SEARCH

SEARCH BY CITATION

Keywords:

  • clonal growth;
  • post-fire resprouting;
  • ramet integration;
  • restinga;
  • self-thinning

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
     Survival, resprouting and growth dynamics of the clonal shrub Andira legalis were studied for 2 years after fire, on a sandy spit on the coastal plain of Brazil. We examined the importance of resprouting for post-fire persistence, the relationship between resprouting patterns and injury suffered and size of individuals, and whether ramet growth was determined by competition and self-thinning.
  • 2
     Post-fire resprouting was responsible for production of new ramets and led to an increase in the mean number of ramets per A. legalis individual.
  • 3
     There was a negative association between resprouting from overground vs. underground. Highly injured individuals showed a significant tendency to sprout new ramets from underground organs whereas less damaged plants produced new branches and leaves from stem buds. Ramet production was related to an individual’s prefire size (basal area), but to a lesser extent than to injury.
  • 4
     Immediately after fire, the G(t,x) function (mean of absolute growth rates of shoots of size x at time t) was size-independent, suggesting an absence of competition. G(t,x) then became size-dependent, while D(t,x) (variance of absolute growth rates of shoots of size x at time t) remained size-independent, indicating that any competition between ramets was symmetric.
  • 5
     Size-dependent mortality and a negative linear relationship between mean ramet size and density indicated self-thinning. However, the low ramet mortality (7.7%) and the absence of size hierarchies (coefficients of variation remained constant) suggest that competition and self-thinning were not intense.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Resprouting allows for the persistence of many perennial shrubs after fire (Matlack et al. 1993; Olson & Platt 1995). Rapid resprouting and high growth rates after fire (Kauffman 1991; de Rouw 1993; Goto et al. 1996; Nascimento 1996) are often associated with protection of aerial and underground buds against high temperatures (Rundel 1982; James 1984), and access to energy reserves stored in underground organs (Bowen & Pate 1993). In the case of clonal plants, allocation of resources to vegetative regeneration depends on functional integration between ramets, a growth behaviour that is not present in non-clonal plants (Hara et al. 1993).

However, ramets are often functionally independent (see Ashmun et al. 1982) and this seems to result in growth patterns similar to those of non-clonal populations (de Kroon et al. 1992), where larger individuals have a greater capacity for obtaining resources, thus reducing resource availability for smaller plants (Hara 1984a; Weiner & Solbrig 1984; Weiner 1988). Competition becomes more intense at higher densities and variability, or inequality of size distribution therefore tends to increase during growth. Size hierarchies develop where a few individuals (dominants) constitute a high proportion of the population biomass, while most plants are relatively small (i.e. suppressed) (Harper 1977; Weiner & Solbrig 1984; Schmitt et al. 1987; Weiner 1988). As population biomass increases, many small individuals tend to die (self-thinning), reducing the density and subsequently the size variability (Weiner 1988; de Kroon et al. 1992).

Although manmade fires are increasingly frequent in the tropics (Nepstad et al. 1995; Pinard & Huffman 1997), the response of the rainforest complex on the Atlantic coast of Brazil to fire has hardly been examined (Araujo & Peixoto 1977; Cirne & Scarano 1996). Fire can promote extinction of both species and local populations, resulting in major losses of genetic information (Harrison 1991). Demographic studies of post-disturbance regeneration may enhance understanding of the functioning of such endangered vegetation and, thus, suggest techniques for its conservation.

This field study investigated the recovery of the clonal shrub Andira legalis following an illegal fire on the Brazilian sandy coastal plain. We examined resprouting, survival and growth dynamics of ramets during 2 years after fire. We aimed to answer the following questions: (i) does resprouting guarantee post-fire persistence of the population of A. legalis?; (ii) does resprouting pattern vary between individuals and, if so, does it depend on size and/or degree of fire injury?; and (iii) is ramet growth pattern determined by competition and self-thinning?

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study site

The study site was located in the Jacarepiá State Ecological Reserve, Saquarema municipality on the northern coast of Rio de Janeiro state, south-eastern Brazil (22°47′–22°57′ S; 42°20′–42°43′ W). Mean annual rainfall is 1000 mm and mean annual temperature ranges from 24 to 26 °C (Sá 1992; also provides a detailed description of the plant communities). The fixed dune forest here is one of the last remnants of such ‘restinga’ vegetation in Rio de Janeiro state. The swale between forest and strand line is covered by marsh vegetation, except where a spit of sandy soil, some 200 m long by 70 m wide (maximum) extends from the forest edge almost to the beach. In January 1995, the shrubby vegetation on this spit was burned by a fire that also spread to parts of the marsh and adjacent forest.

The spit vegetation shows a striking dominance of the shrub A. legalis (Vell.) Toledo (Leguminosae–Papilionoidae) and the geophyte palm Allagoptera arenaria (Gomes) O. Kuntz. Andira legalis, the object of this study, is restricted to coastal environments in Brazil, from the south-eastern states of Rio de Janeiro and Espírito Santo to the north-eastern states of Bahia and Ceará (Mattos 1979). In forest environments, such as that adjacent to the spit, this species becomes a small tree (up to 5 m tall). In the study area, which is fully exposed to sunlight, maximum plant height is 1.5 m, and shoots linked by an extensive underground stem system form monospecific thickets.

Persistence and resprouting patterns

All individuals of A. legalis within a 100 m × 30 m (0.3 ha) grid were labelled prior to the fire. An individual was arbitrarily defined as a group of ramets with overlapping canopies forming a distinct thicket, because for A. legalis, as in many clonal plants, distinction of different hierarchical categories (e.g. ramets, clonal fragments and genets) is difficult in the field (Eriksson 1993; Alpert 1996). The definition was based on previous field observations (Cirne & Scarano 1996), which showed that at least some of the ramets within a thicket showed underground physical connections and, thus, belonged to the same individual. Although this does not rule out the possibility of connections between thickets, spatial proximity of individual ramets will increase their potential for interaction.

To examine whether resprouting contributed to post-fire persistence of the A. legalis population, we counted the number of live ramets per individual before and after fire. Significant differences were tested by a Mann–Whitney Rank Sum test (P < 0.05; Zar 1999).

We performed a multiple regression analysis (Sokal & Rohlf 1995; Zar 1999) of the total number of new stems after fire on the proportion of stems killed by the fire (representing injury) and total basal area of stems in the cluster. This allowed us to examine whether the resprouting pattern varied at the individual level, and determine the relative contribution of fire damage and plant size to this degree of variation. The number of post-fire ramets for each individual was log-transformed before analysis, and the proportional mortality was arcsine-transformed (Zar 1999).

Growth dynamics

To assess whether ramet growth was determined by competition and self-thinning, we performed growth analyses based initially on two functions –G(t,x) and D(t,x) – described in the diffusion model (Hara 1984a,b). G(t,x) is the mean of the absolute growth rates of individuals (in this case, ramets) of size x in time t, and is determined from the absolute difference in ramet size between two time-points (i.e. instantaneous mean of size increase per unit time) whereas D(t,x) is the variance of these absolute growth rates. The G(t,x) function represents an averaged size-dependent trait, while D(t,x) represents trait variation promoted by a range of factors, including environmental heterogeneity, genetic variation, and variation in neighbourhood competition effects (Hara et al. 1993). These functions incorporate size-dependence of individual growth rather than mean plant size values (Hara 1984a), and thus allow us to describe competition processes based on plant performance and to compare the behaviour of different populations and species (Hara et al. 1993).

The G(t,x) and D(t,x) functions were estimated by linear regression. Three months after the fire (April 1995), all the new ramets that had resprouted from underground organs (hereafter called post-fire ramets) were permanently labelled, and height (h) and basal diameter (d) were measured then and every 2 months after. New ramets produced more than 3 months after the fire were not included in the growth analyses. Ramet growth was initially analysed for a 4-month interval (April to August 1995) and then at 6-month intervals until February 1997. We considered only ramets that survived the entire 22-month period. For each time interval, the ramets were divided into eight classes according to their initial size. Each class had a similar range and contained a minimum of seven ramets. The significance of linear regression between the mean size of each class and the mean and variance of absolute growth rates in the respective classes were tested for each time interval (F-test; P < 0.05; Zar 1999).

To examine whether variability in size distribution changed during post-fire growth, the coefficients of variation (CV) of h and d of post-fire ramets were calculated for the same intervals in the growth analysis. Differences between CVs were tested using the Z-test (P < 0.05; Zar 1999).

Post-fire survival of individuals, and resprouting and mortality of post-fire ramets were also checked every 2 months. To examine whether ramet mortality was size-dependent, the ramets were distributed to eight size classes of h and d at each census. Thus, a given ramet could move up one or more size classes from one census to another, depending on the growth rate during that period, with the class ranges varying for each 2-month interval.

Finally, to test whether self-thinning took place during post-fire ramet growth, log of ramet size (h and d) was plotted against log density for the 2 years after fire. Significance of this correlation was tested by the F-test (P < 0.05).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Persistence and resprouting patterns

All individuals (n = 95; comprised of 474 prefire ramets) of A. legalis growing in the grid before fire survived after the disturbance by initiation of new branches and leaves from buds present on pre-existing but damaged shoots, and/or initiation of new shoots from buds on underground stem organs (post-fire ramets). Resprouting through initiation of new branches and leaves was observed in 25.5% prefire ramets whereas initiation of new ramets from underground organs (n = 605, observed up to the fifth month after fire) led to a 27.7% increase in the total number of ramets compared with those found before the fire. The mean number of ramets per individual increased significantly after the fire (8 ± 6 vs. 5 ± 3; n = 95; P < 0.001). In spite of both pre- and post-fire fruit production, no seedlings were observed throughout the 2-year study.

The outstanding regeneration capacity of A. legalis was matched by its post-fire survival at both individual and ramet levels. Of the 339 post-fire ramets that sprouted within 3 months after fire, all but 26 (7.7%) were still alive after 2 years.

The multiple regression model (yPost-fire ramets= 0.593 xInjury+ 0.329 xBasal area; F2,92 = 31.93; P < 0.01; r2 = 0.397) explained more of the variation found in the number of new ramets after fire than did either of the dependent variables alone. Both the level of injury (assessed by the ramet mortality per individual) and individual basal area had a positive effect on the number of new ramets produced after fire. The effects were additive, but injury contributed more to the model than basal area, as seen by the values of the standardized regression coefficient.

Growth dynamics

Between 3 and 7 months after fire, there was no significant linear relationship (P > 0.05) between mean absolute growth rates G(t,x) and height of ramets. There was a significant, positive linear relationship in two of the three subsequent 6-month intervals (Fig. 1). A similar trend was observed in d, with significant linear relationships from 7 to 13 months (r = 0.922; P < 0.01), 13–19 months (r = 0.983; P < 0.01) and 19–25 months (r = 0.885; P < 0.01) after fire. For both h and d, therefore, mean growth rates of ramets tended to be proportional to ramet size, except in the initial 4 months after fire.

image

Figure 1. Mean absolute growth rates G(t,x) of Andira legalis ramets (n = 292) in eight height classes at different time intervals after fire in a restinga vegetation in south-eastern Brazil. (a) 3–7 months, (b) 7–13 months, (c) 13–19 months and (d) 19–25 months. Linear regressions are shown where significant: (b) r = 0.907 (P = 0.0019); (d) r = 0.790 (P = 0.0197).

Download figure to PowerPoint

There was, however, no significant linear relationship (all P > 0.05) between the variance in the absolute growth rates D(t,x), and h (Fig. 2) and d (data not shown) in any of the time intervals considered, which points to a size-independent variance of ramet growth.

image

Figure 2. Variance in the mean rates of absolute growth in height D(t,x) of Andiralegalis (n = 292) in size classes at different time intervals after fire. (a) 3–7 months, (b) 7–13 months, (c) 13–19 months and (d) 19–25 months. No significant linear regressions were found (P > 0.05).

Download figure to PowerPoint

The CV values for ramet h and d remained constant throughout the 2-year period following fire (Fig. 3). For height, none of the CV values was significantly different (Z-test; P < 0.05), varying from 0.41 to 0.47, with identical values in the last two measurements. For basal diameter, CV values varied from 0.30 and 0.35, and significant differences were found only between the seventh month after the fire and each of the last two dates (19 and 25 months). Thus, size variability during the study was nearly constant.

image

Figure 3. Changes in coefficient of variation (CV) of height (cm; ▴) and basal diameter (mm; ▪) of post-fire ramets of Andiralegalis for a 2-year period following a January fire. Values with same letters did not differ significantly from each other (Z-test; P < 0.05). CV values showed negligible variation, especially in the last months.

Download figure to PowerPoint

Mortality of ramets occurred mainly in the smaller size classes (data for h shown in Fig. 4). The significant negative correlation between log of ramet size and log of density for both h and d (Fig. 5) further indicates a self-thinning trajectory during the growth of ramets.

image

Figure 4. Ramet mortality of Andiralegalis (n = 26) per class according to height (▴) or basal diameter (▪) over a 2-year post-fire period. Ramets (n = 339) were distributed between size classes (1, small to 8, large) at each census.

Download figure to PowerPoint

image

Figure 5. Linear regression between mean ramet size of Andiralegalis (height ▴ and basal diameter ▪) and density on a log–log scale. Both linear relationships were significant (P < 0.01).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Post-fire persistence, injury level and resprouting

Fire had a major effect on ramet demography and clonal structure of A. legalis. Although fire is uncommon in restingas, all individuals survived burning, even after intense shoot damage, and resprouting further increased the number of ramets in the population. Vegetative regeneration after fire, as observed here, is a common strategy (Bond & van Wilgen 1996), even in restingas (Araujo & Peixoto 1977), and leads to retention of resources and restriction of losses to the environment (Hutchings & Bradbury 1986). In some fire-prone environments, such as longleaf pine savannas, vegetative regeneration is the only strategy utilized by shrubs after fire (Olson & Platt 1995). This enables rapid recolonization, high ramet production and high growth rates (Brewer & Platt 1994; Olson & Platt 1995; Nascimento 1996), and is also important for A. legalis when there is no new input due to seedling recruitment. Although fruits were produced at the study site and in the adjacent forest both before and after fire, seed germination and establishment did not occur. Thus, as in the case of the shrub Gaylussaciabaccata (Matlack et al. 1993), regeneration here was possible only by resprouting, despite seed production.

Production of new ramets in A. legalis was related both to the level of injury suffered by the individuals due to fire and to the individual size before fire, as reported for other species (e.g. Whelan 1995 and Wright & Bayley 1982; Keeley et al. 1998, respectively). The positive relation between level of injury of prefire ramets and production of post-fire ramets suggests a negative relationship between the proportion of resources allocated to mother and daughter ramets after the fire (Brewer & Platt 1994). Physiological stimulation of one structure or activity may have depressed another in an apparently compensating fashion, thus indicating internal control. Alternatively, the positive correlation could be the result of a correlative inhibition of meristems: in this case shoot meristems of relatively undamaged prefire ramets may produce new branches and leaves and thus inhibit the outgrowth of new ramets from underground meristems. Whatever the cause, regenerated ramets showed some degree of physical and functional interconnection, at least at the beginning of the recovery process. In clonal plants, bud production and resource allocation to shoots can be attributed to physiological integration (Hara et al. 1993). Although the production of new ramets might represent a limitation of parental biomass increase (Wijesinghe & Handel 1994), it leads to horizontal growth (Alpert 1990, 1996; Stuefer & Hutchings 1994) and often results in mutual support between ramets (Stuefer et al. 1994).

Individual size also contributed to the variation in resprouting pattern found. Resprouting vigour was related to prefire individual basal area, suggesting that larger individuals accumulate more reserves and/or have more active underground buds. Auld (1990) demonstrated that resprouting vigour in Angophora hispida is proportional to plant size, which can be regarded as an index of resource accumulation. Bowen & Pate (1993) argued that the amount of carbohydrate in underground organs acts as a primary limiting factor to biomass production after fire. Resource accumulation before fire was related to the number of ramets produced and to the increase in size after fire in G. baccata (Matlack et al. 1993). In the case of the pseudo-annual Helianthus laetiflorus, there was a positive relationship between plant size and the production, and size of structures and vegetative dispersal (Verburg et al. 1996).

Growth dynamics

In the early stages of growth of A. legalis after fire, G(t,x) and D(t,x) functions with respect to h and d were size-independent. This means that small ramets showed, on average, a higher relative growth rate (RGR = G(t,x)/x) than larger ones, suggesting an absence of competition and, possibly, the occurrence of ramet integration (Hara et al. 1993). The size-dependence of the G(t,x) function at later stages means that larger ramets then tend to grow faster, although the size-independence of D(t,x) suggests that they do not necessarily suppress smaller ones (Hartnett & Bazzaz 1985; Hara et al. 1993). Thus, competition, if it develops, will be symmetric. There was no evidence of integration between ramets.

Size hierarchies did not develop, as seen by the similar values of CV throughout all stages of ramet growth, and their absence may reflect the low levels of suppression and dominance, even in the later stages of growth. Weiner (1990), for instance, showed that when competition is symmetric, size variability does not increase because the growth of small and large plants is reduced in a similar manner.

The growth pattern shown by A. legalis in the early growth stages (i.e. size-independent G[t,x] and D[t,x]) is typical of clonal plants (de Kroon et al. 1992; Hara et al. 1993), and is unlikely to occur in predominantly sexual populations (Hara 1984a,b; Hara et al. 1993; Ekstam 1995). The size-dependent G(t,x) of the late growth stages is, however, common to both clonal (Schmitt et al. 1987; Hara et al. 1993) and non-clonal plants (Hara 1984a,b). The size-independent D(t,x) of the late growth stages of A. legalis is more characteristic of sexual populations in later stages of growth, when variance in absolute growth rates may assume near zero values (Hara et al. 1993). The shift from a typical clonal growth behaviour to a pattern more similar to that shown by sexually reproducing plants is consistent with the results showing ramet integration occurring only in the early stages.

The negative linear relationship between total ramet density and size, and the marked decrease in mortality with ramet size, suggest a self-thinning trajectory during ramet growth, as found for other clonal plants (Harper 1977; de Kroon et al. 1992; de Kroon & Kalliola 1995). When the G(t,x) function is linear with regard to size, as in later growth of A. legalis, the CV of size-dependent mortality usually decreases as self-thinning progresses. The absence of such effect was probably due to low mortality (Hara 1986), suggesting that self-thinning was not particularly intense.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank Projeto Restinga (FEEMA/JBRJ/UFRJ/Fundação O Boticário) for sponsorship; Brazilian Research (CNPq) and Education (CAPES) Councils for research grants; D.S.D. Araujo, H.S. Miranda and F. Borghetti for comments; H.C. Lima for information about the species studied; C.F.C. Sá for information about the study area; H.L.T. Zaluar, A. Lobão, R.N. Bastos and F. Reinert for field assistance; H.M. Duarte, E.A. Mattos and K.T. Ribeiro for statistical advice; L. Haddon for linguistic advice; and T. Hara and two anonymous referees for valuable contributions to the final version of this paper.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Alpert, P. (1990) Water sharing among ramets in a desert population of Distichlis spicata. American Journal of Botany, 77, 16481651.
  • Alpert, P. (1996) Nutrient sharing in natural clonal fragments of Fragaria chiloensis. Journal of Ecology, 84, 395406.
  • Araujo, D.S.D. & Peixoto, A.L. (1977) Renovação da comunidade vegetal de restinga após uma queimada. Trabalhos do XXVI Congresso Nacional de Botânica, pp. 117. Academia Brasileira de Ciências, Rio de Janeiro, Brazil.
  • Ashmun, J.W., Thomas, R.J., Pitelka, L.F. (1982) Translocation of photoassimilates between sister ramets in two rhizomatous forest herbs. Annals of Botany, 49, 403415.
  • Auld, T.D. (1990) The survival of juvenile plants of the resprouting shrub Angophora hispida (Myrtaceae) after a simulated low-intensity fire. Australian Journal of Botany, 38, 255260.
  • Bond, W.J. & van Wilgen, B.W. (1996) Fire and Plants. Chapman and Hall, London.
  • Bowen, M.J. & Pate, J.S. (1993) The significance of root starch in postfire shoot recovery of the resprouter Stirlingia latifolia R. Br. (Proteaceae). Annals of Botany, 72, 716.DOI: 10.1006/anbo.1993.1075
  • Brewer, J.S. & Platt, W.J. (1994) Effects of fire season and soil fertility on clonal growth in a pyrophilic forb, Pityopsis graminifolia (Asteraceae). American Journal of Botany, 81, 805814.
  • Cirne, P. & Scarano, F.R. (1996) Rebrotamento após o fogo de Andira legalis (Leguminosae) em restinga fluminense. Impactos de Queimadas em Áreas de Cerrado e Restinga (eds H.S.Miranda, C.H.Saito & B.F.S.Dias), pp. 128136. Anais do Simpósio Impacto de Queimadas sobre os Ecossistemas e Mudanças Globais, UnB, Brasília, Brazil.
  • Ekstam, B. (1995) Ramet size equalisation in a clonal plant, Phragmites australis. Oecologia, 104, 440446.
  • Eriksson, O. (1993) Dynamics of genets in clonal plants. Trends in Ecology and Evolution, 9, 313316.
  • Goto, Y., Yoshitake, T., Okano, M., Shimada, K. (1996) Seedling regeneration and vegetative resprouting after fires in Pinus densiflora forests. Vegetatio, 122, 157165.
  • Hara, T. (1984a) A stochastic model and the moment dynamics of the growth and size distribution in plant populations. Journal of Theoretical Biology, 109, 173190.
  • Hara, T. (1984b) Dynamics of stand structure in plant monocultures. Distribution in plant populations. Journal of Theoretical Biology, 110, 223239.
  • Hara, T. (1986) Effects of density and extinction coefficient on size variability in plant populations. Annals of Botany, 57, 885892.
  • Hara, T., Der Toorn, J.V., Mook, J.H. (1993) Growth dynamics and size structure of shoots of Phragmites australis, a clonal plant. Journal of Ecology, 81, 4760.
  • Harper, J.L. (1977) Population Biology of Plants. Academic Press, London.
  • Harrison, S. (1991) Local extinction in a metapopulation context: an empirical evaluation. Biological Journal of the Linnean Society, 42, 7388.
  • Hartnett, D.C. & Bazzaz, F.A. (1985) The genet and the ramet population dynamics of Solidago canadensis in an abandoned field. Journal of Ecology, 73, 407413.
  • Hutchings, M.J. & Bradbury, I.K. (1986) Ecological perspectives on clonal perennial herbs. When does it pay for individual clone modules to be functionally integrated or independent? Bioscience, 36, 178182.
  • James, S. (1984) Lignotubers and burls – their structure, function and ecological significance in Mediterranean ecosystems. Botanical Review, 50, 227266.
  • Kauffman, J.B. (1991) Survival by sprouting following fire in tropical forests of the eastern Amazon. Biotropica, 23, 219224.
  • Keeley, J.E., Keeley, M.B., Bond, W.J. (1998) Stem demography and post-fire recruitment of a resprouting serotinous conifer. Journal of Vegetation Science, 10, 6976.
  • de Kroon, H., Hara, T., Kwant, R. (1992) Size hierarchies of shoots and clones in clonal herb monocultures: do clonal and non-clonal plants compete differently? Oikos, 63, 410419.
  • de Kroon, H. & Kalliola, R. (1995) Shoot dynamics of the giant grass Gynerium sagittatum in Peruvian Amazon floodplains, a clonal plant that does show self-thinning. Oecologia, 101, 124131.
  • Matlack, G.R., Gibson, D.J., Good, R.E. (1993) Regeneration of the shrub Gaylussacia baccata and associated species after low-intensity fire in an Atlantic coastal plain forest. American Journal of Botany, 80, 119126.
  • Mattos, N.F. (1979) O gênero Andira Lam. (Leguminosae Papilonoideae) no Brasil. Acta Amazonica, 9, 241266.
  • Nascimento, M.T. (1996) Rapid resprouting following fire in juveniles of Metrodorea pubescens (Rutaceae) in a mesophytic forest in central Brazil. Ciência e Cultura, 48, 182183.
  • Nepstad, D.C., Jipp, P., Moutinho, P., Negreiros, G., Vieira, S. (1995) Forest recovery following pasture abandonment in Amazonia: canopy seasonality, fire resistance and ants. Evaluating and Monitoring the Health of Large-Scale Ecosystems (eds D.J.Rapport, C.L.Gaudet & P.Calow), pp. 339349. Springer Verlag, Heidelberg, Germany.
  • Olson, M.S. & Platt, W. (1995) Effects of habitat and growing season fires on resprouting of shrubs in longleaf pine savannas. Vegetatio, 119, 101118.
  • Pinard, M.A. & Huffman, J. (1997) Fire resistance and bark properties of trees in a seasonally dry forest in eastern Bolivia. Journal of Tropical Ecology, 13, 727740.
  • de Rouw, A. (1993) Regeneration by sprouting in slash and burn rice cultivation, Taï rain forest, Côte d’Ivoire. Journal of Tropical Ecology, 9, 387408.
  • Rundel, P.W. (1982) Fire as an ecological factor. Encyclopaedia of Plant Physiology 12A. Physiological Plant Ecology (eds O.L.Lange, P.S.Nobel, C.B.Osmond & H.Ziegler), pp. 501538. Springer Verlag, Heidelberg, Germany.
  • Sá, C.F.C. (1992) A vegetação da restinga de Ipitangas, Reserva Ecológica Estadual de Jacarepiá, Saquarema (RJ): fisionomia e listagem das angiospermas. Arquivos do Jardim Botânico do Rio de Janeiro, 31, 87102.
  • Schmitt, J., Eccleston, J., Ehrhardt, D.W. (1987) Dominance and supression, size-dependent growth and self-thinning in a natural Impatiens capensis population. Journal of Ecology, 75, 651665.
  • Sokal, R.R. & Rohlf, F.J. (1995) Biometry, 3rd edn. W.H. Freeman, New York, NY.
  • Stuefer, J.F., During, H.J., de Kroon, H. (1994) High benefits of clonal integration in two stoloniferous species, in response to heterogeneous light environments. Journal of Ecology, 82, 511518.
  • Stuefer, J.F. & Hutchings, M.J. (1994) Environmental heterogeneity and clonal growth: a study of the capacity for reciprocal translocation in Glechoma hederaceae L. Oecologia, 100, 302308.
  • Verburg, R.W., Kwant, R., Werger, J.A. (1996) The effect of plant size on vegetative reproduction in a pseudo-annual. Vegetatio, 125, 185192.
  • Weiner, J. (1988) Variation in the performance of individuals in plant populations. Plant Population Ecology (eds A.J.Davy, M.J.Hutchings & A.R.Watkinson), pp. 5881. Blackwell Scientific Publications, Oxford.
  • Weiner, J. (1990) Asymmetric competition in plant populations. Trends in Ecology and Evolution, 5, 360364.
  • Weiner, J. & Solbrig, O.T. (1984) The meaning and measurement of size hierarchies in plant populations. Oecologia, 61, 334336.
  • Whelan, R.J. (1995) The Ecology of Fire. Cambridge University Press, Cambridge.
  • Wijesinghe, D.K. & Handel, S.N. (1994) Advantages of clonal growth in heterogeneous habitats: an experiment with Potentilla simplex. Journal of Ecology, 82, 495502.
  • Wright, H.A. & Bayley, A.W. (1982) Fire Ecology: United States and Southern Canada. Wiley-Interscience, New York, NY.
  • Zar, J.H. (1999) Biostatistical Analysis, 4th edn. Prentice Hall, Upper Saddle River, NJ.

Received 18 January 2000 revision accepted 26 September 2000