We have carried out a multiphase analysis of samples from ODP Site 177-1092, Meteor Rise, subantarctic South Atlantic. Samples were analyzed for ice-rafted debris (IRD) and stable isotopes from benthic foraminifera. Both analyses were performed on the same samples. Additional work was performed to identify the paleomagnetic stratigraphy. The analyzed samples range in age from about 2.6(?) Ma to 4.6 Ma, a time span that saw considerable global warmth, but witnessed overall global refrigeration and the transition to truly bipolar glaciations. A tentative oxygen isotopic stratigraphy was established by comparison with Shackleton et al. [R. Soc. Edinburgh Trans. Earth Sci. 8 (1990) 251–261] and shackleton et al. [Proc. ODP Sci. Results 138 (1995) 337–355]. Paleomagnetic results show that the Gauss Normal Chron, including subchrons, is identified, although uncertainties plague the exact definitions of the reversals. The subchrons of the Gilbert Reversed Chron, unfortunately, could not be identified. IRD arrived frequently during the Early and early Late Pliocene, but only as ‘background rafting’ (occasional grains per sample). The first identifiable IRD above background rafting is associated with marine isotope stage (MIS) KM4 (∼3.18 Ma). Successive IRD peaks become larger, the same pattern as noted at nearby Site 114-704. A very large peak near the top of the record, approximately 2.8 Ma, is considered to represent a hiatus. Peaks below 51.3 meters composite depth (mcd) coincide with positive excursions of the oxygen isotopic record, and with negative excursions of the carbon isotopic curve, a pattern also noted at Site 114-704. However, the reasonably large IRD peak at 51 mcd (tentatively identified with MIS G11) coincides with a positive excursion on the carbon isotopic curve and negative excursion on the oxygen isotopic curve. This relationship suggests a northern hemisphere interglacial, rising sea level, destabilization of the Antarctic margin, and delivery of Antarctic icebergs to the Southern Ocean. Such a mechanism has recently been suggested by Kanfoush et al. [Science 288 (2000) 1815–1818] for latest Pleistocene stadial/interstadial oscillations. Here we suggest that such a mechanism may have been in place on glacial/interglacial time scales as early as the Late Pliocene. One interval in the lowermost Gauss Normal Chron and several short intervals in the upper Gilbert Reversed Chron have no IRD. However, oxygen isotopic values of benthic foraminifera are only about 0.62‰ lighter than modern, and must be ascribed to temperature effects in the area of water-mass formation [Hodell and Warnke, Quat. Sci. Rev. 10 (1991) 205–214; Hodell and Venz, Antarctic Research Series 56 (1992) 265–310; Warnke et al., Mar. Micropaleontol. 27 (1996) 237–251]. The East Antarctic Ice Sheet was therefore stable – but the stability of the West Antarctic Ice Sheet may have been compromised.