(Table 1) AMS-14C ages determined in pairs of spreiten and directly adjacent host sediment of cores M39029-4 and M29039-8

he downward transport of surface sediment deep into the sediment column by the Zoophycos-producing animal leads not only to large age differences between the Zoophycos structure and surrounding host sediment but also to large differences in age between different foraminifer species found inside the trace fossil. In the late Quaternary material from the southwestern Portuguese continental slope examined in this study, age differences of up to 2590 years were observed between the planktic foraminifer species Globigerinoides ruber and Globigerina bulloides. These differences are caused by the mixing of surface and host material with different abundances of the two species. If there are differences in the abundance of the two species at the surface and/or in the host sediment, plenty of relatively young foraminifers may be mixed with few relatively old ones, or vice versa. The age differences between species caused by the combination of deep-reaching bioturbation by the Zoophycos producer and abundance variations may be considerably larger than the age differences caused by the homogenizing bioturbation in the mixed layer.

(Table 1) Strontium concentration and isotope ratios of carbonates from DSDP Holes 69-504B as well as 83-504B and fish debris samples from Indian and Pacific Ocean sediments

Recent revisions of the geological time scale by Kent and Gradstein (in press) suggest that, on the average, Cretaceous magnetic anomalies are approximately 10 m.y. older than in Larson and Hilde's (1975) previous time scale. These revised basement ages change estimates for the duration of alteration in the ocean crust, based on the difference between secondary-mineral isochron ages and magnetic isochron-crustal ages, from 3 to approximately 13 m.y. In addition to the revised time scale, Burke et al.'s (1982) new data on the temporal variation of 87Sr/86Sr in seawater allow a better understanding of the timing of alteration and more realistic determinations of water/rock ratios during seawater-basalt interaction. Carbonates from all DSDP sites which reached Layer 2 of Atlantic crust (Sites 105, 332, 417, and 418) are deposited within 10-15 m.y. of crustal formation from solutions with 87Sr/86Sr ratios identical to unaltered or contemporaneous seawater. Comparisons of the revised seawater curve with the 87Sr/86Sr of basement carbonates is consistent with a duration of approximately 10-15 m.y. for alteration in the ocean crust. Our preliminary Sr and 87Sr/86Sr data for carbonates from Hole 504B, on 5.9-m.y.-old crust south of the Costa Rica Rift, suggest that hydrous solutions from which carbonates precipitated contained substantial amounts of basaltic Sr. For this reason, carbonate 87Sr/86Sr cannot be used to estimate the duration of alteration at this site. A basalt-dominated alteration environment at Hole 504B is consistent with heat-flow evidence which indicates rapid sediment burial of crust at the Costa Rica Rift, sealing it from access by seawater and resulting in unusually low water/rock ratios during alteration.

(Table S2) The reconstructed sea surface temperature (SST) over the eastern equatorial Pacific

Abundant hydroclimatic evidence from western Amazonia and the adjacent Andes documents wet conditions during Heinrich Stadial 1 (HS1, 18-15 ka), a cold period in the high latitudes of the North Atlantic. This precipitation anomaly was attributed to a strengthening of the South American summer monsoon due to a change in the Atlantic interhemispheric sea surface temperature (SST) gradient. However, the physical viability of this mechanism has never been rigorously tested. We address this issue by combining a thorough compilation of tropical South American paleorecords and a set of atmosphere model sensitivity experiments. Our results show that the Atlantic SST variations alone, although leading to dry conditions in northern South America and wet conditions in northeastern Brazil, cannot produce increased precipitation over western Amazonia and the adjacent Andes during HS1. Instead, an eastern equatorial Pacific SST increase (i.e., 0.5-1.5 °C), in response to the slowdown of the Atlantic Meridional Overturning Circulation during HS1, is crucial to generate the wet conditions in these regions. The mechanism works via anomalous low sea level pressure over the eastern equatorial Pacific, which promotes a regional easterly low-level wind anomaly and moisture recycling from central Amazonia towards the Andes.