The decline of seagrass communities worldwide has sparked an urgent need for effective restoration strategies, which require a working knowledge of population genetic structure. Halodule wrighti is a common seagrass of the Caribbean region that is being restored to areas of the Gulf of Mexico, yet little is known of its population genetics. This study provides an assessment of individual, clonal and population effects on the genetic structure of 4 natural H. wrightii populations occupying 170 km of coastline in and around Galveston Bay, Texas, for comparison with 7 restored populations ranging in age from 2 to 7 yr. By using molecular markers, in the form of amplified fragment length polymorphisms (AFLPs), we found considerable variation in clonal richness at the population scale (from 0.54 to 0.82), with the restored populations occupying an intermediate to high position within this range. Replicate sampling within individual seagrass beds of 3 to 5m diameter generally revealed higher levels of clonal richness, elevated by 4 to 22% over that at the population scale, suggesting that seed recruitment is more important at the local scale than at distances of >10 m. Genetic diversity was 2 to 3 times less than that expected for a widespread, outcrossing species like H. wrightii, although a 170% increase in the frequency of variable markers relative to the mean for all other populations was noted for a volunteer population that had recruited from a mixture of donor materials planted at a nearby restoration site. Within the spatial extent of this study, natural populations adhered to a model of isolation-by-distance, whereas donor materials from these same natural populations were undergoing a rapid genetic convergence within a restored site where they had been planted together. ?? Inter-Research 2006.