Between antagonism and mutualism: costs and benefits in a nursery pollination system

Abstract

A hundred years after Darwin predicted the existence of a specific hawkmoth that could drink the nectar and pollinate the Malagasy star orchid, coevolution was defined as “an evolutionary change in a trait of the individuals in one population in response to a trait of the individuals of a second population, followed by an evolutionary response by the second population to the change in the first”. Since then, the demonstration that adaptations in both partner species are a result of reciprocal selection became a requirement to associate a certain pattern to coevolution. These relationships between species are dynamic and continually evolve as natural selection reshapes them. In this way, natural selection and coevolution have led to different types of specialised interactions between species, depending on the effects on fitness of the partners involved. When the net effect is negative to one or both partners the relationship is an antagonism. Antagonisms are diverse and include predation and parasitism, among others. In commensalistic interactions, one species benefits and the other remains unaffected, while in mutualistic interactions both species benefit. Mutualistic interactions vary greatly in the degree of obligacy and specificity, even between mutualistic partners. The study of mutualisms has gained popularity among ecologists, and they are now recognised as providing essential ecosystem services. Several ecological and evolutionary patterns that are shared between diverse forms of mutualisms have been identified. First of all, one of the partners performs a service that benefits its associate, and in turn it receives a reward. However, services provided and rewards produced can also be costly. Hence, natural selection will favour traits that minimise such costs without interfering with the mutualism itself. As long as the benefits outweigh the costs, the association will last. The second pattern observed is conditionality: a change in the outcome of the interaction as a result of variations in the given local biotic or abiotic conditions. Such variation can change the benefit and cost ratio, shifting the interaction along a continuum from mutualism towards antagonism, and vice versa. Conditionality can occur when the services or rewards provided by partners change with age, size, or developmental stage, as well as with environmental factors such as resource availability, presence of competitors and natural enemies, or population densities. Lastly, evidence suggests that partners may be able to reinforce mutualistic actions of their associates to reduce defection of the mutualism and decrease the probability that their partner will become an exploiter or a cheater. Plant-pollinator interactions are some of the most well studied examples of coevolution and mutualism. Plants will experiment natural selection in those traits that will allow them to be located more easily by their partners, such as colourful and scented flowers. In exchange for pollination services, plants provide a nutritious reward, usually nectar or pollen. A specific type of pollination interactions are nursery pollination systems. In these interactions the pollinator also lays eggs in or on the host plant, which becomes a breeding site for its offspring. As the offspring develop and feed from reproductive structures of its host plant, very often seeds, the costs associated may be high. For the interaction to remain mutualistic the benefits must balance out the costs of the interaction, otherwise it would be evolutionarily unstable and eventually turn into a parasitism. Nursery pollination systems vary in the degree of specialisation. Highly specialised, coevolved and obligate mutualisms include the interactions between Ficus plants-fig wasps, and Yucca plants-yucca moths. Other interactions such as the partnership between Greya moth and its host plant show conditionality, with the interaction shifting from mutualistic to commensalistic depending on the abundance of copollinators. Interactions between plants of the Caryophyllaceae family and Hadena moths range from antagonisms to facultative mutualisms. One of the interactions belonging to this complex is formed by host plant Silene latifolia and its pollinator/herbivore partner Hadena bicruris, which is often referred to as parasitic due to the extent of seed predation, and even as a “primitive” stage of nursery pollination mutualisms. Being a facultative mutualism with high conditionality, together with the high fitness costs, makes the S. latifolia-H. bicruris interaction a perfect model system to study the early stages of evolution in nursery pollination mutualisms. The aim of this thesis is to explore under which ecological circumstances interactions may behave as a facultative mutualism, creating the possibility for them to evolve into more specialised interactions. To be able to evaluate the nature of the interaction and understand what are the causes leading to conditional outcomes, it is necessary to know the costs and benefits in the system. The positive contributions of male and female adult moths to seed production in S. latifolia plants were quantified through the examination of moth pollinating behaviour and efficiency with observational assays. This was followed by field data collections and the analysis of pollination, infestation, and parasitism rates by a natural enemy in order to investigate conditionality under field conditions. A theoretical model was applied to determine interaction outcomes and the possible role of copollinators. The results show that H. bicruris male pollination benefits are essential for the stability of the system, but pollination, infestation and parasitism rates highly varied between populations. The amount of copollination was a good predictor for the outcome of the system, being clear that third parties associated to the S.latifolia–H. bicruris system play an important role modifying the outcome of the interaction. Following the field investigation, the potential role of a natural enemy of H. bicruris, the ectoparasitoid Bracon variator, in modifying the interaction outcome between the plant and the herbivore was investigated. In particular, I examined whether the effect of the predator acted as a stabilising mechanism, reducing the level of seed consumption and therefore the costs caused to the plant, and how this in turn influenced individual plant fitness. With a series of laboratory and greenhouse experiments germination, survival, and flower production were measured as proxies for individual plant fitness. The results show that seed dormancy increased significantly in seeds from plants infested with H. bicruris, which could act as a short-term strategy to reduce costs in the interaction. Silene latifolia plants also had a higher seed output in the presence of B. variator, suggesting that B. variator may act as a regulator in the system. However, plant survival and flower production also decreased with higher seed densities. Therefore, an increase in seed output may be less beneficial for plant fitness than estimated from the number of seeds alone. There is a need to discuss whether taking simple proxies of fitness such as seed output is sufficient to determine the net effect of multitrophic interactions. To conclude, the causes leading to conditionality in the S. latifolia-H. bicruris system are linked to male moth availability and abundance of copollinators, as well as plant population size. Moreover, the presence or absence of parasitoids can deeply impact conditionality in the S. latifolia - H. bicruris interaction, although the overall effects for plant fitness are not so clear. Further research needs to be carried out to properly assess the impact of the pollinator/herbivore and parasitoid partners on host plant population dynamics through exclusion experiments and modelling approaches at the tritrophic population level. On the other hand, it is expected that natural selection favours traits that limit predation costs inflicted by the pollinator/herbivore to avoid over-exploitation. In line with this idea, inducing seed dormancy and even investing in high amounts of seed production to compensate for the feeding of its pollinator/herbivore partner, could be possible mechanisms that have been selected for in S. latifolia plants to lower the costs of the interaction.