Delimiting coral species using alternative techniques: Montipora digitata (Dana, 1846), a case study

High levels of intra-specific skeletal variability and considerable overlap of skeletal characters between species pose problems for the application of traditional taxonomic methods. Such problems can be overcome using alternative techniques. In this study, four alternative techniques were used to investigate the species status of two sympatrically occurring morphs of Montipora digitata (Dana 1846) found on inshore reefs along the Queensland coast, Australia.\ud \ud Allozyme electrophoresis indicated that gene flow does not occur between the two morphs, which were distinguished by one fixed gene difference at locus LT-2 and frequency differences at three other loci. These differences were consistent at three locations separated by up to 300 km. The genetic distinctiveness of the two morphs is further supported by the comparatively high values of Nei's genetic distance between morphs (0.24), in contrast to low values within populations of the two morphs (0.01). Populations of both morphs were generally in Hardy-Weinberg equilibrium verifying that sexual recruits contribute to the genetic structure of the populations. Low genetic diversity ratios (Go:GE), despite sampling designed to reduce the chance of collecting clonemates, indicated that asexual reproduction also occurs.\ud \ud Breeding experiments demonstrated that fertilisation occurs readily among colonies of the same morph, but that there is virtually no fertilisation between morphs. Reproductive hierarchies detected within the two morphs may indicate a more complex breeding structure within each population. Gamete interaction experiments revealed that the block to fertilisation between morphs occurs before egg activation. It is suggested that incompatibility of egg-sperm binding proteins is the most likely reason for reproductive isolation between the morphs.\ud \ud Breeding experiments involving nine species of Montipora demonstrated that hybridisation occurs within the genus Montipora in vitro, but not at high levels, and survival of hybrid juveniles was poor. Hybridisation is therefore unlikely to play a major role in the evolution of the genus. Univariate and multivariate morphometric analyses based on five skeletal characters revealed that there are significant differences in skeletal morphology between the two morphs. Considerable overlap in these characters, however, makes them unsuitable for species identification. Septal shape was found to reliably separate the two morphs. In view of the concurrence among genetic, breeding and morphological data, and examination of holotypes, the two morphs were renamed M. tortuosa (Dana 1846) and M. digitata (Dana 1846). \ud \ud Comparison of a number of reproductive characters indicates that the two species, M. tortuosa and M. digitata, differ in their reproductive biology. Montipora tortuosa produced more eggs per polyp and larger testes, and a greater percentage of polyps within the colonies produced eggs. However, M. digitata produced larger eggs, a strategy that made the total reproductive output between the two species equivalent. The reproductive study also revealed that these two species, and several other species within the genus Montipora, spawn in both spring and autumn. Reproductive outputs differed between the two breeding seasons. Differences in reproductive output and contrasting environmental conditions during the two spawning seasons are discussed in relation to ultimate and proximate cues governing coral spawning.\ud \ud Values for Nei's genetic distance suggest that the two species evolved 3.5 to 7.1 million years ago. Morphological and ecological similarities between M. tortuosa and M. digitata indicate that they have not diverged substantially in this time, they probably evolved in the same habitat in which they are found today, and they are most likely monophyletic in origin. It is argued that speciation of M. tortuosa and M. digitata may have resulted from stochastic changes to egg-sperm binding proteins on the egg surface. Reproductive hierarchies within the species support the existence of molecular variability in these proteins which may have led to reproductive isolation between M. tortuosa and M. digitata. Studies on mass spawning corals so far have found high levels of hybridisation suggesting that they do not conform to most species concepts. In contrast, M. tortuosa and M. digitata show most of the qualities of "good" morphological, biological, ecological and evolutionary species