PERSPECTIVE From Individual Dispersal to Species Ranges: Perspectives for a Changing World Hanna Kokko 1 * and Andre ´s Lo ´ pez-Sepulcre 1,2 Dispersal is often risky to the individual, yet the long-term survival of populations depends on having a sufficient number of individuals that move, find each other, and locate suitable breeding habitats. This tension has consequences that rarely meet our conservation or management goals. This is particularly true in changing environments, which makes the study of dispersal urgently topical in a world plagued with habitat loss, climate change, and species introductions. Despite the difficulty of tracking mobile individuals over potentially vast ranges, recent research has revealed a multitude of ways in which dispersal evolution can either constrain, or accelerate, species’ responses to environmental changes. F rom sticky seeds to efficient flight machin- ery with complex navigation systems, ani- mals and plants have evolved an impressive variety of dispersal mechanisms. Through the sim- ple act of moving individuals from one area to another, dispersal has important ecological and evolutionary consequences ( 1, 2), including the ability of species to change or expand their ranges (3). The distribution of species we observe today reflects a long history of alternating episodes of dispersal and isolation. Fluctuations in sea level that opened and closed land corridors, the splitting of continents, and the rising of mountain ranges and islands have all left their mark on the where- abouts of extant species. Nowadays, humans are creating new processes that isolate, connect, and shift landscapes at a much higher speed: Anthro- pogenic habitat fragmentation, transport of inva- sive species, and climate change are among them. How will species ranges react to them? How will their dispersal behavior change? Are physical barriers and open corridors all there is to explain species distributions? To predict whether species can shift to new areas requires, on the one hand, understanding the colonization process at the expanding edge of the species range. On the other hand, the possible range contraction where habitat is deteriorating depends crucially on whether individuals simply leave poor-quality habitat or attempt to stay. Local adaptation to changing conditions is also possible, yet strong dispersal can swamp local genetic change and so prevent adaptation from happening (4). Current anthropogenic environmental changes make the study of the evolution of dispersal a requirement for predictive ecology (5). Dispersal is an important determinant of gene spread and is thus subject to strong natural se- lection ( 1, 2, 6). Even if dispersal is risky, it can evolve to avoid the detrimental effects of crowding and competing with kin ( 6). An evolved willingness of individuals to move about is an obvious prereq- uisite for the spread of a species to different parts of its fundamental niche. But how well do organisms actually achieve this? Evolutionary ecologists are increasingly aware that not only limitations of cog- nitive abilities, but also selective pressures them- selves, can cause severe constraints on the habitat use of a given species. In other words, intrinsic spe- cies properties, rather than just limitations imposed by the landscape, can have profound effects on the ability of a species to colonize new areas ( 3). Unraveling these mechanisms is important if we are to understand the evolution of species ranges and how they respond to environmental change. Adaptation Does Not Predict Optimal Space Use The ability to distinguish between suitable and less suitable habitats is an obvious first limitation to colonization of new areas. Dispersal and habitat settlement cues are often based on ‘‘rules of thumb’’ that can be breathtakingly simple. This can lead to ecological traps, where environmental change dissociates habitat quality from the cues and causes individuals to prefer suboptimal hab- itats (7). For example, dragonflies have been observed to patrol asphalt roads instead of rivers, which results from their use of polarized light as a cue for still water and can also lead to a strong preference for landing on oil (8). Another simplistic cue is conspecific attraction, where in- dividuals use the presence of others as an in- dicator of good habitat: Playbacks of song attract young bobolinks Dolichonyx oryzivorus to settle in habitats of random quality ( 9). If individuals simply copy others’ choices, one can envisage a self-reinforcing and intrinsically random compo- nent to habitat use. But attraction to conspecifics can be more than just an easy cue indicating suitable habitat. In many species of animals, individuals directly benefit from living in groups; philopatry (i.e., staying in the natal patch) can be selected for, particularly if local habitats are worth clinging to 1 Laboratory of Ecological and Evolutionary Dynamics, Depart- ment of Biological and Environmental Science, University of Helsinki, Post Office Box 65 (Viikinkaari 1), FIN–00014 Helsinki, Finland. 2 Evolutionary Ecology Unit, Department of Biological and Environmental Science, University of Jyva ¨skyla ¨, FI-40014, Finland. *To whom correspondence should be addressed. E-mail: hanna.kokko@helsinki.fi Fig. 1. Dispersing individuals are a nonrandom subset of the population. (Left) Dispersing females of the Glanville fritillary, Melitaea cinxia, have a higher flight metabolic rate and are more fecund than sedentary ones. [photograph: Anne Holma] (Right) In Siberian jays, Perisoreus infaustus, the subordinate individuals disperse, whereas the heavier and more dominant remain in their natal territories. [photograph: Hannu Siitonen] Differences in selective pressures on dispersal may have profound consequences for the stability of newly founded populations and, consequently, for the ability of the species to spread and react to environmental change. SPECIAL SECTION www.sciencemag.org SCIENCE VOL 313 11 AUGUST 2006 789