Mortality rate in young plants of Aegiphila sellowiana after different periods of soil flooding. D = drained; F15 = flooding for 15 days; F18 = flooding for 18 days; F25 = flooding for 25 days; F50 = flooding for 50 days; F80 = flooding for 80 days; REA = flooding for 50 days followed by re-aeration for 30 days; AVE = average mortality considering all treatments. 

Contexts in source publication

Context 1

... more elevated adjacent areas, such as dikes along the riverbanks. In order to address the issues raised, we determined the influence of different flooding regimes on sapling mortality and analyzed the genetic diversity between surviving and non-surviving plants. Seeds of A. sellowiana , collected in natural environments in the municipalities of Londrina, Arapongas, Nova Santa Barbara and São Jerônimo da Serra, Paraná, Brazil, from five matrices at each locality, were germinated in a nursery in tubes containing moist sub - strate (80% soil and 20% mixture of straw, chicken manure, lime, and coffee dregs). After six months, plants were taken to a greenhouse and transplanted in 4-L plastic pots, containing substrate made with soil and sand (3:1). The experiments were started three months after ac- climation of the plants and lasted 80 days. Survival was determined using plants divided into seven groups: D - 10 plants kept in drained soil; F15 - 10 plants kept in flooded soil for 15 days; F18 - 10 plants kept in flooded soil for 18 days; F25 - 10 plants kept in flooded soil for 25 days; F50 - 15 plants kept in flooded soil for 50 days; F80 - 15 plants kept in flooded soil for 80 days; REA (re-aerated) - 15 plants kept under flooding conditions for 50 days followed by drainage starting on the 51st day and lasting for 30 days. The plants under flooding conditions were kept in soil with 2 cm water above the substrate surface. Plants in each treatment were duly numbered, and at the end of the experiments, dead individuals were recorded. Before the beginning of the flooding experiments, a leaf from each plant utilized in the treatments was collected, numbered and stored in a -80°C freezer. After the end of the experi- ments, the leaves collected were separated into two groups: 1) leaves from plants surviving the stress of flooding and 2) leaves from plants not surviving. Genetic analysis was carried out on randomly selected samples, 30 plants of each group. Extraction of genomic DNA from leaves was performed based on the method described by Doyle and Doyle (1987), except that CTAB was substituted by MATAB (Sigma) in the ex- traction buffer. The concentration of DNA was estimated using a fluorometer (DyNA Quant 200, Höfer-Pharmacia), in accordance with manufacturer instructions. DNA samples obtained from 30 plants of each group were adjusted to 10 ng/μL for polymerase chain reaction (PCR) use. A total of 19 primers (Operon Technologies; Table 1) were used in DNA amplifications in a volume of 15 μL containing 1X PCR buffer (75 mM Tris-HCl, 50 mM KCl, 2.0 mM MgCl 2 , 20 mM (NH 4 ) 2 SO 4 , 0.2 mM dNTPs, 0.4 μM primer, 0.9 U Taq DNA polymerase (Biotools), and 20 ng DNA). DNA amplification was performed using a PTC 100 thermocycler (MJ Research), programmed for 3 min at 94°C for initial denaturation, followed by 48 cycles of 1 min at 94°C, 1 min 45 s at 38°C, and 2 min at 72°C. The last cycle was followed by 7 min of extension at 72°C. The amplification products were separated on 1.2% agarose gels immersed in TAE (40 mM Tris-acetate, 1 mM EDTA, pH 8.0) and stained with ethidium bromide. Electrophoretic gels were run at 120 V for 2 h. The random amplified polymorphic DNA (RAPD) profiles were visualized under UV light and photographed for data analysis. The presence/absence of RAPD bands was recorded in a binary format statistical anal- ysis. Only well-amplified molecular fragments were considered. Bands of similar molecular weight and migration pattern among individuals were considered to be homologous. Control samples containing all the reaction materials except DNA were run to avoid the utilization of bands produced from self-amplification or DNA contamination. The calculation of allele frequencies and genetic diversity using dominant markers can be problematic (Zhivotovsky, 1999). However, Krauss (2000) demonstrated that biased data produced due to the dominant nature of RAPD markers can be eliminated in a set of highly polymorphic data. It has been demonstrated that accurate estimates of genetic param- eters of populations require the use of a large number of RAPD loci and about 30 individuals per population (Tero et al., 2003). The bootstrap method was used with the DBOOT program version 1.1 (Coelho, 2000), for the purpose of measuring the coefficient of variation related to the number of markers used. The percentage of polymorphic loci within populations and the Shannon index were calculated using the POPGENE program (Yeh et al., 2000). Genetic variation within and be- tween populations was estimated by analysis of molecular variance (AMOVA), utilizing the ARLEQUIN program 2.0 (Schneider et al., 2000). Principal coordinate analysis based on genetic distances (Nei, 1978) was carried out using the NTSYS-PC (Numerical Taxonomy and Multivariate Analysis System for personal computer) software, version 2.1 (Rohlf, 2000). In total, 75 plants were submitted to different periods of soil flooding. Of these, 35 (46.7%) died (Figure 1). The treatments with the longest duration of flooding did not show a greater percentage of mortality (Figure 1). In general, the death of the plants occurred early, in the first week of flooding. Hence, those plants that had the capacity of surviving in this period continued living for the rest of the treatment period, even for the longest periods, 50 and 80 days. The surviving plants showed morphological responses to flooding, such as the presence of hyper - trophied lenticels, stem fissures, and diageotropic and adventitious roots (data not shown). The 19 primers that were selected produced 123 consistent and well-defined bands. The bootstrap results showed a coefficient of variation of 4.67%, indicating that the number of markers obtained was suitable for reliable analysis of the results. A variation was observed in the percent- age of polymorphic loci among the primers used. The percent of polymorphic loci was practically equal between the groups of surviving plants (67.48%) and non-surviving plants (66.67%) (Table 1). However, both indices of genetic diversity showed slightly higher values for the surviving group (Table 1). AMOVA showed that approximately 6% of genetic variability was due to differences be - tween groups of plants surviving and not surviving flooding, while 94% was due to genetic differ - ences between individuals within these groups (Table 2). Principal coordinate analysis showed that the surviving plants formed a group in one of the coordinate quadrants, while the plants not surviving flooding were dispersed, where the individuals mixed with those of the surviving group (Figure 2). Flooding is certainly an important threat to A. sellowiana , given the considerable mortal- ity rate (46.7%; Figure 1) and the short time in which it occurred. However, A. sellowiana can be considered tolerant to this stress, since the majority of the individuals tested were able to withstand even 80 days of flooding. Hence, if the water tables rise to higher levels, reaching plant communi- ties that are seldom challenged by flooding, we can expect a reasonable tolerance to it and the sur - vival of a considerable portion of the population. In times of global climate changes, abrupt rises in river levels are a growing concern. According to Silva and Guetter (2003), global temperature has been rising since the 1970s, and the hydrological cycle is accompanying this change. For instance, in Paraná State, South Brazil, the enhancement of the hydrological cycle boosted the frequency and intensity of rains, the river levels, and, paradoxically, the frequency and intensity of drought. These effects are expected to be intensified as a consequence of global temperature rise, amplifying the occurrence of extreme events, such as droughts and floods. A mortality rate similar to the one found in A. sellowiana was reported for P. rigida : 40% of the young plants died following flooding stress (Silva et al., 2010). However, P. rigida is not vulnerable to low-intensity waterlogging, where mortality is associated with long periods of a combination of soil flooding and complete submergence of saplings. Ferreira et al. (2007) found a higher degree of resistance in the Amazonian tree H. sucuuba , where seedlings from flooded habitats died only after several days of complete submergence. In another study, L. divaricata plants, on the other ha nd, were totally tolerant to one month of flooding, but were completely sus - ceptible to an additional month of submergence. These species show different degrees of flooding tolerance, beginning with the most susceptible, A. sellowiana , followed by L. divaricata , P. rigida , and finally the most resistant, H. sucuuba . Certainly, the differences in tolerance seen among them result from a variation in the degree of historical exposure to flooding episodes. The fact that surviving plants showed morphological responses to flooding indicates that a low oxygen escape syndrome strategy is favored in this species, leading to an improvement in oxy- gen supply to the tissues underwater, consequently ensuring the survival of plants (Bailey-Serres and Voesenek, 2008). The study of these adaptations to flooding is the focus of a future study. The principal coordinate analysis results concur with the AMOVA results, which show a genetic difference of only about 6% between the groups of plants surviving and not surviving flooding (Table 2; Figure 2). Although the genetic variation between populations was much smaller than the variation within populations and the coefficient of variation and percentage of polymorphic loci were mostly the same between samples, the percentage of variation was highly significant, indicating that the two samples analyzed are different (Table 2). Similar results were obtained by Carvalho et al. (2002) for L. divaricata (Tiliaceae), in which AMOVA showed a difference of 10.39% between plants from areas submitted to flooding and those from areas that did not ...

Context 2

... the influence of different flooding regimes on sapling mortality and analyzed the genetic diversity between surviving and non-surviving plants. Seeds of A. sellowiana , collected in natural environments in the municipalities of Londrina, Arapongas, Nova Santa Barbara and São Jerônimo da Serra, Paraná, Brazil, from five matrices at each locality, were germinated in a nursery in tubes containing moist sub - strate (80% soil and 20% mixture of straw, chicken manure, lime, and coffee dregs). After six months, plants were taken to a greenhouse and transplanted in 4-L plastic pots, containing substrate made with soil and sand (3:1). The experiments were started three months after ac- climation of the plants and lasted 80 days. Survival was determined using plants divided into seven groups: D - 10 plants kept in drained soil; F15 - 10 plants kept in flooded soil for 15 days; F18 - 10 plants kept in flooded soil for 18 days; F25 - 10 plants kept in flooded soil for 25 days; F50 - 15 plants kept in flooded soil for 50 days; F80 - 15 plants kept in flooded soil for 80 days; REA (re-aerated) - 15 plants kept under flooding conditions for 50 days followed by drainage starting on the 51st day and lasting for 30 days. The plants under flooding conditions were kept in soil with 2 cm water above the substrate surface. Plants in each treatment were duly numbered, and at the end of the experiments, dead individuals were recorded. Before the beginning of the flooding experiments, a leaf from each plant utilized in the treatments was collected, numbered and stored in a -80°C freezer. After the end of the experi- ments, the leaves collected were separated into two groups: 1) leaves from plants surviving the stress of flooding and 2) leaves from plants not surviving. Genetic analysis was carried out on randomly selected samples, 30 plants of each group. Extraction of genomic DNA from leaves was performed based on the method described by Doyle and Doyle (1987), except that CTAB was substituted by MATAB (Sigma) in the ex- traction buffer. The concentration of DNA was estimated using a fluorometer (DyNA Quant 200, Höfer-Pharmacia), in accordance with manufacturer instructions. DNA samples obtained from 30 plants of each group were adjusted to 10 ng/μL for polymerase chain reaction (PCR) use. A total of 19 primers (Operon Technologies; Table 1) were used in DNA amplifications in a volume of 15 μL containing 1X PCR buffer (75 mM Tris-HCl, 50 mM KCl, 2.0 mM MgCl 2 , 20 mM (NH 4 ) 2 SO 4 , 0.2 mM dNTPs, 0.4 μM primer, 0.9 U Taq DNA polymerase (Biotools), and 20 ng DNA). DNA amplification was performed using a PTC 100 thermocycler (MJ Research), programmed for 3 min at 94°C for initial denaturation, followed by 48 cycles of 1 min at 94°C, 1 min 45 s at 38°C, and 2 min at 72°C. The last cycle was followed by 7 min of extension at 72°C. The amplification products were separated on 1.2% agarose gels immersed in TAE (40 mM Tris-acetate, 1 mM EDTA, pH 8.0) and stained with ethidium bromide. Electrophoretic gels were run at 120 V for 2 h. The random amplified polymorphic DNA (RAPD) profiles were visualized under UV light and photographed for data analysis. The presence/absence of RAPD bands was recorded in a binary format statistical anal- ysis. Only well-amplified molecular fragments were considered. Bands of similar molecular weight and migration pattern among individuals were considered to be homologous. Control samples containing all the reaction materials except DNA were run to avoid the utilization of bands produced from self-amplification or DNA contamination. The calculation of allele frequencies and genetic diversity using dominant markers can be problematic (Zhivotovsky, 1999). However, Krauss (2000) demonstrated that biased data produced due to the dominant nature of RAPD markers can be eliminated in a set of highly polymorphic data. It has been demonstrated that accurate estimates of genetic param- eters of populations require the use of a large number of RAPD loci and about 30 individuals per population (Tero et al., 2003). The bootstrap method was used with the DBOOT program version 1.1 (Coelho, 2000), for the purpose of measuring the coefficient of variation related to the number of markers used. The percentage of polymorphic loci within populations and the Shannon index were calculated using the POPGENE program (Yeh et al., 2000). Genetic variation within and be- tween populations was estimated by analysis of molecular variance (AMOVA), utilizing the ARLEQUIN program 2.0 (Schneider et al., 2000). Principal coordinate analysis based on genetic distances (Nei, 1978) was carried out using the NTSYS-PC (Numerical Taxonomy and Multivariate Analysis System for personal computer) software, version 2.1 (Rohlf, 2000). In total, 75 plants were submitted to different periods of soil flooding. Of these, 35 (46.7%) died (Figure 1). The treatments with the longest duration of flooding did not show a greater percentage of mortality (Figure 1). In general, the death of the plants occurred early, in the first week of flooding. Hence, those plants that had the capacity of surviving in this period continued living for the rest of the treatment period, even for the longest periods, 50 and 80 days. The surviving plants showed morphological responses to flooding, such as the presence of hyper - trophied lenticels, stem fissures, and diageotropic and adventitious roots (data not shown). The 19 primers that were selected produced 123 consistent and well-defined bands. The bootstrap results showed a coefficient of variation of 4.67%, indicating that the number of markers obtained was suitable for reliable analysis of the results. A variation was observed in the percent- age of polymorphic loci among the primers used. The percent of polymorphic loci was practically equal between the groups of surviving plants (67.48%) and non-surviving plants (66.67%) (Table 1). However, both indices of genetic diversity showed slightly higher values for the surviving group (Table 1). AMOVA showed that approximately 6% of genetic variability was due to differences be - tween groups of plants surviving and not surviving flooding, while 94% was due to genetic differ - ences between individuals within these groups (Table 2). Principal coordinate analysis showed that the surviving plants formed a group in one of the coordinate quadrants, while the plants not surviving flooding were dispersed, where the individuals mixed with those of the surviving group (Figure 2). Flooding is certainly an important threat to A. sellowiana , given the considerable mortal- ity rate (46.7%; Figure 1) and the short time in which it occurred. However, A. sellowiana can be considered tolerant to this stress, since the majority of the individuals tested were able to withstand even 80 days of flooding. Hence, if the water tables rise to higher levels, reaching plant communi- ties that are seldom challenged by flooding, we can expect a reasonable tolerance to it and the sur - vival of a considerable portion of the population. In times of global climate changes, abrupt rises in river levels are a growing concern. According to Silva and Guetter (2003), global temperature has been rising since the 1970s, and the hydrological cycle is accompanying this change. For instance, in Paraná State, South Brazil, the enhancement of the hydrological cycle boosted the frequency and intensity of rains, the river levels, and, paradoxically, the frequency and intensity of drought. These effects are expected to be intensified as a consequence of global temperature rise, amplifying the occurrence of extreme events, such as droughts and floods. A mortality rate similar to the one found in A. sellowiana was reported for P. rigida : 40% of the young plants died following flooding stress (Silva et al., 2010). However, P. rigida is not vulnerable to low-intensity waterlogging, where mortality is associated with long periods of a combination of soil flooding and complete submergence of saplings. Ferreira et al. (2007) found a higher degree of resistance in the Amazonian tree H. sucuuba , where seedlings from flooded habitats died only after several days of complete submergence. In another study, L. divaricata plants, on the other ha nd, were totally tolerant to one month of flooding, but were completely sus - ceptible to an additional month of submergence. These species show different degrees of flooding tolerance, beginning with the most susceptible, A. sellowiana , followed by L. divaricata , P. rigida , and finally the most resistant, H. sucuuba . Certainly, the differences in tolerance seen among them result from a variation in the degree of historical exposure to flooding episodes. The fact that surviving plants showed morphological responses to flooding indicates that a low oxygen escape syndrome strategy is favored in this species, leading to an improvement in oxy- gen supply to the tissues underwater, consequently ensuring the survival of plants (Bailey-Serres and Voesenek, 2008). The study of these adaptations to flooding is the focus of a future study. The principal coordinate analysis results concur with the AMOVA results, which show a genetic difference of only about 6% between the groups of plants surviving and not surviving flooding (Table 2; Figure 2). Although the genetic variation between populations was much smaller than the variation within populations and the coefficient of variation and percentage of polymorphic loci were mostly the same between samples, the percentage of variation was highly significant, indicating that the two samples analyzed are different (Table 2). Similar results were obtained by Carvalho et al. (2002) for L. divaricata (Tiliaceae), in which AMOVA showed a difference of 10.39% between plants from areas submitted to flooding and those from areas that did not undergo flooding, and Silva et al. (2010), who found a genetic differ - ence of 6.27% between P. rigida plants ...

Context 3

... collected were separated into two groups: 1) leaves from plants surviving the stress of flooding and 2) leaves from plants not surviving. Genetic analysis was carried out on randomly selected samples, 30 plants of each group. Extraction of genomic DNA from leaves was performed based on the method described by Doyle and Doyle (1987), except that CTAB was substituted by MATAB (Sigma) in the ex- traction buffer. The concentration of DNA was estimated using a fluorometer (DyNA Quant 200, Höfer-Pharmacia), in accordance with manufacturer instructions. DNA samples obtained from 30 plants of each group were adjusted to 10 ng/μL for polymerase chain reaction (PCR) use. A total of 19 primers (Operon Technologies; Table 1) were used in DNA amplifications in a volume of 15 μL containing 1X PCR buffer (75 mM Tris-HCl, 50 mM KCl, 2.0 mM MgCl 2 , 20 mM (NH 4 ) 2 SO 4 , 0.2 mM dNTPs, 0.4 μM primer, 0.9 U Taq DNA polymerase (Biotools), and 20 ng DNA). DNA amplification was performed using a PTC 100 thermocycler (MJ Research), programmed for 3 min at 94°C for initial denaturation, followed by 48 cycles of 1 min at 94°C, 1 min 45 s at 38°C, and 2 min at 72°C. The last cycle was followed by 7 min of extension at 72°C. The amplification products were separated on 1.2% agarose gels immersed in TAE (40 mM Tris-acetate, 1 mM EDTA, pH 8.0) and stained with ethidium bromide. Electrophoretic gels were run at 120 V for 2 h. The random amplified polymorphic DNA (RAPD) profiles were visualized under UV light and photographed for data analysis. The presence/absence of RAPD bands was recorded in a binary format statistical anal- ysis. Only well-amplified molecular fragments were considered. Bands of similar molecular weight and migration pattern among individuals were considered to be homologous. Control samples containing all the reaction materials except DNA were run to avoid the utilization of bands produced from self-amplification or DNA contamination. The calculation of allele frequencies and genetic diversity using dominant markers can be problematic (Zhivotovsky, 1999). However, Krauss (2000) demonstrated that biased data produced due to the dominant nature of RAPD markers can be eliminated in a set of highly polymorphic data. It has been demonstrated that accurate estimates of genetic param- eters of populations require the use of a large number of RAPD loci and about 30 individuals per population (Tero et al., 2003). The bootstrap method was used with the DBOOT program version 1.1 (Coelho, 2000), for the purpose of measuring the coefficient of variation related to the number of markers used. The percentage of polymorphic loci within populations and the Shannon index were calculated using the POPGENE program (Yeh et al., 2000). Genetic variation within and be- tween populations was estimated by analysis of molecular variance (AMOVA), utilizing the ARLEQUIN program 2.0 (Schneider et al., 2000). Principal coordinate analysis based on genetic distances (Nei, 1978) was carried out using the NTSYS-PC (Numerical Taxonomy and Multivariate Analysis System for personal computer) software, version 2.1 (Rohlf, 2000). In total, 75 plants were submitted to different periods of soil flooding. Of these, 35 (46.7%) died (Figure 1). The treatments with the longest duration of flooding did not show a greater percentage of mortality (Figure 1). In general, the death of the plants occurred early, in the first week of flooding. Hence, those plants that had the capacity of surviving in this period continued living for the rest of the treatment period, even for the longest periods, 50 and 80 days. The surviving plants showed morphological responses to flooding, such as the presence of hyper - trophied lenticels, stem fissures, and diageotropic and adventitious roots (data not shown). The 19 primers that were selected produced 123 consistent and well-defined bands. The bootstrap results showed a coefficient of variation of 4.67%, indicating that the number of markers obtained was suitable for reliable analysis of the results. A variation was observed in the percent- age of polymorphic loci among the primers used. The percent of polymorphic loci was practically equal between the groups of surviving plants (67.48%) and non-surviving plants (66.67%) (Table 1). However, both indices of genetic diversity showed slightly higher values for the surviving group (Table 1). AMOVA showed that approximately 6% of genetic variability was due to differences be - tween groups of plants surviving and not surviving flooding, while 94% was due to genetic differ - ences between individuals within these groups (Table 2). Principal coordinate analysis showed that the surviving plants formed a group in one of the coordinate quadrants, while the plants not surviving flooding were dispersed, where the individuals mixed with those of the surviving group (Figure 2). Flooding is certainly an important threat to A. sellowiana , given the considerable mortal- ity rate (46.7%; Figure 1) and the short time in which it occurred. However, A. sellowiana can be considered tolerant to this stress, since the majority of the individuals tested were able to withstand even 80 days of flooding. Hence, if the water tables rise to higher levels, reaching plant communi- ties that are seldom challenged by flooding, we can expect a reasonable tolerance to it and the sur - vival of a considerable portion of the population. In times of global climate changes, abrupt rises in river levels are a growing concern. According to Silva and Guetter (2003), global temperature has been rising since the 1970s, and the hydrological cycle is accompanying this change. For instance, in Paraná State, South Brazil, the enhancement of the hydrological cycle boosted the frequency and intensity of rains, the river levels, and, paradoxically, the frequency and intensity of drought. These effects are expected to be intensified as a consequence of global temperature rise, amplifying the occurrence of extreme events, such as droughts and floods. A mortality rate similar to the one found in A. sellowiana was reported for P. rigida : 40% of the young plants died following flooding stress (Silva et al., 2010). However, P. rigida is not vulnerable to low-intensity waterlogging, where mortality is associated with long periods of a combination of soil flooding and complete submergence of saplings. Ferreira et al. (2007) found a higher degree of resistance in the Amazonian tree H. sucuuba , where seedlings from flooded habitats died only after several days of complete submergence. In another study, L. divaricata plants, on the other ha nd, were totally tolerant to one month of flooding, but were completely sus - ceptible to an additional month of submergence. These species show different degrees of flooding tolerance, beginning with the most susceptible, A. sellowiana , followed by L. divaricata , P. rigida , and finally the most resistant, H. sucuuba . Certainly, the differences in tolerance seen among them result from a variation in the degree of historical exposure to flooding episodes. The fact that surviving plants showed morphological responses to flooding indicates that a low oxygen escape syndrome strategy is favored in this species, leading to an improvement in oxy- gen supply to the tissues underwater, consequently ensuring the survival of plants (Bailey-Serres and Voesenek, 2008). The study of these adaptations to flooding is the focus of a future study. The principal coordinate analysis results concur with the AMOVA results, which show a genetic difference of only about 6% between the groups of plants surviving and not surviving flooding (Table 2; Figure 2). Although the genetic variation between populations was much smaller than the variation within populations and the coefficient of variation and percentage of polymorphic loci were mostly the same between samples, the percentage of variation was highly significant, indicating that the two samples analyzed are different (Table 2). Similar results were obtained by Carvalho et al. (2002) for L. divaricata (Tiliaceae), in which AMOVA showed a difference of 10.39% between plants from areas submitted to flooding and those from areas that did not undergo flooding, and Silva et al. (2010), who found a genetic differ - ence of 6.27% between P. rigida plants from areas that were flooded and not flooded. For some species, the ability of surviving waterlogging is not widespread in the popula- tion, as we found for A. sellowania (Ferreira et al., 2007; Silva et al., 2010). One explanation to this assumes that tolerance is a part of the genomic plasticity of the species and that variable response is caused by microhabitat variations. Alternatively, these species coped with variable duration or pressure of waterlogging, insufficient to convert the whole population to flood-tolerant. If so, not all individuals show resistance to flooding as a constitutive feature and the population is walking the evolutionary path to becoming tolerant. In these cases, we should expect to find distinct allele frequencies as well as genetic diversity between them. The results found for A. sellowiana indi- cate that certainly the two groups possess different alleles, selected by the flooding process over time, that are expressed when the plants are submitted to stress from flooding, thereby allowing these plants to survive. Thus, we can assume that flood tolerance is genetically determined in A. sellowiana and that part of its population has this character as a constitutive feature. We can conclude that A. sellowiana is tolerant to flooding due to genetically determined factors. However, the lack of a strong genetic differentiation between saplings surviving and not surviving flooding can be attributed to high levels of gene flow. In addition to using molecular markers to study the pattern of genetic variation of A. sellowiana , it would be interesting to in- vestigate the variation in ...