The role of larval dispersal in ecology and evolution is still largely undescribed in many marine organisms. Larval recruitment has long been viewed as being dominated by unpredictable, highly variable processes but also as a significant driver of population ecology and genetics. Without any knowledge about the directionality and magnitudes of larval recruitment, understanding marine species population dynamics remains a highly uncertain endeavor. Unfortunately, given the need for extensive management and conservation of many fishery species, this uncertainty can directly affect our ability to implement effective management policy. Gag, Mycteroperca microlepis, in the Gulf of Mexico are a perfect example of this conundrum where larval dispersal appears to be important, but we have a very limited understanding of how it is affecting populations. In this dissertation, I describe a series of studies pointed at describing the role of larval dispersal in patterns of population genetics of gag across the Gulf of Mexico as well as examining its relationship to the population ecology of the system. In Chapter Two, I examined patterns of genetic variation across life stages (adult and juvenile), space, and time to explore the role of larval dispersal in the effective population size of gag. Typically, marine systems are believed to exhibit low effective size to census size ratios due to large variation among individuals in reproductive success. While gag do exhibit this low ratio, there is little evidence to support "sweepstakes"-like reproductive success. I found little support for significant genetic differentiation among post-settlement juveniles both in space and time or for juvenile gag being a sample of a subset of adult genetic diversity. Also, adult gag did not exhibit any consistent spatial bias in contributions to juvenile cohorts. Overall, across the entire West Florida Shelf, juvenile gag appeared to be a well-mixed sample of adult gag offspring. The results suggest that larval dispersal has little effect on the population genetics of gag and the patterns of reproductive success among individuals. In Chapter Three, I examined the hypothesis that populations of gag on Campeche Bank off the Yucatan Peninsula are connected to populations of gag on the West Florida Shelf via larval dispersal. Additionally, I determined whether it was likely that this connectivity was due to on-going or historical gene flow. Ecological evidence suggests on-going connectivity. Genetic evidence showed significant, but low levels, of genetic differentiation between these two regions, consistent with a hypothesis of on-going gene flow among populations. Simulations of gag population genetics supported this hypothesis. Analysis of migration directionality indicated one-way migration from Campeche Bank to the West Florida Shelf, congruent with Gulf Loop Current flow. The timing of population divergences also suggested that current population genetic diversity could not be explained solely by historical population dynamics. These results indicate that population of reef fishes in the Gulf of Mexico may be connected across under-appreciated spatial scales on both ecological- and evolutionary-relevant time scales. In Chapter Four, I integrated population genetics data with population dynamics data to test for the support of data for different models of reproductive success and the role of demographic changes versus skewed reproductive success among individuals contributing to low effective size to census size ratios. Using an individual-based simulation model and Approximate Bayesian Computations, I evaluated different models and parameter sets for reproductive success and age-determined sex change using Bayes Factors to assess which factors are likely contributing to the observed patterns of genetic variation given the population ecology and biology of gag. Data generally supported models of skewed reproductive success among individuals, but the level of skew was not high enough to attribute "sweepstakes"-like variation among individuals in reproductive success to the pattern observed in the data. Instead, changes in demography had a much greater effect on support for models, indicating that documented changes in gag population sex ratio may have a large role in the observed patterns in the data.