POPULATION STRUCTURE AND SPECIATION IN TROPICAL SEAS: GLOBAL PHYLOGEOGRAPHY OF THE SEA URCHIN DIADEMA

Abstract.— The causes of speciation in the sea are rarely obvious, because geographical barriers are not conspicuous and dispersal abilities or marine organisms, particularly those of species with planktonic larvae, are hard to determine. The phylogenetic relations of species in cosmopolitan genera can provide information on the likely mode of their formation. We reconstructed the phylogeny of the pantropical and subtropical sea urchin genus Diadema, using sequences of mitochondrial DNA from 482 individuals collected around the world, to determine the efficacy of barriers to gene flow and to ascertain the history of possible dispersal and vicariance events that led to speciation. We also compared 22 isozyme loci between all described species except D. palmeri. The mitochondrial DNA data show that the two deepest lineages are found in the Indian and West Pacific Oceans. (Indo‐Pacific) Diadema setosum diverged first from all other extant Diadema, probably during the initiation of wide fluctuations in global sea levels in the Miocene. The D. setosum clade then split 3‐5 million years ago into two clades, one found around the Arabian Peninsula and the other in the Indo‐West Pacific. On the lineage leading to the other species of Diadema, the deepest branch is composed of D. palmeri, apparently separated when the climate of New Zealand became colder and other tropical echinoids at these islands went extinct. The next lineage to separate is composed of a currently unrecognized species of Diadema that is found at Japan and the Marshall Islands. Diadema mexicanum in the eastern Pacific separated next, whereas D. paucispinum, D. savignyi, and D. antillarum from the western and central Atlantic, and (as a separate clade) D. antillarum from the eastern Atlantic form a shallow polytomy. Apparently, Indo‐Pacific populations of Diadema maintained genetic contact with Atlantic ones around the southern tip of Africa for some time after the Isthmus of Panama was complete. Diadema paucispinum contains two lineages: D. paucispinum sensu stricto is not limited to Hawaii as previously thought, but extends to Easter Island, Pitcairn, and Okinawa; A second mitochondrial clade of D. paucispinum extends from East Africa and Arabia to the Philippines and New Guinea. A more recent separation between West Indian Ocean and West Pacific populations was detected in D. setosum. Presumably, these genetic discontinuities are the result of water flow restrictions in the straits between northern Australia and Southeast Asia during Pleistocene episodes of low sea level. Diadema savignyi is characterized by high rates of gene flow from Kiribati in the central Pacific all the way to the East African Coast. In the Atlantic, there is a biogeographic barrier between the Caribbean and Brazil, possibly caused by fresh water outflow from the Amazon and the Orinoco Rivers. Diadema antillarum populations of the central Atlantic islands of Ascension and St. Helena are genetically isolated and phylogenetically derived from Brazil. Except for its genetic separation by the mid‐Atlantic barrier, Diadema seems to have maintained connections through potential barriers to dispersal (including the Isthmus of Panama) more recently than did Eucidaris or Echinometra, two other genera of sea urchins in which phylogeography has been studied. Nevertheless, the mtDNA phylogeography of Diadema includes all stages expected from models of allopatric differentiation. There are anciently separated clades that now overlap in their geographic distribution, clades isolated in the periphery of the genus range that have remained in the periphery, clades that may have been isolated in the periphery but have since spread towards the center, closely related clades on either side of an existing barrier, and closely related monophyletic entities on either side of an historical barrier that have crossed the former barrier line, but have not attained genetic equilibrium. Except for D. paucispinum and D. savignyi, in which known hybridization may have lodged mtDNA from one species into the genome of the other, closely related clades are always allopatric, and only distantly related ones overlap geographically. Thus, the phylogenetic history and distribution of extant species of Diadema is by and large consistent with allopatric speciation.