Long-term changes in coral colony size distributions on Kenyan reefs under different management regimes and across the 1998 bleaching event

Colony size is an important life-history characteristic of corals and changes in colony size will have significant effects on coral populations. This study summarizes ∼21,000 haphazard colony size measurements of 26 common coral taxa (mostly coral genera) collected annually between 1992 and 2006 in seven Kenyan reef lagoons. There was a major coral bleaching and mortality event in early 1998 and all seven reefs were affected. The seven locations include two long-protected Marine National Parks (Malindi and Watamu), one relatively recently established park (Mombasa), and four unprotected locations (Vipingo, Kanamai, Ras Iwatine, and Diani). They span about 150 km and represent three distinct fishery management regimes: old protected (OP), newly protected (NP), and unprotected (UP). Seventeen taxa had statistically significant different sizes for comparisons of the management regimes, with only one genus, Pavona, having larger sizes in the unprotected reefs. The size of eight coral genera showed a significant time and management interaction, and size frequency differences that existed in management areas prior to 1998 were further increased after the bleaching event. Time alone was a significant factor for eleven genera, and in all cases colonies were smaller after 1998. For most taxa, colony size distributions were significantly skewed and had right-tailed distributions. After 1998, the right-tailed distributions of Acropora, Hydnophora, and Montipora were significantly reduced. Most taxa had peaky distributions and only Acropora experienced a statistically significant change from peaky to flat. The mean sizes of taxa were not related to their mortality across 1998, which indicates that the size effect was within rather than between taxa. Astreopora and Platygyra were well-sampled taxa that did not show an effect of management, but had reduced median sizes across 1998. Consequently, no taxa were tolerant of both fishing and bleaching disturbances and the combined effect was to reduce the size of all corals.     

Sponge-inhabiting barnacles on Red Sea coral reefs

In this study eight different species of barnacles were found within nine species of sponges from the Red Sea. This brings to 11 the number of sponge-symbiotic barnacles reported from the Red Sea, two of these are new Acasta species (not described herein) and one (A. tzetlini Kolbasov) is a new record for this sea. This number is much higher than that of symbiotic barnacles found within sponges from either the N. Atlantic (2) or the Mediterranean (4). Two possible explanations for this are the presence of numerous predators in coral reefs and scarcity of available substrate for settlement. These factors can lead to high incidence of symbiotic relationships. Of the nine sponge species, only one (Suberites cf. clavatus) had previously been known to contain barnacles. Even at the family level, this is the first record of symbiotic barnacles in two out of the seven sponge families (Latrunculiidae, Theonellidae). Our present findings strengthen the apparent rule that the wider the openings in a barnacle shell, the fewer the host taxa with which it will associate, usually from one or two closely related families, and the more frequent it will associate with elastic sponges. Most Neoacasta laevigata found on Carteriospongia foliascens were located on the same side as the sponge's ostia, i.e. facing the incoming water. This adaptation allows the barnacles to catch more suspended particles from the water, provides them with more oxygen and prevents their exposure to discharged sponge waste. The highest density of barnacles observed on one face of a “leaf ” (with ostia) was 0.389 barnacles cm⁻² (one barnacle per 2.57 cm²) and on average 0.181 ± 0.68, while the average on the other side was only 0.068 ± 0.52 barnacles cm⁻². As indicated by the Morisita index, these barnacles most frequently (58%, n = 12) had a clumped spatial distribution (while the rest were randomly distributed), as is to be expected from such sessile organisms with internal fertilization via copulation. The presence of N. laevigata induced the growth of secondary perpendicular projections of its host C. foliascens. Of the N. laevigata examined, 17% brooded 324 ± 41 embryos each, of 286 ± 17 μm total length; only 5.7% (n = 123) were found to be dead. Size distribution analysis of skeletal elements from dead barnacles showed them to be significantly larger than the skeletal elements of the population of live barnacles ( p < 0.05). Received: 26 June 1998 / Accepted: 1 December 1998     

Expanding marine protected areas to include degraded coral reefs

Marine protected areas (MPAs) are a commonly applied solution to coral reef degradation, yet coral reefs continue to decline worldwide. We argue that expanding the range of MPAs to include degraded reefs (DR‐MPA) could help reverse this trend. This approach requires new ecological criteria for MPA design, siting, and management. Rather than focusing solely on preserving healthy reefs, our approach focuses on the potential for biodiversity recovery and renewal of ecosystem services. The new criteria would help identify sites with the highest potential for recovery and the greatest resistance to future threats (e.g., increased temperature and acidification) and sites that contribute to MPA connectivity. The DR‐MPA approach is a compliment rather than a substitute for traditional MPA design approaches. We believe that the DR‐MPA approach can enhance the natural, or restoration‐assisted, recovery of DRs and their ecosystem services; increase total reef area available for protection; promote more resilient and better‐connected MPA networks; and improve conditions for human communities dependent on MPA ecosystem services.     

Diversity and stability of herbivorous fishes on coral reefs

Biodiversity may provide insurance against ecosystem collapse by stabilizing assemblages that perform particular ecological functions (the "portfolio effect"). However, the extent to which this occurs in nature and the importance of different mechanisms that generate portfolio effects remain controversial. On coral reefs, herbivory helps maintain coral dominated states, so volatility in levels of herbivory has important implications for reef ecosystems. Here, we used an extensive time series of abundances on 35 reefs of the Great Barrier Reef of Australia to quantify the strength of the portfolio effect for herbivorous fishes. Then, we disentangled the contributions of two mechanisms that underlie it (compensatory interactions and differential responses to environmental fluctuations ["response diversity"]) by fitting a community-dynamic model that explicitly includes terms for both mechanisms. We found that portfolio effects operate strongly in herbivorous fishes, as shown by nearly independent fluctuations in abundances over time. Moreover, we found strong evidence for high response diversity, with nearly independent responses to environmental fluctuations. In contrast, we found little evidence that the portfolio effect in this system was enhanced by compensatory ecological interactions. Our results show that portfolio effects are driven principally by response diversity for herbivorous fishes on coral reefs. We conclude that portfolio effects can be very strong in nature and that, for coral reefs in particular, response diversity may help maintain herbivory above the threshold levels that trigger regime shifts.     

Measuring settlement intensity of echinoderms on coral reefs

A pilot study was conducted at Davies Reef on the central Great Barrier Reef between 20 November 1991 and 20 January 1992 to assess the suitability of settlement collectors for measuring larval settlement rates of echinoderms on coral reefs. The collectors were deployed for two months during summer when many echinoderm species are known to spawn. A total of 657 newly settled echinoderms were recovered from just 47 collectors, each having a volume of just 0.005 m³ but with a utilisable surface area of about 1.44 m², indicating that competent-to-settle larvae were very abundant during the sampling period. Echinoids (7.8 sampler^-1) and ophiuroids (5.6 sampler^-1) were the most abundant groups on the collectors. Asteroids (0.2 sampler^-1) and crinoids (0.3 sampler^-1 were less common. The asteroids were all identified to species and included the crown-of-thorns starfish Acanthaster planci, Choriaster granulatus and Culcita novaeguineae. The newly settled echinoids could not be distinguished from each other but included Echinometra mathaei and Mespilia globulus. The abundance of each of the five different classes of echinoderms on the samplers was correlated with their abundance on the natural substratum. Significant spatial variability was found in settlement rates of echinoids over 1000s of metres and ophiuroids over 100s of metres, but not over smaller spatial scales. It is concluded that the collectors can provide a useful tool for monitoring spatial and temporal variability in settlement rates of echinoderms on coral reefs and for testing hypotheses about patterns of larval dispersal.     

The trophic status of herbivorous fishes on coral reefs

Dietary analyses of 17 species of nominally herbivorous fishes on the northern Great Barrier Reef were carried out to resolve the questions to what extent diet in nominally herbivorous reef fishes is dominated by living plant material and whether short-chain fatty acid profiles in the alimentary tract predict diet. The fishes included members of the families Acanthuridae, Scaridae and Kyphosidae. The analyses revealed consumption of a wide range of food resources including macroscopic algae, turfing algae, pelagic and planktonic animal matter, organic detritus and sediments. In only 7 of the 17 species was living algae the dominant dietary item. Gastrointestinal short-chain fatty acids (SCFA) profiles reflecting the fermentative activities of gut symbionts were used as a framework for the initial analysis of dietary patterns. The feeding patterns implied by the SCFA profiles were validated by direct observation of food material through gut content analysis. We identified four main trophic groupings differentiated by fermentation profiles and diet. These were: (1) species with low levels of SCFA in gut samples (mean mM 9.2ǂ.8) but a high proportion of isovalerate (mean 15.4%dž.7): Acanthurus nigricauda, A. olivaceus, Ctenochaetus striatus, Chlorurus microrhinos, C.sordidus, Scarus schlegeli; (2) species that fed on turfing, filamentous and small thallate algae and displayed moderate levels of SCFA (mean mM 21.9DŽ.9): Acanthurus nigricans, A.lineatus, Naso tuberosus; (3) species that fed on large thallate (macroscopic) algae with high levels of SCFA (mean mM 41.2Dž.5) with very low proportions of isovalerate (mean 0.5%ǂ.1): Kyphosus cinerascens, K.vaigiensis, Naso unicornis, Zebrasoma scopas; and (4) species with diets dominated by planktonic animal material with moderate levels of SCFA (mean mM 21.3ǃ.5): Naso annulatus, N. brevirostris, N. hexacanthus, N. vlamingii. We conclude that a wide range of dietary items are taken by the 17 nominal herbivores, and that dietary groupings do not reflect taxonomic relationships. There was evidence of convergence in feeding modes and diet between phylogenetically distinct taxa and divergence within particular lineages.     

Parrotfish predation on cryptic sponges of Caribbean coral reefs

Some sponge species that live in crevices in the reef frame appear to be restricted to their cryptic habitat by predation. When cryptic sponges were excavated, on Guigalotupo reef, San Blas, Panama, exposing them to potential predators, they were eaten by fishes that are generally considered to be herbivores, primarily parrotfishes of the genus Sparisoma: S.aurofrenatum (Cuvier & Valenciennes), S.viride (Bonnaterre), and S.chrysopterum (Bloch & Schneider). Of the 9150 bites observed to be taken by these species during paired (i.e., with sponges versus without sponges) trials conducted in defined feeding areas during 1986, 1987, and 1988, 72% (i.e. 6581 bites) were on cryptic sponges, even though these were only offered during half of the total observation time and never constituted more than 7% of the cover of the feeding observation areas. Individual parrotfish returned over and over to take bites of the exposed cryptic sponges until they were entirely consumed. They vigorously chased each other away from the sponges, but exhibited no such defense of their usual algal foods. A total of 18 sponge species were tested. Of the cryptic and semi-cryptic sponge species tested, only one of six was rejected by the parrotfish. Two of these six sponge species were consistently consumed entirely, and two were consumed entirely whenever their surfaces were sliced off with a razor blade, demonstrating that these sponges concentrate defenses against predators in their surfaces. One semi-cryptic species and one semi-exposed species were fed upon, but not entirely consumed. By contrast, 11 of 12 of the exposed and semi-exposed species were rejected. Cryptic sponges grew out of their cavities in the reef only when protected by seasonally thick mats of macroalgae or by cages that excluded fish. Received: 10 January 1997 / Accepted: 4 February 1997     

The trophic status of herbivorous fishes on coral reefs

Estimates of feeding rates, alimentary tract structure and temporal patterns of food processing obtained from twelve species of nominally herbivorous fishes on the northern Great Barrier Reef were compared. These included members of the families Acanthuridae, Scaridae and Kyphosidae. Based on an analysis of diet and short-chain fatty acid (SCFA) profiles from a previous study we initially partitioned the twelve species into four dietary categories, as follows: (a) Category 1: herbivores with a diet of macroscopic brown algae and high SCFA profiles in the hindgut region (Naso unicornis, Kyphosus vaigiensis); (b) Category 2: herbivores feeding on turfing and filamentous red and green algae with moderate SCFA profiles in the hindgut region (N. tonganus, K. cinerascens, Zebrasoma scopas, Acanthurus lineatus); (c) Category 3: zooplankton feeders with moderate SCFA profiles (N. vlamingii, N. brevirostris); (d) Category 4: species feeding on detrital and sedimentary materials with low levels of SCFA (Chlorurus microrhinos, Scarus schlegeli, Ctenochaetus striatus, A. olivaceus). The purpose of this comparison was to determine whether measures of feeding activity, alimentary tract structure, and food processing were concordant with diet. A dichotomy in feeding rates was observed. Species with a diet of algae and zooplankton (categories 1–3) had slower feeding rates than those feeding on detrital aggregates and sediment (category 4). The pattern of food processing also followed the same dichotomy with species of categories 1–3 retaining food in the alimentary tract overnight and commencing the feeding day with substantial amounts of food in the intestine and hindgut. Category-4 species commenced the feeding day with empty alimentary tracts suggesting a rapid turnover of gut contents. Within the herbivorous and zooplankton-feeding species neither alimentary tract structure nor food processing mode were predicted by diet or SCFA profiles. A hindgut fermentation chamber was present in K. vaigiensis but not in N. unicornis, a species with high levels of SCFA in the hindgut region and a diet of brown macroscopic algae. In contrast N. vlamingii, with a diet dominated by animal matter, retained large amounts of food material in a hindgut chamber over the entire feeding cycle. In tropical perciform fishes, herbivory and fermentation are not associated with the alimentary tract structures that characterise herbivorous terrestrial vertebrates. Estimates of the abundance of the different groupings of nominally herbivorous fishes indicated that the dominant elements in the reef grazing and browsing fauna were consumers of detrital and sedimentary materials. These could not be classified as herbivores. Members of this group were dominant in all habitats investigated. Explicitly herbivorous taxa were a minority component in all habitats investigated.Communicated by G.F. Humphrey, Sydney     

Catastrophe and the life span of coral reefs

A strong earthquake in the western Caribbean in 2009 had a catastrophic impact on uncemented, unconsolidated coral reefs in the central sector of the shelf lagoon of the Belizean barrier reef. In a set of 21 reef sites that had been observed prior to the earthquake, the benthic assemblages of 10 were eradicated, and one was partially damaged, by avalanching of their slopes. Ecological dynamics that had played out over the previous 23 years, including the mass mortalities of two sequentially dominant coral species and a large increase in the cover of an encrusting sponge, were instantaneously rendered moot in the areas of catastrophic reef-slope failure. Because these prior dynamics also determined the benthic composition and resilience of adjacent sections of reef that remained intact, the history of disturbance prior to the earthquake will strongly influence decadal-scale recovery in the failed areas. Geological analysis of the reef framework yielded a minimum return time of 2000-4000 years for this type of high-amplitude event. Anthropogenic degradation of ecosystems must be viewed against the backdrop of long-period, natural catastrophes, such as the impact of strong earthquakes on uncemented, lagoonal reefs.     

Epiphytic cyanobacteria maintain shifts to macroalgal dominance on coral reefs following ENSO disturbance

Macroalgal dominance of some tropical reef communities in the Eastern Pacific after coral mortality during the 1997-1998 El Ni?o Southern Oscillation (ENSO) was facilitated by protection from herbivory by epiphytic cyanobacteria. Our results do not support that reduction in number of herbivores was a necessary precursor to coral reef decline and shifts to algal reefs in this system. Rather, macroalgae dominated the community for several years after this pulse disturbance with no concurrent change in herbivore populations. While results of microcosm experiments identified the importance of nutrients, especially phosphorus, in stimulating macroalgal growth, nutrient supply alone could not sustain macroalgal dominance as nutrient-stimulated growth rates in our in situ experiments never exceeded consumption rates of unprotected thalli. In addition, thalli with nutrient-enriched tissue were preferentially consumed, possibly negating the positive effects of nutrients on growth. These tropical reefs may be ideal systems to conduct experimental tests distinguishing phase shifts from alternative stable states. Shifts were initiated by a large-scale disturbance with no evidence of a changing environment except, perhaps, dilution in herbivory pressure due to increased algal cover. Community establishment was most likely stochastic, and the community was likely maintained by strongly positive interaction between macroalgal hosts and cyanobacterial epiphytes that uncoupled consumer control of community structure.     

Temporal and spatial variability in coral recruitment on two Indonesian coral reefs: consistently lower recruitment to a degraded reef

Corals are the primary reef-building organisms, therefore it is key to understand their recruitment patterns for effective reef management. Coral recruitment rates and juvenile coral abundance were recorded in the Wakatobi National Marine Park, Indonesia, on two reefs (Sampela and Hoga) with different levels of environmental degradation (12.5 vs. 44 % coral cover with high and low sedimentation rates, respectively) to examine consistencies in recruitment patterns between years and seasons. Recruitment was measured on multiple panels at two sites on each reef (6–7 m depth) and cleared areas of natural reef. Although coral recruitment was twofold higher in 2008–2009 than in 2007–2008, and seasonal differences were identified, consistent significant differences in recruitment rates were found between the two reefs even though they are separated by only ~1.5 km. Recruitment rates and juvenile abundance were lower on the more degraded reef. These patterns are likely a consequence of differential pre- and post-settlement mortality as a result of the high sedimentation rates and degraded conditions and possibly reduced larval supply.     

Observations on organic aggregates in the vicinity of coral reefs

The abundance of organic aggregates is reported for a variety of coral-reef sites and surrounding waters. There is no net increase of aggregates forming from dissolved organic matter and streaming off the reefs into the lagoons, though the organic components and the aeration of this water might be conducive to aggregate formation. Also, although the waters flowing from the surrounding oceans across the reefs and into the lagoons contribute organic components, the concentration of organic aggregates in the latter habitats is not in excess of quantities in surrounding oceanic waters.     

Herbivory by the Caribbean king crab on coral patch reefs

Caribbean coral reefs are increasingly dominated by macroalgae instead of corals due to several factors, including the decline of herbivores. Yet, virtually unknown is the role of crustacean macrograzers on coral reef macroalgae. We examined the effect of grazing by the Caribbean king crab (Mithrax spinosissimus) on coral patch reef algal communities in the Florida Keys, Florida (USA), by: (1) measuring crab selectivity and consumption of macroalgae, (2) estimating crab density, and (3) comparing the effect of crab herbivory to that of fishes. Mithrax prefers fleshy macroalgae, but it also consumes relatively unpalatable calcareous algae. Per capita grazing rates by Mithrax exceed those of most herbivorous fish, but Mithrax often occurs at low densities on reefs and its foraging activities are reduced in predator-rich environments. Therefore, the effects of grazing by Mithrax tend to be localized and when at low density contribute primarily to spatial heterogeneity in coral reef macroalgal communities.     

A model-based approach to determine the long-term effects of multiple interacting stressors on coral reefs

The interaction between multiple stressors on Caribbean coral reefs, namely, fishing effort and hurricane impacts, is a key element in the future sustainability of reefs. We develop an analytic model of coral-algal interactions and explicitly consider grazing by herbivorous reef fish. Further, we consider changes in structural complexity, or rugosity, in addition to the direct impacts of hurricanes, which are implemented as stochastic jump processes. The model simulations consider various levels of fishing effort corresponding to' several hurricane frequencies and impact levels dependent on geographic location. We focus on relatively short time scales so we do not explicitly include changes in ocean temperature, chemistry, or sea level rise. The general features of our approach would, however, apply to these other stressors and to the management of other systems in the face of multiple stressors. It is determined that the appropriate management policy, either local reef restoration or fisheries management, greatly depends on hurricane frequency and impact level. For sufficiently low hurricane impact and macroalgal growth rate, our results indicate that regions with lower-frequency hurricanes require stricter fishing regulations, whereas management in regions with higher-frequency hurricanes might be less concerned with enhancing grazing and instead consider whether local-scale restorative activities to increase vertical structure are cost-effective.     

A geological perspective on the degradation and conservation of western Atlantic coral reefs

Continuing coral‐reef degradation in the western Atlantic is resulting in loss of ecological and geologic functions of reefs. With the goal of assisting resource managers and stewards of reefs in setting and measuring progress toward realistic goals for coral‐reef conservation and restoration, we examined reef degradation in this region from a geological perspective. The importance of ecosystem services provided by coral reefs—as breakwaters that dissipate wave energy and protect shorelines and as providers of habitat for innumerable species—cannot be overstated. However, the few coral species responsible for reef building in the western Atlantic during the last approximately 1.5 million years are not thriving in the 21st century. These species are highly sensitive to abrupt temperature extremes, prone to disease infection, and have low sexual reproductive potential. Their vulnerability and the low functional redundancy of branching corals have led to the low resilience of western Atlantic reef ecosystems. The decrease in live coral cover over the last 50 years highlights the need for study of relict (senescent) reefs, which, from the perspective of coastline protection and habitat structure, may be just as important to conserve as the living coral veneer. Research is needed to characterize the geological processes of bioerosion, reef cementation, and sediment transport as they relate to modern‐day changes in reef elevation. For example, although parrotfish remove nuisance macroalgae, possibly promoting coral recruitment, they will not save Atlantic reefs from geological degradation. In fact, these fish are quickly nibbling away significant quantities of Holocene reef framework. The question of how different biota covering dead reefs affect framework resistance to biological and physical erosion needs to be addressed. Monitoring and managing reefs with respect to physical resilience, in addition to ecological resilience, could optimize the expenditure of resources in conserving Atlantic reefs and the services they provide.     

Appraisal of visual assessments of habitat complexity and benthic composition on coral reefs

Visual assessments of topographic habitat structure and benthos on coral reefs were appraised using quantitative data collected from 16 replicate surveys within each of 21 sites on Seychelles reefs. Results from visual assessments of reef benthos were similar to those obtained using techniques frequently used to assess benthic complexity and composition. Visual estimates of habitat topography were correlated with rugosity, reef height and holes of 10–70 cm diameter, whilst visual estimates of benthic composition were very similar to those obtained from line intercept transects. Visual estimates of topography correlated strongly with species richness of fish communities and explained 42% of the variation in these data. The relationship between visual estimates of topography and species richness is strongest with fish 10–30 cm total length (TL), abundance of fish within this size category also correlating positively with topographic visual assessments. Visual techniques are prone to observer bias, however with regular training they can be used to quickly provide a reliable and effective means of assessing habitat complexity and benthos on coral reefs.     

Diversity and community structure of symbiotic dinoflagellates from Caribbean coral reefs

A community ecology approach to the study of the most common group of zooxanthellae, dinoflagellates in the genus Symbiodinium, was applied to symbiotic invertebrate assemblages on coral reefs in the western Caribbean, off the Yucatan peninsula (Puerto Morelos, Mexico) and over 1000 km away in the northeastern Caribbean, at Lee Stocking Island, Bahamas. Sequence differences and intragenomic variation, as determined by denaturing gradient gel electrophoresis and sequencing of the internal transcribed spacer 2 (ITS 2) region, were used to classify these symbionts. Twenty-eight genetically distinct Symbiodinium types were identified, eleven of which were found in hosts from both Caribbean locations. A single symbiont population was detected in 72% of hosts from the Yucatan and 92% of hosts from the Bahamas. The reef-wide community distribution of these symbionts is dominated by a few types found in many different host taxa, while numerous rare types appear to have high specificity for a particular host species or genus. Clade or lineage A Symbiodinium spp. was restricted to compatible hosts located within 3-4 m of the surface, while Symbiodinium spp. types from other lineages displayed differences in vertical zonation correlated with ITS type but were independent of clade designation. A comparison of the symbiont types found in field-collected hosts with types previously cultured from these hosts indicates the existence of low density or "background"-symbiont populations and cryptic, potentially non-mutualistic types in some hosts.     

Transmitter attachment and release methods for short-term shark and stingray tracking on coral reefs

This study details a simple and cost-effective means of attaching acoustic transmitters to coral reef sharks and stingrays, which potentially allows for retrieval and reuse on completion of tracks. Between 2008 and 2011, galvanised timed releases were trialled in both static field tests and on blacktip reef sharks Carcharhinus melanopterus, cowtail Pastinachus atrus, and porcupine Urogymnus asperrimus rays in Coral Bay (?23°08′41″, 113°45′53″), Western Australia. The timed releases remained attached to animals for the duration required for tracking and in four out of five deployments transmitters were recovered after release from the animals. The use of modified Rototags for sharks and stainless steel darts for stingrays allowed rapid and effective attachment to animals, with limited impact on their welfare in the short term. External attachment for short-term tracking of coral reef-associated elasmobranchs should be considered as a complementary option to internal placement of transmitters in animals either by surgery or by ingestion.     

Habitat biodiversity as a determinant of fish community structure on coral reefs

Increased habitat diversity is often predicted to promote the diversity of animal communities because a greater variety of habitats increases the opportunities for species to specialize on different resources and coexist. Although positive correlations between the diversities of habitat and associated animals are often observed, the underlying mechanisms are only now starting to emerge, and none have been tested specifically in the marine environment. Scleractinian corals constitute the primary habitat-forming organisms on coral reefs and, as such, play an important role in structuring associated reef fish communities. Using the same field experimental design in two geographic localities differing in regional fish species composition, we tested the effects of coral species richness and composition on the diversity, abundance, and structure of the local fish community. Richness of coral species overall had a positive effect on fish species richness but had no effect on total fish abundance or evenness. At both localities, certain individual coral species supported similar levels of fish diversity and abundance as the high coral richness treatments, suggesting that particular coral species are disproportionately important in promoting high local fish diversity. Furthermore, in both localities, different microhabitats (coral species) supported very different fish communities, indicating that most reef fish species distinguish habitat at the level of coral species. Fish communities colonizing treatments of higher coral species richness represented a combination of those inhabiting the constituent coral species. These findings suggest that mechanisms underlying habitat-animal interaction in the terrestrial environment also apply to marine systems and highlight the importance of coral diversity to local fish diversity. The loss of particular key coral species is likely to have a disproportionate impact on the biodiversity of associated fish communities.