Sunday, May 11, 2014

Mutualism and exploitation in angiosperm pollination

This is a literature review I wrote a few years ago, which I have revisited because I am thinking anew about symbiosis in the context of my novel-in-progress, one major element of which is parasitology and host-switching. Over on my new writing blog, I am tinkering with ideas about utopia and dystopia as analogous with mutualism and parasitism. Since this is a science paper, it seems most appropriate to put it here, rather than there!


Mutualism, the cooperative interaction between species, is ubiquitous and fundamental to life on earth (Bronstein 1994). Mutualistic symbioses are vital for a range of crucial mechanisms ensuring reproduction and nutrient acquisition, as well being critical to broader environmental functions such as the nitrogen cycle and carbon sequestration (Wilson et al. 2009). One of the most important mutualisms is the relationship between angiosperms and their biotic pollinators. As with all symbioses, these are in a state of flux over evolutionary time, with the balance between cooperation and exploitation changing according to environmental variants and arms races between the participants, as natural selection pressures them to maximise their own reproductive success.

While there is a growing and substantial body of research examining the natural history of mutualism and increasing empirical knowledge of costs and benefits to symbionts, evolutionary models are limited and frequently in conflict (Bronstein 2001; Bshary & Bronstein 2004).

In this review, I will discuss the history of biological exploration of interspecies cooperation and consider the diversity of mutualism among flowering plants and their biotic pollinators and the exploitation that can lead to its destabilisation. In addition, I will survey mechanisms thought to contribute to the maintenance of mutualistic interactions, and finally, examine some of the models of mutualism that have been posited.

The evolution of cooperation: a history

Since Darwin’s initial formulation of evolutionary theory (Darwin 1859), cooperation within and between species has presented a challenge to the idea of natural selection favouring individuals which maximise their own fitness through “selfish” behaviour. An explanation of intraspecific cooperation and altruism emerged with Hamilton’s proposal of kin selection (Hamilton 1964), whereby it is in the interest of an individual to foster the well-being of close kin, as these share a high proportion of genes with that individual. This concept was extended and popularised by Dawkins, who emphasised the importance of the gene as the unit of selection, contextualising it in the idea of inclusive fitness (Dawkins 2006).

These rationales for cooperation do not pertain to interspecific mutualism, however, since the genetic relatedness of different species that interact to benefit each other can only be comparatively small. Clearly, kin selection and inclusive fitness do not provide explanations for diverse species providing each other with services and food. Other explanations needed to be sought.

Although Darwin was well aware of interspecific mutuality, and expounded upon examples such as the dispersal of seeds by the mistletoe bird (Darwin 1859), his essential focus was upon explaining evolutionary change by the effects of conflict and competition (Sapp 1994). “Darwin’s bulldog”, Thomas Huxley, emphasised the idea of the Hobbesian war occurring among both humans and other animals (Huxley 1888), as did the staunch laissez-faire liberal Herbert Spencer who coined the term, “survival of the fittest” (Spencer 1904).

The Russian Peter Kropotkin recoiled at the idea of competition and conflict being the essence of evolution as proposed by Darwin and Huxley, and perhaps influenced by his political ideology of anarchist communism and his own natural history experiences in eastern Siberia (Sapp 1994), argued that cooperation and altruism are fundamental to both human nature and species in general (Kropotkin 1902). While Kropotkin himself made little reference to interspecific mutualism, his ideas profoundly influenced subsequent zoologists researching this field who found the perceived social implications of Darwinism repugnant (Sapp 1994). It is perhaps not coincidental that the originator of the principle of symbiogenesis, Boris Mikhaylovich Kozo-Polyansky (Kozo-Polyansky 1924), was a countryman of Kropotkin and educated in Soviet Russia.

In modern times, explanations for biological cooperation are less infused by political ideology, and are based in evolutionary ecology and mathematical models of cost and benefit to the participants. Nevertheless, while the mechanisms of interspecies competition for resources have been studied in depth, those regulating cooperative behaviour between species have received far less attention (Bronstein 1994; Stadler & Dixon 2008) and the dynamics of mutualistic interaction are subject to continuing debate (Bronstein 2001; Bshary & Bronstein 2004; Hoeksema & Bruna 2000; Weyl et al. 2010).
The pervasiveness of mutualism

It has been estimated that every species on earth is involved in at least one cooperative symbiotic relationship with another species (Bronstein 2001). These mutualisms are found between all kingdoms of life and exist in every ecosystem thus explored (Paracer & Ahmadjian 2000). Mutualisms made possible the migration of plants onto land (Pirozynski & Malloch 1975) and are considered to have been a crucial element in the evolution of eukaryotic cells (Margulis 1967). The ubiquity of mutualisms and symbiosis generally has been described as presenting a unifying concept in biological sciences (Paracer & Ahmadjian 2000), without which life as it has evolved on earth would be impossible.

The cooperative association between angiosperms and their biotic pollinators is ancient and fundamental to the diversity of flowering plants. Biotic pollination is the transfer by an organism of a male gamete (in pollen) from the anther of a flower to the stigma of another flower, either of the same plant (or indeed flower) or a separate individual. Abiotic pollination, involving transportation of pollen by wind, rain or gravity is also employed by many angiosperms, and is the primary mechanism for gamete transfer in gymnosperms such as cycads and conifers.

The success of angiosperms in colonising terrestrial ecosystems throughout the planet is due in large measure to the relative precision of biotic pollination, as it requires the production of fewer gametes for a greater reward. Pollen is high in proteins and sugars, and has been an insect food source since before the evolution of angiosperms, as evidenced by gymnosperm pollen found preserved in insect guts from the Permian (Labandeira 1997). This suggests that the evolution of insect pollination occurred as an exploitation of this behaviour (Bronstein, Alarcón & Geber 2006). Angiosperms and their pollinating insects radiated in parallel during the Cretaceous (Grimaldi & Engel 2005). Traits of both plants and their pollinators are subject to selection by this interaction, resulting in increasing diversity mainly in response to multispecies interactions. Most angiosperms are pollinated by multiple pollinators, and most pollinators visit multiple angiosperms (Fenster et al. 2004).

The evolution of nectar as a means of attracting pollinators occurred prior to the Late Jurassic (Labandeira 1997). A large proportion of Australian angiosperms are pollinated by birds, in contrast with plants in other continents, with plants having evolved specialised morphological adaptations that facilitate bird feeding and pollination (Ford, Paton & Forde 1979). Many of these species are also pollinated by mammals. That Australia should be a hotspot for bird pollination has raised a number of hypotheses. It has been suggested that the high abundance of sunlight ensures that photosynthesis and the production of sugars is not a limiting factor in the physiology of many Australian plants, making nectar a relatively cheap product to exchange for pollination. Soil nutrients, by contrast, are particularly limiting in areas of Australia where bird pollination is most common (Ford, Paton & Forde 1979). This is significantly the case with phosphorus, essential for the synthesis of proteins (Atwell, Kriedemann & Turnbull 1999). Producing excess pollen as a reward for pollinators would thus be more expensive for these plants than producing ample nectar to attract them. 

Birds might be a more effective pollinator because they travel further than insects, increasing outcrosses and adding to the advantage conferred on plants using them as preferred pollinators. In addition, it is possible that birds act as defenders of plants against insect herbivory (Ford, Paton & Forde 1979).

Research has suggested that nectar-feeding birds have evolved spatial memory that helps them avoid returning to flowers from which they have fed until sufficient time has passed for the nectar supply to be replenished (Burke & Fulham 2003), demonstrating coevolution of traits of the partners in this mutualism.

The allure of cheating

For a mutualistic relationship to be of use to an individual, the benefits must outweigh the costs [Figure 1]. When each partner obtains a net benefit and barring any extrinsic disruptive events, the mutualism should persist (Bshary & Bronstein 2004). Nevertheless, most mutualisms are asymmetric, with one species benefiting more than the other (Bronstein 1994). If this asymmetry becomes pronounced over time due to environmental changes, the mutualism can transform into a parasitic relationship (Hoeksema & Bruna 2000) defined as occurring when the net cost exceeds the net benefit of the symbiosis. The same two species in different environments can have a parasitic or a mutualistic relationship, as with, for example, mycorrhizal symbioses under different nutrient regimes (Johnson, NC, Graham & Smith 1997).

The benefit to a symbiont of bypassing cooperation with its partner and becoming parasitic is clear when assessed in terms of the cost it pays in mutualism. To efficiently obtain the nutrients or services of its partner without “payment” allows it to retain resources for itself. This process of “cheating” is considered in a number of models of symbiosis that have been developed, which will be reviewed later in this paper. Cheating is commonplace among mutualisms (Bronstein 2001) and this can have a selective pressure upon both the cheater to maintain or increase the benefit obtained as well as on the cheated party to minimise the costs of the cheating (Bronstein, Alarcón & Geber 2006). Such asymmetric symbiosis thus need not automatically result in pure parasitism, as evidenced by the persistence of mutualisms despite cheating (Ferriere et al. 2002).

Both pollinators and plants have been demonstrated to employ cheating behaviour [Figure 2]. “Floral larceny” occurs when animal cheaters obtain nectar without pollination, either as “nectar thieves” which enter flowers in the usual manner but fail to brush against pollen, or as “nectar robbers” which damage the flower corolla and extract nectar bypassing reproductive parts (Irwin, Brody & Waser 2001). These exploiters tend to be opportunistic, rather than evolving from parasitic lineages (Bronstein, Alarcón & Geber 2006). The effect on the fitness of the plants is not always negative, but sometimes neutral or even positive (Irwin, Brody & Waser 2001). For example, a legitimate pollinator visiting a robbed flower will often travel further in pursuit of more nectar, increasing the distance the pollen is taken and thus improving outcrossing levels (Maloof & Inouye 2000). Nevertheless, on balance, floral larceny tends on average have a negative effect on the success of plant reproduction (Irwin, Brody & Waser 2001).

Plants can cheat by way of unrewarding flowers. These attract pollinators by deception (Renner 2006), which is particularly common among orchids (Jersáková, Johnson & Kindlmann 2006; Schiestl 2005). This deception can take a number of forms. Sexual deception has been documented in the genus Cryptostylis (Gaskett & Herberstein 2010). The sole pollinator of Cryptostylis is the male dupe wasp, Lissopimpla excelsa, Ichneumonidae. The male mistakes the Cryptostylis flower [Fig.3] for a female, due to mimicry of both colour (Gaskett & Herberstein 2010) and pheromone (Schiestl, Peakall & Mant 2004), and attempts to mate with it. The subterfuge is so successful that the wasp can even ejaculate during this process (Gaskett, Winnick & Herberstein 2008).

Figure 3: Inflorescence of Cryptostylis subulata.  Scale: 1mm.
Image: Margaret Morgan.

Another form of deception in rewardless flowers is floral mimicry. Here, the plant species produces flowers that closely resemble the rewarding flowers of another species (Johnson, SD 1994), creating convergent phenotypes. The effectiveness of this strategy is strengthened by the physical proximity of high-rewarding “magnet” plants (Johnson, SD et al. 2003).

Rewardless flowers can also attract pollinators by pseudoantagonism, where the flower is “attacked” by an insect fooled into thinking it is another insect encroaching on its territory. Pollen attaches to the insect, and is transferred to another flower during subsequent territorial defensive attacks (Jersáková, Johnson & Kindlmann 2006).

Others attract pollinators by resembling the ovipositioning sites of various insects by mimicking the dung, carrion or fungi in which the larvae grow (Jersáková, Johnson & Kindlmann 2006).
There is strong selection pressure upon pollinators to learn to avoid rewardless flowers and this in turn provides pressure on the plants to finesse their deception. This mechanism is thought to be a driver in the high level of species diversity among orchids, around one third of which pollinate through deception (Cozzolino & Widmer 2005).

Less clear is what the overall advantage to rewardless flower species might be for the cheating strategy. As previously noted, relatively, nectar is not an expensive product for plants to make. Further, rewardless plants generally have low pollinator visitor rates (Jersáková, Johnson & Kindlmann 2006). Rewardless flowers tend to occur in species where there are low population densities and where a single visit is enough to remove all pollen from the flower. Thus it is possible that the mechanism is sufficient to ensure higher rates of cross-fertilisation than self-fertilisation, although there is as yet no evidence suggesting that out-breeding is more common in deceptive orchids (Jersáková, Johnson & Kindlmann 2006).

The resilience of mutualisms

There is a significant asymmetry in cheating rates of plants in comparison to their putative pollinators, with far more foraging insect species displaying exploitation of plants than vice versa. This is thought to reflect the difference in mobility between the partners, with plants being sedentary in contrast with highly mobile insects. Mobility in insects increases the ability to locate suitable alternative food sources or brood sites. Some plants have entirely dispensed with the requirement of biotic pollinators, instead relying on self-fertilisation or abiotic pollination through wind or water (Bronstein, Alarcón & Geber 2006), and it is possible that this change in mechanism is a result of growing loss of reproductive fitness due to cheating. The interaction of symbionts evolves over time, with new associations forming, while others are lost (Kiers et al. 2010). Overall, however, mutualism is evolutionarily stable (Ferriere et al. 2002)

Resilient mutualisms tend to share certain characteristics (Kiers et al. 2010). Facultative interactions, for example, allow participants more options in obtaining services and products (Bronstein 1994). If the mutualism is essential for short term survival, its absence will have profoundly disruptive effects on the symbiont.

Protection from extreme changes in environment also lends stability to mutualisms. Leaf cutter ants, for example, have sustained their symbiosis with the fungus they farm for over 50-million years, and it has been suggested that this is in part because of the protection of the ant nest itself, where environmental conditions are kept in an optimal state (Mueller et al. 2005).

The strict control of symbionts by their partners to ensure “loyalty” is considered a vital element for the persistence of some mutualisms, and models of the methods by which this is policed are considered in the next section of this review. Examples of this control are the propagation by African fungus-growing termites of a single variant of its fungal “crop” (Aanen et al. 2009), and similar behaviour by the leaf cutter ant Acromyrmex octospinosus (Barke et al. 2010). It is suggested that his limitation of the genetic diversity in the symbiont restricts its ability to evolve the ability to cheat (Kiers et al. 2010).

Models of mutualism

A number of models have been developed or applied to explain the maintenance of mutualisms and methods by which exploitation of the relationship is minimised.

The Prisoner’s Dilemma, Iterated Prisoner’s Dilemma and Tit for Tat.

The prisoner’s dilemma is a model derived from Game Theory, which explores simplified options and their potential benefit or loss for the participants. The iterated version of the problem is the same scenario multiply repeated, and was formalised by Axelrod and Hamilton (1981).  In its original formulation, it functions as follows. Two suspects are arrested by police and interrogated separately. Neither will confess. The police have insufficient evidence to charge them, and need at least one to “grass” on the other. Each prisoner has two options:

·         *  Defect from each other (agreeing to give evidence for the prosecution)

·         *  Cooperate with each other (remaining silent)

If one prisoner defects and the other cooperates, the first prisoner will be released and the second will receive the maximum penalty of ten years’ imprisonment. If the first prisoner defects and so does the other, they will each receive five year sentences. If both cooperate (remain silent), the police will not be able to prove their case, and each prisoner receives six months on a minor charge.
Defecting is the best option for each prisoner, given that their goal is to limit the length of time they will be imprisoned. With the iterated version of the game, however, the prisoners’ are able to punish their co-accused for previous choices. In this case, it was found using computer modelling that the optimal strategy was “tit for tat”, in which a player initially cooperates, but then makes a choice each time thereafter which repeats the opponent’s last move. Under this scenario, both parties ultimately benefit the most. This idea was applied as a possible mechanism explaining intraspecies cooperation, (Axelrod & Hamilton 1981), and has since been extended to explore interspecies cooperation. Unfortunately, the model has failed to replicate what is seen in the natural world in symbiotic relationships (Bshary & Bronstein 2004). One explanation for this is that the model has an all-or-nothing structure which does not reflect the complex gradients of real biological interactions (Hoeksema & Bruna 2000). A multiscale approach that incorporates classes of individuals (species) and population fluctuations into the iterated prisoner’s dilemma has been suggested (Doebeli & Knowlton 1998). This version allows for a range of possible responses, based on the variable amount of benefit a symbiont is able to provide according to its resources. This is compatible with Partner Fidelity Feedback (see below), and it appears to predict at least some real-world phenomena with accuracy, such as strong asymmetry in benefits given (Bshary & Bronstein 2004).


Biomarkets, or biological market theory, is another game theoretic model, and uses as its basis the idea that mutualism can be analysed as a trading of goods or commodities in a market where the parties differ in their ease of producing the benefit sought by the other. Partner choice is an element in this model, and it is posited that such choice is a mechanism by which investment is made in the other symbiont, explaining the persistence of mutualism. Failure to provide risks a cheater being abandoned by the choosing partner (Bshary & Bronstein 2004; Noë 2001).  This is consistent with the host sanctions model, below.

By-Product Mutualism

This is an unusual model, because it repudiates the concept usually thought to be at the heart of mutualism: cooperation. By-product mutualism suggests that the only benefits that a symbiont provides to its partner are simply unwanted by-products of its own function. There is no cost in such an arrangement (Brown 1983).

Host Sanctions and Partner Fidelity Feedback

Host Sanctions (HS) and Partner Fidelity Feedback (PFF) are models which seek to explain how a mutualist responds to a cheating partner, with the effect of ameliorating such cheating behaviour in the future (Bull & Rice 1991; West et al. 2002). They have been compared to two competing systems of criminal justice (Weyl et al. 2010): HS operates by punishment as a deterrent against future bad behaviour (by selecting for “evolutionary improvement” in the pool of potential symbionts). The analogy for PFF, by contrast, is a more nuanced, non-retributive treatment of wrong-doers, incorporating a sort of “harm minimisation” approach. In biological terms, this operates by the wellbeing of one symbiont being adversely affected by the exploitation of the other, leading to fewer rewards being physically able to be forthcoming to the latter.

An analysis of these competing models has been undertaken using economic contract theory (Weyl et al. 2010). Weyl et al. tested two commonly cited examples of mutualisms (yucca plants and yucca moths; and legume plants and rhizobia bacteria) and concluded that contrary to the common suggestion that these associations incorporate HS, they in fact demonstrate PFF. This research provides a useful means by which to quantify factors previously somewhat nebulous in PFF models.

Criticism of the models

Bronstein makes substantial criticisms of many of the current models (Bronstein 1994, 2001; Bronstein, Alarcón & Geber 2006), not least because the language of “cheating”, “temptation” and “punishment” inadvertently implies an intention and subterfuge on the part of (usually) non-sentient symbionts. Rather than the term “cheater” as suggested by previous researchers (Soberon & Martinez del Rio 1985), she proposes “exploiter” and argues that without careful distinction between types of exploiters, their evolutionary and ecological consequences will be conflated (Bronstein 2001).

“Exploiter species”, according to this proposal, should be divided into “pure exploiters” and “conditional exploiters”. The former entails a pure strategy that encompasses life-long behaviour within a mutualistic species. Sexual deception in orchids, described above, would fall into this category. “Conditional exploiters”, by contrast, can and do act mutualistically, depending on environmental conditions. This includes nectar robbing insects which also can pollinate plants, and ant defenders which sometimes attack their mutualists.

“Conditional exploiters” represent the majority of “cheaters” in mutualistic relationships and are those which are most likely to shift along the spectrum between complete mutualism and unilateral antagonism, or parasitism.


Mutualism is a fundamentally important element of evolution and the development of biodiversity, acting as a driver for both radiation and speciation. It sits on a transitional spectrum of interactions which ranges from pure exploitation to balanced cooperation, and its position is never static. It engages species from all kingdoms and all ecosystems and is undeniably crucial in the evolution and survival of angiosperms, a group pivotal for our species’ survival and that of many others.

While many mutualistic interactions have been identified and to some extent empirically quantified, a single unifying theory explaining the mechanisms leading to the maintenance of mutualisms is lacking to an extent that is disproportionate to the undeniable importance of mutualisms in evolution and ecology. Some of the current models seeking to explain the exploitation of mutualisms suffer from insufficient regard to evolutionary origin, and few succeed in accurately predicting actual behaviours.

Multiple challenges lie ahead in this field. Why, for example, do some species act as exploiters in some environmental contexts and mutualists in others? At what point will a symbiont abandon a mutualistic relationship? How will climate change and increasing habitat fragmentation impact upon the maintenance of mutualisms?

Theory in this area of biological study needs to meld more completely with natural history and empirical observation, and with substantial further research, these questions might be answered.

Aanen, DK, de Fine Licht, HH, Debets, AJM, Kerstes, NAG, Hoekstra, RF & Boomsma, JJ 2009, 'High symbiont relatedness stabilizes mutualistic cooperation in fungus-growing termites', Science, vol. 326, no. 5956, p. 1103.

Addicott, JF, Bronstein, J & Kjellberg, F 1990, 'Evolution of mutualistic life-cycles: yucca moths and fig wasps', Genetics, evolution, and coordination of insect life cycles, pp. 143–61.

Atwell, B, Kriedemann, P & Turnbull, C 1999, Plants in Action: adaptation in nature, performance in cultivation, MacMillan, Melbourne.

Axelrod, R & Hamilton, WD 1981, 'The evolution of cooperation', Science, vol. 211, no. 4489, p. 1390.

Barke, J, Seipke, RF, Gruschow, S, Heavens, D, Drou, N, Bibb, MJ, Goss, RJM, Yu, DW & Hutchings, MI 2010, 'A mixed community of actinomycetes produce multiple antibiotics for the fungus farming ant Acromyrmex octospinosus', BMC biology, vol. 8, no. 1, p. 109.

Bronstein, JL 1994, 'Our current understanding of mutualism', Quarterly Review of Biology, vol. 69, no. 1, pp. 31-51.

—— 2001, 'The exploitation of mutualisms', Ecology Letters, vol. 4, no. 3, pp. 277-87.

Bronstein, JL, Alarcón, R & Geber, M 2006, 'The evolution of plant–insect mutualisms', New Phytologist, vol. 172, no. 3, pp. 412-28.

Brown, JL 1983, Cooperation: a biologist’s dilemma, vol. 13, Advances in the Study of Behavior, Academic Press, New York.

Bshary, R & Bronstein, JL 2004, 'Game Structures in Mutualistic Interactions: What Can the Evidence Tell Us About the Kind of Models We Need?', Advances in the Study of Behavior, vol. 34, pp. 59-101.

Bull, JJ & Rice, WR 1991, 'Distinguishing mechanisms for the evolution of co-operation', Journal of theoretical biology, vol. 149, no. 1, pp. 63-74.

Burke, D & Fulham, BJ 2003, 'An evolved spatial memory bias in a nectar-feeding bird?', Animal Behaviour, vol. 66, no. 4, pp. 695-701.

Cozzolino, S & Widmer, A 2005, 'Orchid diversity: an evolutionary consequence of deception?', Trends in Ecology & Evolution, vol. 20, no. 9, pp. 487-94.

Darwin, C 1859, On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life, New York: D. Appleton.

Dawkins, R 2006, The selfish gene, Oxford University Press, USA.

Doebeli, M & Knowlton, N 1998, 'The evolution of interspecific mutualisms', Proceedings of the National Academy of Sciences of the United States of America, vol. 95, no. 15, p. 8676.

Fenster, CB, Armbruster, WS, Wilson, P, Dudash, MR & Thomson, JD 2004, 'Pollination syndromes and floral specialization', Annual Review of Ecology, Evolution, and Systematics, vol. 35, pp. 375-403.

Ferriere, R, Bronstein, JL, Rinaldi, S, Law, R & Gauduchon, M 2002, 'Cheating and the evolutionary stability of mutualisms', Proceedings of the Royal Society of London. Series B: Biological Sciences, vol. 269, no. 1493, p. 773.

Ford, HA, Paton, DC & Forde, N 1979, 'Birds as pollinators of Australian plants', New Zealand Journal of Botany, vol. 17, pp. 509-19.

Gaskett, AC & Herberstein, ME 2010, 'Colour mimicry and sexual deception by Tongue orchids (Cryptostylis)', Naturwissenschaften, vol. 97, no. 1, pp. 97-102.

Gaskett, AC, Winnick, CG & Herberstein, ME 2008, 'Orchid sexual deceit provokes ejaculation', Natural History, vol. 171, no. 6.

Grimaldi, DA & Engel, MS 2005, Evolution of the Insects, Cambridge Univ Pr.

Hamilton, WD 1964, 'The genetical evolution of social behaviour. I and II.', Journal of theoretical biology, vol. 7, no. 1, pp. 17-52.

Hoeksema, JD & Bruna, EM 2000, 'Pursuing the big questions about interspecific mutualism: a review of theoretical approaches', Oecologia, vol. 125, no. 3, pp. 321-30.

Huxley, TH 1888, 'The struggle for existence and its bearing upon man.', Nineteenth Century, vol. 23, pp. 163–86.

Irwin, RE, Brody, AK & Waser, NM 2001, 'The impact of floral larceny on individuals, populations, and communities', Oecologia, vol. 129, no. 2, pp. 161-8.

Jandér, KC & Herre, EA 2010, 'Host sanctions and pollinator cheating in the fig tree–fig wasp mutualism', Proceedings of the Royal Society B: Biological Sciences, vol. 277, no. 1687, p. 1481.

Jersáková, J, Johnson, SD & Kindlmann, P 2006, 'Mechanisms and evolution of deceptive pollination in orchids', Biological Reviews, vol. 81, no. 02, pp. 219-35.

Johnson, NC, Graham, JH & Smith, FA 1997, 'Functioning of mycorrhizal associations along the mutualism-parasitism continuum', New Phytologist, vol. 135, no. 4, pp. 575-86.

Johnson, SD 1994, 'Evidence for Batesian mimicry in a butterfly pollinated orchid', Biological Journal of the Linnean Society, vol. 53, no. 1, pp. 91-104.

Johnson, SD, Peter, CI, Nilsson, LA & Ågren, J 2003, 'Pollination success in a deceptive orchid is enhanced by co-occurring rewarding magnet plants', Ecology, vol. 84, no. 11, pp. 2919-27.

Kiers, TE, Palmer, TM, Ives, AR, Bruno, JF & Bronstein, JL 2010, 'Mutualisms in a changing world: an evolutionary perspective', Ecology Letters, pp. no-no.

Kozo-Polyansky, BM 1924, Symbiogenesis: a new principle of evolution., Harvard University Press, Cambridge, Massachusetts.

Kropotkin, P 1902, 'Mutual Aid: A Factor of Evolution', in History of Economic Thought Books.

Labandeira, CC 1997, 'Insect mouthparts: ascertaining the paleobiology of insect feeding strategies', Annual Review of Ecology and Systematics, vol. 28, no. 1, pp. 153-93.

Maloof, JE & Inouye, DW 2000, 'Are nectar robbers cheaters or mutualists?', Ecology, vol. 81, no. 10, pp. 2651-61.

Margulis, L 1967, 'On the origin of mitosing cells', Journal of theoretical biology, vol. 14, no. 3, pp. 225-74.

Mueller, UG, Gerardo, NM, Aanen, DK, Six, DL & Schultz, TR 2005, 'The evolution of agriculture in insects', Annual Review of Ecology, Evolution, and Systematics, vol. 36, no. 1, p. 563.

Noë, R 2001, Biological markets: partner choice as the driving force behind the evolution of mutualisms, Economics in nature: Social dilemmas, mate choice and biological markets, Cambridge University Press, Cambridge.

Paracer, S & Ahmadjian, V 2000, Symbiosis: an introduction to biological associations, Oxford University Press, USA.

Pirozynski, KA & Malloch, DW 1975, 'The origin of land plants: a matter of mycotrophism', BioSystems, vol. 6, no. 3, pp. 153-64.

Renner, SS 2006, 'Rewardless flowers in the angiosperms and the role of insect cognition in their evolution', in NM Waser & J Ollerton (eds), Plant-pollinator interactions: from specialization to generalization, University of Chicago Press, Chicago, pp. 123–44.

Sapp, J 1994, Evolution by association : a history of symbiosis, Oxford University Press, New York.

Schiestl, FP 2005, 'On the success of a swindle: pollination by deception in orchids', Naturwissenschaften, vol. 92, no. 6, pp. 255-64.

Schiestl, FP, Peakall, R & Mant, J 2004, 'Chemical communication in the sexually deceptive orchid genus Cryptostylis', Botanical Journal of the Linnean Society, vol. 144, no. 2, pp. 199-205.

Soberon, JM & Martinez del Rio, C 1985, Cheating and taking advantage in mutualistic interactions, The biology of mutualism Oxford, New York.

Spencer, H 1904, The principles of biology, D. Appleton and company.

Stadler, B & Dixon, AFG 2008, Mutualism: ants and their insect partners, Cambridge Univ Pr.

Tarachai, Y, Compton, SG & Trisonthi, C 2008, 'The benefits of pollination for a fig wasp', Symbiosis, vol. 45, no. 1-3, pp. 29-32.

West, SA, Kiers, ET, Pen, I & Denison, RF 2002, 'Sanctions and mutualism stability: when should less beneficial mutualists be tolerated?', Journal of Evolutionary Biology, vol. 15, no. 5, pp. 830-7.

Weyl, EG, Frederickson, ME, Yu, DW & Pierce, NE 2010, 'Economic contract theory tests models of mutualism', Proceedings of the National Academy of Sciences.

Wilson, GWT, Rice, CW, Rillig, MC, Springer, A & Hartnett, DC 2009, 'Soil aggregation and carbon sequestration are tightly correlated with the abundance of arbuscular mycorrhizal fungi: results from long term field experiments', Ecology Letters, vol. 12, no. 5, pp. 452-61.