During the Mid-Pleistocene, c.900 - 600 ka BP, the global climate system underwent a number of significant changes, most clearly represented by the emergence and subsequent dominance of glacial-interglacial oscillations at the 100-kyr frequency. Ice sheet dynamics, ocean circulation changes, and the carbon cycle have all been cited as potential key components in both the regulation and development of the 100-kyr cycles, but there are a limited number of reconstructions of surface ocean circulation patterns and carbon cycle components during this period. Here, we present results from marine sediment cores located in the Atlantic and equatorial Pacific Oceans, from the time interval 1500 - 500 ka BP. Biomarker reconstructions of sea-surface temperatures and carbon inputs to the ocean floor have been generated at a resolution of c.5-kyr. We compare records from the mid-high latitudes of the northern (ODP Site 983, 60 degrees N) and southern Atlantic (ODP Site 1087, 31 degrees S), to tropical records from the east (ODP Site 849, 110 degrees W) and west equatorial Pacific (ODP Site 806, 159 degrees E). The results indicate that significant synchronous changes in surface ocean circulation took place at all sites during the Middle Pleistocene, accompanied by enhanced carbon inputs to the ocean floor, prior to many previously published records of Mid-Pleistocene climate changes. Between c.1180 - 900 ka BP, surface ocean cooling is evident at all sites, accompanied by equatorward movements of the polar fronts and a possible upwelling increase within the Benguela system. This cooling phase is accompanied by enhanced inputs of marine carbon to the ocean floor, suggesting that the strength of the biological pump increased during this time. The surface ocean cooling culminates in a pronounced episode of SST fall at all sites, centred at c.900 ka BP, after which the polar fronts retreat poleward. A second phase of enhanced carbon input begins at this time, its termination marked by the onset of the dominant 100-kyr cycles in the ice volume records. These results highlight the potential significance of both surface ocean circulation and the carbon cycle in the development of the MPR.