, p.1-15, 2012.         [ Links ]

SILESHI, G.W. Evaluation of statistical procedures for efficient analysis of insect, disease and weed abundance and incidence data. East African Journal of Science, v.1, p.1-9, 2007.         [ Links ]

SILVA, A.; FIGLIOLIA, M.B.; AGUIAR, I.B. Germinação de sementes de Acacia polyphylla DC. (monjoleiro) e de Aspidosperma ramiflorum Müll. Arg. (guatambu). Floresta, v.37, n.3, p.353-361, 2007.         [ Links ]

SILVA, M.M.F.; BASTOS, M.N.C.; GURGEL, E.S.C. Aspectos taxonômicos e morfológicos do processo germinativo e da plântula de Peltogyne venosa subsp. Densiflora (Spruce ex Benth.) M.F. Silva (Leguminosae - Caesalpinioideae). Boletim do Museu Paraense Emílio Goeldi, v.4, n.3, p.291-302, 2009.        [ Links ]

SILVA, M.S.; SANTOS, S.R.G. Tratamentos para superar dormência em sementes de Enterolobium contortisiliquum (Vell.) Morang - tamboril. Instituto Florestal, n.40, p.161-165, 2009.        [ Links ]

SMIDERLE, O.J.; SOUZA, R.C.P. Dormência em sementes de paricarana (Bowdichia virgilioides Hunth Fabaceae-Papilionidae). Revista Brasileira de Sementes, v.25, n.1, p.72-75, 2003.        [ Links ]

SOUZA, S.M.; LIMA, P.C.F. Maturação de sementes de angico (Anadenanthera macrocarpa (Benth.) Brenan). Revista Brasileira de Sementes, v.7, n.2, p.93-99, 1985.        [ Links ]

STEEL, R.G.D.; TORRIE, J.H. Principles and procedures of statistics: a biometrical approach. 2.ed. New York: McGraw-Hill, 633p. 1980.         [ Links ]

VAZ MONDO, V.H.; BRANCALION, P.H.S.; CÍCERO, S.M.; NOVEMBRE, A.D.L.C.; DOURADO-NETO, D. Teste de germinação de sementes de Parapiptadenia rigida (Benth.) Brenan (Fabaceae). Revista Brasileira de Sementes, v.30, n.2, p.177-183, 2008.        [ Links ]

WAENY, J.C.C. Repetitividade e reprodutividade II. São Paulo: IPT. 1980. 14p.         [ Links ]

WARTON, D.; HUI, F. The arcsine is asinine: the analysis of proportions in ecology. Ecology, v.92, p.3-10, 2011.        [ Links ]

WIELEWICKI, A.P.; LEONHARDT, C.; SCHLINDWEIN, G.; MEDEIROS, A.C. Proposta de padrões de germinação e teor de água para sementes de algumas espécies florestais presentes na região sul do Brasil. Revista Brasileira de Sementes, v.28, n.3, p.191-197, 2006.        [ Links ]



Submitted on 05/28/2012.
Accepted for publication on 02/12/2012.



* Corresponding author: <>

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons