The geographic mosaic

One particular empirical observation has dominated much recent thinking on co-evolutionary dynamics: the observation that species interactions are spatially differentiated between subpopulations. For example, in the South African Rediviva bees and Diascia flowers (Figure 11.4), across populations of bees there is variation in the length of the bee legs that matches the length of the local floral spurs. Thus, traits have become geographically differentiated across space. This is what John Thompson (1994, 2001) has called a selection mosaic. There may also exist across space, areas where co-evolution is more intense than in other areas, perhaps simply because of variation in

Fig. 11.5 A crossbill, Loxia curvirostra, from South Hills, Idaho, south of the Rockies, feeding on the cone of a lodgepole pine, Pinus contorta. In this area, devoid of red squirrels, lodgepole pine cones differ from those found in the Rockies, where red squirrels are major seed predators. The cones from South Hills are more cylindrical, and have thicker scales as a defence against crossbills. The South Hills crossbills have large, stout bills as a counter-adaptation. Photo courtesy of Craig Benkman.

Fig. 11.5 A crossbill, Loxia curvirostra, from South Hills, Idaho, south of the Rockies, feeding on the cone of a lodgepole pine, Pinus contorta. In this area, devoid of red squirrels, lodgepole pine cones differ from those found in the Rockies, where red squirrels are major seed predators. The cones from South Hills are more cylindrical, and have thicker scales as a defence against crossbills. The South Hills crossbills have large, stout bills as a counter-adaptation. Photo courtesy of Craig Benkman.

the degree of overlap of geographic range. An example of such hot spots comes from a study by Benkman (1999) on lodgepole pine and crossbills in the Rocky Mountains (Figure 11.5). In areas where red squirrels occur, the pines have developed rounded cones that deter predation of seeds by squirrels. In areas lacking red squirrels, the pines have developed longer cones that are an effective defence against predation by crossbills. In these areas, the crossbills have developed larger stouter bills. Thus, these areas are co-evolutionary hot spots between pines and crossbills.

Subpopulations with divergent traits may also differ in the degree of population mixing. The crossbill populations have different songs which are likely to increase reproductive isolation. Intermediate bill morphologies caused by hybridization should also be selected against, because each population lies at its own adaptive peak (Benkman 2003). In the field, the birds from different populations mate assortatively and are beginning to differentiate genetically (Benkman 2003). Over time this may lead to speciation. These results are intuitively pleasing because crossbills seem to have differentiated across much of the northern hemisphere into a number of incipient species, and locally adaptive diversifying co-evolution provides a potentially widespread mechanism. It is additionally exciting that without considering spatial structure it would be more difficult to imagine that particular mode of co-evolution occurring.

Observations such as these led Thompson to suggest in his 'geographic mosaic theory of coevolution' that geographic structure is influential in the evolution of species interactions: 'Any theory of coevolutionary dynamics must therefore take into account this geographic structuring of most taxa and interactions' (Thompson 2001, p. 332). This stimulated many workers to examine whether linking subpopulations together in a spatial structure can lead to different predictions about co-evolution in those subpopulations than would be predicted if one just considered a single subpopulation in isolation.

So far, several studies suggest that geographic structuring can be influential at that local population scale. For example, Hochberg et al. (2000) developed a model of an obligate symbiont and its host, and investigated how productivity differences across landscapes influence the evolution of virulence in the symbiont. When productivity is high, the host population is high in the absence of the symbiont (a source population) and when productivity is low the host population is low (a sink population). When virulent strains of the symbiont have a competitive advantage over avirulent strains (perhaps via increased transmission), the virulent strain is most likely to be found in the source population, and the avirulent strain persists in marginal habitats that do not favour cheaters or exploiters so much. Rather interestingly, if parasites are a strong selective pressure on the evolution of sex, as suggested by Hamilton, then asexual forms should be relatively favoured in marginal habitats where virulent symbionts cannot persist. This matches empirical observations on many taxa that asexual forms are commonest in marginal habitats (Chapter 2).

It is pleasing when the addition of a single assumption, such as spatial structure, can serve to alter local predictions in the direction indicated by data. However, there is a potentially much greater prize at stake; if the addition of spatial structure can also change the global direction or mode of co-evolution over all subpopulations combined. So far, work on the geographic mosaic has not addressed this in earnest, but observations, such as seen in Rocky Mountain crossbills, suggests that it might do so, at least sometimes. The case of local co-evolution of CMS mutants and nuclear genes in Plantago lanceolata, leading to reproductive isolation between subpopulations (Chapter 5), provides another potential example. It is also interesting that the general theory of the evolution of mutualisms increasingly involves consideration of spatial structure (Chapter 10). Perhaps this represents some convergence of the theory of species interactions from two different perspectives.

Throughout the last three chapters it has been clear that the evolution of species interactions can result in speciation. This is a logically consistent concept, for as the next chapter will show, speciation requires rather specific ecological and environmental conditions, and other species comprise an important part of the ecological environment. The next chapter looks at the mechanisms of speciation more generally.

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