D. Ross Robertson

Ecological speciation in tropical reef fishes.

The high biodiversity in tropical seas provides a long-standing challenge to allopatric speciation models. Physical barriers are few in the ocean and larval dispersal is often extensive, a combination that should reduce opportunities for speciation. Yet coral reefs are among the most species-rich habitats in the world, indicating evolutionary processes beyond conventional allopatry. In a survey of mtDNA sequences of five congeneric west Atlantic reef fishes (wrasses, genus Halichoeres) with similar dispersal potential, we observed phylogeographical patterns that contradict expectations of geographical isolation, and instead indicate a role for ecological speciation. In Halichoeres bivittatus and the species pair Halichoeres radiatus/brasiliensis, we observed strong partitions (3.4% and 2.3% divergence, respectively) between adjacent and ecologically distinct habitats, but high genetic connectivity between similar habitats separated by thousands of kilometres. This habitat partitioning is maintained even at a local scale where H. bivittatus lineages are segregated between cold- and warm-water habitats in both Bermuda and Florida. The concordance of evolutionary partitions with habitat types, rather than conventional biogeographical barriers, indicates parapatric ecological speciation, in which adaptation to alternative environmental conditions in adjacent locations overwhelms the homogenizing effect of dispersal. This mechanism can explain the long-standing enigma of high biodiversity in coral reef faunas.

Adult habitat preferences, larval dispersal, and the comparative phylogeography of three Atlantic surgeonfishes (Teleostei: Acanthuridae).

Although many reef fishes of the tropical Atlantic are widely distributed, there are large discontinuities that may strongly influence phylogeographical patterns. The freshwater outflow of the Amazon basin is recognized as a major barrier that produces a break between Brazilian and Caribbean faunas. The vast oceanic distances between Brazil and the mid‐Atlantic ridge islands represent another formidable barrier. To assess the relative importance of these barriers, we compared a fragment of the mitochondrial DNA (mtDNA) cytochrome b gene among populations of three species of Atlantic surgeonfishes: Acanthurus bahianus, A. chirurgus and A. coeruleus. These species have similar life histories but different adult habitat preferences. The mtDNA data show no population structure between Brazil and the mid‐Atlantic islands, indicating that this oceanic barrier is readily traversed by the pelagic larval stage of all three surgeonfishes, which spend ~45–70 days in the pelagic environment. The Amazon is a strong barrier to dispersal of A. bahianus (d = 0.024, ΦST = 0.724), a modest barrier for A. coeruleus (ΦST = 0.356), and has no discernible effect as a barrier for A. chirurgus. The later species has been collected on soft bottoms with sponge habitats under the Amazon outflow, indicating that relaxed adult habitat requirements enable it to readily cross that barrier. A limited ability to use soft bottom habitats may also explain the low (but significant) population structure in A. coeruleus. In contrast, A. bahianus has not been collected over deep sponge bottoms, and rarely settles outside shallow reefs. Overall, adult habitat preferences seem to be the factor that differentiates phylogeographical patterns in these reef‐associated species.

Recent Region-wide Declines in Caribbean Reef Fish Abundance

Profound ecological changes are occurring on coral reefs throughout the tropics, with marked coral cover losses and concomitant algal increases, particularly in the Caribbean region. Historical declines in the abundance of large Caribbean reef fishes likely reflect centuries of overexploitation. However, effects of drastic recent degradation of reef habitats on reef fish assemblages have yet to be established. By using meta-analysis, we analyzed time series of reef fish density obtained from 48 studies that include 318 reefs across the Caribbean and span the time period 1955-2007. Our analyses show that overall reef fish density has been declining significantly for more than a decade, at rates that are consistent across all subregions of the Caribbean basin (2.7% to 6.0% loss per year) and in three of six trophic groups. Changes in fish density over the past half-century are modest relative to concurrent changes in benthic cover on Caribbean reefs. However, the recent significant decline in overall fish abundance and its consistency across several trophic groups and among both fished and nonfished species indicate that Caribbean fishes have begun to respond negatively to habitat degradation.

Interference Compoetition Structures Habitat Use in a Local Assemblage of Coral Reef Surgeonfishes

We examined use of food and habitat in relation to pairwise interference interactions among all 13 species of a surgeonfish assemblage that lived in a 0.225-ha section ofthe outer edge of the barrier reef at Aldabra, Indian Ocean. Eleven species defended feeding territories intraspecifically and, in most cases, interspecifically. Surgeonfish species pairs fell into two classes: (1) noninteractors {n = 38 pairs) rarely interacted agonistically, and individuals of the two species peacefully shared feeding areas; (b) interactors {n = 27 pairs) typically interacted agonistically whenever they met, and, except under special circumstances, did not share feeding areas. Dominance relations between inter? actors usually were highly asymmetrical, and in only one pair was there no evidence of a dominant/ subordinate relationship. Between interactors diet overlap was higher and large-scale habitat overlap was lower than between noninteractors. In addition, within pairs of interactors with high large-scale habitat overlap, we found small-scale habitat segregation. Interactors did not differ from noninteractors in the extent of similarity in body size between the species in a pair. Reciprocal removal experiments performed with one trophic subset of the assemblage showed that agonistically subordinate species readily took over vacant habitat previously occupied by agonistically dominant species, but that the reverse rarely occurred. These data support the hypothesis that interference competition for food plays a part in structuring the assemblage, by determining many spatial distribution patterns, and promotes habitat partitioning among food competitors. Our data indicate that the presence or absence of such competition may affect 60-80% of habitat-use relationships in that assemblage.

Temporal Coupling of Production and Recruitment of Larvae of a Caribbean Reef Fish

Variation in larval recruitment is thought to have profound effects on the structure of coral reef fish communities, and planktonic processes often are cited as the major factor controlling the temporal and spatial patterns of such recruitment. We looked at the relationship between temporal patterns of larval production and settlement of plank- tonic larvae of the Caribbean damselfish Stegastes partitus at one site and attributed any differences to processes acting in the plankton. In doing so we assumed that the pattern of production we observed was representative of the regime that produced fish that settled in the study area. We monitored spawning and larval recruitment continuously for 3 yr. Both spawning and settlement followed (unimodal) lunar cycles, and both activities spanned - 3 wk of the lunar month. Although the form of the average settlement cycle matched that of the average production cycle, monthly settlement episodes were shorter and (slightly) more variably timed than equivalent production episodes. Although monthly variability in the magnitude of settlement was fourfold greater than corresponding variability in the magnitude of larval production, monthly settlement success did not vary in an extreme manner. There was no significant correlation between the magnitude of larval production in a month and of settlement the following month. Daily growth increments in the otoliths of settlers indicated that (1) larvae were 5 wk old at settlement, (2) there was low overall variability in age at settlement, (3) there were no differences between the ages of settlers arriving early and late in the monthly settlement period, and (4) age variability among settlers collected on the same day was not different from that among settlers collected on different days. Hence, the basic lunar periodicity of settlement is determined by the periodicity of production of relatively fixed-age settlers. Planktonic processes enhance the temporal vari- ability of settlement, principally by affecting the magnitude of settlement events, but also by influencing the duration and precise timing of monthly settlement episodes. Planktonic processes also determine that most of a months successful settlers arrive (and are produced) over a few consecutive days and mix cohorts of larvae that are produced on different days. We conclude that the timing and magnitude of settlement are strongly influenced by both production and planktonic processes, and the latter only partly decouple settlement and production.

Lunar Reproductive Cycles of Benthic-Brooding Reef Fishes: Reflections of Larval Biology or Adult Biology?

Lunar reproductive cycles are common among marine animals. Most hy- potheses concerning the adaptive nature of such cycles consider the effects of the tidal regime or moonlight on the dispersal of planktonic eggs or hatchlings away from adult habitats. We determined the reproductive periodicity of 17 paternal-brooding reef fishes at three neotropical sites with different tidal regimes. We used these and published data to assess the explanatory potential of hatchling-dispersal hypotheses, and of other larval- biology and adult-biology hypotheses. Among the 17 species (15 damselfishes and two blennies), two of which we studied at two sites, we found 14 with lunar and semilunar spawning cycles, one sporadically syn- chronized spawner, two acyclically variable spawners, and one continuous spawner. In 16 species individual nests experienced frequent alternation of short brood-care and rest pe- riods (brood cycling). Simple hatcThling-biology hypotheses are unable to account readily for the range of variation in the types, precision, and lunar timing of spawning cycles among species within and between depth zones at the same site, or for intra- and interspecific variation in spawning patterns in relation to seasonal and geographic differences in tidal regimes. The hypothesis that lunar spawning cycles have evolved to maximize the availability of relatively uniform-age larvae that settle during preferred lunar settlement times is sup- ported by some but not all data. Adult-biology hypotheses can account for the higher frequency of lunar spawning cycles among brooders than nonbrooders, and can accommodate various patterns of reproductive cyclicism and synchronization in brooders. Synchronized activity may be favored in some brooders because colonial spawning provides enhanced defense against egg predators. Fur- ther, the cost of brood care may result in reduced egg survivorship, and both brood cycling and the temporal concentration of spawning may reduce egg losses by allowing males to recuperate and increasing benefits to brood-care for them. Adult-biology constraints may produce variability in spawning patterns. Differences in social systems may determine the ability to form nesting colonies, and to strongly syn- chronize spawning independently of lunar cues. Changes in short-term food availability may influence the regularity of spawning. Adult-biology hypotheses represent generally underemphasized alternatives that have at least as much explanatory potential as larval-biology hypotheses. If hatchling biology is of general importance to paternal brooders, there must be considerably more interspecific variability in hatchling ecology than is generally assumed.

CHAPTER 3 – Age-Based Studies

[Extract] T here exists a vast literature on the age and growth of fishes (Hilborne and Waiters, 1992; Secor et al., 1995a). Much of this reflects a century of research on the major stocks of exploited fishes, primarily those of temperate water, open-ocean, and deep-water habitats. In contrast, there is relatively little age-based demographic information available on coral reef fishes. There are a number of reasons for advocating a more comprehensive understanding of the age-based demography of coral reef fishes. The primary reason relates to the increasing exploitation of reef fish stocks by both artisanal and commercial fisheries, some of the latter (e.g., the live fish trade) being internationally based (Jennings and Polunin, 1996; Birkeland, 1997a; Lee and Sadovy, 1998). Demographic information will provide a better basis for reef fish conservation and management. Assembling age-based data for coral reef fishes will be a complex task. Most species have wide distributions spanning significant geographic gradients (Myers, 1999). Within localized areas a species will occur in a variety of habitats. As poikilotherms, fish are sensitive to prevailing environmental conditions (Atkinson, 1994). Growth and life history patterns of reef fishes can be expected to vary, at geographic, latitudinal, and habitat scales. An emerging picture of the demography of reef fishes suggests three things. Many taxa will be relatively long-lived (Munro and Polunin, 1997), with life-spans exceeding 15 years; they will have highly distinctive patterns of growth, and age-based demographic features will show a strong phylogenetic structure (Hart and Russ, 1996; Newman et al., 1996; Choat and Axe, 1996; Choat et al., 1996). Most importantly, it is becoming evident that size and age in many taxa of reef fishes may be decoupled (Hart and Russ, 1996; Newman et al., 1996, 2000b). Although reef fishes occur in a wide range of sizes, large size may not correlate with longevity. In addition, many species of coral reef fishes may have complex sexual ontogenies, which can in turn influence size and sex-specific growth patterns (Sadovy and Shapiro, 1987; Choat et al., 1996).

The behavioral ecology of three Indian Ocean surgeonfishes (Acanthurus lineatus, A. leucosternon and Zebrasoma scopas): their feeding strategies, and social and mating systems

SynopsisThe relationship between the morphology, feeding strategies and social and mating systems of three surgeonfishes was investigated. Adults of each defend feeding territories, intra-and interspecifically. The largest species, because of its morphological limitation, relies on food that has to be defended against many other species. It forms large colonies in which fishes singly defend small territories containing high standing crop algal mats. Colony formation is a mechanism by which the efficiency and effectiveness of interspecific territory defense is increased. The smallest species, because of its morphological adaptations, is able to rely most on food that other species cannot efficiently exploit. It forms pairs that defend large territories containing a thin algal mat. It is restricted to the poorest quality habitat by the aggressive activities of more dominant species. The third species, which also forms pairs, has an intermediate feeding strategy. The local coexistence of these three and other surgeonfishes results from a combination of (i) their partitioning both habitat and food resources, and (ii) the populations of two of the most dominant species apparently being below the carrying capacity. Territoriality and the absence of parental care facilitates pair formation in surgeonfishes. Permanently territorial species usually form pairs. The colonial species does not form pairs because the colonial habit facilitates interference of males in each others spawnings.

Recent invasion of the tropical Atlantic by an Indo-Pacific coral reef fish

The last tropical connection between Atlantic and Indian–Pacific habitats closed c. 2 million years ago (Ma), with the onset of cold‐water upwelling off southwestern Africa. Yet comparative morphology indicates more recent connections in several taxa, including reef‐associated gobies (genus Gnatholepis). Coalescence and phylogenetic analyses of mtDNA cytochrome b sequences demonstrate that Gnatholepis invaded the Atlantic during an interglacial period ∼145 000 years ago (d = 0.0054), colonizing from the Indian Ocean to the western Atlantic, and subsequently to the central (∼100 000 years ago) and eastern Atlantic (∼30 000 years ago). Census data show a contemporary range expansion in the northeastern Atlantic linked to global warming.

Cohabitation of Competing Territorial Damselfishes on a Caribbean Coral Reef

Individual adults of three Caribbean damselfishes defend feeding areas that do not overlap with those ofconspecific adults. The feeding areas of adults of the two smaller species, Stegastes dorsopunicans and S. planifrons, do not overlap interspecifically. However, feeding areas of adults of the largest species, Microspathodon chrysurus, are superimposed on feeding areas of adults and juveniles of the two Stegastes species. The large species aggressively dominates its smaller interspecific cohabi- tants. Cohabitant individuals of each species eat the same types of benthic microalgae. M. chrysurus has the same daily cycle of feeding activity as the two small species, uses the same feeding microhabitats as they do, and feeds almost exclusively in their feeding areas. Cohabitants defend their feeding areas against the same set of other herbivorous fishes, but M. chrysurus is involved in defensive actions much less frequently than is either of its small cohabitants. M. chrysurus also is much less effective at such defense than at least one cohabitant, S. dorsopunicans. S. dorsopunicans on a reef from which M. chrysurus were removed increased their body mass more and had larger fat deposits than conspe- cifics on a control reef. Thus adults of M. chrysurus appear to be dependents that use their size-based dominance ability to obtain food from their cohabitants. Juveniles of M. chrysurus and S. dorsopun- icans exhibit small-scale habitat segregation and segregated use of feeding substrates in a shared microhabitat, which may facilitate the recruitment of both species.