ScienceWeek

ECOLOGY: ON MUTUALISTIC SPECIES WEBS

The following points are made by John N. Thompson (Science 2006 312:372):

1) No multicellular eukaryotic organism is capable of surviving and reproducing using only its nuclear genes and the gene products it makes. Species coopt the genomes of other species by forming mutualistic, but inherently selfish, alliances. One can grasp the central importance of mutualistic associations in the diversification of life through a simple thought experiment. Try to imagine a plant that can survive and reproduce in a real ecosystem without using, in addition to its nuclear genome, most of the following: a mitochondrial genome (to convert energy); a chloroplast genome (to regulate photosynthesis); one or more mycorrhizal fungal genomes (to improve nutrient and water uptake); the genomes of pollinators (to assist in reproduction); and the genomes of a few birds, mammals, or ants (to move seeds around the ecosystem). Each plant is part of a complex web of interacting mutualists.

2) One of the major challenges for evolutionary biology is to understand how species coevolve and shape complex webs of mutualistic interaction. New work [1] addresses an important component of this problem by asking if mutualistic interactions involving dozens or even hundreds of plant and animal species coevolve in a way that leads to a predictable pattern of links among species. The new work focuses on some of the most visible, diverse, and quantifiable mutualistic interactions found within terrestrial communities -- those between plants and their free-living pollinators and seed-dispersal agents. Some ecosystems, such as tropical rain forests, rely so heavily on these interactions that they would collapse in their absence, because plant reproduction would cease.

3) Within these webs, it is rare for a local plant species and animal species to be so reciprocally specialized that neither interacts with other species [2]. Instead, species differ greatly within webs in the number of links to other species. For example, some bee species are extreme specialists that visit the flowers of only one or two plant species, but other bee species are generalists that visit the flowers of dozens of plant species. In a previous analysis [3], the authors used network theory [4] to show that specialization within these mutualistic webs tends to be nested. In a nested web, a core group of generalists all interact with each other, but extreme specialists interact only with the generalist species. The result is a web with many asymmetries in degrees of specialization among the interacting species. In contrast, interactions between predators and prey or herbivores and plants are often more compartmentalized, forming smaller clusters within the broader interaction web [5].

4) The new study adds additional ecological realism to these analyses. Most studies of nested and compartmentalized webs have been based on qualitative data, in which all connections between species are given equal weight. Recent studies of food webs with antagonistic interactions between species have begun to explore webs in which the connections among species are weighted by the relative frequency with which a species interacts with other species. In extending quantitative network analyses to mutualistic webs, Bascompte et al [1] show that the distribution of specialists and generalists within these webs is unlikely to be due to chance. Moreover, they show that asymmetries in specialization among pairs of interacting species are the rule: Strong dependence on a particular interaction in one direction is frequently accompanied by weak dependence in the other direction. Hence, a plant might rely heavily on the seed-dispersal services of a particular frugivore species, but that same frugivore species might consume fruits from multiple plant species.

References (abridged):

1. J. Bascompte, P. Jordano, J. M. Olesen, Science 312, 431 (2006)

2. N. M. Waser, J. Ollerton, Eds., Plant-Pollinator Interactions: From Specialization to Generalization (Univ. of Chicago Press, Chicago, 2006)

3. J. Bascompte, P. Jordano, C. J. Meli n, J. M. Olesen, Proc. Natl. Acad. Sci. U.S.A. 100, 9383 (2003)

4. S. R. Proulx, D. E. L. Promislow, P. C. Phillips, Trends Ecol. Evol. 20, 345 (2005)

5. T. M. Lewinsohn, V. Novotny, Y. Basset, Annu. Rev. Ecol. Syst. 36, 597 (2005)

Science http://www.sciencemag.org

ScienceWeek http://scienceweek.com