Interphase boundary structure developed during precipitation from solid solution and during massive transformations is considered in diverse alloy systems in the presence of differences in stacking sequence across interphase boundaries. Linear misfit compensating defects, including misfit dislocations, structural disconnections, and misfit disconnections, are present over a wide range of crystallographies when both phases have metallic bonding. Misfit dislocations have also been observed when both phases have covalent bonding (e.g., US: US 2 by Sole and van der Walt). These defects are also found when one phase is ionic and the other is metallic (Nb:Al 2 O 3 by Rühle et al.), albeit when the latter is formed by vapor deposition. However, when bonding is metallic in one phase but significantly covalent in the other, the structure of the interphase boundary appears to depend upon the strength of the covalent bonding relative to that in the metallically bonded phase. When this difference is large, growth can take place as if it were occurring at a free surface, resulting in orientation relationships that are irrational and conjugate habit planes that are ill matched (e.g., ZrN:␣ Zr-N by Li et al. and Xe(solid):Al-Xe by Kishida and Yamaguchi). At lower levels of bonding directionality and strength, crystallography is again irrational, but now edge-to-edge-based low-energy structures can replace linear misfit compensating defects (␥ m :TiAl:␣ Ti-Al by Reynolds et al.). In the perhaps still smaller difference case of Widmanstätten cementite precipitated from austenite, one orientation relationship yields plates with linear misfit compensating defects at their broad faces whereas another (presumably nucleated at different types of site) produces laths with poorly defined shapes and interfacial structures. Hence, Hume-Rothery-type bonding considerations can markedly affect interphase boundary structure and thus the mechanisms, kinetics, and morphology of growth.