We examine the effects of different biogeographic histories on assemblage composition in three major marine habitats in two biogeographically distinct marine realms. Specifically, we quantify the taxonomic and functional composition of fish assemblages that characterise coral reef, seagrass and mangrove habitats, to explore the potential effects of biogeographic history and environment on assemblage composition. The three habitats were surveyed in the Caribbean and on the Great Barrier Reef using a standardised underwater visual census method to record fish size and abundance data. The taxonomic composition of assemblages followed biogeographic expectations, with realm-specific family-level compositions. In marked contrast, the functional composition of assemblages separated habitats regardless of their biogeographic locations. In essence, taxonomy characterises biogeographic realms while functional groups characterise habitats. The Caribbean and Indo-West Pacific have been separated for approximately 15 million years. The two realms have different taxonomic structures which reflect this extended separation, however, the three dominant shallow-water marine habitats all retain distinct functional characteristics: seagrass fishes are functionally similar regardless of their taxonomic composition or biogeographic location. Likewise, for coral reefs and mangroves. The results emphasise the advantages and limitations of taxonomic vs. functional metrics in evaluating patterns. Taxonomy primarily reflects biogeographic and evolutionary history while functional characteristics may better reflect ecological constraints. IntroductionHistorical and ecological biogeographic processes have a profound effect on the distribution of organisms and how they interact with their environment (Wiens and Donoghue 2004). The species-rich tropics offer an ideal setting to study how these mechanisms have shaped the structure of the communities we recognise today (Floeter et al. 2008, Mouillot et al. 2013a. Coral reefs, in particular, have triggered the interest of researchers as they provide an opportunity to study a high-diversity system which operates at global biogeographic scales (Mouillot et al. 2014, Leprieur et al. 2016. This research has underpinned a fast-growing ResearchBiogeographic patterns in major marine realms: function not taxonomy unites fish assemblages in reef, seagrass and mangrove systems

Signal divergence is an important process underpinning the diversification of lineages. Research has shown that signal divergence is greatest in species pairs that possess high geographic range overlap. However, the influence of range‐size differences within pairs is less understood. We investigated how these factors have shaped signal divergence within brightly coloured coral reef butterflyfishes (genus: Chaetodon). Using a novel digital imaging methodology, we quantified both colouration and pattern using 250 000 sample points on each fish image. Surprisingly, evolutionary age did not affect colour pattern dissimilarity between species pairs, with average differences arising in just 300 000 years. However, the effect of range overlap and range symmetry was significant. Species‐pair colour patterns become more different with increasing overlap, but only when ranges are similar in size. When ranges differ markedly in area, species‐pair colour patterns become more similar with increasing overlap. This suggests that species with small ranges may maintain non‐colour‐based species boundaries.

Every animal dies. In nature, mortality usually occurs due to predation by other animals. One of the fundamental consequences of mortality is the transfer of energy and nutrients from one organism (prey) to another (predator). On coral reefs, these key interactions and processes, that are essential for ecosystem functioning, are primarily mediated by fishes; up to 53% of fishes on coral reefs can be regarded as piscivorous. To date, piscivory on coral reefs has been primarily studied with regard to the species piscivores feed on, and how piscivores control populations. Consequently, understanding prey selectivity by piscivorous fishes has been a major goal. However, prey functional traits may also be important in understanding these ‘energy transactions’, especially in complex ecosystems such as coral reefs. Our goal, therefore, was to quantify—at a community level—functional traits of prey that have been shown to influence predator–prey interactions. We found that, on average, deep‐bodied, social fishes occupy higher positions in the water column, whereas solitary species are usually elongate and more closely associated with the benthos. On closer examination, we found that solitary species have a size‐dependent relationship, with substratum associations shifting to water column associations, at approximately 50 mm body length. Our results reveal three distinct prey functional groups: cryptobenthic substratum dwellers, solitary epibenthics and social fishes. These groups display significant differences in their morphologies and behaviours. Furthermore, based on a meta‐analysis of published mortality rates of small‐bodied (<100 mm TL) reef fishes, we show that the three groups display different mortality rates, possibly due to differential exposure to, and potential to be captured by, different predator types. Although fishes are widely available on coral reefs, they may not be equally available as prey to all piscivore types. Prey are not simply victims; they are capable of influencing potential predation through functional traits. A free Plain Language Summary can be found within the Supporting Information of this article.