This paper describes the microfabrics of a quartzitic mylonite from a shear zone which separates major tectonic units in the Upper Jotun Nappe, a part of the Jotun Nappe Complex. The mylonite is rich in porphyroclasts, consisting of a variety of minerals such as feldspar, garnet, clinopyroxene, epidote and titanite. The porphyroclasts display a large variation in shape from plate and needle with extreme aspect ratios to ellipsoid and sub-spherical. Many of the porphyroclasts are asymmetric, fish-shaped, demonstrating a top-to-ESE transport direction in the shear zone. Relict subgrain microstructures in the largest porphyroclasts indicate that the precursor was a gneiss, which was deformed by dislocation creep. Compositional zoning is common among the porphyroclasts. In titanite this zoning is clear evidence of diffusive mass transfer, which indicates that dissolution-precipitation creep was a dominant process in the shaping of the porphyroclasts during mylonitisation. A change in the metamorphic environment is demonstrated by the microstructures. These include (1) larger garnet porphyroclasts are enclosed by a corona-like mineral aggregate, which includes a new generation of garnet, and (2) a silicification of plagioclase porphyroclasts, which suggests that the mylonitisation was accompanied by an influx of silica-rich fluids. The crystallographic preferred orientation of garnet porphyroclasts adds no support to recent reports that garnet plasticity may explain, at least partly, the elongate shapes. Misorientation angles of the garnet porphyroclasts demonstrate a close-range orientation relationship, which is ascribed to fragmentation and drifting apart processes during mylonitisation. The quartz matrix is medium- to coarse-grained. A large fraction of the porphyroclasts is completely included in these quartz grains, which indicates that the formation of porphyroclasts was followed by annealing and grain growth in the quartz matrix. Quartz displays no, or a very weak, shape-preferred orientation but a very strong crystallographic preferred orientation, compatible with glide on the {m} system. This, together with interlobate grain boundaries, lattice bending which grades into subgrain microstructures, and evidence of deformation lamellae, is taken to indicate that the annealing period was followed by a combination of dislocation creep and dislocation glide under medium- to high-temperature conditions. Arrays of transverse fracture systems in the elongate porphyroclasts are ascribed to tensile stresses set up during exhumation and cooling. Contrary to an earlier interpretation it is argued that the origin of this stress system is not distant but of a local nature. Because of petrological similarities with nearby plutonic rocks it is considered unlikely that the quartzite has a supracrustal origin, but is rather a mangeritic gneiss, which was highly enriched in silica. Comparison with recent geochronological data leads to the conclusion that the most likely time of formation of the shear zone was during the Sveconorwegian Orogeny.