Background: The blue mussel, Mytilus edulis L., is an important contributor to the shellfish sector of Scottish aquaculture, with 7,270 tonnes worth £8.8 million being produced for the year 2015. Since 2010, production values have fluctuated as a result of inconsistent spat settlement, several business closures, and heightened levels of marine toxins in some areas. On Scotland’s west coast, some farms (most notably Loch Etive) have suffered production losses from the appearance of non-marketable mussels with particularly fragile shells and poor quality meat. Recent research has demonstrated that these undesirable traits have a genetic factor, linked to the presence of a non-native but related species Mytilus trossulus (Gould, 1850) and often its hybrids with the native M. edulis. M. trossulus has been classed as a commercially damaging species under Scottish law, but there is insufficient data on hybridisation and introgression patterns in Scottish mussel populations to evaluate any possible impacts this could have on production. Existing research has focused on single locus genotyping to identify Mytilus spp. and their hybrids in Scotland. By instead utilising multilocus genotyping, introgression could be identified and a better understanding of population structure could be gained, with implications for management to maintain productivity and profitability. The aim of the research presented here was to develop and validate a suite of new species diagnostic markers for multilocus genotyping of field populations of Scottish mussels, thereby establishing a more complete picture of the taxonomic relationships between species than previous studies have permitted. Results: Analysis of SNPs identified with RADseq confirmed the presence of three genetically distinct Mytilus species in Scotland: M. edulis, M. galloprovincialis and M. trossulus. RADseq and KASP genotyping technology successfully identified and validated a suite of 12 highly robust diagnostic SNP markers for multilocus genotyping of Mytilus mussel populations. These markers permitted more comprehensive genotyping than previous studies had, allowing presumed pure species individuals to be distinguished from first generation (F1) hybrids and introgressed (FX) genotypes in reference populations, and subsequently presented the possibility of exploring introgression in a wider scale study. Multilocus genotyping of mussel populations from around Scotland revealed widespread introgression of M. edulis with both M. galloprovincialis and M. trossulus. No pure M. galloprovincialis was identified and pure M. trossulus was restricted to a single site in Loch Etive, possibly part of a relict population. F1 hybrids between M. edulis and M. trossulus were identified in Loch Etive and in Loch Fyne on the west coast. This was evidence of ongoing hybridisation and suggested an active hybrid zone existed in Scotland, something that previous single locus genotyping studies had not acknowledged. A link between shell fragility and M. trossulus introgression was recognised at a single site outside of Loch Etive, but this was not apparent anywhere else and the actual causes of shell fragility remain unevaluated. There was a clear difference between the genetics of most farmed stock and wild populations, which indicated an anthropogenic effect on introgression and subsequent species composition, and had implications for future farm site selection and broodstock sourcing. Temporal species composition in Loch Etive differed over a short time period, but high proportions of M. trossulus alleles were observable some 25 months after a major fallowing event had taken place. Pure M. trossulus was also identifiable, which was consistent with the presence of an established population of M. trossulus existing in this area. Conclusion: Multilocus genotyping has produced a more in depth picture of species diversity in Scottish mussel populations. SNP assays revealed widespread introgression between three genetically distinct species – M. edulis, M. galloprovincialis and M. trossulus – and furthermore recognised that, to date, single locus genotyping has overestimated the abundance of pure Mytilus mussels in Scottish waters. However, this hitherto unidentified genetic complexity does not appear disadvantageous to mussel production, despite the prevalence of M. trossulus introgression among farmed populations, and it is somewhat unlikely that genetics are the sole cause of undesirable shell characteristics among Mytilus spp. mussels.

This thesis focuses on toxicity and occurrence of hydroxylated polybrominated diphenyl ethers (OH-PBDEs) in Baltic Sea biota. The aims were to assess OH-PBDEs potency for disruption of oxidative phosphorylation (OXPHOS) and determine their and related compounds exposure in Baltic blue mussel, herring and long-tailed duck. A method for analysis of OH-PBDEs in herring and long-tailed duck plasma was also evaluated. Relevant OH-PBDEs were tested in vitro for OXPHOS disruption, using a classic rat mitochondrial respiration assay and a cell mitochondrial membrane potential assay. All compounds were found to disrupt OXPHOS either by protonophoric uncoupling and/or via inhibition of the electron transport chain. 6-OH-BDE47 and 6-OH-BDE85, were identified as particularly potent OXPHOS disruptors. Strong synergism was observed when OH-PBDEs were tested as a mixture corresponding to what is present in Baltic blue mussels. Baltic blue mussel is main feed for several species of mussel feeding sea ducks which have decreased dramatically in numbers. To assess long-tailed ducks exposure to brominated substances, liver tissue from long-tailed ducks wintering in the Baltic Sea and blue mussels were analysed. The result confirms that long-tailed duck are exposed to OH-PBDEs via their diet. However, low concentrations were found in the duck livers, which suggest low retention of these compounds despite daily intake. How the nutritional value of blue mussels as feed for sea ducks are affected by OH-PBDE exposure still needs further studies. Other species of sea ducks foraging on Baltic blue mussels during summer months can also be more exposed due to seasonal variation in primary production. Herring sampled in the Baltic Proper and Bothnian Sea, were found to contain OH-PBDEs and high levels of their methylated counterpart, MeO-PBDEs. As demethylation of MeO-PBDEs is known to occur in fish, MeO-PBDEs may pose as additional source for more toxic OH-PBDEs in herring and their roe. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 2: Manuscript. Paper 3: Manuscript.</p>