Jonathan P. A. Gardner

To better understand the distribution of native smooth-shelled blue mussels (genus Mytilus ) in the North Pacific region (USA, Canada, Russia, China, Korea) and to quantify the geographic extent of invasion and the degree of hybridisation between native and non-native mussels we analysed mussels from 15 North Pacific sites and 16 global reference sites for variation at 59 single nucleotide polymorphism (SNP) loci. Samples were collected between 2006 and 2014, and some sample sizes were small (< 10 mussels) with the result that some site-specific interpretations are limited. Overall, our SNP-based results confirmed earlier findings based on other marker types and revealed (1) the likelihood of multiple independent invasion events of M. galloprovincialis , (2) the spatially patchy nature of bioinvasion in the USA, Canada, and Russia, (3) that sites in China and Korea exhibited only invasive M. galloprovincialis , and (4) the limited extent of hybridisation between M. trossulus × M. galloprovincialis beyond the F ₁ generation. The regions of origin of invasive Mediterranean M. galloprovincialis are primarily the central Mediterranean Sea and secondarily the Black Sea and Sea of Azov. The source sites of origin of invasive Atlantic M. galloprovincialis cannot be determined. Whilst we cannot categorically rule out alternative hypotheses, these results support the interpretation of multiple independent invasion events. Our results provide the first large geographic scale SNP-based assessment of smooth-shelled blue mussel genetic variation in the North Pacific region and highlight the positive (development of aquaculture) and negative (hybridisation with the endemic species) impacts of invasive M. galloprovincialis in the North Pacific region.

Describing patterns of spatially‐explicit population genetic structure and connectivity is vital for well‐designed conservation or management policies for marine species. However, for species having high dispersal capabilities and large population sizes, describing significant genetic structure may be difficult. Perna canaliculus , the endemic New Zealand green‐lipped mussel, is the cornerstone species of this country's aquaculture industry. Several studies performed with various genetic marker types have revealed two distinct genetic groups (northern and southern), but no further genetic differentiation. Here, we describe the first use of ddRAD‐seq‐derived SNPs to examine the genetic structuring of P. canaliculus . Using 185 neutral and 28 outlier SNPs from 32 sampling sites, two major genetic groups (northern and south‐west—with a genetic break at ~42° S, below the Cook Strait/Raukawa Moana region) were confirmed. The use of the outlier panel revealed stronger genetic differences in allelic frequencies than markers under neutral expectations. Moreover, the SNP data provided, for the first time, evidence of substructure over the extensive range of the population that was sampled: six further genetic subgroups were identified within the two main groups. Seascape analysis revealed that genetic variation correlates strongly with variables related to food availability, habitat composition and current speed, reflecting selective pressure and genetic signals of local adaptation, suggesting these processes play a key role modulating the genetic structure of the green‐lipped mussel. Gene flow networks suggest migration from the south‐west group into the northern group. This pattern of migration is also revealed in contemporary oceanographic models of larval dispersal, which indicate high historical gene flow between the two groups, and a minor influence of genetic drift. These results offer valuable insights into the genetic structure and factors influencing genetic divergence in P. canaliculus and how genetic stocks may be managed.

Understanding patterns of connectivity and adaptive divergence is crucial for supporting conservation and sustainable management of harvested species. This study utilised single nucleotide polymorphisms (SNPs) to investigate spatially explicit patterns of genetic structure, gene flow and adaptive divergence in the commercially important southern rock lobster Jasus edwardsii across New Zealand (NZ), a species with a 2‐year pelagic larval duration. Using neutral (1608 SNPs) and outlier (250 SNPs) panels, genetic analyses revealed significant population differentiation between J. edwardsii individuals collected from one site in Tasmania (Australia) and those from 22 sites in NZ, supporting earlier findings indicating substantial genetic variation between the two stocks. Within NZ, neutral markers revealed a high degree of genetic connectivity of J. edwardsii , but also subtle genetic differentiation of one northern site only. In contrast, analysis of outlier markers identified three genetically differentiated regions within NZ: the Northeast (NE), Northwest (NW) and Southern (S) groups. Migration models indicated bidirectional but asymmetrical gene flow amongst these groups (S → NW → NE), consistent with larval dispersal and settlement models. Patterns of genetic structure at the outlier loci were strongly correlated with the environmental variables ‘carbonate concentration’, ‘roughness at the seafloor’ and ‘sea surface temperature’ across NZ, which may contribute to differential settlement and survival of J. edwardsii . These findings demonstrate complex patterns of connectivity and the influence of environment factors on genetic structure and local genetic adaptative divergence of J. edwardsii despite its very high dispersal potential. The population genomic and connectivity findings support the current spatially explicit management regime of the rock lobster fishery in New Zealand, and the seascape genomic findings may be of value to the fishery in responding to new or enhanced patterns of environmental variability driven by climate change or to the development of a new aquaculture industry.

The core–periphery hypothesis (CPH) predicts that genetic diversity is greatest at the centre and lowest at the edges of a species' distribution because genetic diversity is a function of a species' abundance, which is also expected to be greatest at the centre and lowest at the edges of the distribution. Variants of the CPH include the ‘Ramped North' (greatest variation in the north), the ‘Ramped South' (greatest in the south), and the ‘Abundant Edge' (greatest at the distributional edges). Here, we present the first standardised multi‐phylum analysis of the CPH using nine indices of genetic diversity for New Zealand's marine biota, covering 52 species. Based on 80 studies across eight phyla, spatial variation in the genetic indices was tested against four models (Normal (N), Ramped North (RN), Ramped South (RS), Abundant Edge (AE)). Only 22.7% of all individual taxon‐specific tests were statistically significant: Ramped North (10.5%), Ramped South (7.4%), Abundant Edge (2.6%) and Normal (2.2%). Nonetheless, amongst the Chordata (Ramped North and Ramped South), Arthropoda (Ramped South) and Mollusca (Ramped North), a reasonably consistent pattern of genetic variation was observed within each phylum. Spatially‐explicit genetic diversity of the remaining taxa fitted different models but without any obvious pattern across the phyla. Generalised binomial testing of observed p‐values for each genetic index across all studies revealed that 10 of 29 tests were significant (5 RN, 2 N, 2 RS, 1 AE). Overall, our meta‐analysis revealed no real support for the CPH and only limited support for a Ramped model (either Ramped North or Ramped South) of spatially‐explicit genetic diversity. For New Zealand coastal marine taxa, we conclude that consistently strong patterns of genetic variation across multiple taxa do not exist and the CPH requires extensive testing from multiple other regions before we can say that such patterns exist, let alone explain them.

Geometric morphometry is effective in distinguishing bivalve species and populations, including the economically and environmentally important Mediterranean mussel, Mytilus galloprovincialis. Although widely distributed, shell shape variation in M. galloprovincialis along the eastern Adriatic Sea has been infrequently studied. Farming practices and environmental conditions may affect the development of shell shape, as has been reported in the Mytilus genus from numerous locations globally. Building on earlier genetic analyses of mussels collected along a natural environmental gradient, this study aimed to identify shell landmark differentiation between wild and cultured populations and among northern, middle, and southern populations of the eastern Adriatic Sea using a geometric morphometric approach. Samples from 12 sites in Croatia, Montenegro, and Albania, including 4 aquaculture farms, were examined for variation in 9 internal shell landmarks. Wild populations exhibited a more extended posterior adductor muscle scar and a more elongated shape compared to farmed populations. Mussels from low salinity environments in the north and south exhibited an elongated shell shape compared to high salinity environments. Southern populations exhibited an extended posterior adductor muscle scar, along with an elongation of the lateral ligament and ventral umbo orientation that caused a concave shape of the ventral shell border compared to northern populations. The differences in environmental conditions in the Adriatic Sea, such as reduced salinity in Boka Kotorska Bay (Montenegro) in the south and Limski Bay (Croatia) in the north, likely play a role in influencing the variability of shell landmarks. These results may be applied to farming practices so that high-quality spat are collected from source sites with environmental conditions that match the farm site to which the spat are transferred. Overall, these results provide valuable insight into how M. galloprovincialis shell landmarks respond to environmental variation at large (hundreds of kilometres) spatial scale.

Geometric morphometry is effective in distinguishing bivalve species and populations, including the economically and environmentally important Mediterranean mussel, Mytilus galloprovincialis . Although widely distributed, shell shape variation in M. galloprovincialis along the eastern Adriatic Sea has been infrequently studied. Farming practices and environmental conditions may affect the development of shell shape, as has been reported in the Mytilus genus from numerous locations globally. Building on earlier genetic analyses of mussels collected along a natural environmental gradient, this study aimed to identify shell landmark differentiation between wild and cultured populations and among northern, middle, and southern populations of the eastern Adriatic Sea using a geometric morphometric approach. Samples from 12 sites in Croatia, Montenegro, and Albania, including 4 aquaculture farms, were examined for variation in 9 internal shell landmarks. Wild populations exhibited a more extended posterior adductor muscle scar and a more elongated shape compared to farmed populations. Mussels from low salinity environments in the north and south exhibited an elongated shell shape compared to high salinity environments. Southern populations exhibited an extended posterior adductor muscle scar, along with an elongation of the lateral ligament and ventral umbo orientation that caused a concave shape of the ventral shell border compared to northern populations. The differences in environmental conditions in the Adriatic Sea, such as reduced salinity in Boka Kotorska Bay (Montenegro) in the south and Limski Bay (Croatia) in the north, likely play a role in influencing the variability of shell landmarks. These results may be applied to farming practices so that high‐quality spat are collected from source sites with environmental conditions that match the farm site to which the spat are transferred. Overall, these results provide valuable insight into how M. galloprovincialis shell landmarks respond to environmental variation at large (hundreds of kilometres) spatial scale.

The development of molecular methods to detect Mytilus hybrids is important for food authentication, conservation, and the sustainable management of shellfish aquaculture as accurate food labeling is a legal requirement, and because introgression may promote undesirable phenotypes or displace native species. However, nuclear and mitochondrial markers can segregate independently, compromising diagnostic congruence between these markers. This study aimed to detect hybrids in the Mytilus edulis species complex using a multi-locus approach, including RFLP-PCR assays for Me 15/16, 16S rRNA, and COIxba, in samples collected from five continents. We used a model-based Bayesian method for hybrid and pure species detection to analyze the diagnostic potential of nuclear and mitochondrial markers in mussel samples from South America, North America, Africa, Oceania, and Europe. Our results showed that the combined use of markers can differentiate between M. trossulus, M. edulis, M. galloprovincialis, and M. chilensis. The combined use of nuclear and mitochondrial molecular markers also improves hybrid detection and allows us to identify introgression using Bayesian analysis.

The genetic characteristics of invasive species have a significant impact on their ability to establish and spread. The blue mussel (Mytilus galloprovincialis), native to the Mediterranean Sea, is a leading invasive species of intertidal coasts throughout much of the world. Here, we used mitochondrial DNA sequence data to investigate the genetic diversity and phylogeographic structure of invasive (M. galloprovincialis) versus native (Mytilus chilensis) populations of blue mussels in Chile. We evaluated whether genetic diversity in invasive populations could be explained by the genetic characteristics of the native sources from which they might be derived. A phylogenetic analysis confirmed two lineages of the invasive M. galloprovincialis, i.e., the NW Atlantic and the Mediterranean lineages. We found no evidence of genetic structure in the invasive range of M. galloprovincialis in Chile, most probably because of its recent arrival. We did, however, detect a spatial mixture of both M. galloprovincialis lineages at sampling locations along the Chilean coast, giving rise to higher levels of genetic diversity in some areas compared to the population of native M. chilensis. The coastal area of the invasion is still small in extent (~100 km on either side of two large ports), which supports the hypothesis of a recent introduction. Further expansion of the distribution range of M. galloprovincialis may be limited to the north by increasing water temperatures and to the south by a natural biogeographic break that may slow or perhaps stop its spread. The use of internal borders as a tool to minimise or prevent M. galloprovincialis spread is therefore a genuine management option in Chile but needs to be implemented rapidly.

The Asian green mussel (AGM), Perna viridis, is a mussel of ecological and economic importance throughout much of the tropical and subtropical regions of Asia, from China and Korea in the east to Oman on the Arabian Peninsula in the west. We collated published DNA sequence data and then analysed 467 bp of cytochrome c oxidase I gene (COI) sequence variation from 649 mussels (576 from SE Asia and 73 from India) to test for regional differentiation. Analysis of molecular variation, a haplotype network and a neighbour-joining tree all revealed significant differentiation between mussels from India and those from SE Asia. We observed a greater number of segregating sites (haplotypes) and private haplotypes than expected for Indian mussels compared to SE Asian mussels, based on the respective sample sizes. Tests of neutrality revealed population expansion or a recent selective sweep for only the mussels from India. We suggest that the differentiation of the Indian from the SE Asian mussels is explained by the pronounced seasonal input of freshwater into the northern Bay of Bengal area, which acts as a (semi-permeable) barrier to gene flow between the two regions (i.e., east versus west of the Bay). This suggestion is based on the oceanography of the region and is consistent with the biogeography and reports of genetic breaks in a range of taxa in this region. Further targetted sampling of AGMs from the east coast of India, Bangladesh and Myanmar is required to test this hypothesis, and additional sampling of AGMs from west of India (e.g., Pakistan, Iran, the Arabian Peninsula) will also be informative. Finally, this study is based on the analysis of a single marker (COI) because this is all that is presently available from published sources but we note that the application of new molecular markers such as single nucleotide polymorphisms to newly collected mussel samples will greatly advance our understanding of the AGM genetic discontinuity and its age, and help test its geographical location and its nature (e.g., a steep/shallow cline or a stepped cline).

In New Zealand, Kotatea aurantiaca and Kotatea lobata are two common, endemic, co-occurring, and morphologically similar soft coral species that currently cannot be distinguished without microscopic examination of sclerites and of which little is known regarding any aspect of their ecology or biology. The aim of the present study is to ascertain if, and in what ways, their colony growth forms differ, and to test the taxonomic value of macroscopic morphological character measurements using statistical discrimination analyses. A binary logistic regression model is developed whereby macroscopic characters of colony morphology can be used to assign specimens to either species with ~ 90% accuracy. Species assignment accuracy is greatest when ratios formed from morphological measurements are used rather than direct measurements. Here, these ratios are used for the first time to account for the appearance-altering habit among soft corals of hydrostatically expanding and contracting their coelenteron with seawater. Relationships between colony morphology and depth are also examined, and it is suggested that phenotypic plasticity detected in K. lobata, causing it to resemble K. aurantiaca more closely at greater depths, may contribute to their morphological overlap. It is anticipated that this discrimination technique will facilitate future research on the ecology and biology of these species and will be replicated on other sets of morphologically similar soft corals for which species discrimination has been problematic.