Printer Friendly


JOHAN EHRL[acute{E}]N [1,3]


Abstract. The distribution of species depends on the availability of suitable habitats, the capacity to disperse to these habitats, and the capacity of populations to persist after establishment. Dispersal limitation implies that not all suitable habitat patches will be occupied by a species. However, the extent to which dispersal limits local distribution is poorly known. In this study, we transplanted seeds, bulbils, and juvenile plants to examine patterns of dispersal limitation and patch occupancy in seven temperate-forest herbs. Recruitment was recorded during four years in 48 patches. The investigated species varied considerably in their natural abundance in the patches. Patterns of seedling emergence and establishment among patches were not related to any of nine investigated abiotic factors. In contrast, the availability of seeds or bulbils was found to limit recruitment in six of the investigated species. Establishment was also successful in many patches where the species did not occur naturally. Es timated patch occupancy in the investigated species ranged from 17.2% to 94.6%. Seed size was positively correlated with the probability of successful establishment of seeds and negatively correlated with patch occupancy. The results suggest that dispersal limitation is an important structuring factor in temperate-forest herb communities. The distribution of species can be perceived as the result of processes operating both among and within patches. Seed size is a key trait in these processes.

Key words: colonization; deciduous forest; dispersal; establishment experiments; metapopulation dynamics; parch occupancy; perennial herbs; recruitment limitation; seed size; seed sowing; soil nutrients.


The distribution of species can be perceived at many different spatial scales, from local populations to their geographical range. Typically, locally abundant species also have wide regional and geographical distributions (Gaston 1996, Gaston and Curnutt 1998). Brown (1984, 1995) suggested that both local abundance and regional distribution of species reflect their niche requirements; the more specialized a species is, in terms of its niche, the less abundant it is and the more restricted is its distribution range. Hanski et al. (1993) argued that metapopulation dynamics, i.e., variation in colonization and extinction rates, may be the cause behind the positive correlation between abundance and distribution. If both local and regional distributions of plants basically reflect habitat requirements (e.g., Harper 1977, Crawley 1990), it follows that recruitment would be limited by availability of suitable microsites rather than by seeds. It has also been argued that patterns of species frequency and local abund ance in terrestrial vascular plants meet the predictions of niche-based models; but not those of metapopulation models (Scheiner and Rey-Benayns 1997). However, recent studies based on seed-sowing experiments (e.g., Eriksson and Ehrl[acute{e}]n 1992, van der Meijden et al. 1992, Crawley and Brown 1995, Ackerman et al. 1996), seed predator exclusions (Louda 1982, Louda and Potvin 1995), and analyses of seed rain vs. recruitment (Clark et al. 1998) have demonstrated that recruitment rate in many plant populations is enhanced by increasing seed availability.

Distributions over larger spatial scales have also been found to be limited by seed dispersal; i.e., seed sowing at unoccupied sites results in recruitment (e.g., Primack and Miao 1992, Losos 1995, Burke and Grime 1996, Ehrl[acute{e}]n and Eriksson 1996, Tilman 1997, Eriksson 1998). If seed addition results in recruitment at sites that are not occupied, a corollary is that larger regions may include unoccupied, but suitable, sites. Metapopulation theory predicts that unoccupied patches exist (Levins 1969, Hanski 1997). Because local colonization and extinction are ongoing dynamics, there should be an equilibrium fraction of suitable patches left unoccupied. Thus, patch occupancy patterns have been used to infer aspects of the underlying colonization and extinction dynamics (Hanski 1994a, b).

One problem is identifying what constitutes a suitable patch. Patch suitability is often defined, subjectively or with the help of statistical methods, on the basis of habitat characteristics (Hanski and Simberloff 1997). A more accurate method is to employ experimental introductions to assess patch suitability. One factor hampering the identification of unoccupied but suitable patches after seed-sowing experiments is that the subsequent survivorship of emerged seedlings may not have been recorded over a sufficiently long period of time. There might be population bottlenecks occurring at later stages of recruitment (Losos 1995). Examination of survivorship of experimentally introduced plants is therefore essential when assessing factors that limit regional distributions. By combining information on natural patch occupancy and abiotic conditions with emergence and survivorship of recruits after experimental seed sowing, we may more reliably detect dispersal limitation and assess patch occupancy.

Many studies have shown that seed size is a key factor for colonization in closed plant communities. Westoby et al. (1996) reviewed a number of experiments that demonstrated that larger seed mass does convey benefits in seedling establishment under a wide variety of circumstances. Fr[dott{o}]borg and Eriksson (1997) found that seed size was positively related to local colonization rates, defined as dispersal and establishment, and negatively related to local extinction rates in deciduous forest herbs and grasses. However, this advantage may be counteracted by a trade-off between seed size and seed number, making smaller seeded species, due to their large seed production, more efficient dispersers (Clark et al. 1998, Eriksson and Jakobsson 1998). Large-seeded species are often restricted to late-successional habitats (Salisbury 1942, 1974, Baker 1972, Foster and Janson 1985, Rydin and Borgeg[dot{a}]rd 1991). One explanation may be that large-seeded species are successful at all successional stages once seeds arrive, but that colonization is delayed by ineffective seed dispersal (Harper et al. 1970, Foster et al. 1986, Wood and Moral 1987).

In this study, we used a seed-sowing approach to examine occupancy patterns in seven perennial forest herbs. We addressed the following specific questions: (1) Are local and regional distributions limited by dispersal? (2) Can suitable patches be identified by habitat characteristics, such as vegetation and soil chemistry? (3) How is the emergence rate of seedlings related to their subsequent survivorship? (4) What is the patch occupancy of the individual species? (5) Is interspecific variation in establishment and patch occupancy related to seed size?


The study was performed at Tullgarn in southeastern Sweden, [sim]45 km south-southwest of Stockholm, in an area covering 5 km2 of forests and agricultural land. Seven perennial forest herbs, typical of deciduous and coniferous forests on rich soils, were chosen for the study (Table 1). Within the area, 48 patches were identified based on a subjective criterion of being potentially suitable to host populations of the study species, i.e., forest patches where the field layer was dominated by grasses and herbs. We excluded coniferous forests with a field layer dominated by lichens, mosses, and shrubs. Twenty-six patches were chosen from mixed-deciduous forest and the remaining 22 from coniferous forest. The experimental patches occurred as more or less isolated fragments within a matrix of unsuitable habitat types (open agricultural land or coniferous forests on poor soils). In July 1992, a 2 X 2 m quadrat was placed randomly in each of these 48 patches. Each quadrat was divided into 16 adjacent 0.5 X 0.5 m plo ts.

Seeds of six species and bulbils of one species were sown into the 0.5 X 0.5 m plots (Table 1). Bulbils refer to the bulblets born at leaf axils in Dentaria bulbifera. The sowing took place in July and August 1992 for six of the seven species, and in August 1993 for one species (Lathyrus). A given plot was planted with a single species. For Convallaria, Denraria, Lathyrus, and Polygonarum, a single plot was planted at each site. Although this protocol enabled studies of among-patch variation, we expected that environmental heterogeneity also occurs within patches. To examine variation within patches at the scale of decimeters to meters, seeds of Paris and Campanula were sown into two plots each and Actaea was sown into three plots per site. The choice of species for these different treatments was guided by the availability of seeds after collection.

In order to further examine the possibility that factors acting on subsequent stages of the life cycle limit recruitment, we transplanted 24 juveniles into one plot per site for one of the species, Convallaria. Thus, 12 of 16 plots per site were used for sowings or transplantations. The four remaining plots were left as controls to record natural recruitment. Experimental plots were arranged so that multiple replicates for one species were not adjacent to each other. Identical sowing protocols were used at all 48 sites. Altogether 79440 seeds, 7920 bulbils, and 1152 juveniles were sown or transplanted into the experimental plots.

Seeds and bulbils were collected at maturation just before natural dispersal and were sown into plots as soon as possible after collection. All seeds were checked for seed predation and other types of damage, and such damaged seeds were removed before sowing. Extensive seed damage was observed only in Lathyrus, where 55% of the seeds were so affected. Seeds were evenly dispersed within plots. A 10-cm margin along the borders was left without treatment. The soil was not disturbed, but the vegetation in each plot was lightly shaken to ensure that seeds settled to the soil surface.

Convallaria juveniles were raised in the greenhouse from seeds collected and sown in 1990. Seeds germinated in May 1992 and by the time of transplantation in July 1992, they had reached a size of [sim]10 cm, corresponding to the typical aboveground size of 3-yr-old Convallaria seedlings in the field.

An area within a 50 m radius around each experimental site was surveyed for the presence of species included in the experiment. Soil cores 2.5 cm in diameter and 20 cm deep were collected from each site in June 1993. The soil samples were analyzed for pH and concentrations of nitrate, ammonium, phosphate, magnesium, calcium, and potassium. Analyses were performed with standard methods at the Chemical Laboratory of the Swedish University of Agricultural Sciences. Nitrate, ammonium, and [H.sup.+] were measured after extraction with KC1, and other analyses were done after Al extraction. Additional soil samples were taken in May and June 1993 to determine the soil moisture content. Samples were taken after five days without precipitation and moisture content was estimated as (soil fresh mass - soil dry mass)/soil fresh mass.

From 1993 and onward, the experimental and control plots were visited two times annually, in May and in June. At each visit the entire plot area was thoroughly examined, the position of each emerged individual was recorded, and the presence or absence of previously emerged individuals was noted. In the last experimental year (1996), survival could not be recorded at one site because the markings had been lost.


For each species, the patches were assigned to one of the three following categories: occupied, unoccupied suitable, and unoccupied unsuitable. "Occupied" means that the species was present within 50 m of the experimental site. "Unoccupied suitable" was defined as patches in which a species was not found naturally but recruited successfully. All plots in which seeds or bulbils germinated and some seedlings survived to their third season were considered as representing such a successful recruitment. Hence, the proportion of unoccupied patches that were potentially suitable was calculated on the basis of successful establishment at unoccupied sites. Patches where a species was not present initially and where recruitment did not take place were denoted "unoccupied unsuitable." Establishment of populations may fail for other reasons than patch unsuitability, for example demographic and environmental stochasticity. Therefore, this measure of patch suitability, based on recruitment success, is a conservative estim ate.

From these data, we calculated a measure of species "patch occupancy," defined as the ratio of occupied patches to the sum of occupied and unoccupied but suitable patches.

In the analyses of emergence of seedlings and emergence of shoots from transplanted bulbils and juvenile plants, we included only plots where the respective species was able to survive to the third year.

In cross-species comparisons, only one of the replicates (randomly chosen) for Actaea, Campanula, and Paris, respectively, was used in the analyses. The among-replicate variation within sites was assessed in separate analyses. One species, Campanula, completely failed to emerge and establish, and was therefore omitted from analyses of patch occupancy and demographic properties.

Statistical analyses were performed with the SYSTAT 6.0 statistical package (SYSTAT 1996). Frequencies of emergence and establishment between plots at sites with and without the natural presence of established individuals were analyzed using chi-square tests after Yate's correction.

The influence of nine abiotic factors on natural abundance, seedling emergence, and establishment (seedling survival up to the third year), respectively, was examined for each species using logistic regression models. In these analyses, natural abundance was categorized as 0 or 1, depending on whether populations were present in the patches naturally. Emergence was categorized as 0 or 1, depending on whether any seedlings or juveniles emerged in plots after experimental transplantations. Lastly, establishment was categorized as 0 or 1, depending on whether any such emerging individuals survived within the plot up to the third season. All nine predictors and a constant were considered for the initial models. Predictors not significant at [alpha] = 0.15 were removed from the model in a stepwise procedure. Estimates of soil nutrient contents (in milligrams per 100 g dry soil) and soil moisture (percentage of soil fresh mass) were log-transformed before analysis. Convallaria seeds germinated in all but one plot, Convallaria juveniles emerged and established in all but two plots, and Campanula failed to germinate in all plots. Therefore, these responses were not examined by the regression models.

Differences in emergence and survival rates between plots at sites with and without natural presence of established individuals were examined by t tests of un-weighted plot averages. Before analysis, the number of emerging individuals was log-transformed and the proportion of individuals surviving was arcsine square-root transformed.


Recruitment among sites

The percentage of sites where a species was present naturally varied among species. Dentaria, Polygonatum, and Campanula had the lowest presence (10.4%, 10.4%, and 16.7%, respectively), whereas Actaea and Convallaria had the highest (72.9% and 64.6%; Table 2). Paris and Lathyrus had intermediate presence (43.8% and 37.5%).

For six of the seven species, recruitment was limited by the availability of seeds or bulbils. No seedlings were found in plots without sowing. For all species except Campanula, recruits emerged in most of the experimentally sowed plots. The first seedlings of Actaea, Convallaria, Paris, and Polygonatum emerged in the second year after sowing (1994), whereas Lathyrus, which was sown one year later, appeared in the first year after it was sown (1994). Transplanted bulbils of Dentaria and juvenile plants of Convallaria were first recorded in 1993, the year after transplantation. No seedlings of Campanula were recorded during the entire study. In the remaining species, the fraction of plots with seedling emergence ranged from 72.9% of plots for Lathyrus to 97.9% for Convallaria (Table 2). For bulbils of Dentaria, emergence occurred in 81.2% of the plots.

Survival of Convallaria individuals to the third season occurred in nearly all plots with transplanted seeds and juveniles (87.2% and 95.8%, respectively). Establishment was much less frequent for the other species (Table 2). Surviving Dentaria and Polygonatum recruits were found in [sim]60% of the plots, and for Actaea, Paris, and Lathyrus, [less than] 30% of the plots contained recruits surviving to their third year.

For only one species, Paris, establishment was significantly larger in patches where it was present naturally (Table 2). However, although not significantly different, establishment of Actaea, Dentaria, and Polygonatum was two to three times higher in patches where they were present naturally than in unoccupied patches. In contrast, Lathyrus showed the opposite trend. Convallaria established extremely well from seeds and juveniles in both occupied and unoccupied patches.

Establishment following experimental seed sowing suggested that more than half of the identified unoccupied patches were suitable for three of the investigated species (Convallaria, Dentaria, and Polygonatum; Fig. 1). For Actaea and Paris, only a small percentage of the unoccupied patches were suitable and all patches appeared to be unsuitable for Campanula recruitment. Lathyrus was intermediate in this respect.

The results of the multiple sowing plots suggested that establishment success was often different between replicates within patches. Both establishment and failure within the same site occurred in 34.0% of the Actaea plots and in 17.0% of the Paris sites.

Abiotic factors influencing recruitment

Multiple logistic regression was used to determine the plot-wise relationships among nine abiotic factors and natural abundance, seedling emergence, and survival up to the third year. None of these relationships was significant after Bonferroni correction (lowest P value for predictors: emergence, P = 0.017; establishment, P = 0.011). Hence, the results suggest that the investigated abiotic factors are not important in explaining differences in emergence and establishment of species among plots.

Emergence and survival within plots

In plots where at least one individual survived up to the third year, the emergence rates for Actaea, Convallaria, Paris, and Polygonatum were relatively high and similar among species, ranging from 24.3% to 33.8% of the seeds sown. Only Lathyrus had a considerably lower emergence rate (10.3%).

Survival of seedlings to the second year was 40.8-77.2% for all species except Paris, which had a survival that year of only 15.1%. Between the second and the third year, more than half of the individuals (51.4-80.7%) survived in all species. If this trend of increased survival with age continues, plants are likely to remain in a large percentage of the plots where they had survived up to the third year.

The suggested pattern of increased survival with increased age is further supported by results of the transplantations of Convallaria juveniles. Their survival was 73.5% during the first year, but increased to 84.6-84.8% during the two subsequent seasons. These numbers were higher than emergence (26.7%) and survival (67.7-74.5%) of Convallaria seedlings. Under the assumption that transplanted juveniles have survival probabilities that are similar to those of seedlings in their third year, as suggested by the similarity in size, then the fate of juveniles during the three experimental years approximately resembles that of seedlings from their third to sixth year. In no case was there a significantly higher emergence or a higher survival in patches where natural populations were present than in with patches without natural populations.

The correlation between the number of seedlings emerging in a plot and the survival of these seedlings to the third year was significantly negative only in Paris. However, the sign of this relationship was negative for all six species, which is unlikely to occur by chance (P = 0.016). A negative relationship between seedling emergence and subsequent survival was also present among species (Spearman r = -0.943, N 6, P = 0.02). The species that had the lowest emergence rate, Lathyrus and Polygonatum, had the highest survival up to the second and the third year.

Patch occupancy

Patch occupancy in species ranged from 17.2% to 94.6% (Fig. 1). The highest patch occupancies were found in Actaea (94.6%) and Paris (91.3%). Two species, Polygonatum (17.2%) and Dentaria (17.2%), occupied less than one-fifth of the suitable patches. Lathyrus (62.1%) and Convallaria (67.4%) were intermediate in this respect. Patch occupancy for Campanula was not interpretable, as it had zero germination in all patches.

Seed size, establishment, and patch occupancy

The probability of establishment from seeds, estimated by the percentage of the sown seeds that emerged and survived until their third season, averaged over all plots, was positively correlated with seed size (Fig. 2). The larger the seeds, the larger was the probability of successful colonization, given that a patch was reached by a seed. Dentaria, which has bulbils that are larger than the largest seeds in the study, also fits into this picture; 8.8% of sown bulbils survived until the third season (Table 2).

The estimated patch occupancy was negatively correlated with seed size (Fig. 3), suggesting that the distribution of large-seeded species was more limited by dispersal than the distribution of species with smaller seeds.


Experimental sowing resulted in successful recruitment in six of the seven investigated forest herbs. Recruitment occurred in both occupied and unoccupied patches. As survival of seedlings increases with their age, experimental establishment most likely represents true recruitment. Because no recruitment was observed in control plots, these results suggest that both local population size and regional distribution are highly limited by seed availability (in the following text, we use the term "seeds" to also include the bulbils of Dentaria). The basic reason for seed limitation within occupied patches is that the local seed source does not saturate available microsites with seeds, e.g., due to a low seed production, or (as in Lathyrus) due to seed predation (Ehrl[acute{e}]n 1996, 1997). On an among-patch scale, the available seed pool is not efficient in reaching potential patches for recruitment, i.e., the species are dispersal limited.

The seven investigated species represent a natural patch abundance spectrum of forest herbs, from the comparatively rare Dentaria, Polygonatum, and Campanula, to the more common Actaea and Convallaria. Paris and Lathyrus are intermediate in this respect. One of the species, Convallaria, was remarkable in the sense that it managed to recruit at almost all patches with and without natural populations. In contrast, Campanula did not recruit anywhere. For the remaining species, except Lathyrus, establishment was higher in occupied than in unoccupied patches, although a significant difference was demonstrated only for Paris.

It is important to remember that recruitment was not successful in all patches with natural populations. This suggests that establishment of populations may fail for reasons other than patch unsuitability. Therefore, a percentage of unoccupied patches that were classified as unsuitable in this study may actually be suitable. It is possible to correct for this underestimation and calculate the percentage of unoccupied patches that were suitable by assuming a similar probability of recruitment success at suitable patches, unoccupied as well as occupied. For five of the species, such corrected estimates were slightly higher (results not shown) but qualitatively identical (rank order among species) to the uncorrected ones. For Lathyrus, the establishment pattern suggests that unoccupied patches were actually more suitable than occupied patches.

Considering only patches where at least one seedling survived to the third year, no species exhibited a higher emergence and survival in occupied than in unoccupied suitable patches. This indicates that among suitable patches, the pattern of natural abundance does not reflect any difference in patch quality for recruitment. This conclusion gains further strength from the soil analysis. Overall, there were not more than possibly weak relationships between abiotic soil factors and natural abundance or recruitment over the range of investigated habitats. Although this study did not enable us to identify the environmental factors responsible for variation in patch suitability, the difference in establishment success between replicates for Paris and Actaea shows that such variation also occurs within patches. This suggests that the decisive factors determining establishment also vary at a small scale.

Previous studies of patch occupancy in plants are rare and have been based on descriptive data (e.g., Ouborg 1993, Eriksson 1996, Husband and Barrett 1996, Quintana-Ascencio and Menges 1996). Purely descriptive approaches to assessing patch occupancy are associated with weaknesses because they only use indirect measures of patch suitability, such as habitat characteristics. An experimental approach is needed to estimate patch suitability directly. If the present study had been purely descriptive, we probably would have considered all 48 patches as suitable for all of the species, as visual inspection and the investigated abiotic factors did not discriminate between suitable and unsuitable patches. Moreover, recruitment studies performed during too short a time period may overestimate patch suitability because the bottleneck of recruitment occurs after seedling emergence. The present study also illustrates this problem; overall, 43% of the patches where emerging seedlings were found did not harbor any survivor s in the third season. Our study does not completely eliminate this problem, as plants did not reproduce. However, the increasing survival rates indicate that it was considerably reduced. Hence, our estimates of patch occupancy are likely to be more reliable than those based on descriptive or short-term studies.

In animals, assessments of patch suitability are mostly based on descriptive information (Hanski and Simberloff 1997). However, experimental testing by releasing individuals in unoccupied patches has sometimes been carried out (e.g., Crowell 1973, Schoener and Schoener 1983, Harrison 1989). Harrison (1989) transplanted larvae of the butterfly Euphydryas editha bayensis to vacant habitats and recorded their performance during two seasons. Butterflies were able to survive and produce adults one year later in 25% of the cases, and Harrison concluded that the present distribution of the butterfly was limited by an inability to disperse to distant patches.

Patch occupancy varies considerably between species in our study. Two species occupy less than one-fifth of the available suitable patches, whereas two species exhibit an almost total patch occupancy. The variation in seed size among species provides a possible explanation for the differences in patch occupancy. Establishment success is positively related to seed size among the investigated species. The fact that species with larger seeds have a lower patch occupancy indicates that they are less able than small-seeded species to disperse to available suitable patches. A corollary is that the regional distribution of small- vs. large-seeded forest herbs is limited by different factors.

Seed size is negatively correlated with patch occupancy among the investigated deciduous-forest herb species. Other experimental tests of the mechanisms underlying the relationship between propagule size and habitat distribution indicate that seed size may affect establishment in many, but not all, habitats. Rabinowitz (1978) demonstrated that mangrove species with larger progagules survived better in habitats dominated by species with smaller propagules, and concluded that differential dispersal contributed heavily to mangrove zonation. In early primary successions of subalpine habitats, larger seeded species were capable of establishing in barren substrates but were incapable of reaching them, whereas smaller seeded species were capable of reaching open sites but were incapable of growing there (Wood and del Moral 1987). Moreover, Burke and Grime (1996) showed that limestone grassland species with large seeds established evenly across a productivity--disturbance matrix, whereas smaller seeded species were more dependent on disturbance for establishment. In contrast, there was no evidence that interspecific variation in seed size influenced the probability of establishment among North American grassland species (Tilman 1997).

In conclusion, the results suggest that dispersal efficiency is an important structuring factor in temperate-forest herb communities. The investigated species are limited within local populations by seed availability, and on a larger scale by seed dispersal to suitable, but unoccupied, patches. The distribution of species, as well as species richness and composition of plant communities, therefore cannot be explained without accounting for dispersal processes occurring at both local and regional scales.


We are grateful to A. Jakobsson, K. Lehtil[ddot{a}], P. Redbo-Torstensson, and H. Rydin for helpful comments on the manuscript. This study was supported financially from the Swedish Natural Science Research Council (to J. Ehrl[acute{e}]n and O. Eriksson) and the Swedish Council for Forestry and Agricultural Research (to O. Eriksson).

(1.) Department of Ecological Botany, Uppsala University, Villa[dot{v}][ddot{a}]gen 14, S-752 36 Uppsala, Sweden

(2.) Department of Botany, Stockholm University, S-106 91 Stockholm, Sweden

(3.) Present address: Department of Botany, Stockholm University, S-106 91, Stockholm, Sweden.



Ackerman, J. D., A. Sabat, and J. K. Zimmerman. 1996. Seedling establishment in an epiphytic orchid: an experimental study of seed limitation. Oecologia 106:192-198.

Baker, H. G. 1972. Seed weight in relation to environmental conditions in California. Ecology 53:997-1010.

Brown, J. H. 1984. On the relationship between abundance and distribution of species. American Naturalist 124:253-279.

Brown, J. H. 1995. Macroecology. University of Chicago Press, Chicago, Illinois, USA.

Burke, M. J. W., and J. P. Grime. 1996. An experimental study of plant community invasibility. Ecology 77:776-790.

Clark, J. S., E. Macklin, and L. Wood. 1998. Stages and spatial scales of recruitment limitation in southern Appalachian forests. Ecological Monographs 68:213-235.

Crawley, M. J. 1990. The population dynamics of plants. Philosophical Transactions of the Royal Society of London 330:125-140.

Crawley, M. J., and S. L. Brown. 1995. Seed limitation and the dynamics of feral oilseed rape on the M25 motorway. Proceedings of the Royal Society of London B, Biological Sciences 259:83-88.

Crowell, K. L. 1973. Experimental zoogeography: Introduction of mice to small islands. American Naturalist 107:535-558.

Ehrl[acute{e}]n, 3. 1996. Spatiotemporal variation in predispersal seed predation intensity. Oecologia 108:708-713.

Ehrl[acute{e}]n, J. 1997. Risk of grazing and flower number in a perennial plant. Oikos 80:428-434.

Ehrl[acute{e}]n, J., and O. Eriksson. 1996. Seedling recruitment in the perennial herb Lathyrus vernus. Flora 191:377-383.

Eriksson, [dot{A}]. 1998. Regional distribution of Thymus serpyllum; management history and dispersal limitation. Ecography 21:35-43.

Eriksson, O. 1996. Regional dynamics of plants: a review of evidence for remnant, source-sink and metapopulations. Oikos 77:248-258.

Eriksson, O., and J. Ehrl[acute{e}]n 1992. Seed and microsite limitation of recruitment in plant populations. Oecologia 91:360-364.

Eriksson, O., and A. Jakobsson. 1998. Abundance, distribution and life histories of grassland plants: a comparative study of 81 species. Journal of Ecology 86:922-933.

Foster, R. B., B. J. Arce, and T. S. Wachter. 1986. Dispersal and the sequential plant communities in Amazonian Peru floodplain. Pages 363-370 in A. Estrada and T. H. Fleming, editors. Frugivores and seed dispersal. Dr. W. Junk, Dordrecht, The Netherlands.

Foster, S. A., and C. H. Janson. 1985. The relationship between seed size and establishment conditions in tropical woody plants. Ecology 66:773-780.

Fr[ddot{o}]borg, H., and O. Eriksson. 1997. Local colonization and extinction of field layer plants in a deciduous forest and their dependence upon life history features. Journal of Vegetation Science 8:395-400.

Gaston, K. J. 1996. The multiple forms of the interspecific abundance-distribution relationship. Oikos 75:211-220.

Gaston, K. J., and J. L. Curnutt. 1998. The dynamics of abundance-range size relationship. Oikos 81:38-44.

Hanski, I. 1994a. A practical model of metapopulation dynamics. Journal of Animal Ecology 63:151-162.

Hanski, I. 1994b. Patch-occupancy dynamics in fragmented landscapes. Trends in Ecology and Evolution 9:131-135.

Hanski, I. 1997. Metapopulation dynamics: from concepts and observations to predictive models. Pages 69-91 in I. Hanski and M. Gilpin, editors. Metapopulation biology. Academic Press, San Diego, California, USA.

Hanski, I., J. Kouki, and A. Halkka. 1993. Three explanations of the positive relationship between distribution and abundance of species. Pages 108-116 in R. E. Ricklefs and D. Schluter, editors. Species diversity in ecological communities: historical and geographical perspectives. University of Chicago Press, Chicago, Illinois, USA.

Hanski, I., and D. Simberloff. 1997. The metapopulation approach, its history, conceptual domain, and application to conservation. Pages 5-26 in I. Hanski and M. Gilpin, editors. Metapopulation biology. Academic Press, San Diego, California, USA.

Harper, J. L. 1977. The population biology of plants. Academic Press, London, UK.

Harper, J. L., P. H. Lovell, and K. G. Moore. 1970. The shapes and sizes of seeds. Annual Review of Ecology and Systematics 1:327-356.

Harrison, S. 1989. Long-distance dispersal and colonization in the Bay checkerspot butterfly. Ecology 70:1236-1243.

Husband, B. C., and S. C. H. Barrett. 1996. A metapopulation perspective in plant population biology. Journal of Ecology 84:461-469.

Levins, R. 1969. Some demographic and genetic consequences of environmental heterogeneity for biological control. Bulletin of the Entomological Society of America 15:237-240.

Losos, E. 1995. Habitat specificity of two palm species: experimental transplantation in Amazonian successional forests. Ecology 76:2595-2606.

Louda, S. M. 1982. Limitation of the recruitment of the shrub Haplopappus squarrosus (Asteraceae) by flower- and seedfeeding insects. Journal of Ecology 70:43-53.

Louda, S. M., and M. A. Potvin. 1995. Effect of inflorescence-feeding insects on the demography and lifetime fitness of a native plant. Ecology 76:229-245.

Ouborg, N. J. 1993. Isolation, population size and extinction: The classical metapopulation approaches applied to vascular plants along the Dutch Rhine-system. Oikos 66:298-308.

Pontis, C. 1992. Om groddknoppar och tandr[ddot{o}]tter. Meddelande 92:2 Stockholms Universitet, Botaniska Institutionen, Avdelningen f[ddot{o}]r Ekologi. (In Swedish).

Primack, R. B., and S. L. Miao. 1992. Dispersal can limit local plant distribution. Conservation Biology 6:513-519.

Quintana-Ascencio, P. F., and E. S. Menges. 1996. Inferring metapopulation dynamics from patch-level incidence of Florida scrub plants. Conservation Biology 10:1210-1219.

Rabinowitz, D. 1978. Early growth of mangrove seedlings in Panama, and an hypothesis concerning the relationship of dispersal and zonation. Journal of Biogeography 5:113-133.

Rydin, H., and S.-O. Borgegard. 1991. Plant characteristics over a century of primary succession on islands: Lake Hjalmaren. Ecology 72:1089-1101.

Salisbury, E. J. 1942. The reproductive capacity of plants. Bell, London, UK.

Salisbury, E. J. 1974. Seed size and mass in relation to environment. Proceedings of the Royal Society of London B 186:83-88.

Scheiner, S. M., and J. M. Rey-Benayas. 1997. Placing empirical limits on metapopulation models for terrestrial plants. Evolutionary Ecology 11:275-288.

Schoener, T. W., and A. Schoener. 1983. The time to extinction of a colonizing propagule of lizards increases with island area. Nature 302:332-334.

SYSTAT. 1996. SYSTAT for Windows. Statistics, version 6.0. SPSS, Chicago, Illinois, USA.

Tilman, D. 1997. Community invasibility, recruitment limitation, and grassland biodiversity. Ecology 78:81-92.

van der Meijden, E., P. G. Klinkhamer, T. J. de Jong, and C. A. M. van Wijk. 1992. Metapopulation dynamics of biennial plants: how to exploit temporary habitats. Acta Botanica Neerlandica 41:249-270.

Westoby, M., M. Leishman, and J. Lord. 1996. Comparative ecology of seed size and dispersal. Philosophical Transactions of the Royal Society of London B 351:1309-1318.

Wood, D. M., and R. del Moral. 1987. Mechanisms of early primary succession in subalpine habitats on Mount St. Helens. Ecology 68:780-790.
COPYRIGHT 2000 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Geographic Code:1USA
Date:Jun 1, 2000

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters