Dispersal, and the underlying movement behaviour, are processes of pivotal importance for understanding and predicting metapopulation and metacommunity dynamics. Generally, dispersal decisions are non-random and rely on information, such as the presence of conspecifics. However, studies on metacommunities that include interspecific interactions generally disregard information use. Therefore, it remains unclear whether and how dispersal in metacommunities is informed and whether rules derived from single-species contexts can be scaled up to (meta-)communities. Using experimental protist metacommunities, we show how dispersal and movement are informed and adjusted by the strength of inter-specific interactions. We found that predicting informed movement and dispersal in metacommunities requires knowledge on behavioural responses to intra-and inter-specific interaction strength. Consequently, metacommunity dynamics inferred directly from single-species metapopulations without taking inter-specific interactions into account are likely flawed. Our work identifies the significance of information use for understanding metacommunity dynamics, stability and the coexistence and distribution of species.

Plant-animal mutualistic networks are highly diverse and structured. This has been explained by coevolution through niche based processes. However, this explanation is only warranted if neutral processes (e.g. limited dispersal, genetic and ecological drift) cannot explain these patterns. Here we present a spatially explicit model based on explicit genetics and quantitative traits to study the connection between genome evolution, speciation and plant-animal network demography. We consider simple processes for the speciation dynamics of plant-animal mutualisms: ecological (dispersal, demography) and genetic processes (mutation, recombination, drift) and morphological constraints (matching of quantitative trait) for species interactions, particularly mating. We find the evolution of trait convergence and complementarity and topological features observed in real plant-animal mutualistic webs (i.e. nestedness and centrality). Furthermore, the morphological constraint for plant reproduction generates higher centrality among plant individuals (and species) than in animals, consistent with observations. We argue that simple processes are able to reproduce some well known ecological and evolutionary patterns of plant-animal mutualistic webs. IntroductionSince Darwin's book "On The Origin of Species" (Darwin, 1862b), the idea of coevolution 1 has sparked interest from biologists trying to understand how species interactions generate trait changes. The first clear indication of coevolution was Darwin's moth example (Darwin, 1862a) showing that the long corolla from the orchid Angraecum sesquispedale could only be reached by a pollinator species with a similar proboscis length. However, much later Janzen (1980) argued that this amazingly high specialization between plants and animals was not the only example of coevolution. He explained that coevolution can also be the product of multiplespecies interactions, a term that he coined "diffuse coevolution". Diffuse coevolution means that selection on traits is determined by the interaction of all species in the community and not only * Corresponding author: franencinas@gmail.com 1 defined as reciprocal evolutionary change between species 1. CC-BY-NC-ND 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint . http://dx.doi.org/10.1101/007393 doi: bioRxiv preprint first posted online Jul. 23, 2014; based on pair-wise interactions. This is based on the idea of pollination or dispersal "syndromes", where plants have a set of traits that attract a specific group of pollinator or animal seed-disperser species.Later, the idea of "diffuse coevolution" was related to patterns of nestedness detected across biogeographic regions in mutualistic networks. Nestedness, defined as a non-random pattern of interactions where specialist species interact with proper subsets of more generalist species puts the concept of "diffuse co...