Articulated and disarticulated shells of Anadara ovalis, Anomia simplex, Argopecten irradians, Astarte castanea, Crassostrea virginica, Divalinga quadrisulcata, Donax variabilis, Ensis directus, Geukensia demissa, Mercenaria mercenaria, Mya arenaria, Mytilus edulis, Petricola pholadiformis, Spisula solidissima, and Tagelus plebeius collected from New Jersey intertidal life and death assemblages were measured, and the frequency, type (scalloped, divoted, cleft, embayed), location (ventral, anterior, posterior), and shell size at inception of repair determined. Repair frequency ranges from zero (D. quadrisulcata) to 0.30 (M. arenaria). Size refuge from sublethal breakage was attained by S. solidissima and C. virginica. Posterior repairs necessitated by siphon-nipping characterize M. mercenaria and E. directus. Ventral repairs associate with species subjected to valve-wedging (S. solidissima, M. mercenaria) or with ventral egress of the foot (A. ovalis). Ventral repairs among deep infaunal clams (M. arenaria, E. directus, T. plebeius) may have been induced by sediment-loading stress during burrowing. Anterior repairs are characteristic of deep infaunal species with foot exposure (M. arenaria, T. plebeius). Repair frequencies were compared statistically with valve surface area, ventral margin thickness, burrowing rate, depth of sediment penetration, shell microstructure, shell ornament, and tissue exposure when valves adducted. Of these comparisons, the only significant correlation was between frequency of embayed repairs and valve surface area. The complex relationship between repair frequency and these variables is underscored by factor analysis. New quantitative approaches are employed that advocate phyletic rather than assemblage-level tests of escalation involving Tertiary congeners from the Atlantic Coastal Plain.