The Zambales ophiolite of western Luzon, Philippines, exposes a typical succession of basalt flows, diabasic dikes, gabbro and tectonized harzburgite. The age established by limiting strata is late Eocene. Lack of evidence of thrust faulting and the general domal disposition of the lithologie units indicate that the ophiolitic rocks are exposed by uplift. Highly complex internal layered structures within the complex are related to processes developed during formation of the ophiolite and the Zambales ophiolite may be one of the least disturbed (by emplacement) ophiolitic masses known. The exposed mass trends north and the upper surface plunges at low angles (a few degrees) to the north and south. The chemistry and composition of the rocks in the northwest part of the Zambales area (Acoje block) is distinct from that in the southeastern segment (Coto block). The Acoje block, according to Evans (1983) and Hawkins and Evans (1983), resembles (on a chemical basis) arc-tholeiite series rocks from intra-island arcs and the rocks in the Coto block are typical back-arc basin rock series. The present writer believes that the ophiolite composes a single genetic unit and that the changes in composition are the result of changes that took place during the initial formation. The gabbro probably formed below a spreading center in an elongate, in cross section, V-shaped, magma chamber. The gabbro is estimated by the writer to be less than 2 km thick and may be less than 1 km in places. Numerous erosional windows through the gabbro in the northern and eastern side of the Zambales area show that the gabbro remaining in those areas is likely to be only a few hundred meters thick. Harzburgite is exposed to a depth of about 800 m in the Bagsit River area and this may be the deepest part of the ophiolite accessible for study on which there is any control on depth. A transitional zone, about 200 m thick lying between the gabbro and harzburgite, is composed of serpentinized dunite. Commonly the dunite contains disseminated sulfide minerals and at the Acoje Mines, platinum-group elements. A compositional layering within the gabbro is in places cumulate in the lower part of the unit but may have formed by nucleation higher up on the relatively steep sides of the magma chamber. A widespread gneissic banding in the gabbro forms large mappable structures which are many times more complex than is the disposition of the major rock units. These structures are believed to be the result of extensive slumping in the magma chamber. The structure produced by the cumulate layering merges with the gneissic banding, commonly without discernible change in attitude. This tectonic layered structure crosses the gabbro-peridotite boundary at any angle without seeming to disturb the original rock distribution. At greater depths below the boundary (ca. 800 m), the harzburgite contains low dipping banding, which probably reflects the result of differential movement within the mantle. Chromite occurs almost exclusively in a zone that generally lies no more than 200-300 m below the gabbro-peridotite boundary. Refractory-grade chromite is found in this zone below the olivine gabbro in the Goto block and as low-grade metallurgical grade chromite below norite in the Acoje block. At Acoje Mines the chromite is present in layers in dunite, which the writer interprets as being distributed in a zone along the gently dipping (ca. 25°) gabbro-peridotite boundary. The steeply dipping (ca. 60-80 ° ) individual layers lie en echelon along the boundary at an angle (ca. 50 ° ) to the contact. At Coto the chromite forms large discontinuous masses in the lowest dunite and in the uppermost harzburgite. Except for the chromite present as layers at Acoje, the regional tectonic layering crosses the chromite deposits without structural deviation. The chromite deposits and associated peridotite may be cumulate in origin, but have been modified to such an extent that cumulate textures are generally obliterated. The angle of repose of cumulate layers in the Acoje area and in the Coto block dip towards each other raising the possibility that the Zambales area may contain the relics of a spreading center. Initial emplacement of the Zambales ophiolite took place by uplift and the ultramafic portion was exposed to erosion in the earliest Miocene or late Oligocene. Submergence of some of the ophiolite followed the previous uplift and on the west side of the Zambales Range submergence of several kilometers is indicated. Final emergence appears to have taken place in Pliocene or Pleistocene time by block uplift and areas of greatest uplift closely conform to the present topographic surface.