AT CERTAIN POINTS, the disciplines of history and archaeology converge; and some of the closest rapprochements have traditionally taken place in the field of Classics. I am concerned here with one quite specific form of close relationship that can exist only between certain kinds of historical and archaeo logical approach. These must be defined at the outset. The historian who bases his account fairly and squarely on the ancient sources will tend to construct a narrative account that deals in concepts similar to theirs: an account, that is, couched mainly in terms of political, constitu tional, and military events. To this view of history, archaeological work for the most part can make little or no contribution. But there is one large area of exception. The careful excavation of a site-particularly of a settlement-site-is likely to reveal episodes in the site's history: e.g., the construction, the exten sion, the rebuilding, or the destruction of some part or parts of a settlement; the deposition of a hoard; the reinforcement of a fortification or perhaps the failure to build one at all. Other kinds of site, though less obviously, can yield evidence for their own types of "episode": the appearance of a burial or burials in a cemetery in circumstances that invite speculation as to the occasion on which they took place or even as to the identity of the deceased, or a change in the nature or quantity of dedications at a sanctuary. To bring together this type of historical approach and this category of