We developed an isotope (87Sr/86Sr, delta18O) reference section for the uppermost Oligocene to lower upper Miocene (ca. 25-8 Ma) at Site 608 in the northeastern North Atlantic. This site contains the least ambiguous magnetostratigraphic record of Miocene polarity changes available, providing direct correlations to the Geomagnetic Polarity Time Scale (GPTS). We integrate biostratigraphic, magnetostratigraphic, Sr isotope, and stable isotope data to provide a reference section for Miocene isotope fluctuations. The direct correlation of isotopes and biostratigraphy to the Geomagnetic Polarity Time Scale (GPTS) provides relatively precise age estimates. We use these age estimates to evaluate the timing of first and last occurrences of planktonic foraminifera, and conclude that many of these are synchronous within a 0.5 m.y. resolution between subtropical Site 563 (33°N) and high-latitude Site 608 (43°N). In addition, we use this chronology to estimate the ages of previously established Miocene oxygen isotope Zones Mi 1 through Mi 7 and to compare the Sr isotope record at Site 608 with previously published 87Sr/86Sr records. We approximate latest Oligocene to early late Miocene (25-8 Ma) Sr isotope changes with two linear regressions. The rate of increase of 87Sr/86Sr was high from the latest Oligocene (~25 Ma) to earliest middle Miocene (~15 Ma), with an estimated rate of 0.000059/m.y. Our ability to reproduce Sr isotope measurements is +/-0.000030 or better, yielding a stratigraphic resolution of as good as +/-0.5 m.y. for this interval. The rate of change was much lower from about 15 to 8 Ma (on average, 0.000013/m.y.), yielding Sr isotope stratigraphic resolution of worse than +/-2.3 m.y. The causes of the late Eocene to Miocene 87Sr/86Sr increases are not known. We speculate that a moderate 87Sr/86Sr increase (0.000030/m.y) which occurred during the late Eocene-latest Oligocene can be explained by intermittent glaciations and deglaciations of the Antarctic continent. These pulse-like changes in the input of glacial weathering products yield what appears to be a monotonic, linear increase. The increase in the frequency of glaciations during the latest Oligocene-early Miocene can explain the higher rate of change of 87Sr/86Sr at this time. We speculate that by the middle Miocene, the development of a permanent east Antarctica ice sheet resulted in decreased input of glacial weathering products and a lower rate of 87Sr/86Sr change.
