Evolutionary stable dispersal strategies

One of the great developments in evolutionary biology that resulted from Wynne-Edwards's challenge was the notion of the evolutionary stable strategy or ESS (Chapter 5). These are strategies that, when adopted by a population, cannot be invaded by any alternative strategy. ESSs result from the relative fitness of individuals displaying alternative strategies. They are used to predict evolutionary outcomes when the fitness of a given strategy depends on the strategies adopted by the remainder of the population.

Hamilton (1967), in his paper on extraordinary sex ratios, is normally attributed with the first explicit ESS model in evolutionary biology. It was Hamilton again who, together with Bob May, first applied the approach to the evolution of dispersal (Hamilton and May 1977). Hamilton and May imagined a very simple scenario in which the environment consisted of a number of identical habitat patches that were each occupied by a single individual. These adults died every year (an annual species), and the offspring then competed to exploit the vacant patches. Adults would give rise parthenogenetically to dispersive offspring, with a given frequency. Dispersing offspring were then allocated evenly across the patches, while non-dispersing offspring remained in the natal patch to compete. The outcome of competition within a patch was random, and dispersive offspring suffered a mortality cost before arriving at their new patch.

Given these conditions, what is the ESS rate of dispersive offspring? The solution turns out to be remarkably simple: 1/(2 — s) where s is the survival of dispersers from their natal patch to the new patch. Thus, if there is no mortality cost to dispersal, all individuals disperse. If the mortality cost is extremely high (s ^0) over 50% of individuals should still disperse! Why should this happen? Imagine a scenario in which the resident strategy is for zero dispersal (Figure 6.2). A mutant individual with some level of dispersal will displace this strategy: its offspring will compete for, and some of the time gain, new patches which the resident non-dispersers can never regain. The specific genetic advantage from dispersive offspring is that offspring compete less among themselves and more among alternative strategies which they can displace; there by reducing competition between kin that share the same genes.

When they added in the possibility that patches could become totally vacant (by distributing offspring randomly, not evenly among patches), Hamilton and May noticed an interesting point: the ESS dispersal strategy was always greater than that which would maximize site occupancy. The implications are, first that the predominance of individual over group selection increases dispersal, and second that what is best for the individual is not always best for the population. This second theme is actually one of the most

No dispersal

Mutant disperser

No dispersal

Mutant disperser

Fig. 6.2 Hamilton and May's model of the evolution of dispersal. In the left hand column, individuals

(small circles) are distributed among four patches (large circles). After reproduction (b), offspring compete and mortality acts (c). In the right hand column, a mutant disperser (small open circle) arises in one patch (a). Some of its offspring disperse to neighbouring patches (b), and some of them are successful. After competition, its patch occupancy has increased, hence, dispersal tendency spreads through the population.

Fig. 6.2 Hamilton and May's model of the evolution of dispersal. In the left hand column, individuals

(small circles) are distributed among four patches (large circles). After reproduction (b), offspring compete and mortality acts (c). In the right hand column, a mutant disperser (small open circle) arises in one patch (a). Some of its offspring disperse to neighbouring patches (b), and some of them are successful. After competition, its patch occupancy has increased, hence, dispersal tendency spreads through the population.

important messages of the evolution of dispersal, one that has been reinforced by subsequent work. In fact dispersal evolution can theoretically both increase the risk of population extinction and reduce it (see Chapter 13).

Since then, a number of authors have extended Hamilton and May's assumptions, normally recovering their basic result as a special case (see Johnson and Gaines 1990). It was clear, however, that the mathematical complexities of the ESS approach would sometimes become insurmountable, and many subsequent authors have taken alternative modelling approaches, such as evolutionary simulations (see Chapter 1). Comfortingly these models have tended to reinforce the findings of the ESS models, such that dispersal theorists can present to the world an almost united front. What have they found?

First, variation in site suitability over time tends to favour dispersal. This suggestion preceded the advent of formal modelling (e.g. Southwood 1962) but has been confirmed by it. Temporal variation rewards individuals that hedge their bets by placing offspring in a variety of patches in case the current patch deteriorates. A special case of this is the so-called 'Janzen-Connell' hypothesis (Janzen 1970; Connell 1971), usually used to explain species coexistence. In some organisms local habitat patches may become unsuitable simply because it is already occupied by the species. One reason may be because parents harbour natural enemies that can affect offspring, and the effect of those enemies decreases with distance from the parent. This selects strongly for dispersal between individuals of the same species, and hence, also generates a species-rich local mix of individuals. Chaotic local population dynamics, which also make patch quality unpredictable in time, can select for dispersal in the same way.

Spatial, rather than temporal, variation in the environment, has the opposite effect on dispersal evolution: it will tend to reduce it because most individuals will be in the best sites, so fewer individuals benefit from moving from poor to good sites than lose from moving from good to poor sites. This spatial variation includes habitat fragmentation, which can reduce dispersal rates severely (Travis and Dytham 1999). Clearly, in a situation where populations persist through migration between subpopulations (a metapopulation), habitat fragmentation will select for reduced dispersal. This will enhance the probability of extinction, a nasty side effect of selection acting on individuals rather than groups. This phenomenon, termed evolutionary suicide, has been explicitly modelled by Gyllenberg et al. (2002). They show that under a range of conditions in which habitat patches can become unsuitable, evolutionary suicide can occur.

The type of competition per patch is also predicted to influence dispersal. One of the benefits of dispersal is reduced competition among residents that are kin (Hamilton and May 1977). Inbreeding and population structures that promote kin-groupings should increase those benefits. Given these expectations, interesting predictions can be made about the sex that is expected to disperse under different mating systems (Perrin and Mazalov 2000). The sex that is expected to disperse the most, depends on which sex experiences the most severe form of competition and for what resource. When competition affects both sexes equally, no sex bias in dispersal is expected. In contrast, under polygyny or promiscuity, competition for mates among males exceeds competition for resources among females, thus males should disperse more than females. Male dispersal reduces the chances of females dispersing,because related females are unlikely to become inbred. In many monogamous breeding systems, however, males must defend resources to attract females, so females might be expected to disperse more.

A final set of predictions relates to populations not at equilibrium. Many populations show transient fluxes in geographic range, either because of changing climate or invasions into new habitats, or because they are declining (see Chapter 13). Populations whose ranges are expanding are selected for an increase in dispersal at the expanding range front due to the local appearance of newly suitable patches (Travis and Dytham 2002).

Waste Management And Control

Waste Management And Control

Get All The Support And Guidance You Need To Be A Success At Understanding Waste Management. This Book Is One Of The Most Valuable Resources In The World When It Comes To The Truth about Environment, Waste and Landfills.

Get My Free Ebook


Post a comment