Mt. Morning is a 2,732 m high, Cenozoic, alkaline eruptive centre situated in the south-west corner of McMurdo Sound in the Ross Sea, Antarctica. Mt. Morning is approximately 100 km south-west of Mt. Erebus, the world's southernmost active volcano. Several Cenozoic, alkali eruptive centres in this region make up the Erebus Volcanic Province. The region is currently undergoing continental extension. Regional-scale, north-striking faulting on the northern flank of Mt. Morning has offset vertical dykes, as young as 3.9 Ma, by up to 6 m dextrally. This is consistent with the trans-extensional regime in the region. The faults also have a dip-slip component, downthrown to the east. These faults define part of the western boundary of the West Antarctic Rift System. Mt. Morning straddles the boundary between the continental rift shoulder of the Transantarctic Mountains in Southern Victoria Land, and the perceived oceanic crust of the Ross Sea. Age determination of the youngest offset dyke constrains movement in the last 3.88 � 0.05 m.y., to an average rate of 0.0015 mm per year. Volcanism on Mt. Morning is divided into two phases. Phase I was erupted between 18.7 � 0.3 and 114 � 0.2 Ma and Phase II between 6.13 � 0.20 and 0.15 � 0.01 Ma. The two phases are separated by a 5.3 m.y. period of quiescence. The geochemistry of Phase I is mildly alkaline; it is composed of volcaniclastic deposits, dykes, sills, and volcanic plugs of nepheline-basanite, nepheline-trachyte, quartz-mugearite, quartz-trachyte, and rhyolite. Phase I rocks evolved along at least two trends: a quartz normative trend, and a nepheline normative trend. Chemical variation in Phase I can be explained in part by crystal fractionation, which has been modelled using major element multiple linear regression. Phase I quartz-mugearite can fractionate to quartz-trachyte after 44% crystallisation. Quartz-trachyte can fractionate to rhyolite after a further 6% erystallisation. The models indicate that clinopyroxene + plagioclase + opaque oxides � alkali feldspar � apatite are the dominant fractionated phases. Many of the Phase I quartz normative volcanic rocks have relatively high ⁸⁷Sr/⁸⁶Sr ratios (0.70501), suggesting that assimilation, most likely of crustal material, has modified them. Phase I nepheline-basanite can fractionate to nepheline-trachyte after 68% crystallisation. Modelling indicates clinopyroxene + nepheline + olivine + opaque oxides are the dominant fractionated phases. Phase II volcanic rocks are strongly alkaline and are mapped as flows, volcaniclastic deposits, dykes, and sills. They have been erupted mainly from parasitic scoria vents and rarely from fissure vents. Rock types include picrobasalt, basalt, basanite, tephrite, hawaiite, mugearite, phonotephrite, tephriphonolite, benmoreite, and phonolite. Chemical variations in the Phase II volcanic rocks can be explained by simple fractionation. Phase II picrobasalt can fractionate to phonotephrite after 78% crystallisation. Phonotephrite can fractionate to phonolite after at least 35% crystallisation, depending on which of several multiple linear regression models are selected. Fractionation is dominated by the removal of clinopyroxene + plagioclase + nepheline + olivine + opaque oxides � apatite � kaersutite. Volcanic rocks in the Erebus Volcanic Province are strongly alkaline on a silica versus total alkalis plot, similar to the Phase II volcanic rocks from Mt. Morning. Mildly alkaline rocks of Phase I are, to date, unique within the Erebus Volcanic Province. Bulk rock isotope ratios of ⁸⁶Sr/⁸⁷Sr (0.70307 - 0.70371 and 0.70498 - 0.70501), �⁴�Nd/�⁴⁴Nd (0.512650 - 0.512902), and �⁰⁶Pb/�⁰⁴Pb (18.593 -20.039) show that the majority of Mt. Morning volcanic rocks lie on a mixing line between HIMU (high-[mu]; enriched in �⁰⁶Pb and �⁰⁸Pb and relatively depleted in ⁸⁶Sr/⁸⁷Sr values) and DM (depleted mantle; high �⁴�Nd/�⁴⁴Nd, low ⁸⁶Sr/⁸⁷Sr, and low �⁰⁶Pb/�⁰⁴Pb). This is similar to the majority of volcanic rocks from the SW Pacific, including Antarctica and New Zealand. Mt. Morning volcanic rocks have tapped this broadly common mantle reservoir. There are variations in radiogenic isotope ratios between Mt. Morning and Mt. Erebus. There are also differences between the incompatible element ratios in volcanic rocks from Mt. Morning, Mt. Erebus, and White Island (a third eruptive centre in the Erebus Volcanic Province), suggesting heterogeneity in the mantle beneath the Erebus Volcanic Province. Significant chemical differences are also noted between ultramafic xenoliths collected from Mt. Morning and from Foster Crater only 15 km away. This suggests a deca-kilometre, possibly even kilometre-scale, heterogeneity in the mantle. Such small-scale chemical differences appear difficult to reconcile with large-scale plume hypotheses for the initiation of volcanism in the Erebus Volcanic Province. Instead, volcanism is much more likely to be related to numerous small plumes, or the preferred hypothesis, metasomatism and amagmatic rifting, followed by decompression melting of upwelling mantle and volcanism during transtensional lithospheric rifting. This latter model is supported by a lack of regional updoming expected with a plume(s), and fits models of localised extension proposed in this thesis. Calc-alkaline and alkaline igneous xenoliths, of felsic to mafic crustal material, have been collected from Mt. Morning. U-Pb geochronology (545.4 � 3.7 Ma and 518.6 � 4.4 Ma) on crustal xenoliths from Mt. Morning illustrate that the basement is Cambrian. Bulk rock chemistry of crustal xenoliths has similarities to compositions reported for Ross Orogen rocks, suggesting the Mt. Morning volcanic edifice is built on a basement that is composed of Cambrian Ross Orogen rock types. Quartz-bearing felsic granulite xenoliths with greater than 70 weight percent silica, collected from Mt. Morning, suggest that part of the basement is felsic. This is the only occurrence of felsic xenoliths reported to date east of the present day coastline of Victoria Land. Mt. Morning crops out less than 25 km from the known northern end of the Koettlitz Glacier Alkaline Province in the Transantarctic Mountains. The partially alkaline basement beneath Mt. Morning suggests the province may continue beneath part of Mt. Morning. The mantle beneath Mt. Morning can be characterised as anhydrous and otherwise largely unmetasomatised, which is atypical of xenoliths collected from the western Ross Sea. Only a handful of Mt. Morning xenoliths show petrographic evidence of metasomatism, these include modal phlogopite, apatite, Fe-Ni sulphide, and plagioclase (in pyroxenite xenoliths), suggesting metasomatising fluids occur discretely in this region. Where present, the metasomatic fluid(s) beneath Mt. Morning are enriched in Ba, LREEs, Th, U, P, Fe, Ni, S, and K, and depleted in Ti relative to the metasomatic fluid composition described at nearby Foster Crater. Oxygen fugacity (fO₂) of the Antarctic shallow mantle has been measured from xenoliths collected from Mt. Morning, where fO₂ was demonstrated to be strongly dependant upon spinel Fe�⁺ content that was measured using Mössbauer spectroscopy, and calculated from the olivine-orthopyroxene-spinel oxybarometer. fO₂ in the rifted Antarctic mantle varies between 0.1 and -1 log units relative to the fayalite-magnetite-quartz buffer and is coupled to melt depletion, with increasing degrees of melt extraction resulting in a more oxidised mantle. This range of upper mantle fO₂ is commonly observed in continental rift settings worldwide. The mantle beneath Mt. Morning is composed of, in increasing degree of fertility, dunite, harzburgite, and lherzolite. Xenoliths representing discrete samples of this mantle have mostly crystallised in the spinel stability field of the mantle at pressures of approximately 15 kb and temperatures between 950 - 970 �C. Symplectites of spinel and pyroxene have been interpreted as petrographic evidence that some of the spinel peridotite originated in the garnet stability field of the mantle. Rare plagioclase-bearing spinel lherzolite (plagioclase lherzolite) is also present in the mantle beneath Mt. Morning, which crystallised at temperatures of between 885 and 935 �C at 5 kb. The Mt. Morning peridotite xenoliths plot along the pre-defined geotherm for the Erebus Volcanic Province, strongly supporting it as the appropriate choice of geothermal gradient for south-west McMurdo Sound. Mineral and bulk rock compositions are nearly identical between the plagioclase lherzolite xenoliths and spinel lherzolite xenoliths. Mineral and bulk rock chemistry suggest it is unlikely that the plagioclase is due to metasomatism. Petrographic evidence and mass balance calculations suggest that the plagioclase lherzolite has crystallised via a sub-solidus (metamorphic) transition from spinel lherzolite upon decompression and upwelling of the mantle. The occurrence of plagioclase lherzolite beneath Mt. Morning could be explained by lithospheric scale uplift along faults that define the western margin of the West Antarctic Rift System. Plagioclase lherzolite has also been collected and described from White Island. White Island is also interpreted to straddle lithospheric scale faults. Rifting and buoyant uplift is sufficient to explain the presence of plagioclase lherzolite in the Erebus Volcanic Province. Plagioclase lherzolite has also been described from Mt Melbourne, an eruptive centre in Northern Victoria Land. Known occurrences of plagioclase lherzolite from the western shoulder of the Ross Sea now cover an area 430 km long and 80 km wide. This is one of the largest provinces of plagioclase peridotite worldwide so far reported.

Mt. Morning is a 2,732 m high, Cenozoic, alkaline eruptive centre situated in the south-west corner of McMurdo Sound in the Ross Sea, Antarctica. Mt. Morning is approximately 100 km south-west of Mt. Erebus, the world's southernmost active volcano. Several Cenozoic, alkali eruptive centres in this region make up the Erebus Volcanic Province. The region is currently undergoing continental extension. Regional-scale, north-striking faulting on the northern flank of Mt. Morning has offset vertical dykes, as young as 3.9 Ma, by up to 6 m dextrally. This is consistent with the trans-extensional regime in the region. The faults also have a dip-slip component, downthrown to the east. These faults define part of the western boundary of the West Antarctic Rift System. Mt. Morning straddles the boundary between the continental rift shoulder of the Transantarctic Mountains in Southern Victoria Land, and the perceived oceanic crust of the Ross Sea. Age determination of the youngest offset dyke constrains movement in the last 3.88 � 0.05 m.y., to an average rate of 0.0015 mm per year. Volcanism on Mt. Morning is divided into two phases. Phase I was erupted between 18.7 � 0.3 and 114 � 0.2 Ma and Phase II between 6.13 � 0.20 and 0.15 � 0.01 Ma. The two phases are separated by a 5.3 m.y. period of quiescence. The geochemistry of Phase I is mildly alkaline; it is composed of volcaniclastic deposits, dykes, sills, and volcanic plugs of nepheline-basanite, nepheline-trachyte, quartz-mugearite, quartz-trachyte, and rhyolite. Phase I rocks evolved along at least two trends: a quartz normative trend, and a nepheline normative trend. Chemical variation in Phase I can be explained in part by crystal fractionation, which has been modelled using major element multiple linear regression. Phase I quartz-mugearite can fractionate to quartz-trachyte after 44% crystallisation. Quartz-trachyte can fractionate to rhyolite after a further 6% erystallisation. The models indicate that clinopyroxene + plagioclase + opaque oxides � alkali feldspar � apatite are the dominant fractionated phases. Many of the Phase I quartz normative volcanic rocks have relatively high ⁸⁷Sr/⁸⁶Sr ratios (0.70501), suggesting that assimilation, most likely of crustal material, has modified them. Phase I nepheline-basanite can fractionate to nepheline-trachyte after 68% crystallisation. Modelling indicates clinopyroxene + nepheline + olivine + opaque oxides are the dominant fractionated phases. Phase II volcanic rocks are strongly alkaline and are mapped as flows, volcaniclastic deposits, dykes, and sills. They have been erupted mainly from parasitic scoria vents and rarely from fissure vents. Rock types include picrobasalt, basalt, basanite, tephrite, hawaiite, mugearite, phonotephrite, tephriphonolite, benmoreite, and phonolite. Chemical variations in the Phase II volcanic rocks can be explained by simple fractionation. Phase II picrobasalt can fractionate to phonotephrite after 78% crystallisation. Phonotephrite can fractionate to phonolite after at least 35% crystallisation, depending on which of several multiple linear regression models are selected. Fractionation is dominated by the removal of clinopyroxene + plagioclase + nepheline + olivine + opaque oxides � apatite � kaersutite. Volcanic rocks in the Erebus Volcanic Province are strongly alkaline on a silica versus total alkalis plot, similar to the Phase II volcanic rocks from Mt. Morning. Mildly alkaline rocks of Phase I are, to date, unique within the Erebus Volcanic Province. Bulk rock isotope ratios of ⁸⁶Sr/⁸⁷Sr (0.70307 - 0.70371 and 0.70498 - 0.70501), �⁴�Nd/�⁴⁴Nd (0.512650 - 0.512902), and �⁰⁶Pb/�⁰⁴Pb (18.593 -20.039) show that the majority of Mt. Morning volcanic rocks lie on a mixing line between HIMU (high-[mu]; enriched in �⁰⁶Pb and �⁰⁸Pb and relatively depleted in ⁸⁶Sr/⁸⁷Sr values) and DM (depleted mantle; high �⁴�Nd/�⁴⁴Nd, low ⁸⁶Sr/⁸⁷Sr, and low �⁰⁶Pb/�⁰⁴Pb). This is similar to the majority of volcanic rocks from the SW Pacific, including Antarctica and New Zealand. Mt. Morning volcanic rocks have tapped this broadly common mantle reservoir. There are variations in radiogenic isotope ratios between Mt. Morning and Mt. Erebus. There are also differences between the incompatible element ratios in volcanic rocks from Mt. Morning, Mt. Erebus, and White Island (a third eruptive centre in the Erebus Volcanic Province), suggesting heterogeneity in the mantle beneath the Erebus Volcanic Province. Significant chemical differences are also noted between ultramafic xenoliths collected from Mt. Morning and from Foster Crater only 15 km away. This suggests a deca-kilometre, possibly even kilometre-scale, heterogeneity in the mantle. Such small-scale chemical differences appear difficult to reconcile with large-scale plume hypotheses for the initiation of volcanism in the Erebus Volcanic Province. Instead, volcanism is much more likely to be related to numerous small plumes, or the preferred hypothesis, metasomatism and amagmatic rifting, followed by decompression melting of upwelling mantle and volcanism during transtensional lithospheric rifting. This latter model is supported by a lack of regional updoming expected with a plume(s), and fits models of localised extension proposed in this thesis. Calc-alkaline and alkaline igneous xenoliths, of felsic to mafic crustal material, have been collected from Mt. Morning. U-Pb geochronology (545.4 � 3.7 Ma and 518.6 � 4.4 Ma) on crustal xenoliths from Mt. Morning illustrate that the basement is Cambrian. Bulk rock chemistry of crustal xenoliths has similarities to compositions reported for Ross Orogen rocks, suggesting the Mt. Morning volcanic edifice is built on a basement that is composed of Cambrian Ross Orogen rock types. Quartz-bearing felsic granulite xenoliths with greater than 70 weight percent silica, collected from Mt. Morning, suggest that part of the basement is felsic. This is the only occurrence of felsic xenoliths reported to date east of the present day coastline of Victoria Land. Mt. Morning crops out less than 25 km from the known northern end of the Koettlitz Glacier Alkaline Province in the Transantarctic Mountains. The partially alkaline basement beneath Mt. Morning suggests the province may continue beneath part of Mt. Morning. The mantle beneath Mt. Morning can be characterised as anhydrous and otherwise largely unmetasomatised, which is atypical of xenoliths collected from the western Ross Sea. Only a handful of Mt. Morning xenoliths show petrographic evidence of metasomatism, these include modal phlogopite, apatite, Fe-Ni sulphide, and plagioclase (in pyroxenite xenoliths), suggesting metasomatising fluids occur discretely in this region. Where present, the metasomatic fluid(s) beneath Mt. Morning are enriched in Ba, LREEs, Th, U, P, Fe, Ni, S, and K, and depleted in Ti relative to the metasomatic fluid composition described at nearby Foster Crater. Oxygen fugacity (fO₂) of the Antarctic shallow mantle has been measured from xenoliths collected from Mt. Morning, where fO₂ was demonstrated to be strongly dependant upon spinel Fe�⁺ content that was measured using Mössbauer spectroscopy, and calculated from the olivine-orthopyroxene-spinel oxybarometer. fO₂ in the rifted Antarctic mantle varies between 0.1 and -1 log units relative to the fayalite-magnetite-quartz buffer and is coupled to melt depletion, with increasing degrees of melt extraction resulting in a more oxidised mantle. This range of upper mantle fO₂ is commonly observed in continental rift settings worldwide. The mantle beneath Mt. Morning is composed of, in increasing degree of fertility, dunite, harzburgite, and lherzolite. Xenoliths representing discrete samples of this mantle have mostly crystallised in the spinel stability field of the mantle at pressures of approximately 15 kb and temperatures between 950 - 970 �C. Symplectites of spinel and pyroxene have been interpreted as petrographic evidence that some of the spinel peridotite originated in the garnet stability field of the mantle. Rare plagioclase-bearing spinel lherzolite (plagioclase lherzolite) is also present in the mantle beneath Mt. Morning, which crystallised at temperatures of between 885 and 935 �C at 5 kb. The Mt. Morning peridotite xenoliths plot along the pre-defined geotherm for the Erebus Volcanic Province, strongly supporting it as the appropriate choice of geothermal gradient for south-west McMurdo Sound. Mineral and bulk rock compositions are nearly identical between the plagioclase lherzolite xenoliths and spinel lherzolite xenoliths. Mineral and bulk rock chemistry suggest it is unlikely that the plagioclase is due to metasomatism. Petrographic evidence and mass balance calculations suggest that the plagioclase lherzolite has crystallised via a sub-solidus (metamorphic) transition from spinel lherzolite upon decompression and upwelling of the mantle. The occurrence of plagioclase lherzolite beneath Mt. Morning could be explained by lithospheric scale uplift along faults that define the western margin of the West Antarctic Rift System. Plagioclase lherzolite has also been collected and described from White Island. White Island is also interpreted to straddle lithospheric scale faults. Rifting and buoyant uplift is sufficient to explain the presence of plagioclase lherzolite in the Erebus Volcanic Province. Plagioclase lherzolite has also been described from Mt Melbourne, an eruptive centre in Northern Victoria Land. Known occurrences of plagioclase lherzolite from the western shoulder of the Ross Sea now cover an area 430 km long and 80 km wide. This is one of the largest provinces of plagioclase peridotite worldwide so far reported.