Speciation, Connectivity and Self-Recruitment Among Mollusc Populations from Isolated Oceanic Islands

The conventional view that marine populations are demographically ‘open’ and exchange migrants (juveniles or adults, but mostly larvae) has been challenged by recent genetic studies and the discovery of significant genetic subdivision among populations on small geographic scales. Despite the numerous publications on the matter, the extent to which some/all marine populations rely on self-recruitment and whether this reliance is stable in time and space currently remains unanswered. This is particularly true for populations from isolated oceanic archipelagos, such as the New Zealand (NZ) subantarctic islands and the Kermadec Islands. The specific objectives of this thesis were to: 1) assess the genetic diversity, phylogeography and contemporary levels of dispersal and self-recruitment in populations of the Cellana strigilis limpet complex, endemic to the NZ subantarctic islands; 2) conduct a morphometric analysis of the C. strigilis complex to complement its molecular investigation; 3) develop and optimize specific microsatellite markers for Nerita melanotragus, a marine gastropod of the Kermadec Islands and New Zealand North Island rocky shores; 4) assess the genetic structuring and levels of connectivity of N. melanotragus populations within the Kermadec Islands, within NZ North Island, and between the Kermadec Islands and NZ; and 5) compare the genetic structuring of N. melanotragus at the Kermadec Islands to that of NZ North Island populations, to test for any “island effect” on connectivity levels, and test for possible gene flow between the two groups. Genetic investigation of the C. strigilis complex confirmed the presence of two distinct lineages, separated by their sister species Cellana denticulata. Morphometric analyses were congruent with molecular analyses, and were used to provide a new taxonomic description of the C. strigilis limpet complex: two species were recognized, Cellana strigilis and Cellana oliveri. The role of the subantarctic islands during the last glacial maximum was highlighted, and the colonisation history of the islands by the two Cellana species was explained. Contemporary levels of connectivity (gene flow) among the different populations of the two lineages were low, or non-existant, revealing their high reliability on self-recruitment. However, the analysis detected a recent migration event in one of the two lineages. Considering the geographical distance of the islands and the life history of the Cellana species, the use of mediated dispersal means (e.g., rafting on a natural substrate such as kelp) seems very likely. Ten novel polymorphic microsatellite loci were developed for N. melanotragus, and seven of those were used to investigate the levels of connectivity and self-recruitment in six populations from the Kermadec Islands, and nine populations from the east coast of NZ North Island. According to what can be expected for a species with a long pelagic larval duration (PLD), genetic homogeneity was recorded for the Kermadec Islands populations. A lack of genetic structuring was also found for the nine populations on the NZ North Island, which is congruent with the literature in this geographic area. However, what was surprising was the high level of genetic homogeneity found between the Kermadec Islands and the NZ North Island, meaning that the two groups are effectively exchanging individuals. Hence, the Kermadec archipelago can be considered “open” at the scale of the South Pacific, for N. melanotragus populations. This Ph.D. highlights the importance of having the correct taxonomy for conservation and connectivity studies, and gives a better understanding of the historical and contemporary patterns of genetic connectivity in the NZ offshore islands. It illustrated how historical events, such as the last glacial maximum, can shape local genetic diversity, and how this historical pattern can be maintained because of limited contemporary gene exchange. Also, this thesis demonstrated that remote populations could be strongly connected to mainland populations, contributing to the resilience of both systems and confirming the necessity of integrating remote oceanic habitats in the creation of effective Marine Protected Areas (MPA) networks to protect the marine environment