Intracrystalline distortions (like undulose extinction, dislocations, and subgrain boundaries) in olivine from naturally-deformed peridotites are generally taken as signs of dislocation creep. However, similar features in olivine phenocrysts that have been found in basaltic magmas are still not well understood. In particular, whether subgrain boundaries in olivine phenocrysts arise from plastic deformation or grain growth is still debated (in the latter case, they are essentially grain boundaries but not subgrain boundaries. Therefore, we used hereinafter subgrain-boundary-like structures instead of subgrain boundaries to name this kind of intracrystalline distortion). Here we carried out a detailed study on dislocations and subgrain-boundary-like (SG-like) structures in olivine phenocrysts from two Hawaiian basaltic lavas by means of petrographic microscopy, scanning electron microscopy, and transmission electron microscopy (TEM). Abundant and complex dislocation substructures (free dislocations, dislocation walls, and dislocation tangles) were observed in the decorated olivine grains, similar to those in olivine from peridotite xenoliths entrained by the Hawaiian basalts. The measured average dislocation density is (2.9±1.3)×10¹¹ m⁻², and is three to five orders of magnitude higher than that in laboratory-synthesized, undeformed olivine. TEM observations on samples cut across the SG-like structures by FIB (focused ion beam) demonstrated that this kind of structures is made of an array of dislocations. These observations clearly indicate that these structures are real subgrain boundaries rather than grain boundaries. These facts suggest that the observed high dislocation densities and subgrain boundaries cannot result from crystal crystallization/growth, but can be formed by plastic deformation. These deformation features do not prove that the olivine phenocrysts (and implicitly mantle xenoliths) were deformed after their capture by the basaltic magmas, but can be ascribed to a former deformation event in a dunitic cumulate, which was formed by magmatic fractionation, then plastically deformed, and finally disaggregated and captured by the basaltic magma that brought them to the surface.

通常将天然变形橄榄岩中橄榄石的晶内形变（例如过度消光，位错和亚晶界）视为位错蠕变的迹象。然而，在玄武岩浆中发现的橄榄石表晶的相似特征仍未得到很好的理解。特别是，橄榄石中的隐晶界的亚晶界是由于塑性变形还是晶粒生长引起的（在后一种情况下，它们本质上是晶界而不是亚晶界。因此，我们在下文中使用亚晶界状结构代替亚晶界）来命名这种晶内畸变）。在这里，我们通过岩相显微镜，扫描电子显微镜和透射电子显微镜（TEM）进行了详细研究，研究了两个夏威夷玄武岩熔岩中橄榄石表晶中的位错和亚晶界样（SG-like）结构。在装饰的橄榄石晶粒中观察到丰富而复杂的位错亚结构（自由位错，位错壁和位错缠结），类似于夏威夷玄武岩夹带的橄榄岩异岩中的橄榄石。测得的平均位错密度为（2.9±1.3）×10 类似于橄榄岩中由橄榄岩玄武岩夹带的橄榄石。测得的平均位错密度为（2.9±1.3）×10 类似于橄榄岩中由橄榄岩玄武岩夹带的橄榄石。测得的平均位错密度为（2.9±1.3）×10¹¹ m ^-2，比实验室合成的未变形橄榄石高出三到五个数量级。TEM对通过FIB（聚焦离子束）切割成SG状结构的样品的观察表明，这种结构是由位错阵列构成的。这些观察清楚地表明，这些结构是真实的亚晶界而不是晶界。这些事实表明，观察到的高位错密度和亚晶界不能由晶体结晶/生长引起，而可以由塑性变形形成。这些变形特征不能证明橄榄石的隐晶（以及隐含的地幔异种岩）在被玄武岩浆捕获后就发生了变形，但是可以归因于以前的变形事件，它是由岩浆分馏形成的双元堆积物，