Testing the coral fragment size-dependent survivorship hypothesis for the calcareous hydrozoan Millepora complanata

Abstract

The distribution of living colonies of the calcareous hydrozoan Millepora complanata and broken and reattached millepore fragments was examined in m* quadrats along line transects on fringing reefs at Barbados, West Indies. Of the broken fragments 35 % reattached, but survival was not found to be size-dependent, in contrast to reports of breakage in Scleractinia. The number of reattached fragments was positively correlated with the total number of fragments and with the number of quadrats containing mlllepore colonies. It is proposed that fragmentation has a selective advantage because of the high survival rate of fragments and that there should be an advantage to increasing colony size in the population because the risk of breakage increases with colony size. Breakage of coral colonies during storms has been widely reported, and resulting damage to reefs is frequently on a large scale (Highsmith et al. 1980, Tunnicliffe 1980, Woodley et al. 1981, Dollar 1982, Rogers et al. 1982). Breakage appears to be most common in branching species (Highsmith et al. 1980, Tunnicliffe 1980, 1981) but has also been reported in massive and solitary corals (Highsmith 1980, 1982). Reattachment and regeneration of fragments occur frequently and ameliorate the destructive effects of storms (Highsmith 1982). Indeed, fragmentation is regarded as a natural process (Hughes 1989) and an important mode of asexual reproduction in some species (Highsmith 1980, 1982, Highsmith et al. 1980), and it may enhance propagation over broader habitat ranges (Tunnicliffe 1981). Survival of coral fragments is strongly size-dependent; large fragments have a greater chance of success than do small fragments (Loya 1976, Highsmith et al. 1980, Hughes 1989). Similar predictions that mortality rates decline with increasing colony size have been made for the survival of sexually produced coral colonies (Connell 1973, Hughes and Jackson 1985, Lewis 1989a). The size-dependent survival hypothesis thus appears to apply for fragments, larval colonizing stages O Inter-ResearchlPrinted in Germany and regenerating older colonies. An opportunity for testing the generality of the survival hypothesis was recognized from observations of loose and attached fragments of the calcareous hydrozoan Millepora complanata Lamarck on reefs at Barbados, West Indies. Millepora complanata is a common, calcareous hydrozoan on Caribbean reefs and flourishes in high wave-energy environments (Boschma 1948, Lewis 1989b). These millepores tend to grow so that the faces of the blades or plate-like colonies are presented towards waves and currents (Stearn & k d i n g 1973, Lewis 1989b). This appears to be the least mechanically stable orientation for rigid, plate-like forms and likely to increase risk of breakage. Risk of breakage also increases with size (Denny et al. 1985). Thus, breakage of M. complanata colonies is likely to occur frequently, and broken fragments were routinely observed in field populations on reefs in Barbados during all seasons. If reattachment of millepores occurs, it might be expected that survival of the fragments would, like hermatypic corals, be size-dependent. Methods. The abundance and spatial distribution of Millepora complanata were estimated from contiguous m2 quadrats laid out along line transects on 3 reefs on the west coast of Barbados. The reefs were situated within a 3 km stretch of the northern sector of the west coast which is least affected by eutrophication (Tomascik & Sander 1985). For each reef, 25 transects were established on the surface of the ridges or spurs making up the reefs (Stearn & Riding 1973, Stearn et al. 1977). The transects were aligned perpendicular to the shore and spaced a t suitable intervals so as to include the full length of the reefs. Five adchtional transects were sampled for each reef along the grooves or channels between the ridges (Stearn & Riding 1973, Mah &