As co-chief scientist on DSDP Leg 35 in 1973, Cam Craddock (1930-2006) produced the first useful information on Cenozoic Antarctic Peninsula glaciation - an early middle Miocene (15-17 Ma) glacial onset. Subsequent work, onshore and offshore, has greatly extended our knowledge but that early conclusion stands today. Initial glacial onset was within the Eocene-Oligocene boundary interval (although earlier, short-lived glaciations have been proposed, from indirect measurements) and the peninsula probably became deglaciated in the earliest Miocene (ca. 24 Ma). The renewed middle Miocene glaciation probably continued to the present and, for the last 9 Myr at least, has persisted through glacial (orbital) cycles, with grounded ice advance to the shelf edge during maxima. Although orbital cyclicity affected earlier AP palaeoclimate also, the level of glaciation through a complete cycle is uncertain.