Successful dispersal can enhance both individual fitness and population persistence, but the process of dispersal is often inherently risky. The interplay between the costs and benefits of dispersal are poorly documented for species with complex life histories due to the difficulty of tracking dispersing individuals. Here we investigate variability in dispersal histories of a freshwater fish, Awaous stamineus, across the species' entire geographic range in the Hawaiian archipelago. Like many animals endemic to tropical island streams, these gobies have an amphidromous life cycle in which a brief marine larval phase enables dispersal among isolated freshwater habitats. Using otolith microchemistry, we document three distinct marine dispersal pathways, all of which are observed on every island. Surprisingly, we also find that 62% of individuals complete their life cycle entirely within freshwater, in contrast to the assumption that amphidromy is obligate in Hawaiian stream gobies. Comparing early life history outcomes based on daily otolith growth rings, we find that individuals with marine dispersal have shorter larval durations and faster larval growth, and their growth advantage over purely freshwater counterparts continues to some degree into adult life. These individual benefits of maintaining a marine dispersal phase presumably balance against the challenge of finding and reentering an island stream from the ocean. The facultative nature of amphidromy in this species highlights the selective balance between costs and benefits of dispersal in life history evolution. Accounting for alternative dispersal strategies will be essential for conservation of the amphidromous species that often dominate tropical island streams, many of which are at risk of extinction.

The persistence and resilience of marine populations in the face of disturbances is directly affected by connectivity among populations. Thus, understanding the magnitude and pattern of connections among populations and the temporal variation in these patterns is critical for the effective management and conservation of marine species. Despite recent advances in our understanding of marine connectivity, few empirical studies have directly measured the magnitude or pattern of connections among populations of marine fishes, and none have explicitly investigated temporal variation in demographic connectivity. We use genetic assignment tests to track the dispersal of 456 individual larval fishes to quantify the extent of connectivity, dispersal, self-recruitment and local retention within and among seven populations of a coral reef fish (Stegastes partitus) over a three-year period. We found that some larvae do disperse long distances (*200 km); however, self-recruitment was a regular phenomenon. Importantly, we found that dispersal distances, self-recruitment, local retention and the pattern of connectivity varied significantly among years. Our data highlight the unpredictable nature of connectivity, and underscore the need for more, temporally replicated, empirical measures of connectivity to inform management decisions.

Despite substantial advances in our understanding of marine population dynamics, there is still much uncertainty as to what processes influence connectivity, gene flow and population structure. To explore this, we examined the spatial and temporal variation in population genetic structure of adult and recently settled bicolor damselfish Stegastes partitus, a coral reef fish. We genotyped adult and juvenile fish from 10 sites over 4 sample years at 9 microsatellite loci. We show spatial heterogeneity in adult and juvenile population structure; however, we found no evidence of a pattern of spatial genetic divergence. Furthermore, genetic structure changed through time and between life stages in an unpredictable manner. Using these data, we test whether pre-or postsettlement selection, sweepstakes effects or variability in connectivity can explain the observed chaotic genetic patchiness. Our results indicate that the contributions of various larval sources likely change through time as a result of stochastic processes such as oceanographic flow. Our results have implications for the management of marine populations, as spatial and temporal variability in connectivity may act to promote long term stability of populations. Therefore it is important that marine management efforts account for such heterogeneity in the design of protected areas.KEY WORDS: Connectivity · Temporal variability · Larval dispersal · Chaotic genetic patchiness · Genetic structure · Coral reef fish Resale or republication not permitted without written consent of the publisherMar Ecol Prog Ser 417: [263][264][265][266][267][268][269][270][271][272][273][274][275] 2010 through time and space (Larson & Julian 1999). This fine-scale genetic heterogeneity, termed chaotic genetic patchiness (Johnson & Black 1982), is characterized by low level genetic differentiation among and between adult and recruit populations (i.e. low F ST ) that is not consistent in space or time (Johnson & Black 1984).There are 4 main hypotheses put forth to explain chaotic genetic patchiness (Larson & Julian 1999); each infers a different mechanism by which this pattern can be explained. These hypotheses are (1) localized postsettlement selection resulting from microgeographic variation in environmental conditions, (2) variable local natural selection on pre-settlement individuals generating variability in cohorts through space and time, (3) 'sweepstakes chance-matching' (Hedgecock 1994) created by variable reproductive success of the adult source populations as a result of stochastic processes. This causes a genetic drift effect during the larval stage and a subsequent reduction in genetic variability in the recruit populations, and (4) spatial and temporal variability in the genetic composition of recruits caused by fluctuations in the source of larvae (Selkoe et al. 2006).Post-settlement natural selection has been invoked to explain spatial genetic heterogeneity in a marine snail (Johannesson et al. 1995). Generally, however, most studies have shown that gen...

The phenomenon of chaotic genetic patchiness is a pattern commonly seen in marine organisms, particularly those with demersal adults and pelagic larvae. This pattern is usually associated with sweepstakes recruitment and variable reproductive success. Here we investigate the biological underpinnings of this pattern in a species of marine goby Coryphopterus personatus. We find that populations of this species show tell-tale signs of chaotic genetic patchiness including: small, but significant, differences in genetic structure over short distances; a non-equilibrium or “chaotic” pattern of differentiation among locations in space; and within locus, within population deviations from the expectations of Hardy-Weinberg equilibrium (HWE). We show that despite having a pelagic larval stage, and a wide distribution across Caribbean coral reefs, this species forms groups of highly related individuals at small spatial scales (<10 metres). These spatially clustered family groups cause the observed deviations from HWE and local population differentiation, a finding that is rarely demonstrated, but could be more common than previously thought.