Challenges in directly measuring dispersal have long impeded tests of theory that larval development mode should affect realized gene flow by determining pelagic larval duration (PLD), the swimming phase of the life cycle. Meta-analyses of available genetic studies have been equivocal as to the strength of the relationship for swimming lecithotrophs and planktotrophs, in particular. To understand how larval dispersal affects population structure, we are assessing phylogeography and population genetic structure in sacoglossans by quantifying PLD in the lab, assessing gene flow among populations for co-occurring species, and correcting for phylogenetic effects. Preliminary results support the hypothesis that species with non-planktonic development have more divergent, geographically restricted lineages, but some lecithotrophs are over-dispersed relative to the mean, and some planktotrophic species are surprisingly under-dispersed. These findings suggest larvae of some sea slugs species, despite being small and slow-swimming, have evolved behavioral mechanisms that may alter expected PLD-Fst relationships by increasing or decreasing local retention; or alternatively, that PLD is more variable within species than is presently appreciated. These results have implications for the design of marine protected areas in threatened shallow-water ecosystems, revealing regions that harbor unique lineages and retain ancestral polymorphism.