Comparative studies on the distribution of archaeal versus bacterial communities associ- ated with the surface mucus layer of corals have rarely taken place. It has therefore remained enigmatic whether mucus-associated archaeal and bacterial communities exhibit a similar specificity towards coral hosts and whether they vary in the same fashion over spa- tial gradients and between reef locations. We used microbial community profiling (terminal- restriction fragment length polymorphism, T-RFLP) and clone library sequencing of the 16S rRNA gene to compare the diversity and community structure of dominant archaeal and bacterial communities associating with the mucus of three common reef-building coral spe- cies (Porites astreoides, Siderastrea siderea and Orbicella annularis) over different spatial scales on a Caribbean fringing reef. Sampling locations included three reef sites, three reef patches within each site and two depths. Reference sediment samples and ambient water were also taken for each of the 18 sampling locations resulting in a total of 239 samples. While only 41% of the bacterial operational taxonomic units (OTUs) characterized by T- RFLP were shared between mucus and the ambient water or sediment, for archaeal OTUs this percentage was 2-fold higher (78%). About half of the mucus-associated OTUs (44% and 58% of bacterial and archaeal OTUs, respectively) were shared between the three coral species. Our multivariate statistical analysis (ANOSIM, PERMANOVA and CCA) showed that while the bacterial community composition was determined by habitat (mucus, sediment or seawater), host coral species, location and spatial distance, the archaeal com- munity composition was solely determined by the habitat. This study highlights that mucus- associated archaeal and bacterial communities differ in their degree of community turnover over reefs and in their host-specificity. 

Rehabilitating populations of Caribbean coral species that have declined in recent decades has become a management priority throughout the region, stimulating the development of new methodologies to arti cially reseed degraded reefs. Rearing lar- vae of ecologically important coral species appears a particularly attractive method to aid the recovery of degraded populations because genetic recombination could yield new genotypes better capable of coping with the altered conditions on modern Caribbean reefs. Well-developed elkhorn coral (Acropora palmata Lamarck, 1816) populations form dense thickets that contribute to the maintenance of healthy and productive reefs by providing shelter to a variety of other reef organisms (Gladfelter and Gladfelter 1978). After >95% of A. palmata populations were decimated by a disease beginning in the mid-1970s, this species was listed as critically endangered under the Red List of threatened Species (IUCN 2013) and restoration e orts were initiated throughout the region to assist its recovery (Young et al. 2012). In 2011, we collected gametes from eight A. palmata colonies in situ o Curaçao, which were subsequently cross-fertilized to generate larvae. Competent larvae were settled on clay tiles (Panel A) and reared in a ow-through land-based nursery for one year (Panels B–C), after which they were outplanted to a breakwater at 2–5 m depth (Panel D) [refer to Chamberland et al. (2015) for details on methodology]. Seven out of nine outplanted colonies survived and continued to grow in situ (Panels D–E), reaching a size of 30–40 cm diameter and 20–30 cm height after 4 yrs (Panel F). On 8 and 10 September, 2015, nine and 11 d after the full moon, two colonies were ob- served releasing gametes between 155 and 175 min after sunset (Panels G–H). is is the rst time that an endangered Caribbean Acropora coral species was raised from larvae and grown to sexual maturity in the eld. Indeed, only one other study has documented age and colony size at reproductive onset in a broadcast spawning scler- actinian coral reared from larvae (Baria et al. 2012). e relatively short time until onset of spawning (≤4 yrs) observed for A. palmata shows that recovery of degraded coral populations by enhancing natural recruitment rates may be practicable if out- planted colonies are able to rapidly contribute to the natural pool of larvae.