Current Projects: Theoretical and Conceptual Studies
Mutualisms
| Mutualisms have been heavily studied for decades, but in an unsystematic way that has obscured features shared interactions differing greatly in natural history, as well as differences among superficially similar interactions. Thus, while the ecological importance of mutualism is frequently asserted and the evolutionary puzzle it poses frequently pointed out, most of the most basic questions about its ecology and evolution have yet to be addressed. Much of my research career has been spent in trying to synthesize a unified view of mutualism and to identify research priorities for the future. I have written a lot of reviews dealing with various aspects of the evolutionary ecology of mutualism. The most recent general one is in a book called Evolutionary Ecology: Concepts and Case Studies (Bronstein, 2001 in Fox et al., 2001). Other reviews on particular aspects of mutualism include Bronstein, 1994 on conditional outcomes in mutualism; Bronstein 1998 on ant/plant mutualisms; Bronstein 2001 on the cost of mutualism; Bronstein and Barbosa 2002 on multitrophic effects on mutualism; Bronstein et al. 2006 on insect/plant mutualisms; and Bronstein et al. 2007 on mutualistic seed dispersal by vertebrates in desert ecosystems. These studies have motivated my recent field studies of a pollination mutualism in which the plant’s primary pollinator is also a devastating herbivore ; an obligate pollination mutualism that is reassembling after an anthropogenically caused fire ; and a generalized ant/plant protection mutualism in a highly seasonal environment. |
The costs of cooperating (and the benefits of defecting)
| The net effect of mutualism is highest when individuals are able to maximize the benefits they receive from their partners for the least possible investment. As a consequence, conflicts of interest between mutualists seem unavoidable, because the benefit accrued by one partner often translates into the cost experienced by the other. Such conflicts have been repeatedly argued to destabilize mutualisms, as non-cooperative individuals should fare best; their apparent advantage has been thought to necessitate adaptations to punish or exclude cheaters, if mutualism is to persist. My research in this area has been devoted to critically evaluating this general argument, and to developing alternatives that take into account the empirical realities of how mutualisms function in nature.
My approach has included field measures of the costs and benefits of cooperation in certain mutualisms, comparative analyses, and theoretical models. Much of my early work on the fig/pollinator interaction has focused on documenting costs that fig wasps inflict by consuming fig seeds, and searching for mechanisms by which figs might limit those costs (Bronstein 1988, 1992, Anstett et al. 1996, Bronstein and Hossaert-McKey 1996, Hossaert-McKey and Bronstein 2001, Kjellberg et al. 2001). I have extended this inquiry to other mutualisms involving pollinators whose offspring feed on plant tissue (Addicott et al. 1990, Ziv and Bronstein 1996, Bronstein 2001b, Holland et al. 2002, Bronstein et al. 2006a). Our models suggest that pollinator populations may be sufficiently self-limiting as to stabilize their own mutualisms without requiring partner control (Morris et al. 2003, Wison et al. 2003, Bronstein et al. 2003a).
Regardless of how high the costs of mutualism are (or are not), individuals that experience greater benefits and reduced costs are expected to leave more offspring, favoring the evolution of cheating, an event widely believed to be disastrous to the persistence of cooperation. For this reason, it is generally thought that either cheating rarely arises, cheating arises and extinguishes mutualism, or cheating is under strict partner control. I have argued that in fact, mutualism is ubiquitous in nature, cheating is ubiquitous within mutualisms, and cheating is very rarely punished (Bronstein 2001c, 2003, Bshary and Bronstein 2004). How can this be? Theory that I am developing with Regis Ferrière offers one answer (Ferrière et al. 2002, 2007). We show that cheating establishes a background against which better mutualists can display any competitive superiority. Competitive asymmetries in effect generate a selective force that can counter the pressure for reducing investment in mutualism. They can lead not only to the coexistence of mutualist and cheater phenotypes (see also Law et al. 2001), but to a progressive increase in the benefits of mutualism. |
Mutualisms in natural and altered community settings
| Mutualisms clearly are not isolated from the communities in which they occur; furthermore, extreme specialization is quite rare in non-symbiotic mutualisms. Highly specialized, pairwise mutualisms offer advantages for studying many ecological and evolutionary phenomena, particularly when compared with diffuse mutualisms characterized by weaker selection pressures and a wider array of partner species. However, both the population dynamics and evolution of the key players in mutualisms are known to be affected by species other than single partners (e.g., Bronstein 1994b, 1998, Bronstein and Barbosa 2002, Holland et al. 2004). Thus, we need to confront the challenge of placing mutualisms back into their community contexts. I am addressing this need by studying both specialized and generalized mutualisms in variable and anthropogenically altered environments, and in the presence of specialized non-mutualists.
I have a longstanding interest in non-cooperative species that usurp the benefits mutualists exchange. I have explored the natural history of exploiters associated with specialized mutualisms (Bronstein 1991b, 1999, Bronstein and Ziv 1997), and have used what I have learned to model the ecological (Bronstein et al. 2003a, Morris et al. 2003, Wilson et al. 2003) and evolutionary (Ferriere et al. 2007) dynamics of similar mutualist/exploiter networks. One of our recent findings is that evolutionary trajectories of pairwise mutualisms are altered by the presence of third-species exploiters. In fact, exploiters can sometimes stabilize mutualisms that would otherwise evolve to extinction (Ferriere et al. 2007).
In nature, mutualisms occur in remarkably variable environments. Habitat variation alters both the benefits and costs of mutualism (e.g., Bronstein and Patel 1992a,b, Bronstein 1994b, 1995, Bronstein and Hossaert-McKey 1995, 1996, Smith and Bronstein 1996, Bronstein and Hoffmann 1997). What are the implications for the persistence of mutualisms in habitats subject to anthropogenic change? A recent postdoc, Josh Ness, and I looked particularly closely at disruption of mutualisms by invasive ant species (Ness et al. 2004, Ness and Bronstein 2004). Anthropogenic change can be expected to affect mutualisms not only ecologically, however, but evolutionarily as well. In the first study of mutualism within the framework of evolutionary conservation biology, Ulf Dieckmann, Regis Ferrière and I used adaptive dynamics theory to explore mutualistic coevolution in altered landscapes (Bronstein et al. 2003b). The conclusions we draw about the relative resilience and resistance of obligate vs. facultative mutualisms subject to different kinds of threats may help in prioritizing conservation efforts; they also open up many new, basic questions about multispecies coevolution. |
Learn more about our empirical studies
Back to the research overview |
