The 3.8 – 3.7 Ga Isua Supracrustal Belt (ISB) contains a great variety of rocks formed on and below the seafloor. However, many of the rock units have experienced a high degree of deformation and metasomatism during amphibolite facies metamorphism. In order to reconstruct the conditions present during the formation of these rock units, and by extension, the processes operating during the Eoarchean, I have located, described and analyzed samples from outcrops that preserve the structures created during their formation. Along the center of the ISB occurs a suite of highly deformed and altered felsic rocks, for which the origin has remained enigmatic. This felsic unit becomes intercalated with mafic lithologies towards the south and is eventually replaced by mafic rocks with crosscutting felsic sheets. Within the intercalating zone, I discovered an outcrop containing pristine contacts between felsic and mafic magma. Samples from this outcrop display lower degrees of deformation and limited metasomatism. Through petrographic analysis I have characterized the contact as a magma mingling feature. In my first manuscript, I have determined the age of mafic intrusion into felsic magma at 3808.0 ± 1.2 Ma by U-Pb dating of zircons from the felsic contact. The zircons have textures that we associate with recrystallization during reheating. This is the first absolute age for mafic magmatism in the outer belt of the ISB. I have also dated zircons from more deformed felsic rocks both within the central felsic unit and within the outer crosscutting sheets and find that these rocks were most likely formed within <10 Ma of the magma mingling zone and each other. Some zircons have ages ca. 10 Ma above this range, which I attribute to a small inherited, xenocrystic or antecrystic component. The zircon age population is skewed towards younger ages indicating that some zircons experienced early Archean Pb-loss. In accordance with previous dating of metamorphic phases, I attribute the majority of this loss to a tectonic event at ca. 3.75 Ga. This event could correspond to the accretionary emplacement of the ISB. εHf(3808 Ma) values measured for zircons range from +2.4 to −2.6. I ascribe higher values to interaction with mafic magma at emplacement level and lower values to inherited or xenocrystic grains for which the U-Pb age has been reset during magmatic transport. This is corroborated by zircon textures and sample context. The average δ18O of zircon from samples that are not in direct contact with mafic rocks is 4.79 ‰ and varies down to 3.60 ‰. This indicates that they were derived from a source with sub-mantle δ18O. I argue that the magmatic source was seafloor gabbros that experienced high temperature hydrothermal alteration by fluids that had previously interacted with crustal rock at lower temperatures. Structural observations combined with zircon ages make it clear that the felsic sheets in the outer belt and the central felsic unit are not a late addition to the ISB but rather formed within the same tectonic configuration as the mafic rocks. This conclusion contrasts with interpretations in earlier work suggesting that the felsic rocks were an expression of the TTG magmatism observed in the surrounding gneiss complex. Models for the formation of the ISB must therefore account for coeval formation of mafic and felsic melts. The Hafnium and oxygen isotopic compositions of zircons are consistent with a melt source partially consisting of supracrustals but dominated by lower oceanic crustal rocks. In the second manuscript, I account for the alteration and magma mingling processes that variably affect parts of the felsic suite. I then make inferences about the melting conditions and composition of the source rock. I present major and trace elemental data and whole rock Hf and Nd isotopic compositions for the felsic suite of rocks and comingled mafic rocks. I also present field and petrographic observations with which I provide context for inferences about both magma mingling processes and metasomatic alteration of the central felsic unit. Unaltered and uncontaminated felsic rocks have granodioritic to trondhjemitic compositions indicating a mafic source composition. High Sr/Y ratios and low fractionated HREE compositions indicate that melting took place within the garnet stability field without stable plagioclase. This is consistent with burial to eclogite facies conditions. Comingling mafic rocks were derived from melting of depleted mantle outside the garnet stability field, possibly indicating that felsic melts were generated below the region of mantle melting. Low Zr/Sm and high Th content in the felsic rocks is indicative of an enriched source. Such a source could have been plume or arc generated basalts or it could have formed through metasomatism preceding melting. A negative Pb anomaly in the pristine felsic samples indicates that their source experienced fluid loss before melting, which is consistent with progressive burial of hydrothermally altered basaltic lithologies. An external fluid source is therefore required to trigger melting. This external fluid might also be responsible for enrichment of the source rock. I propose that dehydration of the lower part of the oceanic crust occurred during its subduction. Melting conditions were not met due to sluggish heating of the inner slab. With deeper burial, dehydration of serpentinite beneath the gabbroic section led to metasomatic migration of fluids and triggering of melting. Whole rock Hf and Nd isotopic compositions of uncontaminated felsic samples are decoupled from the modern mantle array with near chondritic Hf isotopic compositions and εNd(3808 Ma) up to +3.5. These compositions are similar to those of surrounding TTG gneisses and basaltic amphibolites from the ISB, implying a connection between the three groups of samples. Samples whose context and composition indicate contamination by mafic magma have lower εNd(3808 Ma) and higher εHf(3808 Ma). This suggests that mafic magma was derived from depleted mantle, which had not experienced decoupling of the Lu-Hf and Sm-Nd systems. Due to disturbed Hf and Nd isotopic compositions in the mafic samples, this cannot be corroborated by their direct analysis. The derivation of felsic magma from a source with a Hf-Nd isotopic composition outside the terrestrial array could indicate that felsic melts were derived from lithologies that were derived from a different mantle region than the comingling mafic melts. This would be consistent with horizontal movement of the crust, relative to the mantle, as would be necessary for subduction. Our model implies that the mantle source of ISB amphibolites could have been contaminated by felsic melts, which might explain similar Hf and Nd isotopic compositions for some samples. The third manuscript deals with a completely different suite of rocks. Well-preserved amphibolitic pillow basalts occur within the inner belt in the north-western arm. The pillow basalts contain amygdules consisting of various mineral phases, most prominently epidote, oligoclase, quartz, calcite and chalcopyrite. I characterized the textures and mineral compositions of amygdules and measured the sulfur isotopic composition of chalcopyrite as well as minor sulfide phases, pyrrhotite and pyrite. M-shaped REE patterns for whole rock samples indicate the loss of LREE and Eu during >230° C hydrothermal alteration of the pillow. This is supported by the mobilization of copper, which requires high temperature fluids, and by epidote formation in the amygdules. A subsequent lower temperature (<100° C) stage of hydrothermal alteration is indicated by the formation of calcite and of complex intergrowths of quartz and oligoclase that we interpret as forming from metamorphic recrystallization of zeolite. Chalcopyrite and pyrite sulfur isotopic compositions display dominantly near zero δ34S values and positive Δ33S. This indicates that sulfate was not involved in their formation, but that sulfur was partially derived from sediments containing sulfur that had been cycled through the atmosphere under anoxic conditions. I argue that this sulfur was recycled into the mantle source of the basalts during subduction of sediments. Pyrrhotite grains contain a larger range in δ34S of 7.8 ‰ and have negative Δ33S indicating their derivation from sulfate formed through atmospheric photochemical reactions. I posit that sulfate built up in relation to the hydrothermal system during the high temperature hydrothermal stage and was released during the low temperature stage. Lower temperatures also allowed the sulfate to be reduced though the operation of microbial sulfate reduction. The biological reduction of sulfate is supported by Δ36S values that lie beneath the Archean array and by elevated phosphorous concentrations in calcite within the amygdules.