Geese described are non-migratory, free-flying Todd's Canada geese (Branta canadensis interior). The genealogy of 261 of these geese was traced by archival research and three years of field observations. Nest locations and densities, preferences for various types of artificial nest structures, clutch sizes, hatching success, brood survival to flight stage, and food habits were recorded. Resul ts indicate geese may:,pair as yearlings, but these bonds may be broken and re-formed before breeding. Pair bonding generally resulted in geese of similar ages remaining together until the death of one partner, although re-pairing, polygamy, and pairing between broodmates also occurred. The dominance hierarchy of related birds strongly influenced the position of 'outsiders' pairing with indigenous females. Dominant status passed not only from male to male, but, upon the death of the dominant male, in at least one instance, the surviving female retained dominant status. Gang broods were composed of progeny of the rearing pair, plus goslings relinquished by female offspring or siblings of the rearing pair. Among indentifiable geese, gang broods were reared by the dominant pair on each impoundment. Geese retained their family integrity both in flight and during the post-molt dispersion. Female and males paired with local females, nested in their natal areas. No significant relationship (P < 0.05) was found between clutch size and age of the female. Twelve-year productivity of the Patuxent geese appeared related to the reproductive success of a specific resident family. Collars, legbands, and telemetry were initially used to distinguish conspecifics. It was subsequently discovered that individual geese could be recognized by cheek-patch patterns, unusual plumage, or mannerisms. It is suggested that cheek-patch similarities in related Canada geese might be used to trace gene flow within flocks, and may be used for individual recognition by other Canada geese.