| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Tropical Biology |
2Universidade Federal do Rio de Janeiro, CCS, IB, Departamento de Ecologia, Caixa Postal 68020, cep 21941-970, Rio de Janeiro, RJ, Brazil; 3Center for Ecological Research and Forestry Applications (CREAF), Faculty of Sciences, Autonomous University of Barcelona, 08193 Bellaterra (Barcelona), Spain
Received for publication March 2, 2004. Accepted for publication December 9, 2004.
ABSTRACT
Restingas are sandy coastal plains that stand between the sea and the Brazilian Atlantic forest mountains. The predominant restinga vegetation type in northern Rio de Janeiro, Brazil, is characterized by the formation of islands that begins with colonization by some pioneer herbs and/or woody plants. Pioneer plants are stress-resistant and nurse many other less-resistant plant species. Determining the spatiotemporal variation in the dynamics of nurse plants is essential to understand the ecological functioning of restingas as a whole. The goal of this study was to analyze the spatiotemporal variation in population dynamics of the nurse bromeliad Aechmea nudicaulis. We monitored A. nudicaulis ramets in different habitats, microhabitats, and years. We analyzed the spatiotemporal variation in demographic traits and in population growth rate. Results showed young ramet traits were more variable at the microhabitat level, and when variable, vegetative ramet traits varied at all spatiotemporal scales. Overall, 
values indicated that A. nudicaulis basically remained spatiotemporally stable as most of the values did not significantly differ from unity. Hence, the stability of A. nudicaulis in different microhabitats and habitats in the restinga may create several settlement opportunities for many other less-resistant species.
Key Words: Aechmea nudicaulis • Brazil • Bromeliaceae • life-cycle traits • nurse plants • population dynamics • restinga
The Brazilian Atlantic rainforest is considered one of the 25 biodiversity hotspots containing high species diversity and high levels of endemism (Fonseca, 1985
; Myers et al., 2000
). The Atlantic coastal vegetation of Brazil comprises several plant communities, such as different forest types (e.g., coastal and montane rainforests and semi-deciduous forests) and open vegetation (Scarano, 2002
). Plant communities at the periphery of the Atlantic rainforest (e.g., swamp forests, restingas, and high altitude fields) are subjected to more extreme environmental conditions than plants in the mesic rainforest, mainly due to drought, salinity, and/or nutrient deficiency (Scarano et al., 2001
). Restingas are sandy coastal plains formed by a mosaic of plant communities interspersed with beach ridges and lagoon systems that stand between the sea and the Atlantic forest mountain chain (Lacerda et al., 1993
; Scarano, 2002
). The most characteristic vegetation type of the restingas is structured by the formation of islands, and succession is triggered by the ability to colonize bare sand by some pioneer herbs and/or woody plants. In northern Rio de Janeiro, southeastern Brazil, one of the most thoroughly studied restingas in the country, the establishment of small vegetation islands is driven by some palms [e.g., Allagoptera arenaria (Gomes) Kuntze], bromeliads [e.g., Aechmea nudicaulis (L.) Griseb.], and some woody plants (e.g., Clusia hilariana Schlecht.) that act as nurse plants favoring the entry of many other plant species (Zaluar and Scarano, 2000
; Scarano, 2002
). Nurse plants provide shelter from high temperatures that make seedling establishment rare for plant species less resistant to harsh conditions (Flores and Jurado, 2003
).
Interestingly, nurse plants in restingas are often terrestrial forms of typical epiphytes of the neighboring Atlantic rainforest (Scarano, 2002
). Moreover, most nurse plants in restingas exhibit crassulacean acid metabolism, such as some bromeliads and particularly some species of the genus Clusia (Reinert et al., 1997
; Lüttge, 1999
; Scarano, 2002
). This mode of photosynthesis acts as a stress-resistant mechanism maximizing water-use efficiency. Hence, it has been hypothesized that stress-resistant plants from the Atlantic rainforest have migrated to the geologically younger lowlands where they nurse seeds and seedlings of other less-resistant plant species. The exchange of plant species between contrasting habitats (e.g., from rocky habitats to the canopy of neighboring forests and from the canopy of rainforests to restingas) may have been a common process in the Brazilian Atlantic rainforest complex, particularly prior to the current levels of fragmentation. This process has been considered one of the factors accounting for the high plant diversity found in these rainforests (Scarano, 2002
). The extent of the exchange is large since approximately 80% of the plant species occurring in restingas of the state of Rio de Janeiro are also found in montane rainforests (Araujo, 2000
). The ability of some plant species to colonize marginal habitats where they nurse other plant species may be of key relevance in a global change scenario, because it suggests high levels of plant morpho-physiological and ecological plasticity to distinct environmental conditions (Scarano, 2002
).
Determining the spatiotemporal variation in dynamics of nurse plants is essential to understanding the ecological functioning of restingas (Zaluar and Scarano, 2000
; Liebig et al., 2001
; Scarano, 2002
). Given that restingas are a mosaic of plant communities and environments, the establishment probabilities of nurse plants, and therefore of the other plants that depend on nurse plants, may greatly differ depending on the effects of spatial and temporal heterogeneity on their performance. The goal of this study was to determine the spatiotemporal variation in demographic attributes (i.e., survival, growth, and reproduction) and population dynamics of the nurse plant Aechmea nudicaulis. To our knowledge, this is one of the first studies examining the detailed demography of a plant species occurring in Brazilian restingas (but see Cirne and Scarano, 2001
). The bromeliad A. nudicaulis is a clonal plant with a vigorous vegetative reproduction that allows a single fragment to potentially occupy a large area, forming large ramet systems (Sampaio et al., 2002
). It has been shown that germination of Clusia hilariana, the main nurse tree in this vegetation (Zaluar and Scarano, 2000
; Liebig et al., 2001
), takes place predominantly inside the tanks formed by this bromeliad (Scarano, 2002
). However, no aspects related to the growth dynamics of A. nudicaulis are known to date, including flowering phenology and age. We selected three different habitats within a restinga in northern Rio de Janeiro state, Brazil, differing in the distance to the coastline and vegetation cover. Within each habitat, we defined four different microhabitats (from bare sand to mature vegetation islands) where A. nudicaulis can be found. We analyzed the spatial (among habitats and microhabitats) and temporal (across years) variation in all demographic traits of this clonal bromeliad in order to build projection matrices to estimate the spatiotemporal variation in dynamics of A. nudicaulis populations.
MATERIALS AND METHODS
Plant species and study site
Aechmea nudicaulis is a terrestrial, clonal-tank bromeliad (60–90 cm height) widely distributed across Central and South America (Smith and Downs, 1979
). Ramets form tanks that harvest water and litter and usually house animals such as frogs and lizards. Like many other species of the Bromeliaceae family, A. nudicaulis exhibits a semelparous life cycle (Sampaio et al., 2002
). Ramets produce 1–2 new ramets per year, and the distance between ramets is approximately 10 cm. Rhizomes are usually buried at 5 cm in the soil and last longer than the ramets. After the death of the ramet, an underground vestige can be found, indicating the previous existence of a ramet (Sampaio et al., 2002
). Flowering mainly takes place between May and October. Inflorescences (60–100 cm height) are upright panicles and include several yellow flowers.
This study was conducted at Restinga de Jurubatiba National Park (22°23' S, 41°45' W; 14 140 ha), RJNP hereafter. Climate at RJNP is tropical and markedly seasonal with mean minimum temperature of 20°C, mean maximum temperature of 30°C, and annual rainfall of 1164 mm concentrated in summer (November to February). The two study years were similar in regard to rainfall (ca. 1000 mm) and climate as a whole (data from the local climate station). A comprehensive description of vegetation types and main plant species occurring in Brazilian restingas can be found in Lacerda et al. (1993)
.
Population sampling
No seedlings were observed for A. nudicaulis during the period of study. In fact, sexual recruitment seems to be a rare event among many restinga plant species (e.g., Cirne and Scarano, 2001
), especially in the case of A. nudicaulis (Sampaio et al., 2002
). Demographic data were totally obtained from A. nudicaulis clonal fragments located in three habitat types and four microhabitats within each habitat monitored from 2001 to 2003. All fragments within each microhabitat in each habitat were considered as a population. Habitats were beach ridges that differed in their distance from the sea (near, intermediate, and far) and in vegetation cover (low, intermediate, and high). The three habitats were located at approximately 100, 500, and 600 m from the coastline and had low (20%), intermediate (29%), and high (42%) vegetation cover, respectively. Vegetation cover was estimated by Pimentel (2002)
by using line intercepts and measuring the proportional length of the projected vegetation cover vs. open areas upon each of them. The distance between habitats ranged about 1000–2000 m. Four microhabitats were identified within each habitat where A. nudicaulis occurred: (1) bare sand, (2) associated with the palm Allagoptera arenaria, (3) neighboring zones at the border of vegetation islands, and (4) shaded zones inside vegetation islands. A total of two plots (100 x 100 m) were laid down within each habitat including all four microhabitats. A total of 96 A. nudicaulis fragments occurring in the four microhabitats were monitored within each habitat (overall 288 fragments; see Table 1). The number of live ramets per fragment remained fairly constant between habitats, microhabitats, and years, and a total of 1149, 1116, and 1173 ramets were tagged and monitored in 2001, 2002, and 2003, respectively. Censuses were conducted in April/May. Given that we could not determine whether different fragments also corresponded to different genets, we used individual ramets as demographic units. Ramet demography of A. nudicaulis was therefore determined by pooling all ramets from all fragments of each microhabitat within each habitat, and monitoring their demographic attributes over time. At each census, height (the variable that best represents the growth of A. nudicaulis; Sampaio et al., 2002
), status (vegetative/reproductive), and fate (dead/alive) of each ramet were recorded. New ramets produced by extant ramets were also tagged and measured. Ramets always remained connected to their mother plants. These basic data were used to build population projection matrices for 2 yr (from 2001–2002 to 2002–2003) for each microhabitat within each habitat. No seedlings were observed during the two years of study. Although we did not conduct germination experiments in the field, Pinheiro and Borghetti (2003)
carried out laboratory experiments for seed germination and verified that the germination rate of A. nudicaulis seeds is very high (>90%), but seeds are very sensitive to heat and drought. Pinheiro and Borghetti (2003)
showed that seeds did not germinate after being exposed to 50°C, and 4 d of imbibition were necessary to germinate. Given that in restingas the temperature at the sand surface in summer can reach up to 70°C at the peak of the radiation (Scarano, 2002
), A. nudicaulis seed germination in natural conditions may only take place in very specific and infrequent conditions and seed banks are not formed.
|
). The projection matrix model used had the form: n(t + 1) = An(t), where n(t) and n(t + 1) are vectors whose elements, ai, are the number of individuals that belong to the ith category at time t and t + 1, respectively. A is the population projection matrix, whose elements, aij, represent the transitions or contributions from individuals in the jth category to the ith category over one time step. The dominant eigenvalue of the matrix gives the population growth rate. The associated right and left eigenvectors w and v give the stable-stage distribution and the reproductive values of each stage class, respectively (Caswell, 2001
). The life-cycle graph and its corresponding population projection matrix for A. nudicaulis had three stages: young ramets, vegetative ramets, and reproductive ramets (Fig. 1). Vegetative ramets were easily differentiated from reproductive ramets that produced inflorescences and generally died after flowering. Young ramets were differentiated from vegetative ramets using two biological criteria: (1) young ramet height was set below 36 cm, which was the average height of 1-yr-old ramets, and (2) flowering probabilities of ramets below 36 cm were nil. Vegetative ramets were all classified in one category because once a ramet reaches the vegetative stage its growth rate turns out to be very low so that size differences between years are very small. Using these criteria, all ramets were classified into their corresponding size/stage categories, and their fate (i.e., stasis, change of category or death) was also recorded from one year to another. As a result, we built 24 matrices (2 yr x 3 habitats x 4 microhabitats).
|
2 distributed, with the number of degrees of freedom equal to the difference between models (Caswell, 2001
). The probability of ramet production by flowering ramets was very low and was not analyzed.
The effects of year, habitat, and microhabitat on the fate of young (growth to vegetative or death) and vegetative (survival, growth to flowering or death) ramets were tested with log linear analyses. The analysis started with the null model YHM, F, which indicates that year (Y), habitat (H), and microhabitat (M) had an independent effect on fate (F). New models were created to test the effect of each factor on fate. The difference between each new model and the null model gave the unexplained variance between models that is 2 distributed, with the number of degrees of freedom equal to the difference between models (Caswell, 2001
).
We used Monte Carlo simulations to generate confidence intervals for population growth rate, , according to the method developed by Alvarez-Buylla and Slatkin (1993)
. The method basically assumes that estimated demographic parameters are the sum of the actual value plus an error term that represents the error made in estimating demographic parameters. In order to include the error term in the simulations, the method takes into account the observed variance estimates for ramet production, whereas the sampling variances for transition probabilities were estimated according to a binomial distribution. Variability in ramet production was also sampled from a binomial distribution because A. nudicaulis ramets only produced a range of 0–3 new ramets a year. The method also accounts for the effects of truncating the distribution of matrix errors so that only biologically possible values occur. For each one of the 24 observed matrices, we conducted 100 simulations obtaining 100 new simulated matrices. As a result, 100 new population growth rates were produced from which the standard error (
) was used to calculate approximate 95% confidence intervals (i.e., ± 2).
Elasticity analyses were performed on projection matrices of each site to determine the relative contribution of demographic traits to population growth rate, (Caswell, 2001
). The sensitivity of to a change in the matrix entry aij (sij = 
/aij) measures how changes in individual demographic parameters influence . The elasticity (eij = sij x aij/) of to a change in aij is a measure of proportional sensitivity and quantifies the proportional change in resulting from a proportional change in aij (Caswell, 2001
). The routine to compute confidence intervals as well as population growth rates, sensitivities, and elasticities were all generated with MATLAB (MathWorks, 1997
).
RESULTS
The life cycle of A. nudicaulis was characterized by the fairly high survival rates of young and vegetative ramets (Table 2). Young ramets exhibited high growth rates at the vegetative stage. Flowering rates were low and ramets presented high mortality rates after flowering. Production of new ramets was low for all stages. Elasticity analyses clearly indicated that the young/vegetative ramet loop accumulated almost all the elasticity of the system and that this pattern was highly consistent between habitats, microhabitats, and years (Table 3).
|
|
|
|
|
|
|
, of A. nudicaulis. Most of the values estimated for A. nudicaulis (20 of 24) did not significantly differ from unity, i.e., the estimated confidence intervals encompassed unity (Fig. 5), indicating that A. nudicaulis populations basically remained stable over time. It must be emphasized, however, that estimated confidence intervals for many populations are quite large as a result of the high variability in demographic attributes within A. nudicaulis populations. One of the exceptions where significantly differed from unity was found inside vegetation islands in the sparsely vegetated habitat in 2001–2002 (Fig. 5). The other values significantly different from unity were observed in the highly vegetated habitat (Fig. 5), indicating that in this habitat the dynamics of A. nudicaulis populations were more variable than in the rest of habitats. In highly vegetated habitats, microhabitats in bare sand and inside vegetation islands yielded values with the highest temporal variation whereas those in palms and at the border of vegetation islands exhibited low temporal variation (Fig. 5). Moreover, in highly vegetated habitats, values from different years in bare sand and inside vegetation islands significantly differed from one another, i.e., one did not fall within the confidence interval of the other (Fig. 5). Populations in the habitat with intermediate vegetation cover exhibited the less variable values (Fig. 5).
|
Spatiotemporal variation in demographic attributes
We analyzed the spatiotemporal variation in demographic attributes and population dynamics of the nurse plant Aechmea nudicaulis. Although this study was based on two years of data (three sampling periods), young ramets differed from vegetative ramets of A. nudicaulis in their spatiotemporal pattern of variation. The probability of ramet production from young ramets significantly varied between microhabitats although the pattern of variation between microhabitats also differed between habitats. In contrast, the probability of ramet production from vegetative ramets was totally constant across habitats, microhabitats, and years. Such a difference in the spatiotemporal variation in ramet production between young and vegetative ramets could be attributed to the fact that ramet production tends to be size dependent in clonal plants (Hutchings and Barkham, 1976
; Carlsson and Callaghan, 1990
; Wijesinghe and Whigham, 1997
; Cirne and Scarano, 2001
). Some studies have shown that both the number and size of new ramets increased with ramet size (Wikberg and Svensson, 2003
), whereas others have found no relationship between ramet size and probability of ramet production (Bullock, 1980
; Chazdon, 1992
; Mendoza and Franco, 1998
). In the case of A. nudicaulis, young ramets produce more new ramets than vegetative and flowering ramets although the probability of ramet production is much more variable for young ramets than for vegetative ramets. In fact, A. nudicaulis seems to behave quite differently from other clonal plants that preferentially allocate more resources to developing a storage system and to increasing ramet size during the first stages of the life cycle (De Steven, 1989
; Cheplick, 1995
; Mendoza and Franco, 1998
; Cirne and Scarano, 2001
).
On the other hand, ramet fate was more variable across habitats, microhabitats, and years for vegetative ramets than for young ramets. It must be emphasized that, when variable, the fate of young ramets significantly varied between microhabitats and all interactions including microhabitat and the other factors of study were also significant. Hence, for young ramets, microhabitat was the most important source of variation determining the production and fate of new ramets. In general, however, the spatiotemporal pattern of variation in production and fate of ramets was quite difficult to interpret for both young and vegetative ramets. This result suggests that small-scale population characteristics (e.g., below- and aboveground interactions with other plants) ultimately determined the performance of A. nudicaulis plants. Other studies on other plant species also indicated that small-scale population characteristics could account for a large proportion of unexplained variance in demographic attributes (Quintana-Ascencio and Menges, 2000
; Albert et al., 2001
; Riba et al., 2002
). Further studies focusing on specific microhabitat characteristics are needed to elucidate the factors affecting performance in A. nudicaulis plants. Apart from bare sand, the rest of the microhabitats can be quite heterogeneous, especially the neighboring areas around palms and vegetation islands.
Spatiotemporal variation in population dynamics
Overall, population growth rates, , and their estimated confidence interval around unity clearly indicated the high stability of A. nudicaulis populations across habitats, microhabitats, and years. Given the fact that values were basically determined by ramet fate and ramet production from young and vegetative ramets, the moderate spatiotemporal variation exhibited by these demographic parameters can account for the low variation in . There were, however, two exceptions to this pattern. One was the population in sparsely vegetated habitats that peaked inside vegetation islands in one of the years. An extraordinary peak of ramet production from young ramets accounted for that positive . The other exception was found in highly vegetated habitats where populations growing in bare sand and inside vegetation islands exhibited significant negative and positive values in the same years. Negative and positive values were accounted for by low and high ramet production values, respectively, especially from young ramets. Hence, highly vegetated sites were the habitats that produced more spatiotemporal variation in ramet production that in turn affected the dynamics of A. nudicaulis populations. It has clearly been demonstrated from theoretical and empirical points of view that increasing variation in demographic traits leads to increasing extinction probabilities due to a higher temporal variation in (Tuljapurkar and Orzack, 1980
; Caswell, 2001
). Hence, A. nudicaulis populations in bare sand and inside vegetation islands in highly vegetated habitats might have higher extinction probabilities than populations next to palms and at the border of vegetation islands. The effect of the vegetation driven by interspecific competition for space (bare sand) or resources (inside islands) might account for this result. However, given the long life span of the species and its vigorous clonal reproduction, A. nudicaulis might easily cope with temporal variation in as shown by other clonal and long-lived plant species (Eriksson, 1996
; Picó and Riba, 2002
).
The other demographic trait that characterized the dynamics of A. nudicaulis populations is the low impact of flowering ramets on , given their low ramet production and high mortality rates. Moreover, no seedlings were observed during the three years of study. Hence, the demographic implications of sexual reproduction of A. nudicaulis in restingas is apparently nil over the time period of study. Such a low importance of sexual reproduction has also been observed in other clonal plants (Wikberg and Svensson, 2003). Although sexual reproduction in restingas might be seen as a developmental constraint, flowering A. nudicaulis plants might play an important ecological role on a larger spatial and temporal scale. In particular, given the fact that A. nudicaulis belongs to the group of plants growing as terrestrials in restingas and as preferential epiphytes in the rainforest, flowering of A. nudicaulis in restingas might be seen as a genetic pool that maintains the gene flow between restinga and rainforest populations. However, this is unlikely to occur because the degree of fragmentation is presently very high, as the Brazilian Atlantic rainforest has been reduced to only 7.5% of its original area (Scarano, 2002
), and the distance between the restinga and the montane rainforest is also very large (approximately 70 km).
Plant communities marginal to the Atlantic rainforests are structurally and functionally dependent on a small number of nurse plants (Scarano, 2002
). Our results suggest that the occurrence of one such plant, A. nudicaulis, in different microhabitats within the restinga, where the plant can maintain stable populations, may create a wide array of settlement opportunities for several other plant species less resistant to the rigorous environmental conditions occurring in restingas.
FOOTNOTES
1 The authors thank F. Lloret for helpful comments on an earlier draft of the manuscript. We are very grateful to Tatiana F. Araujo for field assistance. This research was supported by the Programa de Pesquisas Ecológicas de Longa Duração (PELD), the Brazilian Research Council (CNPq), and the Netherlands Organization for International Cooperation in Higher Education (NUFFIC) through different travel grants. 
4 Author for correspondence (e-mail: michelle_sampaio{at}yahoo.com.br
)
LITERATURE CITED
Albert M. J. A. Escudero J. M. Iriondo 2001 Female reproductive success of narrow endemic Erodium paularense in contrasting microhabitats. Ecology 82: 1734-1747[CrossRef][ISI]
Alvarez-Buylla E. R. M. Slatkin 1993 Finding confidence limits on population growth rates: Monte Carlo test of a simple analytic method. Oikos 68: 273-282[CrossRef][ISI]
Araujo D. S. D. 2000 Análise florística e fitogeográfica das restingas do Estado do Rio de Janeiro. D.Sc. thesis, Universidade Federal do Rio de Janeiro, Brazil
Bullock S. H. 1980 Demography of an undergrowth palm in littoral Cameroon. Biotropica 12: 247-255[CrossRef][ISI]
Carlsson G. A. T. V. Callaghan 1990 Programmed tiller differentiation, intraclonal density regulation and nutrient dynamics in Carex bigelowii. Oikos 58: 219-230[CrossRef][ISI]
Caswell H. 2001 Matrix population models: construction, analysis, and interpretation, 3rd ed. Sinauer, Sunderland, Massachusetts, USA
Chazdon R. 1992 Patterns of growth and reproduction of Geonoma congesta, a clustered understory palm. Biotropica 25: 43-51
Cheplick G. P. 1995 Life history trade-offs in Amphibromus scabrivalvis (Poaceae): allocation to clonal growth, storage, and cleistogamous reproduction. American Journal of Botany 82: 621-629[CrossRef][ISI]
Cirne P. F. R. Scarano 2001 Resprouting and growth dynamics after fire of the clonal shrub Andira legalis (Leguminosae) in a sandy coastal plain in southeastern Brazil. Journal of Ecology 89: 351-357[CrossRef]
de Steven D. 1989 Genet and ramet demography of Oenocarpus mapora ssp. mapora, a tropical palm of Panamanian tropical moist forest. Journal of Ecology 77: 579-596[CrossRef]
Eriksson O. 1996 Regional dynamics of plants: a review of evidence for remnant, source-sink and metapopulations. Oikos 77: 248-258[CrossRef][ISI]
Flores J. E. Jurado 2003 Are nurse-protege interactions more common among plants from arid environments?. Journal of Ecology 14: 911-916
Fonseca G. A. B. 1985 The vanishing Brazilian Atlantic Forest. Biological Conservation 34: 17-43[CrossRef][ISI]
Hutchings M. J. P. B. Barkham 1976 An investigation of shoot interactions in Mercurialis perennis L., a rhizomatous perennial herb. Journal of Ecology 64: 723-743[CrossRef][ISI]
Lacerda L. D. D. S. D. Araujo N. C. Maciel 1993 Dry coastal ecosystems of the tropical Brazilian coast. In E. van der Maarel [ed.], Dry coastal ecosystems: Africa, America, Asia and Oceania, 477–493. Elsevier, Amsterdam, Netherlands
Liebig M. F. R. Scarano E. A. Mattos H. L. T. Zaluar U. Lüttge 2001 Ecophysiological and floristic implications of sex expression in the dioecious neotropical CAM tree Clusia hilariana Schltdl. Trees Structure and Function 15: 278-288[ISI]
Lüttge U. 1999 One morphotype, three physiotypes: sympatric species of Clusia with obligate C3 photosynthesis, obligate CAM and C3-CAM intermediate behaviour. Plant Biology 1: 138-148[CrossRef][ISI]
MathWorks. 1997 MATLAB: the language for technical computing. Prentice Hall, Upper Saddle River, New Jersey, USA
Mendoza A. M. Franco 1998 Sexual reproduction and clonal growth in Reinhardtia gracilis (Palmae), an understory tropical palm. American Journal of Botany 85: 521-527[Abstract]
Myers N. R. A. Mittermeier C. G. Mittermeier G. A. B. Fonseca J. Kent 2000 Biodiversity hotspots for conservation priorities. Nature 403: 853-858[CrossRef][Medline]
Picó F. X. M. Riba 2002 Regional-scale demography of Ramonda myconi: remnant population dynamics in a preglacial relict species. Plant Ecology 161: 1-13[CrossRef][ISI]
Pimentel M. 2002 Variação espacial na estrutura de comunidades vegetais da formação aberta de Clusia no Parque Nacional da Restinga de Jurubatiba, RJ. D.Sc. thesis, Universidade Federal do Rio de Janeiro, Brazil
Pinheiro F. F. Borghetti 2003 Light and temperature requirements for germination of seeds of Aechmea nudicaulis (L.) Griesebach and Streptocalyx floribundus (Martius ex Schultes F.) Mez (Bromeliaceae). Acta Botanica Brasilica 17: 27-35
Quintana-Ascencio P. F. E. S. Menges 2000 Competitive abilities of three narrowly endemic plant species in experimental neighborhoods along a fire gradient. American Journal of Botany 87: 690-699
Reinert F. A. Roberts J. M. Wilson L. Ribas G. Cardinot H. Griffiths 1997 Gradation in nutrient composition and photosynthetic pathways across the restinga vegetation of Brazil. Botanica Acta 110: 135-142[ISI]
Riba M. F. X. Picó M. Mayol 2002 Effects of regional climate and small-scale habitat quality on performance in the relict species Ramonda myconi. Journal of Vegetation Science 13: 259-268[CrossRef][ISI]
Sampaio M. C. L. E. Perissé G. Aguiar de Oliveira R. Iglesias Rios 2002 The contrasting clonal architecture of two bromeliads from sandy coastal plains in Brazil. Flora 197: 443-451[ISI]
Scarano F. R. 2002 Structure, function and floristic relationship of plant communities in stressful habitats marginal to the Brazilian Atlantic rainforest. Annals of Botany 90: 517-524
Scarano F. R. H. M. Duarte K. T. Ribeiro P. J. F. P. Rodrigues E. M. B. Barcellos A. C. Franco J. Brulfert E. Deléens U. Lüttge 2001 Four sites with contrasting environmental stress in southeastern Brazil: relations of species, life form diversity, and geographical distribution to ecophysiological parameters. Botanical Journal of the Linnean Society 136: 345-364[CrossRef]
Smith L. B. R. J. Downs 1979 Bromelioideae (Bromeliaceae). Flora Neotropica Monograph 14: 1493-2142
Tuljapurkar S. D. S. H. Orzack 1980 Population dynamics in variable environments. I. Long-run growth rates and extinction. Theoretical Population Biology 18: 314-342[CrossRef][ISI]
Wijesinghe D. K. D. F. Whigham 1997 Costs of producing clonal offspring and the effects of plant size on population dynamics of the woodland herb Uvularia perfoliata (Liliaceae). Journal of Ecology 85: 907-919[CrossRef]
Wikberg S. B. M. Svensson 2003 Ramet demography in a ring-forming clonal sedge. Journal of Ecology 91: 847-854[CrossRef]
Zaluar H. L. T. F. R. Scarano 2000 Facilitação em restingas de moitas: um século de buscas por espécies focais. In F. A. Esteves and L. D. Lacerda [eds.], Ecologia de Restingas e Lagoas Costeiras, 3–23. NUPEM, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil
This article has been cited by other articles:
|
|
 |
|
  F. F. COELHO, C. CAPELO, A. C. O. NEVES, R. P. MARTINS, and J. E. C. FIGUEIRA Seasonal Timing of Pseudoviviparous Reproduction of Leiothrix (Eriocaulaceae) Rupestrian Species in South-eastern Brazil Ann. Bot., December 1, 2006; 98(6): 1189 - 1195. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
