Diversity and Distributions, (Diversity Distrib.

) (2005) 11, 125130

EDITORIAL

Long-distance dispersal research: building a network of yellow brick roads

Department of Evolution, Systematics and Ecology, Alexander Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Givat Ram, Jerusalem 91904, Israel

ABSTRACT

Correspondence: Ran Nathan, Department of Evolution, Systematics and Ecology, Alexander Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Givat Ram, Jerusalem 91904, Israel. Tel.: + 972 2 6584314; Fax: + 972 2 6584757; E-mail: rnathan@cc.huji.ac.il

This special issue of Diversity and Distributions presents six papers that contribute to the assembly of a general research agenda for studying long-distance dispersal (LDD) across a variety of taxonomic groups (e.g. birds, sh, aquatic invertebrates and plants), ecosystems (e.g. terrestrial and marine ecosystems, wetlands and grasslands) and thematic elds (e.g. biological transport, marine biology, biogeochemistry and biodiversity conservation). This editorial emphasizes the need to develop a network integrating different research approaches (yellow brick roads) to address the great challenge (nding the end of the rainbow) of quantifying, understanding and predicting LDD and its implications. I review the key avenues for future research suggested in the special issue contributions, and stress the critical importance of properly considering the spatial and temporal scales relevant to the process and system of interests. I propose combining absolute and proportional denitions of LDD as a default practice in any investigation of LDD processes. When LDD is dened primarily by an absolute critical distance that characterizes key feature(s) of the system of interest, a quantitative assessment of the proportion of dispersal events expected to move beyond this critical threshold distance should also be provided. When LDD is dened primarily by a certain small fraction of dispersal events that travel longer than all others, an estimate of the absolute distance associated with this high percentile at the tail of the dispersal curve should also be added. Keywords Biodiversity conservation, biological invasions, biological transport, complexity, interdisciplinary research, large-scale movements, long-distance dispersal, scale effects, spatial ecology, spatial spread.

FINDING YOUR WAY(S) IN THE (RE-DISCOVERED) LAND OF OZ Interest in dispersal has risen rapidly over the last 15 years (Nathan, 2003) with a recent disproportionate increase in interest particularly in long-distance dispersal (LDD) (Fig. 1). In fact, the existence of these typically rare events of long movements, and their disproportionate importance to populations, species and communities, has long been recognized by the likes of Darwin (1859) and Wallace (1876). Nevertheless, interest in LDD lay dormant over most of the previous century, especially between the 1960s and 1980s, when LDD was considered unquantiable, unpredictable and unimportant (see Nathan, 2001 for review). An important positive outcome of this period for dispersal research was a clear recognition of the evolutionary and ecological signicance of the vast majority of dispersal events of individuals travelling fairly short distances (e.g. Janzen, 1970; Hamilton & May, 1977; Lidicker & Caldwell, 1982).
2005 Blackwell Publishing Ltd www.blackwellpublishing.com/ddi

Early in this recent LDD renaissance, British plant ecologist Jonathan Silvertown published a short discussion-type article with the captivating title: Dorothys dilemma and the unication of plant population biology (Silvertown, 1991). Dorothys dilemma, to nd a way to the end of the rainbow, was solved by travelling along the yellow brick road. Silvertown used this metaphor to encourage plant ecologists to look beyond their limited vision, bounded by the population edge, to nd the all-important tails of the seed dispersal curve. He emphasized one such yellow brick road for studying LDD: ecologists should look at larger scales than they normally do, and should employ genetic tools to quantify LDD through estimates of gene ow rates and genetic structure differentiation. The potential of this specic methodological approach for studying LDD has been emphasized recently (Ouborg et al., 1999; Cain et al., 2000; Godoy & Jordano, 2001; Spong & Creel, 2001; Nathan et al., 2003; Telfer et al., 2003; He et al., 2004). While acknowledging the great potential of genetic tools for studying LDD, the goal of this special issue is to explore
DOI: 10.1111/j.1366-9516.2005.00159.x

125

R. Nathan I summarize the major avenues for future research and emphasize possible links among the different approaches. Then, I discuss two fundamental problems, the sensitivity of LDD to scale and the denition of LDD, in relation to the diverse opinions presented in this special issue. A NETWORK OF YELLOW BRICK ROADS The six contributions from this special issue cover a broad spectrum of organisms, ecosystems and thematic elds, embracing a variety of spatial scales and methodological approaches (Table 1). All contributions strongly recommend a shift from the rather narrative and qualitative analyses, which are being employed by most researchers today, to more concrete and quantitative examinations of LDD; as well as acknowledging the overwhelming challenge involved with such a demanding shift. To facilitate the journey to the end of the rainbow, this special issue highlights some yellow brick roads that provide new insights and better understanding of LDD patterns, processes and implications. Nathan et al. (2005) advocate the development of a new methodological approach for studying multi-scale effects of atmosphere dynamics on aerial transport processes. This mechanistic approach combines detailed atmospheric models with high-resolution data on the relatively long movements of airborne organisms. Soons & Ozinga (2005) used a mechanistic wind dispersal model to evaluate LDD ability of wind-dispersed grassland plants. This model incorporates the effects of smallscale turbulent eddy motion, which can drastically affect LDD of small airborne organisms (Nathan et al., 2002; Soons et al., 2004), but does not deal with the effects of large-scale atmospheric processes, nor does incorporate effects of structural heterogeneity in the landscape. The atmospheric models recommended by Nathan et al. (2005) can incorporate atmospheric processes operating on a broad range of scales as well as virtually any type of landscape heterogeneity. While admitting the immense challenge

Figure 1 A rapid recent increase in the relative frequency of publications using the word dispersal (open columns) and especially the phrase long-distance dispersal (LDD, lled columns) in their title, keywords or abstract, compared to the smaller increase in the total number of publications (grey columns) included in the Science Citation Index (ISI Web of knowledge), over the years 19912003. Search for LDD was (long-distance or long-range) and dispersal. The number of publications using LDD is multiplied by a factor of 20 to match the number of publications using dispersal. Most publications use dispersal in the ecological sense (see Nathan, 2003).

new avenues for studying LDD and to advance integration among alternative methods, chiey by means of interdisciplinary research. LDD processes are typical examples of complex biological systems, for which single approach solutions are probably not appropriate or realistic. Therefore, using Silvertowns metaphor, this special issue advocates the establishment of a network of yellow brick roads for addressing the high complexity and uncertainty associated with LDD. In this editorial, I attempt to synthesize some of the major topics highlighted in the six contributions from this special issue. First,

Table 1 Summary of the key features of the contributions in this special issue
Contribution Nathan et al. Organisms Airborne biota LDD denition* Spatial scale of LDD log10(meters) Suggested yellow brick road Both 26 Developing mechanistic models of biological transport based on multiscale atmospheric models Distinguishing between dispersal beyond the local population and extreme LDD and their relative importance for different ecological processes Recognizing the importance of a specic LDD mechanism, and how it could be quantied Using stable isotopes to quantify LDD over large scales Integrating mechanistic models and regional monitoring data to evaluate the role of LDD in species regional survival Identifying tools for incorporating LDD in biodiversity conservation

Kinlan et al.

Marine biota

Both

26

Green and Figuerola Hobson Soons and Ozinga Trakhtenbrot et al.

Bird-dispersed aquatic invertebrates Mostly birds Wind-dispersed plants All

Absolute distance Absolute distance Proportional distance Both

46 56 23

25 (6 7 human-mediated LDD)

*Two general types of LDD denitions: absolute distance and proportional distance (see text).

126

Diversity and Distributions, 11, 125130, 2005 Blackwell Publishing Ltd

Editorial involved with developing such an interdisciplinary modelling framework, Nathan et al. (2005) illustrate why this approach could be highly rewarding for quantifying, predicting and understanding aerial LDD processes. Kinlan et al. (2005) laid out an innovative framework for examining the role of LDD in marine systems. They call for reconsidering the LDD denition and placing it in the context of some old chestnuts of marine ecology, including the relative openness of marine demography. They outlined an ambitious agenda that should provoke new ideas among scholars of LDD in the terrestrial ecosystem as well. Because data and tools for implementing the framework suggested by Kinlan et al. (2005) are still lacking, they illustrate the potential of their approach with some preliminary analysis of data and theory that give insight into marine LDD. They also emphasize the need to rene marine transport models by incorporating realistic modelling of turbulent mixing processes, which may be as inuential in the hydrosphere (Cowen, 2002) as they are in the atmosphere (Nathan et al., 2005). A major challenge will be to integrate realistic hydrodynamic models with the complex behaviours (swimming, vertical shifts) exhibited by many marine species during the dispersal phase (Cowen et al., 2003). Darwin (1859) devoted two chapters (XII and XIII) of his The Origin of Species to discuss a variety of LDD mechanisms and their implications for species geographical distribution. One of the LDD mechanisms he discussed, the passive transport of aquatic plants and invertebrates by migrating waterbirds, has been virtually neglected ever since. Green & Figuerola (2005) argue that waterbirds transport, either internally or externally, enormous numbers of invertebrate propagules over distances of up to thousands of kilometres. They suggest specic hypotheses about the relationship between the main migratory yways and abundance, distribution, and genetic and species diversity of aquatic invertebrates. They also outlined some directions for investigating these hypotheses, including large-scale comparisons of species composition, small-scale eld studies on bird feeding strategies, gut retention times, propagule survival and movements, and tests of genetic differentiation between wetlands on and off migratory yways. Genetic tools hold great promise for LDD research and their integration with other methods has also been suggested by Hobson (2005) and Kinlan et al. (2005). Motivated by recent advances in applying stable isotope techniques to study migratory connectivity (mostly of birds), Hobson (2005) proposed possible ways this technique may be applied to the study of LDD. He particularly emphasized the promise of this technique in identifying new immigrants, presumably arriving through LDD, by detecting isotopic outliers within an isotopically known population. Hobson suggested that the combination of this technique with other methods is the key to improve the rather coarse resolution currently available by stable isotopes (as mentioned below). Articial isotopic enrichment, one of the techniques highlighted by Hobson (2005), could be particularly useful for marking aquatic invertebrates studied by Green & Figuerola (2005). Natural and articial chemical tags are also an emerging tool in marine LDD research (Kinlan et al., 2005), which opens another possibility for cross-system synergy in approaches.
Diversity and Distributions, 11, 125130, 2005 Blackwell Publishing Ltd

Soons & Ozinga (2005) presented a broad comparative study combining three independent data sets, two using long-term and large-scale surveys to estimate regional survival of a large number of wind-dispersed grassland plant species, and the third, using a mechanistic model to estimate their LDD ability. They found that LDD explains a greater proportion of the variance in regional plant survival than short-distance dispersal, and has a roughly equivalent effect to that of seed longevity. However, they also found that in time periods of increasing nutrient availabilities, such as the second half of the 20th century in the Netherlands, LDD was much less important for regional plant survival than plant nutrient requirement. While this special issue generally emphasizes the importance of LDD, an important take-home message from this study is the need to put LDD in the right perspective, and recognize circumstances where postdispersal processes are of overriding importance. LDD plays a key role in determining a variety of ecological processes and therefore, should often be considered in a conservation context. In reality, this is very rarely done, mainly because of the great difculty and high uncertainty inherent in estimating LDD rates. The importance of considering LDD in conservation plans for the design of marine reserves and sustainable harvest of exploited sh species (Kinlan et al., 2005), and the control of invasive exotic species (Green & Figuerola, 2005) and regional survival of plants (Soons & Ozinga, 2005) are emphasized in this special issue. Trakhtenbrot et al. (2005) reviewed the topic and claried the relative importance of LDD for major conservation issues. They discussed the pros and cons of qualitative and quantitative tools and explore how information on LDD can help guide management decisions. They distinguished among cases of excessive LDD, in which high LDD levels of alien invasive species and genetically modied organisms threaten the survival of native species, and cases of insufcient LDD, where low LDD levels threaten the survival of indigenous species inhabiting fragmented landscapes or undergoing range shifts in response to changing environmental conditions. An important challenge in the establishment of a network of yellow brick roads for studying LDD is implementing tools to assess LDD, including humanmediated events, as standard practice in conservation practices. THE CRITICAL IMPORTANCE OF SCALE One of the main insights gained from ecological research over the last two decades has been broad recognition of the importance of considering the spatial and temporal scales of ecological and evolutionary processes (Levin, 1992; Tilman & Kareiva, 1997; Silvertown & Antonovics, 2001; Chave & Levin, 2003). Dispersal in general and LDD in particular are among the most fundamental processes responsible for sensitivity to scale; hence, LDD constitutes a major impetus for growing recognition of the importance of scale-dependent processes. It is therefore not surprising that all the contributions from this issue working on different processes operating over a variety of spatial scales (Table 1) strongly emphasize the need to properly consider scale effects. All together, they highlight different scale problems associated with LDD. 127

R. Nathan Nathan et al. (2005) emphasized the importance of short-term and small-scale events for long-term large-scale aerial biological transport processes. A diverse set of organisms, ranging from tiny spores, seeds and insects to the much larger birds, utilize the atmosphere for passive or active journeys of various lengths and durations. While it is commonly recognized that such aerial transport processes are strongly affected by long-term (> days) and large-scale (> tens of kilometres) variation in atmospheric conditions, the role of atmospheric processes operating at much shorter (< minutes) and smaller (< hundreds of meters) scales has traditionally been neglected. Yet, some small-scale airow characteristics such as turbulent gusts and coherent eddy motion can largely determine the onset and fate of aerial journeys. Kinlan et al. (2005) demonstrated the substantial variation in the typical scale of marine species dispersal, ranging over ve orders of magnitude even when the comparison is restricted to organisms with sedentary adults. They show a considerable discrepancy between estimates of average dispersal distance and estimates of rate of spread across several taxonomic groups. This suggests that information on the mean properties of the dispersal process may not be very informative about extreme events that are important for large-scale colonization. Higgins et al. (2003) reached a similar conclusion for terrestrial plants, demonstrating substantial disparity between the mechanisms inferred from seed morphology, which mediate the short-distance dispersal of most seeds, and the nonstandard mechanisms that actually drive LDD. Kinlan et al. (2005) further analysed the general positive relationship between the scale of the local population of sedentary adults and the level of propagule retention. In conclusion, they argue that any evaluation of the role of LDD in marine systems, as well as any examination of common concepts such as self-recruitment and openness, must be explicitly linked to the scale of the process and the organization unit (population/ species/community) of interest. Green & Figuerola (2005) emphasized the role of spatial scale in LDD of aquatic invertebrates in relation to the major yways of migratory waterbirds. They argue that invertebrate community structure should vary among wetlands not only because of local interactions such as competition and predation (Shurin, 2000), but also as a result of waterbird migratory movements. The spatial scales of LDD as dened by Green & Figuerola (2005) are relatively large (Table 1), because waterbirds can transport invertebrate propagules over very long distances during their migratory ights. Hobson (2005) denes equal or even larger scales (Table 1). Detection of LDD events by stable isotope markers depends rst and foremost on the scale of the isotopic landscape, which is currently rather coarse (Hobson, 2005). Studies using deuterium isotope markers in feathers can distinguish the source location at a resolution of approximately 300 m in altitude and 1.5 degrees (170 km) in latitude (estimated for North America). This coarse resolution renders the stable isotope technique inefcient for most LDD events in most species; yet it offers a unique opportunity to detect extreme LDD events. This opportunity is particularly attractive because organisms are selfmarked as they feed and therefore, the highly demanding mass marking of capture-recapture applications is not necessary. 128 Finally, scale effects are tremendously important for considering LDD in management decisions (Trakhtenbrot et al., 2005). Assessment of conservation practices must consider the huge interspecic variation in LDD; the ve orders of magnitude difference reported by Kinlan et al. (2005) is not unique to marine systems, wherein the difference of six orders of magnitude in LDD ability was estimated by Soons & Ozinga (2005) for 190 wind-dispersed plant species of north-west European grasslands (Table 1). In fact, the concepts of excessive and insufcient LDD of Trakhtenbrot et al. (2005) are inherently scale-dependent, and therefore any implementation of LDD in conservation efforts should entail sensitivity analysis of the scales over which LDD is estimated. Overall, the need to explicitly incorporate LDD in conservation decisions further emphasizes the need to develop a multiscale view in evaluating conservation strategies (Doak et al., 1992; Schwartz, 1999; Levin, 2000; Hartley & Kunin, 2003). LDD DEFINITION Among the multiple methodological challenges involved in studying LDD (Nathan et al., 2003), the question of denition is of primary importance. Although each of the six contributions from this special issue addressed this challenging question in a different (case- and context-specic) way (Table 1), two general types of denitions can be distinguished. First, the absolute distance denition, assigns LDD events as dispersal beyond a certain threshold distance, whereas the second, the proportional distance denition, assigns LDD events as those reaching distances that are longer than those reached by most other dispersing individuals. The exact threshold values selected for the absolute distance LDD denition are typically arbitrary, but their order of magnitude should reasonably reect the scale of the process in question. In some cases, the LDD scales (or the screening capacity of the applied technique) are only partially known, imposing a rather coarse denition; for example, Hobson (2005) denes LDD at regional (hundreds of kilometres) to continental (thousands of kilometres) scales, over which long-distance movements of highly mobile organisms such as birds might be detected by stable isotopes. A more specic denition is given by Green & Figuerola (2005), who dene bird-mediated LDD of aquatic invertebrates using a threshold distance of 10 km between wetlands that are not connected hydrologically. One of the two denitions proposed by Kinlan et al. (2005) (their case-I LDD) refers to the boundary of the area over which adults of the local population interact and reproduce. This type of LDD denition is more commonly applied in the literature; nevertheless, Kinlan et al. (2005) warn that identication of the distance that best characterizes the local scale is not necessarily as trivial as it is often considered. Other natural sources of an absolute distance scale may include the average size of discrete habitat patches, or the length scale of variation in local demographic processes (growth, mortality, and/or fecundity). Studies using the proportional distance LDD denition select an arbitrary high percentile of the dispersal distances distribution to distinguish LDD. Soons & Ozinga (2005), for example, designated the 99th percentile of the simulated dispersal kernels
Diversity and Distributions, 11, 125130, 2005 Blackwell Publishing Ltd

Editorial of wind-dispersed seeds as a measure of LDD ability, whereas Kinlan et al. (2005) (their case-II LDD) use extreme-value distributions to estimate the farthest distance travelled by an individual in a discrete period of time. In practice, this type of LDD denition is not broadly applicable, mostly because it requires the distribution of dispersal distances to be known or to be amenable to reliable estimation. For instance, Green & Figuerola (2005) noted that the paucity of data on bird-mediated dispersal distance distributions of aquatic invertebrates precludes a proportional denition of LDD in this system. However, the combination of process-based models that motivate the form of a dispersal kernel (e.g. Kinlan et al., 2005; Nathan et al., 2005) with improved LDD data may lead to increasing applicability of this measure. These two types of LDD denitions can be interpreted as mutually exclusive if taken to their extremes. On one hand, the absolute distance can be dened independently of the dispersal characteristics of the species or population in question. Kinlan et al. (2005), for example, offer as one denition of a threshold for their case-I LDD the boundaries of the area where (nondispersing) adults interact and breed, irrespective of the area where (dispersing) larvae may occur. On the other hand, the proportional distance can be dened while completely ignoring any characteristic of the system that is not related to dispersal. For example, the 99th percentile threshold used by Soons & Ozinga (2005) to assess LDD ability depends exclusively on dispersalrelated characteristics, regardless of the patchiness of the landscape. Trakhtenbrot et al. (2005) argued that the proportional distance denition should be preferred in conservation contexts in problems of excessive LDD (e.g. if the main concern is the estimate of LDD ability in nonindigenous elements). If the main concern is to alert for cases in which LDD might be insufcient to place individuals beyond a certain distance, such as the typical interpatch distance in fragmented landscape, the absolute distance denition should be the primary rst choice. Determining two mutually exclusive denitions of LDD can therefore be valuable for some theoretical and practical purposes. Nevertheless, the use of one type of LDD denition while completely ignoring the other might be confusing and even misleading. Nathan et al. (2003; p. 263) gave an example of a proportional LDD denition (99th percentile) estimated from simulations of wind-dispersed seeds in a dense forest vs. open landscape, resulting in an LDD threshold of three vs. 200 meters, respectively. While the latter value may correspond to interpatch distance in moderately fragmented forested landscapes, the former is evidently an inappropriate measure, as LDD events would fall within the crown projection of the source tree. Kinlan et al. (2005) raised similar concerns, emphasizing that the distinction they made between their case-I and case-II LDD in marine systems is based on the substantial difference between the typical scale of movements of sedentary adults (used to dene case-I LDD) and larvae (case-II). If the difference between the movement scales of dispersing vs. reproductive (nondispersing) individuals is much less pronounced, as is the case for certain marine species either lacking larvae or with highly mobile adults, and perhaps for many terrestrial species, these two denitions are expected to collapse into a one operative criterion.
Diversity and Distributions, 11, 125130, 2005 Blackwell Publishing Ltd

The overall conclusion emerging from this discussion is that the two types of LDD denitions should be considered as complementary rather than competing alternatives. I propose combining absolute and proportional denitions as a default practice in any investigation of LDD processes. This will provide adequate information for any specic case under study and would help avoid inappropriate or ambiguous denitions. If the absolute distance denition is the logical rst choice, (e.g. when one wishes to assess interpatch connectivity in a fragmented landscape), information on the proportion of individuals that travel equal or longer distances than the specied threshold should be added. For instance, LDD could be dened as 1000 m (0.1%), where 0.1% indicates the 99.9th percentile at the tail of the dispersal curve. If the proportional distance denition is the rst logical choice, (e.g. when we wish to compare LDD ability of invasive species), information on the absolute distance associated with this high percentile should be added. For instance, LDD could be dened as 0.1% (1000 m). Determining an absolute distance LDD denition is considered less difcult than determining a proportional distance denition. Nevertheless, any attempt to resolve these specic problems of denition, and to address other challenges involved with collecting, analysing and interpreting LDD data, is best undertaken by more than one approach. Assembling an interdisciplinary network of yellow brick roads is probably the only way to address the difculties associated with LDD research. ACKNOWLEDGEMENTS I am grateful to Dave Richardson for inviting me to edit this special issue, to all authors and reviewers for their thoughtful work, and to Andy Green, Keith Hobson, Brian Kinlan, Esther Lachman, Dave Richardson, Nir Sapir, Merel Soons, and Ana Trakhtenbrot for their comments on an earlier draft of this article. I kindly acknowledge the support from the National Science Foundation (IBN-9981620), the German-Israeli Foundation (GIF 20061032.12/2000), the Israeli Science Foundation (ISF-474/ 02), the US-Israel Binational Science Foundation (BSF 229/ 2002) and the International Arid Land Consortium (IALC-03R/25). REFERENCES Cain, M.L., Milligan, B.G. & Strand, A.E. (2000) Long-distance seed dispersal in plant populations. American Journal of Botany, 87, 12171227. Chave, J. & Levin, S. (2003) Scale and scaling in ecological and economic systems. Environmental and Resource Economics, 26, 527557. Cowen, R.K. (2002) Oceanographic inuences on larval dispersal and retention and their consequences for population connectivity. In: Coral Reef Fishes (ed. by P.F. Sale), pp. 149170. Academic Press, London. Cowen, R.K., Paris, C.B., Olson, D.B. & Fortuna, J.L. (2003) The role of long distance dispersal versus local retention in replenishing marine populations. Gulf and Caribbean Research, 14, 129137. 129

R. Nathan Darwin, C. (1859) The Origin of Species by Means of Natural Selection. John Murray, London. Doak, D.F., Marino, P.C. & Kareiva, P.M. (1992) Spatial scale mediates the inuence of habitat fragmentation on dispersal success: implications for conservation. Theoretical Population Biology, 41, 315 336. Godoy, J.A. & Jordano, P. (2001) Seed dispersal by animals: exact identication of source trees with endocarp DNA microsatellites. Molecular Ecology, 10, 2275 2283. Green, A.J. & Figuerola, J. (2005) Recent advances in the study of long-distance dispersal of aquatic invertebrates via birds. Diversity and Distributions, 11, 149 156. Hamilton, W.D. & May, R.M. (1977) Dispersal in stable habitats. Nature, 269, 578 581. Hartley, S. & Kunin, W.E. (2003) Scale dependency of rarity, extinction risk, and conservation priority. Conservation Biology, 17, 1559 1570. He, T.H., Krauss, S.L., Lamont, B.B., Miller, B.P. & Enright, N.J. (2004) Long-distance seed dispersal in a metapopulation of Banksia hookeriana inferred from a population allocation analysis of amplied fragment length polymorphism data. Molecular Ecology, 13, 1099 1109. Higgins, S.I., Nathan, R. & Cain, M.L. (2003) Are long-distance dispersal events in plants usually caused by nonstandard means of dispersal? Ecology, 84, 1945 1956. Hobson, K.A. (2005) Using stable isotopes to trace long-distance dispersal in birds and other taxa. Diversity and Distributions, 11, 157164. Janzen, D.H. (1970) Herbivores and the number of tree species in tropical forests. American Naturalist, 104, 501528. Kinlan, B.P., Gaines, S.D. & Lester, S.E. (2005) Propagule dispersal and the scales of marine community process. Diversity and Distributions, 11, 139 148. Levin, S.A. (1992) The problem of pattern and scale in ecology. Ecology, 73, 1943 1967. Levin, S.A. (2000) Multiple scales and the maintenance of biodiversity. Ecosystems, 3, 498 506. Lidicker, W.Z. & Caldwell, R.L. (1982) Dispersal and Migration. Hutchinson. Ross Publishing Co, Stroudsburg, Pennsylvania. Nathan, R. (2001) Dispersal biogeography. In: Encyclopedia of Biodiversity (ed. by S.A. Levin), vol. II, pp. 127152. Academic Press, San Diego. Nathan, R. (2003) Seeking the secrets of dispersal. Trends in Ecology and Evolution, 18, 275 276. Nathan, R., Katul, G.G., Horn, H.S., Thomas, S.M., Oren, R., Avissar, R., Pacala, S.W. & Levin, S.A. (2002) Mechanisms of long-distance dispersal of seeds by wind. Nature, 418, 409 413. Nathan, R., Perry, G., Cronin, J.T., Strand, A.E. & Cain, M.L. (2003) Methods for estimating long-distance dispersal. Oikos, 103, 261273. Nathan, R., Sapir, N., Trakhtenbrot, A., Katul, G.G., Bohrer, G., Otte, M., Avissar, R., Soons, M.B., Horn, H.S., Wikelski, M. & Levin, S.A. (2005) Long-distance biological transport processes through the air: can natures complexity be unfolded in-silico? Diversity and Distributions, 11, 131137. Ouborg, N.J., Piquot, Y. & Van Groenendael, J.M. (1999) Population genetics, molecular markers and the study of dispersal in plants. Journal of Ecology, 87, 551568. Schwartz, M.W. (1999) Choosing the appropriate scale of reserves for conservation. Annual Review of Ecology and Systematics, 30, 83108. Shurin, J.B. (2000) Dispersal limitation, invasion resistance, and the structure of pond zooplankton communities. Ecology, 81, 30743086. Silvertown, J.W. (1991) Dorothys dilemma and the unication of plant population biology. Trends in Ecology and Evolution, 6, 346348. Silvertown, J. & Antonovics, J., eds (2001) Integrating Ecology and Evolution in a Spatial Context. Blackwell Science, Oxford. Soons, M.B., Heil, G.W., Nathan, R. & Katul, G.G. (2004) Determinants of long-distance seed dispersal by wind in grasslands. Ecology, 85, 30563068. Soons, M.B. & Ozinga, W.A. (2005) How important is longdistance seed dispersal for regional survival of plant species? Diversity and Distributions, 11, 165172. Spong, G. & Creel, S. (2001) Deriving dispersal distances from genetic data. Proceedings of the Royal Society of London Series BBiology Sciences, 268, 25712574. Telfer, S., Piertney, S.B., Dallas, J.F., Stewart, W.A., Marshall, F., Gow, J.L. & Lambin, X. (2003) Parentage assignment detects frequent and large-scale dispersal in water voles. Molecular Ecology, 12, 19391949. Tilman, D. & Kareiva, P., eds (1997) Spatial Ecology: the Role of Space in Population Dynamics and Interspecic Interactions. Princeton University Press, Princeton. Trakhtenbrot, A., Nathan, R., Perry, G. & Richardson, D.M. (2005) The importance of long-distance dispersal in biodiversity conservation. Diversity and Distributions, 11, 173181. Wallace, A.R. (1876) The Geographical Distribution of Animals. Harper & brothers, New York.

130

Diversity and Distributions, 11, 125130, 2005 Blackwell Publishing Ltd