Noble gas and halogen data from a suite of Icelandic samples are presented. Iceland combines hotspot volcanism, a spreading ridge and abundant subglacially erupted samples. This combination allows for samples that erupted under high enough pressures to retain a measurable mantle volatile content, and also display signatures representing interaction between ocean island basalt (OIB) and mid-ocean ridge basalt (MORB) mantle sources.Erupted samples used to determine the mantle's halogen and noble gas content have undergone a degassing process that can alter their volatile composition. An existing disequilibrium degassing model is developed with the modified model taking into account the evolution of the major volatiles over a multi-stage process and the different conditions present during magma ascent and quenching. The modified model allows substantially lower elemental noble gas ratios to be reached under disequilibrium conditions than allowed by the original model. Initial CO2 concentrations, pressure, diffusivity, ascent rate and degree of disequilibrium are shown to be critical parameters for this model. Final degassed noble gas concentrations are most affected by the surface quenching stage of an eruption, whereas noble gas elemental ratios can be primarily determined during magma ascent. In applying this model to MORB and OIB sample suites, the 3He/22Ne ratio of the MORB source mantle is constrained to be lower than 4.4, similar to estimates for the OIB source mantle. Additionally the most straightforward match between the degassing model and OIB helium and neon data suggest the OIB source mantle has 3He concentrations similar to or lower than the MORB source mantle. This finding requires a model for the OIB source mantle in which a high 3He/4He component is added to a helium-poor protolith.Noble gas studies are hampered by the large, isotopically atmospheric component typically found in Icelandic subglacial samples, which can swamp other signatures. Detailed analysis of a volatile rich sample from SW Iceland shows evidence for more than one 'contaminant' component and that two component fits used incorrectly can produce misleadingly precise source mantle noble gas ratios. Multi component best fits to noble gas elemental ratios find that four components are present in samples from this region. These components are unfractionated air, fractionated air and a mantle component which shows some variation due to degassing. Combining the disequilibrium degassing model with component resolution allows limits to be placed on the source mantle composition for this sample. The light noble gas source composition is compatible with mixing between a solar ('direct nebula') component and a MORB-like component. This direct nebula signature is at odds with an implanted signature seen in both Ne and Kr for the convecting mantle, and shows that both accretionary volatile origins must have contributed during the Earth's formation. The heavy noble gases show an elemental abundance pattern which is distinct from air and solar patterns, and trends towards seawater. This confirms the presence of a recycled volatile signature in Iceland's mantle but it is not possible to further constrain the origin of this signature.The Icelandic halogen data shows no evidence for significant fractionation during degassing or melt generation. Source estimates for the Br/Cl and I/Cl ratios for Iceland's plume are found to be (1.56±0.03) x 10-3 and (3.1±0.3) x 10-5, compatible with estimates for the MORB source mantle. Halogen source concentrations in central Iceland are found to be approximately three times higher than estimates for the convecting mantle and correlate with the regions of Iceland that show high 3He/4He ratios and high source water contents. This may indicate a recycled halogen signature associated with Iceland's proposed mantle plume.