The solution of the problem whether the ultrabasic inclusions in lavas are accumulation products of early crystallized minerals of the lavas in which they occur, or fragments of the earth’s peridotite shell carried to the surface by the eruptive force of the lavas, largely depends on the answer to the question whether these inclusions are tectonites or not. The structure of the specimens from Auvergne (France), which formed the main subject of this study, has been proved conclusively to be of a tectonic nature, from the macroscopically visible intersecting slip planes which are definitely younger than the banding of the specimen, as well as from microscopic evidence that suggests that both the olivine and enstatite crystals are concentrated in the intersecting slip planes. The fabric of olivine and enstatite proved to be symmetrical in respect of these slip planes, although remnants of an initial, predeformational orientation related with the banding still persist. In order to establish the relation between the old and the new fabric a comparative study of the fabric of a nodule from Dreiser Weiher (chapter II) and the crystal orientation of the banded sample from Auvergne (chapter III) has been made. It was concluded that the tectonic fabric of the sample from Auvergne could be interpreted as the result of a mechanical rotation of the composing crystals around two rotation axes, from their earlier orientation which is observed in the German specimen, into their new orientation which is symmetrical in respect of the intersecting slip planes. The rotation axis of olivine proved to be parallel to the intersection of the banding and a slip plane (S1), the rotation axis of enstatite is parallel to the intersection of the second slip planes (S2) and a plane normal to the banding, which is characterized by the X01—Yen girdle in the pre-deformational fabric. The geometry of the observed fabric further suggests that the crystals rotated in opposite sense over supplementary angles around the rotation axes. The distribution of olivine in the intersecting planes as well as the above mentioned rotation were further proved in a second sample from Auvergne on which an axial distribution analysis was carried out. The results of this analysis confirmed the expectations that crystals of a specific orientation are concentrated in intersecting directions in the plane of observation. The lattice orientation of olivine proved to be dominated by an orientation of {010} parallel to S1 and [010] sub-parallel to S2, while the enstatite crystals showed a strong preferred orientation of [001] sub-parallel to S1 and [100] in S2. The comparison of the preferred orientation of the X01 axes in five samples of different mineralogical composition has demonstrated that the preferred orientation of the X01 axes increases with increasing olivine content. All these results point to a tectonic nature of the structure of the ultrabasic inclusions in the lavas of Auvergne. Since the Mg/Fe distribution in the pyroxenes of the nodules suggests that these minerals crystallized at temperatures well above those of magmatic assemblages, it was concluded that the studied specimens are not derived by crystal fractionation from the lavas in which they occur, but are likely to be fragments of the earth’s peridotite shell. The fabric of the type-locality of the lherzolites in the French Pyrenees proved to be secondary and symmetrical in respect of the local Alpine stress field in such a way that the Z axes of both olivine and enstatite are parallel to the major axis of the stress deviator. The interpretation of this tectonic fabric of olivine has been based on the translation mechanism of olivine, described by Chudoba and Frechen (1950). In the last chapter attention has been paid to some recent theories and experiments which all lead to the conclusion that the orientation of crystals during growth, either in a uniaxial stress field or under a temperature gradient, is governed by the lattice of the crystals involved, a conclusion that might be useful for the interpretation of olivine fabrics, for the results of this study suggest that the crystal structure and not the grain shape governs he fabric of many ultrabasic rocks.