Researchers chose 6 sites, and at each site, selected 10 patch reefs that had established populations of vermetids. On each reef, size (length, width, and water depth), as well as percent cover of corals and non corals were recorded. Of the 10 reefs surveyed at each site, they chose randomly chose 5 to remove vermetids using wire hooks, which removed the snail and left the shells intact. the remaining reefs were unmanipulated and contained vermetids at ambient densities.
Experimental colonies came from three distinct colonies of Pocillopora verrucosa. Colonies were chosen near the MOE site (~600m west). Within the site of collection, colonies were collected from separate patch reefs (separated by greater than 100m) to maximize the probability of different genotypes. Each colony terminal branches were removed in segments of 5cm in length. Each fragment was attached to a plastic base and the skeletal mass was estimated using the bouyant weight technique (Davies 1989). One fragment from each reef was attached to each reef.
Coral fragments were assessed after 155 days. Corals were collected, their final buoyant mass was measured and their proprtion of the coral surface area that was alive was visually estimated.