The Pacific margin of the Antarctic Peninsula was drilled during Ocean Drilling Program (ODP) Leg 178 to understand the past 10 m.y. of its glacial history and to test a drilling strategy that might be applied in other regions of the Antarctic margin. This paper offers a mature view of the achievements of drilling, which succeeded in both aims, and focuses on the lessons for subsequent margin drilling and the results of postcruise studies on samples and data from the leg. One of two complementary depositional environments drilled was the topset component of the glacial prograded wedge of the outer continental shelf. Despite poor recovery, a Pliocene-Pleistocene age was found for the major glacial sedimentary sequence groups S1 and S2 previously defined using seismic reflection survey, and a glacial nature and late Miocene age was established for the sampled part of the underlying sequence group S3. Such broad, low-resolution conclusions are possible from shelf drilling using presently available techniques and are useful in terms of Antarctic glacial history, but may represent the limit of what can be achieved if core recovery is poor. The second depositional environment was the glacially derived fine-grained sediment drifts on the upper continental rise, which gave continuous recovery and a high-resolution record of the past 10 m.y. Biogenic opal concentrations showed that the southeast Pacific Ocean (through variations in temperature and sea-ice cover) behaved compatibly with global climate, with a cool late Miocene and warm early Pliocene, cooling to a cold Pleistocene. Nevertheless, ice-rafted detritus confirmed that the Antarctic Peninsula ice sheet was present throughout this period, and clay mineralogy showed that it remained sufficiently large for regular grounding-line migration to the shelf edge. Spectral analysis of sedimentary parameters at drift Sites 1095 and 1096 did not show the dominance of frequencies usually associated with orbital insolation variation. It is therefore uncertain if the glacial-interglacial cyclicity evident in sampled sediments deposited off the Antarctic Peninsula before the late Pliocene-Pleistocene growth of large Northern Hemisphere ice sheets was caused by an orbitally driven or an essentially autocyclic variation in Antarctic ice sheet volume. The matter needs further investigation. An Antarctic Peninsula ice sheet existed for the entire period sampled by drilling, and glacial onset was earlier. The onshore and offshore Antarctic Peninsula records suggest very different times of onset.