A KNOWLEDGE of the age of marine sediments is necessary to determine the timing of events and rates of processes in the marine realm, and the relationships among marine and other climatically sensitive records. The establishment of an accurate chronology for Pleistocene marine sediments beyond the range of radiocarbon dating (approximately the past 45 kyr) has therefore been a goal of palaeoceanographers for decades. Early attempts(1,2) based on measurements of the radionuclides Th-230 and Pa-231 mere beset with problems, and subsequent studies focused on tying fluctuations in marine sediment oxygen-isotope records to events such as the formation of col al reef terraces and changes in the Earth’s magnetic polarity(3,4), and tuning the resultant chronologies to the Earth’s orbitally driven insolation variations(5-8). But these chronologies (especially the age and duration of the last interglacial period) have been challenged by several studies(9-12), raising questions about the fundamental cause of Pleistocene climate fluctuations. Here we report the direct U-Th dating of aragonite-rich marine sediments from the Bahamas, and present an accurately dated marine oxygen-isotope record for the last two interglacials. We obtain dates of 120-127 kyr BP for the last interglacial and 189-190 kyr np for the late stage 7 interglacial. These dates are in accord with the general theory of orbitally forced climate fluctuations and demonstrate the potential of our direct-dating approach for developing an absolute chronology for the Pleistocene marine oxygen-isotope record.