Hostname: page-component-745bb68f8f-hvd4g Total loading time: 0 Render date: 2025-02-11T09:08:47.182Z Has data issue: false hasContentIssue false
  • English
  • Français

How regional climate and seed traits interact in shaping stress–tolerance of savanna seeds?

Published online by Cambridge University Press:  02 November 2021

Leandro C. Ribeiro ,
Eduardo R. M. Barbosa Open the ORCID record for Eduardo R. M. Barbosa [Opens in a new window]  and
Fabian Borghetti Open the ORCID record for Fabian Borghetti [Opens in a new window]
Leandro C. Ribeiro
Affiliation:
Departamento de Ensino, Instituto Federal Goiano, Rodovia Professor Geraldo Silva Nascimento, Km 2.5, Urutaí, Goiás, Brazil Departamento de Ensino, Instituto Federal do Ceará, Rodovia CE 060, Km 332, Acopiara, Ceará, Brazil Departamento de Botânica, Universidade de Brasília, Campus Universitário Darcy Ribeiro, S/N, Brasília, Distrito Federal, Brazil
Eduardo R. M. Barbosa
Affiliation:
Departamento de Botânica, Universidade de Brasília, Campus Universitário Darcy Ribeiro, S/N, Brasília, Distrito Federal, Brazil
Fabian Borghetti*
Affiliation:
Departamento de Botânica, Universidade de Brasília, Campus Universitário Darcy Ribeiro, S/N, Brasília, Distrito Federal, Brazil
*
*Correspondence: Fabian Borghetti, E-mail: borghetti.fabian@gmail.com
Rights & Permissions [Opens in a new window]

Abstract

Functional traits related to regeneration responses to the environment are highly determinants of distribution patterns of plant communities. A large body of studies on seed traits suggests that regional climate may act as a strong filter of plant recruitment; however, few studies have evaluated the relative importance of seed traits and environmental filters for seed persistence at the population level. We tested the role of seed mass, water content and desiccation tolerance, as well as the germination time as proxies for seed tolerance to environmental filters (water deficit, heat shock and high temperatures) by comparing the response of tree species co-occurring in savannas located in different regions: Cerrado biome of Central Brazil and the Rio Branco savannas of northern Brazil. Seeds collected in savannas of Rio Branco showed a higher tolerance to environmental filters than those collected in savannas of the Cerrado. While the germination percentages largely varied in response to the treatments, the germination times were virtually unaffected by them, irrespective of seed origin, seed mass and water content. At the population level, the regional environment was a key determinant of seed tolerance to stress, irrespective of seed traits. Germination time was shown to represent a conservative seed trait and more linked to a species-specific germination strategy than to regional characteristics. Our results suggest that recruitment patterns of Cerrado savannas may be more impacted than Rio Branco savannas by the climate scenarios predicted for the future.

Keywords

desiccation toleranceenvironmental heterogeneitygermination timehigh temperaturesseed traitstropical savannaswater deficit
Type
Research Paper
Information
Seed Science Research , Volume 31 , Issue 4 , December 2021 , pp. 300 - 310
Check for updates
Copyright
Copyright © The Author(s), 2021. Published by Cambridge University Press

Introduction

Plant species growing in heterogeneous environments may develop a wide variety of functional traits in response to local characteristics (Berg et al., Reference Berg, Becker and Matthies2005). Adaptation to local conditions and response plasticity are essential requirements for the effective establishment of new individuals within a population (Galloway, Reference Galloway2005), directly defining plant fitness, population dynamics and community structure (Donohue et al., Reference Donohue, Casas, Burghart, Kovach and Willis2010; Baskin and Baskin, Reference Baskin and Baskin2014). Thus, unravelling factors that control plant recruitment from seeds represent an essential step for predicting vegetation structure and composition in the future (Leishman et al., Reference Leishman, Wright, Moles, Westoby and Fenner2000; Coomes and Grubb, Reference Coomes and Grubb2003). Since it is difficult to evaluate functional responses for all plant species, the identification of traits that best represent plant responses to environmental conditions represents a promising approach (Chevin et al., Reference Chevin, Lande and Mace2010; Barbosa et al., Reference Barbosa, van Langevelde, Tomlinson, Carvalheiro, Kirkman, de Bie and Prins2014).

Seed germination and post-germination adaptions to environmental filters represent crucial phases in the life cycle of plants (Donohue et al., Reference Donohue, Casas, Burghart, Kovach and Willis2010), and new individuals cannot become part of the community if the local climate is suitable for adults but not for seeds and seedlings (Poschlod et al., Reference Poschlod, Abedi, Bartelheimer, Drobnik, Rosbakh, Saatkamp, van der Maarel and Franklin2013). Regenerative traits strongly influence the distribution and structure of plant communities (Rees and Westoby, Reference Rees and Westoby1997; Moles et al., Reference Moles, Ackerly, Tweddle, Dickie, Smith, Leishman, Mayfield, Pitman, Wood and Westoby2007; Cochrane et al., Reference Cochrane, Yates, Hoyle and Nicotra2015; Jiménez-Alfaro et al., Reference Jiménez-Alfaro, Silveira, Fidelis, Poschlod and Commander2016). For example, morphological traits such as seed surface, coat properties and seed mass are functionally related to regeneration steps as dispersal, soil persistence and responses to biotic and abiotic disturbances (Westoby et al., Reference Westoby, Falster, Moles, Vesk and Wright2002; Fenner and Thompson, Reference Fenner and Thompson2005). Seed mass, in particular, was related to seed tolerance to heat shock (Ribeiro et al., Reference Ribeiro, Barbosa, van Langevelde and Borghetti2015). Biophysical traits, such as seed water content (SWC) and desiccation tolerance, are correlated with each other (Hamilton et al., Reference Hamilton, Offord, Cuneo and Deseo2013) and with seed tolerance to water deficit (Tweddle et al., Reference Tweddle, Dickie, Baskin and Baskin2003). Physiological traits represent parameters that estimate the percent (or proportion) of seeds that germinate in a seed batch and the rate and distribution of germination over time (Jiménez-Alfaro et al., Reference Jiménez-Alfaro, Silveira, Fidelis, Poschlod and Commander2016). Such parameters are strongly affected by environmental filters such as light, moisture and temperature (Fenner and Thompson, Reference Fenner and Thompson2005; Donohue et al., Reference Donohue, Casas, Burghart, Kovach and Willis2010), and they have been used for evaluating germination responses to regional constraints, such as water deficit (Flores and Briones, Reference Flores and Briones2001; Kos and Poschlod, Reference Kos and Poschlod2008, Reference Kos and Poschlod2010; Fay and Schultz, Reference Fay and Schultz2009), high temperatures (Flores and Briones, Reference Flores and Briones2001; Ribeiro and Borghetti, Reference Ribeiro and Borghetti2014) and fire (Dayamba et al., Reference Dayamba, Tigabu, Sawadogo and Oden2008; Ribeiro et al., Reference Ribeiro, Barbosa, van Langevelde and Borghetti2015). As these environmental factors represent major filters in neotropical savannas, physiological traits such as seed dormancy, germination percentage and germination time are expected to play an important role in the vegetation dynamics of these ecosystems. However, the relative importance of morphological, biophysical and physiological traits related to seed tolerance to environmental filters is still an open gap in the literature (Jiménez-Alfaro et al., Reference Jiménez-Alfaro, Silveira, Fidelis, Poschlod and Commander2016). Therefore, the comparison of seed responses of species subjected to different environmental conditions would contribute to unravelling which trait(s) best explain seed tolerance to stress.

In Brazil, there are two large savanna areas separated by the Amazon basin: (i) Rio Branco savannas, occurring in the state of Roraima, northern Brazil and (ii) Cerrado savannas, occurring within the Cerrado, a biome largely distributed over Central Brazil (Borghetti et al., Reference Borghetti, Barbosa, Ribeiro, Ribeiro, Walter, Scogings and Sankaran2019). They share several plant species and are subjected to a similar fire frequency and mean annual precipitation (see Table 1 for more details). However, annual temperature variation is wider in Cerrado savannas, and the soil is usually clayish, while Roraima savannas are subjected to a narrower temperature range and occur over predominantly sandy soils (Table 1).

Table 1. Environmental characteristics reported for Rio Branco and Cerrado savannas (Castro and Kauffman, Reference Castro and Kauffman1998; Gomes et al., Reference Gomes, Curi, Motta, Ker, Marques and Schulze2004; Barbosa and Fearnside, Reference Barbosa and Fearnside2005b; Benedetti et al., Reference Benedetti, Júnior, Schaefer, Melo and Uchôa2011; INMET, 2013, 2014).

Notes: Mean annual precipitation (MAP), mean maximum and minimum temperature and relative humidity averaged for the period from 2004 to 2013. These data represent regional parameters for the regions selected for this study.

Soils with low vegetation covering, such as those found in Rio Branco savannas (Barbosa and Fearnside, Reference Barbosa and Fearnside2005a), are more affected by solar radiation than soils with high vegetation covering (Oliveira and Silva, Reference Oliveira and Silva1994), such as the Cerrado soils (Castro and Kauffman, Reference Castro and Kauffman1998). Maximum air temperatures recorded for Rio Branco savannas are higher than those recorded for Cerrado savannas (Table 1). Considering that soil temperatures can be positively correlated with air temperatures (Ooi et al., Reference Ooi, Auld and Denham2012), we expect that soils of Rio Branco savannas will experience higher temperatures than those of Cerrado savannas. Due to the higher water holding capacity of clayish soils (Gomes et al., Reference Gomes, Curi, Motta, Ker, Marques and Schulze2004; Benedetti et al., Reference Benedetti, Júnior, Schaefer, Melo and Uchôa2011), water availability in the sandy soils of Rio Branco savannas is expected to be lower than that in the clayish soils of Cerrado savannas. Moreover, due to higher amounts of biomass available for burning and a higher fine-fuel combustion factor (Table 1), soils of Cerrado savannas are expected to experience fires of higher severity (Kauffman et al., Reference Kauffman, Cummings and Ward1994; Castro and Kauffman, Reference Castro and Kauffman1998; Barbosa and Fearnside, Reference Barbosa and Fearnside2005a; Miranda et al., Reference Miranda, Nascimento-Neto, Castro-Neves and Miranda2010) and with increased residence time (Bristow, Reference Bristow1998; DeBano et al., Reference DeBano, Neary and Ffolliott1998; Miranda et al., Reference Miranda, Nascimento-Neto, Castro-Neves and Miranda2010; Nwadibia et al., Reference Nwadibia, Ugwu and Aduloju2010) than soils of Rio Branco savannas.

Variations in seed attributes and germination patterns among populations subjected to different environmental conditions are expected (Baskin and Baskin, Reference Baskin and Baskin2014), but they can only be accessed if seeds from populations occurring in different regions and/or subjected to distinct environmental conditions are studied under the same experimental settings. In this respect, the use of conspecific and congeneric species may contribute to unravelling trait–environment interactions, variation in micro-environmental conditions, ecological breadth and geographic ranges (Daws et al., Reference Daws, Burslem, Crabtree, Kirkman, Mullins and Dalling2002; Donohue et al., Reference Donohue, Casas, Burghart, Kovach and Willis2010). Due to their regional particularities with respect to climate and soil properties, savannas of Rio Branco and the Cerrado biome offer an excellent opportunity to conduct studies focusing on intraspecific variation in seed traits. For that, we used congeneric and conspecific tree species occurring in both Rio Branco and Cerrado savannas to compare the variation in physiological (germination time), morphological (seed mass) and biophysical (water content and desiccation tolerance) traits, as well as the extent to which environmental filters and seed origin may influence this variation. Also, considering that physiological traits may be more responsive to, and better reflect seed tolerance to stress than morphological and biophysical traits (Saatkamp et al., Reference Saatkamp, Affre, Dutoit and Poschlod2011; Jiménez-Alfaro et al., Reference Jiménez-Alfaro, Silveira, Fidelis, Poschlod and Commander2016), we made the following predictions:

  1. 1. The physiological trait explains better seed stress–tolerance than morphological and biophysical traits.

  2. 2. Seed origin and environmental filters are better predictors of seed tolerance to stress than seed traits.

  3. 3. Seeds from Rio Branco savannas have a higher tolerance to water deficit, high temperatures and heat shock than seeds from Cerrado savannas.

Materials and methods

Study sites: a general view

Rio Branco savannas are located north of the Amazon basin, covering about 43,000 km2 in the northern and northeastern regions of Roraima State, but they also extend to the country of Guyana (Barbosa and Fearnside, Reference Barbosa and Fearnside2005b; Barbosa et al., Reference Barbosa, Campos, Pinto and Fearnside2007; Meneses et al., Reference Meneses, Costa and Behling2013).

The Cerrado biome covers around 2 million km2 in Central Brazil, representing the largest continuous savanna area in South America (Borghetti et al., Reference Borghetti, Barbosa, Ribeiro, Ribeiro, Walter, Scogings and Sankaran2019). Much of the Cerrado is distributed over the Brazilian Central Plateau (Sano et al., Reference Sano, Rosa, Brito and Ferreira2010). The Cerrado vegetation has a gradient of physiognomies; from pure grasslands to forests, however, savanna predominates (Bueno et al., Reference Bueno, Pennington, Dexter, Kamino, Pontara, Neves, Ratter and Oliveira-Filho2017; Borghetti et al., Reference Borghetti, Barbosa, Ribeiro, Ribeiro, Walter, Scogings and Sankaran2019). Although separated by the Amazon basin, these two regions share several species, suggesting that they were in some way connected in the past (Silva and Bates, Reference Silva and Bates2002; Prance, Reference Prance2006; Furley, Reference Furley2007).

Selected species and seed collection

Seed samples were collected in 2011 from natural populations occurring in the two main Brazilian savanna regions: Rio Branco savannas – one site at the campus of the Federal University of Roraima (UFRR) (02°38′N, 60°49′W), located 15 km north of Boa Vista, in Roraima State, at an elevation of 77 m above sea level (a.s.l); Cerrado savannas – two sites: (i) IBGE Ecological Reserve (15°55′S, 47°52′W), Federal District, at approximately 1100 m a.s.l. and (ii) Experimental Farm of the Goiás State University, campus Ipameri (IPM) (17°41′S and 48°11′W), Goiás State, at approximately 780 m a.s.l. For all these regions, the climate is classified as tropical with a dry winter (Aw) under the Köppen–Geiger classification system (Peel et al., Reference Peel, Finlayson and McMahon2007).

We selected conspecific and congeneric species of common occurrence over both Rio Branco and Cerrado savannas (see Table 2 for more details). For species selection, we also considered their frequency of occurrence in the areas of collection. We collected seeds from at least ten adult individuals of each species, from both savanna sites, except for the two species of Himatanthus (Table 2). After collection, samples of seeds were immediately used for seed morphological and biophysical traits assays, as described in the next section. Another sample of seeds was stored in paper bags under laboratory conditions (22–24°C, 65–85% relative humidity) for a maximum of 4–6 weeks before the germination experiments. Storage under these conditions maintains seed moisture and viability for an extended period of time (Brasil, 2009).

Table 2. Habitat, collection site, seed dispersal period and seed dry mass of the tree species selected for this study

(For dry mass, data are expressed as mean ± standard deviation; P refers to the statistical significance (ANOVA one-way).

Notes: Since neither H. articulatus nor H. obovatus are common to Rio Branco and Cerrado savannas, we worked with different species within this genus. However, phylogenetic studies have shown that Himatanthus has a recent origin, suggesting that morphological and genetic differences within this genus are minimal (Spina, Reference Spina2004; Spina et al., Reference Spina, Bittrich and Kinoshita2013).

Experimental design

Seed morphological and biophysical traits

Samples of 100 seeds of each species were selected to estimate fresh and dry seed mass. Each single seed was weighed before and after being dried in an electric oven, set at 105°C for 24 h (Brasil, 2009). SWC was estimated on a dry weight basis.

For the treatment of desiccation, seed samples of each species had their initial water content reduced to 5% in a hermetically sealed container containing silica gel, which was renewed daily. Seeds were weighed every 2 h during the first 12 h and then every 24 h (Ribeiro and Borghetti, Reference Ribeiro and Borghetti2014) until the mass corresponding to the desired moisture content was achieved. The seed mass corresponding to a water content of 5% was estimated according to Cromarty et al. (Reference Cromarty, Ellis and Roberts1985). The controls consisted of samples of seeds not subjected to desiccation.

Water deficit

For treatments of water deficit, seed samples of each species were placed to germinate in osmotic solutions prepared with polyethylene glycol 6000 P.A. (PEG 6000). This method gives a good estimate of germination behaviour in relation to soil water availability under field conditions (Hardegree and Emmerich, Reference Hardegree and Emmerich1994). We used PEG 6000 solutions at different concentrations to obtain osmotic potentials (OPs) ranging from −0.2 to −1.2 MPa (Villela et al., Reference Villela, Doni-Filho and Siqueira1991). The water potentials chosen for this study represent values registered in the upper soil layers of neotropical savannas, even during the rainy season (Nardoto et al., Reference Nardoto, Souza and Franco1998; Franco, Reference Franco, Oliveira and Marquis2002). The controls consisted of samples of seeds placed to germinate in distilled water (0.0 MPa).

Heat shock

For heat shock treatments, seed samples of each species were placed in a temperature-controlled oven with air circulation set to provide each of the following temperatures: 80, 110, 140, 170 and 200°C. Each heat shock was separately applied for 2.5 or 5.0 min. These combinations of temperatures and times of exposure were chosen based on field recordings taken in the soil of Brazilian savanna ecosystems (surface and/or shallow depths) during a prescribed fire (see Miranda et al., Reference Miranda, Miranda, Dias and Dias1993). The controls consisted of samples of seeds not subjected to heat shock.

Temperature

Seed samples were placed to germinate in chambers set at 20, 25, 30, 35, 40 and 45°C and 12-h photoperiod (white light, 30 μmol m−2 s−1). These temperatures were chosen because they cover a large range of temperature for the germination of savanna tree species of Brazilian ecosystems (Brancalion et al., Reference Brancalion, Novembre and Rodrigues2010; Borghetti et al., Reference Borghetti, Caetano, Colli, Françoso and Sinervo2021).

General procedures

Four replicates of 25 seeds for each species, from each savanna (Rio Branco and Cerrado), were used for each treatment. For the treatments of desiccation, heat shock and temperature, seeds were placed to germinate in 15-cm Petri dishes lined with two sheets of filter paper moistened with distilled water, which was replenished every 2 days. For treatments of OP, PEG solutions were used instead of water, except in the control. To minimize water evaporation from the solutions and, consequently, changes in the PEG concentration, the Petri dishes were sealed with parafilm (Kos and Poschlod, Reference Kos and Poschlod2008). After the treatments of desiccation and heat shock, as well as for the experiments of OP, seeds were placed to germinate in temperature-controlled chambers, set at 30°C and 12-h photoperiod (white light, 30 μmol m−2 s−1), a temperature within the range considered appropriate for the germination of savanna seeds (Zaidan and Carreira, Reference Zaidan and Carreira2008; Brancalion et al., Reference Brancalion, Novembre and Rodrigues2010; Borghetti et al., Reference Borghetti, Caetano, Colli, Françoso and Sinervo2021).

Germination was scored daily by counting and removing the germinated seeds until no signal of germination could be detected. The criterion used for germination was the emergence of the radicle, followed by its geotropic curvature to avoid false germination (Labouriau, Reference Labouriau1983). As the species have different germination kinetics, the experiments ranged from 10 to 40 days, considering species with fastest (e.g. Anadenathera peregrina) and slowest (e.g. Curatella americana) germinations, respectively. After the experiments finished, the viability of non-germinated seeds was tested using 1% w/v tetrazolium chloride solution. Seeds were cut into two halves and the exposed embryos were placed in contact with tetrazolium solution for 24 h in the dark at 30°C. Seeds that had their embryos stained dark pink or red after soaking were considered viable (Moore, Reference Moore and Heydecker1973).

Physiological parameters

The germination percentage was calculated according to Labouriau (Reference Labouriau1983), and the germination time (in hours) according to Farooq et al. (Reference Farooq, Basra, Hafeez and Ahmad2005).

The germination time was the (physiological) trait selected to test for the effects of different environmental filters (desiccation, water deficit, temperature and heat shocks) on seed performance. In those treatments in which the germination percentage was below 10%, the germination time was not calculated.

Statistical analysis

To verify the variation in seed morphological and biophysical traits as well as the effects of different treatments on the germination time, we grouped the species according to their collection site (Cerrado and Rio Branco).

To test variation in seed morphological and biophysical traits between savannas, we used a generalized linear mixed model (GLMM) with log-link function and Gamma distribution of errors. In those analyses, species was included as a random factor and the collection site was included as a fixed factor.

To determine how desiccation, OP, heat shock and temperature treatments affect both germination percentage and germination time (physiological traits), for each species from each collection site, we also performed the GLMM. For germination percentage analyses, the GLMM was performed using germination data (absence or presence) with logit-link function and binomial distribution of errors. For germination time analyses, the GLMM was performed using the germination time with log-link function and Gamma distribution of errors. Fresh mass, dry mass and water content were included as covariates in the models for the analyses of both germination percentage and germination time. Regardless of the treatment, the collection site was included as a fixed factor and species was included as a random factor. Every treatment was included as a fixed variable for analyses of their effects on each germination trait. All statistical analysis was carried out using R v.4.0.2 (R Core Team, 2020).

Results

Morphological and biophysical traits

There were no statistical differences between the parameters fresh and dry seed mass and SWC when comparing species from Rio Branco and Cerrado savannas (Table 3 and Fig. 1). Also, there was no statistical difference in seed tolerance to desiccation when comparing both savannas (Fig. 2A). In this case, the best-fitting model selected took into account only SWC, being the significant factor within the model (Z = −3.256; P < 0.01). Desiccation did not significantly affect the germination times of the seeds from both savannas (Fig. 3A). In this case, the best-fitting model selected took into account the factors collection site (S) and SWC. However, only SWC was significant within the model (t = 3.606; P < 0.001).

Fig. 1. Fresh mass (A), dry mass (B) and water content (C) of seeds of populations of trees from Cerrado and Rio Branco savannas, Brazil. The same letters above the boxes indicate there is no significant difference.

Fig. 2. Effects of desiccation (A), water deficit (B), heat shock (C) and temperature (D) on the germination percentage of populations of trees from Cerrado and Rio Branco savannas, Brazil (*P < 0.05; **P  < 0.01; n.s., not significant).

Fig. 3. Effects of desiccation (A), water deficit (B), heat shock (C) and temperature (D) on the germination rate of populations of trees from Cerrado and Rio Branco savannas, Brazil (*P < 0.05; **P < 0.01; n.s., not significant).

Table 3. GLMMs of morphological and biophysical seed traits for conspecific and congeneric species from Cerrado and Rio Branco savannas, Brazil

Physiological traits and treatments

Water deficit

Seeds collected from species occurring in both Cerrado and Rio Branco savannas showed similar germination percentages at OPs ranging from zero (control) to −0.4 MPa (Fig. 2B). However, seeds from Rio Branco showed statistically higher germination percentages at lower water potentials than seeds from Cerrado (Fig. 2B). No germination was observed at −1.2 MPa, suggesting that for both savannas this treatment would represent the limiting water potential for seed germination. For the germination percentage, the best-fitting model selected took into account the factors collection site (S) and OP, as well as their interaction (S × OP). However, only OP (Z = −9.212; P < 0.0001) and the interaction S × OP (Z = 4.226; P < 0.0001) were significant within the model.

More negative OP did not lead to significant increases in germination times of seeds from Cerrado and Rio Branco savannas (Fig. 3B). The germination times were very similar within OPs ranging from zero (control) to −0.6 MPa (Fig. 3B). Germination times at OPs more negative than −0.8 MPa were not calculated because the germination percentages in these treatments were below 10% for Cerrado species (see Fig. 2B). For the trait germination time, the best-fitting model selected took into account the factors S, OP as well as their interactions (S × OP). However, only S (t = 2.390; P < 0.05) and OP (t = 5.111; P < 0.0001) were significant within the model.

Heat shock

Seeds from both savannas tolerated heat shocks of 80°C, regardless of the exposure time (Fig. 2C). In other words, no significant differences were observed comparing seeds subjected to 80°C and control, within and between savannas. However, higher temperatures for longer times led to a significant reduction in germination percentages, these results being savanna-dependent. Heat shocks of 110 and 140°C for 2.5 min of exposure were more harmful to seeds from Cerrado than those from Rio Branco (Fig. 2C). When the exposure time was increased to 5.0 min, seeds from Cerrado showed a slightly higher tolerance to heat shock than those from Rio Branco (Fig. 2C). Heat shock of 170°C significantly reduced the germination percentages and was lethal for several seeds from Cerrado and Rio Branco (Fig. 2C). No seeds germinated after heat shock of 200°C (data not shown). For the germination percentages, the factors site of seed collection (S), temperature (Tp) and time (Ti), as well as their interactions were all significant within the best-fitting model selected, being S (Statistic = −3.002; P < 0.01), Tp (Z = −4.932; P < 0.0001), Ti (Z = −2.454; P < 0.05), S ×  Tp (Z = 6.110; P < 0.0001), S × Ti (Z = 4.428; P < 0.0001), Tp × Ti (Z = 3.473; P < 0.001) and S × Tp × Ti (Z = −5.900; P < 0.001).

Heat shock treatments did not significantly affect the germination times for seeds of Cerrado and Rio Branco savannas (Fig. 3C). Germination times after heat shocks of 140 and 170°C for 2.5 min were not calculated because the germination percentages were below 10% for Cerrado seeds (see Fig. 2C). Also, the germination times were not calculated for heat shocks of 110, 140 and 170°C for 5 min for both Cerrado and Rio Branco seeds for the same reason (see Fig. 2C). For the trait germination time, the best-fitting model selected took into account only the factor time (t = 11.027; P < 0.0001).

Temperature

There was no significant difference in the germination percentages at 20, 25 and 30°C when comparing species from Rio Branco and Cerrado (Fig. 2D). However, at 35 and 40°C, seeds from Rio Branco had statistically higher germination percentages than those from Cerrado (Fig. 2D). No seed germinated at 45°C, irrespective of seed origin. For the germination percentages, the best-fitting model selected took into account the factors collection site (S) and temperature (Tp), as well as their interaction (S × Tp). However, only Tp (Z = −7.225; P < 0.0001) and the interaction of S × Tp (Z = 2.409; P < 0.05) were significant within the model.

The increasing temperature did not affect the germination times when comparing seeds from Cerrado and Rio Branco (Fig. 3D). The germination time was not calculated for 40°C because the germination percentage in that treatment was below 10% for Cerrado species (see Fig. 2D). For this trait, the best-fitting model selected took into account the factors S (t = −8.558; P < 0.0001) and Tp (t = −4.052; P < 0.0001).

Seed viability

The tetrazolium test revealed zero viability among the non-germinated seeds subjected to any of the treatments and less than 10% of viability among those non-germinated seeds in their respective controls. These results indicate that viability and germination were substantially equivalent, regardless of the treatment applied. In other words, if a seed failed to germinate during or after any of the treatments, it was because they had lost viability, not because they were dormant.

Discussion

In this paper, we tested the role of environmental filters and seed traits on seed stress–tolerance of congeneric and conspecific tree species from Cerrado and Rio Branco savannas.

We found that (i) the germination time was a better parameter to explain seed stress–tolerance than seed mass, water content and desiccation tolerance (Prediction 1); (ii) seed origin and environmental filters were better predictors of seed tolerance to stress than seed mass and water content (Prediction 2); (iii) seeds from Rio Branco showed higher tolerance to water deficit, heat shock, low-severity heat shock and high temperatures than seeds from Cerrado (Prediction 3).

Morphological and biophysical traits

Since the factors seed mass and water content were not significant in the models for desiccation, water deficit, heat shock and temperature treatments, they do not predict stress–tolerance when comparing congeneric and conspecific pairs subjected to different environmental conditions.

Seed mass greatly differed among species (Table 2), but the differences vanished when the masses were averaged for comparison between the Cerrado and Rio Branco savannas (Fig. 1). Since the seeds from Rio Branco species were shown to be more tolerant to the stress treatments than those from Cerrado species, our results suggest that regional environmental characteristics are a better determinant of seed tolerance (and, possibly, seed persistence in soil) than seed mass (Leishman and Westoby, Reference Leishman and Westoby1994; Leishman et al., Reference Leishman, Wright, Moles, Westoby and Fenner2000; Baraloto et al., Reference Baraloto, Forget and Goldberg2005; Moles et al., Reference Moles, Ackerly, Webb, Tweddle, Dickie and Westoby2005; Rees and Venable, Reference Rees and Venable2007; Muller-Landau, Reference Muller-Landau2010).

All the species selected for this study disperse seeds at the end of the dry season (see Table 2), and our study shows that their SWC does not differ when comparing congeneric and conspecific pairs occurring in the different regions of seed sampling (Fig. 1). The SWC of savanna species seems to be more related to the period of seed dispersal than to the region of occurrence, and previous studies with savanna seeds show that it is usually lower among seeds dispersed during the dry season than during the wet season (Gottsberger and Silberbauer-Gottsberger, Reference Gottsberger and Silberbauer-Gottsberger2006).

At the conclusion of their development, seeds of most species have a water content ranging from 5 to 20% of their total mass (Tweddle et al., Reference Tweddle, Dickie, Baskin and Baskin2003). Our study showed that seeds from both savanna regions have water content within this range, as already reported for seeds of other savanna species (Sales et al., Reference Sales, Pérez-García and Silveira2013; Ribeiro and Borghetti, Reference Ribeiro and Borghetti2014). Seeds with a lower water content are prone to exhibit higher desiccation tolerance than those with a higher water content (Pammenter and Berjak, Reference Pammenter and Berjak2000; Tweddle et al., Reference Tweddle, Dickie, Baskin and Baskin2003; Pritchard et al., Reference Pritchard, Daws, Fletcher, Gaméné, Msanga and Omondi2004; Ribeiro and Borghetti, Reference Ribeiro and Borghetti2014), and this pattern was also observed among the species selected for this study (Fig. 2A).

Physiological trait and treatments

Water deficit

Edaphic factors are major determinants of recruitment when discussing the effects of water deficit on seed survival and germination under field conditions (Schütz et al., Reference Schütz, Milberg and Lamont2002; Kos and Poschlod, Reference Kos and Poschlod2008). For example, in a condition of water shortage, sandy soils will retain less water and dry faster than clayish soils (Fidalski et al., Reference Fidalski, Tormena, Alves and Auler2013). Considering that seed banks in savannas tend to be formed close to the soil surface (Scott et al., Reference Scott, Setterfield, Douglas and Andersen2010; Salazar et al., Reference Salazar, Goldstein, Franco and Miralles-Wilhelm2011; Andrade and Miranda, Reference Andrade and Miranda2014), it is reasonable to expect that seeds occurring in sandy soils will face more negative water potentials and, therefore, will be selected for higher tolerance to water deficit than those occurring in clayish soils. Our results show that seeds of Rio Branco species (sandy soil) had higher germination percentages at lower OPs (Fig. 2B) than those from Cerrado species (clayish soil).

Heat shock

Our study revealed that the germination of seeds exposed to 80°C was barely affected, corroborating that heat shocks at temperatures up to 80°C have a minimal impact on seed viability, irrespective of the species’ habitat (Escudero et al., Reference Escudero, Nuñez and Pérez-García2000; Hanley and Lamont, Reference Hanley and Lamont2000; Thomas et al., Reference Thomas, Morris and Auld2007; Ribeiro and Borghetti, Reference Ribeiro and Borghetti2014; Ribeiro et al., Reference Ribeiro, Pedrosa and Borghetti2013, Reference Ribeiro, Barbosa, van Langevelde and Borghetti2015). However, high temperatures combined with long times of exposure can lead to a reduction of the germination percentages, depending on the environmental characteristics of species’ habitat. The effects of high temperatures generated during a fire on the soil seed banks can vary greatly among different ecosystems depending on the severity (a combination of fireline intensity and duration), fire-return interval (Keeley, Reference Keeley2009; Pivello et al., Reference Pivello, Oliveras, Miranda, Haridasan, Sato and Meirelles2010), fine-fuel amount (Kauffman et al., Reference Kauffman, Cummings and Ward1994; Barbosa and Fearnside, Reference Barbosa and Fearnside2005a) and soil properties (DeBano et al., Reference DeBano, Neary and Ffolliott1998). Fire temperatures in Brazilian savannas can reach 350°C at the soil level, and heat pulses can vary from 1 to 3 min (Miranda et al., Reference Miranda, Sato, Neto, Aires and Cochrane2009). Although it rarely occurs, the 5-min exposure time was included as a treatment representing an extreme condition that could be reached in the case of an intense fire in Brazilian savannas.

In our study, we observed that combining high temperatures with short times of exposure (a low-severity fire) was more damaging to seeds of Cerrado than to seeds of Rio Branco species, while combining low temperatures with long times of exposure (a medium-severity fire) was more damaging to seeds of Rio Branco than to seeds of Cerrado species. However, combining high temperatures with long times of exposure was harmful to seeds from Cerrado and Rio Branco species. According to our findings, environmental factors such as soil properties (e.g. texture and thermal conductivity) and amount of fine-fuel biomass (available for burning) emerge as major factors explaining seed tolerance to heat shock among species from Cerrado and Rio Branco.

Temperature

The germination percentages of tropical savanna species are usually reduced by temperatures of 35°C and above (Tambelini and Perez, Reference Tambelini and Perez1999; Sy et al., Reference Sy, Grouzis and Danthu2001; Ribeiro and Borghetti, Reference Ribeiro and Borghetti2014; Borghetti et al., Reference Borghetti, Caetano, Colli, Françoso and Sinervo2021). Our findings corroborated this pattern and also revealed that seeds from Cerrado were more sensitive to higher temperatures than seeds from Rio Branco. Once again, these differences between the population response and temperature seem to be related to their respective regional climates. Interestingly, differences in germination traits explained by environmental gradients (moisture and temperature) were already reported for congeneric and conspecific pairs occurring in other neotropical savannas (Ranieri et al., Reference Ranieri, Lana, Negreiros, Araújo and Fernandes2003; Garcia et al., Reference Garcia, Jacobi and Ribeiro2007; Sales et al., Reference Sales, Pérez-García and Silveira2013). Taken together, our study shows that soil properties are better determinants of seed tolerance to water stress, high temperatures and heat shocks than the vegetation cover itself, although it is likely that the latter also determines microclimate properties of the former.

Germination time

We found that desiccation, water deficit and heat shock did not affect the germination times of Cerrado and Rio Branco seeds (Fig. 3). Even treatments of increasing temperatures (where an increase in the speed of germination would be expected due to the kinetic effect of temperature on physiological processes) were unable to affect the times of germination of the studied species (Fig. 3D).

Previous studies have reported that for tropical species, the germination times represent a phylogenetically conserved trait (Norden et al., Reference Norden, Daws and Antoine2009), and our study shows that the treatments had a minimal impact on this parameter (Fig. 3). The range of values recorded for the germination times (always above 300 h and frequently above 400 h – Fig. 3) reveals a slow germination pattern, which may represent a strategy of species growing in habitats subject to an unpredictable climate such as the tropical savannas. Also recognized as a bet-hedging strategy, this germination pattern has been reported as an important mechanism for seed persistence under variable environments (Clauss and Venable, Reference Clauss and Venable2000; Simons and Johnston, Reference Simons and Johnston2006; Ooi et al., Reference Ooi, Auld and Denham2009; Ooi, Reference Ooi2012; Tielbörger et al., Reference Tielbörger, Petru and Lampei2012; Gremer and Venable, Reference Gremer and Venable2014).

Conclusions

Our results showed that for conspecific and congeneric populations of widespread neotropical savanna trees: (a) environmental characteristics are better determinants of stress–tolerance (and seed persistence in soil) than innate seed traits; (b) the germination time is a better parameter to explain seed stress–tolerance than seed mass, water content and desiccation tolerance, and represents a conservative trait at the population level and (c) seeds from Rio Branco savannas display a higher tolerance to stress than those from Cerrado savannas. Finally, our results suggest that the intensification of stress conditions as predicted by climate change models will very likely act more strongly on seed recruitment in Cerrado than in Rio Branco savannas.

Acknowledgements

The authors thank Stuart Klorfline for reviewing the language.

Financial support

This work was supported by the Coordination for the Improvement of Higher Education – CAPES (CAPES/NUFFIC grant no. 019/2010, CAPES/PNADB grant no. 451/2010); the National Council for Scientific and Technological Development – CNPq (grant no. 476297/2004-4 and grant no. 312152/2018-3); the University of Brasília, through their Deanship of Research and Post-Graduation, for financial support.

Conflicts of interest

The author(s) declare none.

References

Andrade, LAZ and Miranda, HS (2014) The dynamics of the soil seed bank after a fire event in a woody savanna in central Brazil. Plant Ecology 215, 1199–1209.CrossRefGoogle Scholar
Baraloto, C, Forget, P and Goldberg, DE (2005) Seed mass, seedling size and neotropical tree seedling establishment. Journal of Ecology 93, 1156–1166.CrossRefGoogle Scholar
Barbosa, RI and Fearnside, PM (2005a) Above-ground biomass and the fate of carbon after burning in the savannas of Roraima, Brazilian Amazonia. Forest Ecology and Management 216, 295–316.Google Scholar
Barbosa, RI and Fearnside, PM (2005b) Fire frequency and area burned in the Roraima savannas of Brazilian Amazonia. Forest Ecology and Management 204, 371–384.CrossRefGoogle Scholar
Barbosa, RI, Campos, C, Pinto, F and Fearnside, PM (2007) The ‘Lavrados’ of Roraima: biodiversity and conservation of Brazil's Amazonian savannas. Functional Ecosystems and Communities 1, 29–41.Google Scholar
Barbosa, ERM, van Langevelde, F, Tomlinson, KW, Carvalheiro, LG, Kirkman, K, de Bie, S and Prins, HHT (2014) Tree species from different functional groups respond differently to environmental changes during establishment. Oecologia 174, 1345–1357.Google ScholarPubMed
Baskin, CC and Baskin, JM (2014) Seeds: ecology, biogeography, and evolution of dormancy and germination. San Diego, CA, Academic Press.Google Scholar
Benedetti, UG, Júnior, JFV, Schaefer, CEGR, Melo, VF and Uchôa, SCP (2011) Gênese, química e mineralogia de solos derivados de sedimentos pliopleistocênicos e de rochas vulcânicas básicas em Roraima, norte amazônico. Revista Brasileira de Ciência do Solo 35, 299–312.Google Scholar
Berg, H, Becker, U and Matthies, D (2005) Phenotypic plasticity in Carlina vulgaris: effects of geographical origin, population size, and population isolation. Oecologia 143, 220–231.CrossRefGoogle ScholarPubMed
Borghetti, F, Barbosa, ERM, Ribeiro, LC, Ribeiro, JF and Walter, BMT (2019) South American savannas, pp. 77–122 in Scogings, PF and Sankaran, M (Eds) Savanna woody plants and large herbivores. Chichester, John Wiley and Sons.CrossRefGoogle Scholar
Borghetti, F, Caetano, GHO, Colli, GR, Françoso, R and Sinervo, BR (2021) The firewall between Cerrado and Amazonia: interaction of temperature and fire govern seed recruitment in a neotropical savana. Journal of Vegetation Science 32, e12988.CrossRefGoogle Scholar
Brancalion, PHS, Novembre, ADLC and Rodrigues, RR (2010) Temperatura ótima de germinação de sementes de espécies arbóreas brasileiras. Revista Brasileira de Sementes 32, 15–21.CrossRefGoogle Scholar
Brasil (2009) Ministério da Agricultura, Pecuária e Abastecimento. Regras para análise de sementes. Brasília, MAPA/ACS.Google Scholar
Bristow, KL (1998) Measurement of thermal properties and water content of unsaturated sandy soil using dual probe heat-pulse probes. Agricultural and Forest Meteorology 89, 75–84.CrossRefGoogle Scholar
Bueno, ML, Pennington, RT, Dexter, KG, Kamino, LHY, Pontara, V, Neves, DM, Ratter, JA and Oliveira-Filho, AT (2017) Effects of quaternary climatic fluctuations on the distribution of neotropical savanna tree species. Ecography 40, 403–414.CrossRefGoogle Scholar
Castro, EA and Kauffman, JB (1998) Ecosystem structure in the Brazilian Cerrado: a vegetation gradient of aboveground biomass, root mass and consumption by fire. Journal of Tropical Ecology 14, 263–283.CrossRefGoogle Scholar
Chevin, LM, Lande, R and Mace, GM (2010) Adaptation, plasticity, and extinction in a changing environment: towards a predictive theory. PLoS Biology 8, e1000357.CrossRefGoogle Scholar
Clauss, MJ and Venable, DL (2000) Seed germination in desert annuals: an empirical test of adaptive bet hedging. The American Naturalist 155, 168–186.CrossRefGoogle ScholarPubMed
Cochrane, A, Yates, CJ, Hoyle, GL and Nicotra, AB (2015) Will among-population variation in seed traits improve the chance of species persistence under climate change? Global Ecology and Biogeography 24, 12–24.CrossRefGoogle Scholar
Coomes, DA and Grubb, PJ (2003) Colonization, tolerance, competition and seed-size variation within functional groups. Trends in Ecology & Evolution 18, 283–291.CrossRefGoogle Scholar
Cromarty, AS, Ellis, RH and Roberts, EH (1985) Designing of seed storage facilities for genetic conservation. Rome, IPGRI.Google Scholar
Daws, MI, Burslem, DFRP, Crabtree, LM, Kirkman, P, Mullins, CE and Dalling, JW (2002) Differences in seed germination responses may promote coexistence of four sympatric Piper species. Functional Ecology 16, 258–267.CrossRefGoogle Scholar
Dayamba, SD, Tigabu, M, Sawadogo, L and Oden, PC (2008) Seed germination of herbaceous and woody species of the Sudanian savanna-woodland in response to heat shock and smoke. Forest Ecology and Management 256, 462–470.CrossRefGoogle Scholar
DeBano, LF, Neary, DG and Ffolliott, PF (1998) Fire's effects on ecosystems. New York, John Wiley & Sons.Google Scholar
Donohue, K, Casas, KK, Burghart, L, Kovach, LK and Willis, CG (2010) Germination, post germination adaptation, and species ecological ranges. Annual Review of Ecology, Evolution and Systematics 41, 293–319.Google Scholar
Escudero, A, Nuñez, Y and Pérez-García, F (2000) Is fire a selective force of seed size in pine species? Acta Oecologica 21, 245–256.CrossRefGoogle Scholar
Farooq, M, Basra, SMA, Hafeez, K and Ahmad, N (2005) Thermal hardening: a new seed vigour enhancement tool in rice. Journal of Integrative Plant Biology 47, 187–193.CrossRefGoogle Scholar
Fay, PA and Schultz, MJ (2009) Germination, survival, and growth of grass and forb seedlings: effects of soil moisture variability. Acta Oecologica 35, 679–684.CrossRefGoogle Scholar
Fenner, M and Thompson, K (2005) The ecology of seeds. Cambridge, Cambridge University Press.CrossRefGoogle Scholar
Fidalski, J, Tormena, CA, Alves, SJ and Auler, PAM (2013) Influência das frações de areia na retenção e disponibilidade de água em solos das formações Caiuá e Paranavaí. Revista Brasileira de Ciência do Solo 37, 613–621.CrossRefGoogle Scholar
Flores, J and Briones, O (2001) Plant life form and germination in a Mexican inter-tropical desert: effects of soil water potential and temperature. Journal of Arid Environments 47, 485–497.Google Scholar
Franco, AC (2002) Ecophysiology of woody plants, pp. 178–197 in Oliveira, PS and Marquis, RJ (Eds) The cerrados of Brazil: ecology and natural history of a neotropical savanna. New York, Columbia University Press.Google Scholar
Furley, PA (2007) Tropical savannas and associated forests: vegetation and plant ecology. Progress in Physical Geography 31, 203–211.CrossRefGoogle Scholar
Galloway, LF (2005) Maternal effects provide phenotypic adaptation to local environmental conditions. New Phytologist 166, 93–100.CrossRefGoogle ScholarPubMed
Garcia, QS, Jacobi, CM and Ribeiro, BA (2007) Resposta germinativa de duas espécies de Vellozia (Velloziaceae) dos campos rupestres de Minas Gerais, Brasil. Acta Botanica Brasilica 21, 451–456.CrossRefGoogle Scholar
Gomes, JBV, Curi, N, Motta, PEF, Ker, JC, Marques, JJGS and Schulze, DG (2004) Análise de componentes principais de atributos físicos, químicos e mineralógicos de solos do bioma Cerrado. Revista Brasileira de Ciência do Solo 28, 137–153.Google Scholar
Gottsberger, G and Silberbauer-Gottsberger, I (2006) Life in the cerrado: a South American tropical seasonal ecosystem: origin, structure, dynamics and plant use. Ulm, Reta Verlag.Google Scholar
Gremer, JR and Venable, DL (2014) Bet hedging in desert winter annual plants: optimal germination strategies in a variable environment. Ecology Letters 17, 380–387.Google Scholar
Hamilton, KN, Offord, CA, Cuneo, P and Deseo, MA (2013) A comparative study of seed morphology in relation to desiccation tolerance and other physiological responses in 71 Eastern Australian rainforest species. Plant Species Biology 28, 51–62.CrossRefGoogle Scholar
Hanley, ME and Lamont, BB (2000) Heat pre-treatment and the germination of soil- and canopy-stored seeds of south-western Australian species. Acta Oecologica 21, 315–321.CrossRefGoogle Scholar
Hardegree, SP and Emmerich, WE (1994) Seed germination in response to polyetilene glycol solution. Seed Science and Technology 22, 1–7.Google Scholar
INMET (2013) Normais climatológicas do Brasil: 1961–1990. Available at: http://www.inmet.gov.br/portal/index.php?r=clima/normaisClimatologicas (accessed 25 September 2013).Google Scholar
INMET (2014) Estações convencionais – gráficos. Available at: http://www.inmet.gov.br/portal/index.php?r=home/page&page=rede_estacoes_conv_graf (accessed 20 October 2014).Google Scholar
Jiménez-Alfaro, B, Silveira, FAO, Fidelis, A, Poschlod, P and Commander, LE (2016) Seed germination traits can contribute better to plant community ecology. Journal of Vegetation Science 27, 637–645.Google Scholar
Kauffman, JB, Cummings, DL and Ward, DE (1994) Relationships of fire, biomass and nutrient dynamics along a vegetation gradient in the Brazilian Cerrado. Journal of Ecology 82, 519–531.CrossRefGoogle Scholar
Keeley, JE (2009) Fire intensity, fire severity and burn severity: a brief review. International Journal of Wildland Fire 18, 116–126.CrossRefGoogle Scholar
Kos, M and Poschlod, P (2008) Correlates of inter-specific variation in germination response to water stress in a semi-arid savannah. Basic and Applied Ecology 9, 645–652.CrossRefGoogle Scholar
Kos, M and Poschlod, P (2010) Why wait? Trait and habitat correlates of variation in germination speed among Kalahari annuals. Oecologia 162, 549–559.Google ScholarPubMed
Labouriau, LG (1983) A germinação das sementes. Washington, DC, Secretaria Geral da Organização dos Estados Americanos.Google Scholar
Leishman, MR and Westoby, M (1994) The role of large seeds in seedling establishment in dry soil conditions: experimental evidence for semi-arid species. Journal of Ecology 82, 249–258.CrossRefGoogle Scholar
Leishman, MR, Wright, IJ, Moles, AT and Westoby, M (2000) The evolutionary ecology of seed size, pp. 31–57 in Fenner, M (Ed.) Seeds: the ecology of regeneration in plant communities. Wallingford, CABI Publishing.CrossRefGoogle Scholar
Meneses, MENS, Costa, ML and Behling, H (2013) Late Holocene vegetation and fire dynamics from a savanna-forest ecotone in Roraima state, northern Brazilian Amazon. Journal of South American Earth Sciences 42, 17–26.Google Scholar
Miranda, AC, Miranda, HS, Dias, IFO and Dias, BFS (1993) Soil and air temperatures during prescribed Cerrado fires in Central Brazil. Journal of Tropical Ecology 9, 313–320.Google Scholar
Miranda, HS, Sato, MN, Neto, WN and Aires, FS (2009) Fires in the Cerrado, the Brazilian savanna, pp. 427–450 in Cochrane, MA (Ed.) Tropical fire ecology: climate change, land use, and ecosystem dynamics. Chichester, Springer-Praxis Publishing.CrossRefGoogle Scholar
Miranda, HS, Nascimento-Neto, W and Castro-Neves, BM (2010) Caracterização das queimadas de Cerrado, pp. 23–33 in Miranda, HS (Org.)Efeitos do regime do fogo sobre a estrutura de comunidades de cerrado: resultados do projeto fogo, Brasília, Ibama.Google Scholar
Moles, AT, Ackerly, DD, Webb, CO, Tweddle, JC, Dickie, JB and Westoby, M (2005) A brief history of seed size. Science 307, 576–580.Google ScholarPubMed
Moles, AT, Ackerly, DD, Tweddle, JC, Dickie, JB, Smith, R, Leishman, MR, Mayfield, MM, Pitman, A, Wood, JT and Westoby, M (2007) Global patterns in seed size. Global Ecology and Biogeography 16, 109–116.CrossRefGoogle Scholar
Moore, RP (1973) Tetrazolium staining for assessing seed quality, pp. 347–366 in Heydecker, W (Ed.) Seed ecology. London, Butterworths.Google Scholar
Muller-Landau, HC (2010) The tolerance-fecundity trade-off and the maintenance of diversity in seed size. Proceedings of the National Academy of Sciences of the USA 107, 4242–4247.CrossRefGoogle ScholarPubMed
Nardoto, GB, Souza, MP and Franco, AC (1998) Estabelecimento e padrões sazonais de produtividade de Kielmeyera coriacea (Spr) Mart. nos cerrados do Planalto Central: efeitos do estresse hídrico e sombreamento. Revista Brasileira de Botânica 98, 313–319.Google Scholar
Norden, N, Daws, MI, Antoine, C, et al. (2009) The relationship between seed mass and mean time to germination for 1037 tree species across five tropical forests. Functional Ecology 23, 203–210.CrossRefGoogle Scholar
Nwadibia, N, Ugwu, E and Aduloju, K (2010) Theoretical analysis of the influence of the thermal diffusivity of clay soil on the thermal energy distribution in clay soil of Abakaliki, Nigeria. Research Journal of Applied Sciences, Engineering and Technology 3, 216–221.Google Scholar
Oliveira, ME and Silva, IL (1994) Efeitos do fogo sobre o solo. Floresta e Ambiente 1, 142–145.Google Scholar
Ooi, MKJ (2012) Seed bank persistence and climate change. Seed Science Research 22, S53–S60.CrossRefGoogle Scholar
Ooi, MKJ, Auld, TD and Denham, AJ (2009) Climate change and bet-hedging: interactions between increased soil temperatures and seed bank persistence. Global Change Biology 15, 2375–2386.Google Scholar
Ooi, MKJ, Auld, T and Denham, A (2012) Projected soil temperature increase and seed dormancy response along an altitudinal gradient: implications for seed bank persistence under climate change. Plant and Soil 353, 289–303.CrossRefGoogle Scholar
Pammenter, NW and Berjak, P (2000) Evolutionary and ecological aspects of recalcitrant seed biology. Seed Science Research 10, 301–306.CrossRefGoogle Scholar
Peel, MC, Finlayson, BL and McMahon, TA (2007) Updated world map of the Köppen-Geiger climate classification. Hydrology and Earth System Science 11, 1633–1644.CrossRefGoogle Scholar
Pivello, VR, Oliveras, I, Miranda, HS, Haridasan, M, Sato, MN and Meirelles, ST (2010) Effect of fires on soil nutrient availability in an open savanna in Central Brazil. Plant and Soil 337, 111–123.CrossRefGoogle Scholar
Poschlod, P, Abedi, M, Bartelheimer, M, Drobnik, J, Rosbakh, S and Saatkamp, A (2013) Seed ecology and assembly rules in plant communities, pp. 164–202 in van der Maarel, E and Franklin, J (Eds) Vegetation ecology. Chichester, Wiley-Blackwell.Google Scholar
Prance, GT (2006) Tropical savannas and seasonally dry forests: an introduction. Journal of Biogeography 33, 385–386.CrossRefGoogle Scholar
Pritchard, HW, Daws, MI, Fletcher, BJ, Gaméné, CS, Msanga, HP and Omondi, W (2004) Ecological correlates of seed desiccation tolerance in tropical African dry land trees. American Journal of Botany 91, 863–870.CrossRefGoogle Scholar
Ranieri, BD, Lana, TC, Negreiros, D, Araújo, LM and Fernandes, GW (2003) Germinação de sementes de Lavoisiera cordata Cogn. e Lavoisiera francavillana Cogn. (Melastomataceae), espécies simpátricas da Serra do Cipó, Brasil. Acta Botanica Brasilica 17, 523–530.CrossRefGoogle Scholar
R Core Team (2020) R: a language and environment for statistical computing. Vienna, R Foundation for Statistical Computing.Google Scholar
Rees, M and Venable, DL (2007) Why do big plants make big seeds? Journal of Ecology 95, 926–936.Google Scholar
Rees, M and Westoby, M (1997) Game-theoretical evolution of seed mass in multi-species ecological models. Oikos 78, 116–126.CrossRefGoogle Scholar
Ribeiro, LC and Borghetti, F (2014) Comparative effects of desiccation, heat shock and high temperatures on seed germination of savanna and forest tree species. Austral Ecology 39, 267–278.Google Scholar
Ribeiro, LC, Pedrosa, M and Borghetti, F (2013) Heat shock effects on seed germination of five Brazilian savanna species. Plant Biology 15, 152–157.CrossRefGoogle ScholarPubMed
Ribeiro, LC, Barbosa, ERM, van Langevelde, F and Borghetti, F (2015) The importance of seed mass for the tolerance to heat shocks of savanna and forest tree species. Journal of Vegetation Science 26, 1102–1111.CrossRefGoogle Scholar
Saatkamp, A, Affre, L, Dutoit, T and Poschlod, P (2011) Germination traits explain soil seed persistence across species: the case of Mediterranean annual plants in cereal fields. Annals of Botany 107, 415–426.Google ScholarPubMed
Salazar, A, Goldstein, G, Franco, AC and Miralles-Wilhelm, F (2011) Timing of seed dispersal and dormancy, rather than persistent soil seed-banks, control seedling recruitment of woody plants in Neotropical savannas. Seed Science Research 21, 103–116.CrossRefGoogle Scholar
Sales, NM, Pérez-García, F and Silveira, FAO (2013) Consistent variation in seed germination across an environmental gradient in a Neotropical savanna. South African Journal of Botany 87, 129–133.CrossRefGoogle Scholar
Sano, EE, Rosa, R, Brito, JLS and Ferreira, LG (2010) Land cover mapping of the tropical savanna region in Brazil. Environmental Monitoring and Assessment 166, 113–124.CrossRefGoogle ScholarPubMed
Schütz, W, Milberg, P and Lamont, BB (2002) Germination requirements and seedling responses to water availability and soil type in four eucalypt species. Acta Oecologica 23, 23–30.Google Scholar
Scott, K, Setterfield, S, Douglas, M and Andersen, A (2010) Soil seed banks confer resilience to savanna grass-layer plants during seasonal disturbance. Acta Oecologica 36, 202–210.CrossRefGoogle Scholar
Silva, JMC and Bates, JM (2002) Biogeographic patterns and conservation in the South American Cerrado: a tropical savanna hotspot. BioScience 52, 225–233.Google Scholar
Simons, AM and Johnston, MO (2006) Environmental and genetic sources of diversification in the timing of seed germination: implications for the evolution of bet hedging. Evolution 60, 2280–2292.CrossRefGoogle ScholarPubMed
Spina, AP (2004) Estudos taxonômico, micro-morfológico e filogenético do gênero Himatanthus Willd. Ex Schult. (Apocynaceae: Rauvolfioideae - Plumerieae). PhD dissertation, University of Campinas, Campinas, São Paulo.Google Scholar
Spina, AP, Bittrich, V and Kinoshita, LS (2013) Typifications, new synonyms and a new combination in Himatanthus (Apocynaceae). Taxon 62, 1304–1307.Google Scholar
Sy, A, Grouzis, M and Danthu, P (2001) Seed germination of seven Sahelian legume species. Journal of Arid Environments 49, 875–882.CrossRefGoogle Scholar
Tambelini, M and Perez, SCJGA (1999) Temperature limits on germination of Stryphnodendron polyphyllum Mart. Journal of Tropical Forestry Science 11, 630–636.Google Scholar
Thomas, PB, Morris, EC and Auld, TA (2007) Response surfaces for the combined effects of heat shock and smoke on germination of 16 species forming soil seed banks in south-east Australia. Austral Ecology 32, 605–616.CrossRefGoogle Scholar
Tielbörger, K, Petru, M and Lampei, C (2012) Bet-hedging germination in annual plants: a sound empirical test of the theoretical foundations. Oikos 121, 1860–1868.CrossRefGoogle Scholar
Tweddle, JC, Dickie, JB, Baskin, CC and Baskin, JM (2003) Ecological aspects of seed desiccation sensitivity. Journal of Ecology 91, 294–304.CrossRefGoogle Scholar
Villela, FA, Doni-Filho, L and Siqueira, EL (1991) Tabela de potencial osmótico em função da concentração de polietilenoglicol 6000 e da temperatura. Pesquisa Agropecuária Brasileira 26, 1957–1968.Google Scholar
Westoby, M, Falster, DS, Moles, AT, Vesk, PA and Wright, IJ (2002) Plant ecological strategies: some leading dimensions of variation between species. Annual Review of Ecology and Systematics 33, 125–159.CrossRefGoogle Scholar
Zaidan, LBP and Carreira, RC (2008) Seed germination in Cerrado species. Brazilian Journal of Plant Physiology 20, 167–181.CrossRefGoogle Scholar
Figure 0

Table 1. Environmental characteristics reported for Rio Branco and Cerrado savannas (Castro and Kauffman, 1998; Gomes et al., 2004; Barbosa and Fearnside, 2005b; Benedetti et al., 2011; INMET, 2013, 2014).

Figure 1

Table 2. Habitat, collection site, seed dispersal period and seed dry mass of the tree species selected for this study

Figure 2

Fig. 1. Fresh mass (A), dry mass (B) and water content (C) of seeds of populations of trees from Cerrado and Rio Branco savannas, Brazil. The same letters above the boxes indicate there is no significant difference.

Figure 3

Fig. 2. Effects of desiccation (A), water deficit (B), heat shock (C) and temperature (D) on the germination percentage of populations of trees from Cerrado and Rio Branco savannas, Brazil (*P < 0.05; **P  < 0.01; n.s., not significant).

Figure 4

Fig. 3. Effects of desiccation (A), water deficit (B), heat shock (C) and temperature (D) on the germination rate of populations of trees from Cerrado and Rio Branco savannas, Brazil (*P < 0.05; **P < 0.01; n.s., not significant).

Figure 5

Table 3. GLMMs of morphological and biophysical seed traits for conspecific and congeneric species from Cerrado and Rio Branco savannas, Brazil