The object of the present study is to introduce a means of comparing the Vostok and marine chronologies. Our strategy has been to use the Δ18O of atmospheric O2 (denoted Δ18Oatm) from the Vostok ice core as a proxy for the Δ18O of seawater (denoted Δ18Osw). Our underlying premise in using Δ18Oatm as a proxy for Δ18Osw is that past variations in Δ18Osw (an indicator of continental ice volume) have been transmitted to the atmospheric O2 reservoir by photosynthesizing organisms in the surface waters of the world's oceans. We compare our record of Δ18Oatm to the Δ18Osw record which has been developed from studies of the isotopic composition of biogenic calcite (Δ18Oforam) in deep-sea cores. We have tied our Δ18Oatm record from Vostok to the SPECMAP timescale throughout the last 135 kyr by correlating Δ18Oatm with a Δ18Osw record from V19--30. Results of the correlation indicate that 77% of the variance is shared between these two records. We observed differences between the Δ18Oatm and Δ18Osw records during the coldest periods, which indicate that there have been subtle changes in the factors which regulate Δ18Oatm other than Δ18Osw. Our use of Δ18Oatm as a proxy for Δ18Osw must therefore be considered tentative, especially during these periods.

By correlating Δ18Oatm with Δ18Osw, we provide a common temporal framework for comparing phase relationships between atmospheric records (from ice cores) and oceanographic records constructed from deep-sea cores. Our correlated age-depth relation for the Vostok core should not be considered an absolute Vostok timescale. We consider it to be the preferred timescale for comparing Vostok climate records with marine climate records which have been placed on the SPECMAP timescale. We have examined the fidelity of this common temporal framework by comparing sea surface temperature (SST) records from sediment cores with an Antarctic temperature record from the Vostok ice core. We have demonstrated that when the southern ocean SST and Antarctic temperature records are compared onthis common temporal framework, they show a high degree of similarity. We interpret this result as supporting our use of the common temporal framework for comparing other climate records from ¿ American Geophysical Union 1