Opposing fatigue fracture surfaces originate from the same advancing crack and may retain a conjugate geometrical relationship. In real fatigue, however, this relationship is modified by crack-tip plasticity, crack closure, mixed-mode displacement, repeated crack-face contact, environmental interaction and local material loss. This review critically examines how such processes affect the preservation and interpretation of fracture-surface complementarity and how the resulting correspondence can be quantified using three-dimensional surface data. A transformation-aware framework is developed that distinguishes three progressively more specific levels of evidence: scalar descriptor similarity, descriptor complementarity under inversion of the surface-height field, and direct spatial geometrical complementarity of registered mating surfaces. The framework also separates physical mismatch from differences introduced by measurement, preprocessing and registration. Retrospective analysis of three previously published paired-surface datasets showed strong specimen-specific spatial correspondence despite weaker agreement of selected scalar descriptors: median spatial correlation coefficients were 0.990 for 7075-T651 low-cycle-fatigue pairs, 0.924 for 6082 aluminium pairs and 0.983 for pre-corroded 7075-T651 pairs. An additional within-material comparison of one monotonic and nine fatigue-fracture pairs showed variable spatial correspondence and exposed non-unique geometrical matches in auxiliary non-mating controls. These values are illustrative and should not be interpreted as universal thresholds. The reviewed evidence shows that strong global mating can coexist with pronounced local asymmetry and that residual mismatch should be interpreted as a diagnostic field, not as a direct label of a fracture mechanism. The proposed framework provides a physically grounded basis for FRASTA-type reconstruction, quantitative fractography and post-failure fracture-mechanics analysis.
