Resolution of the chronology of marine terrace sequences is critical to determine uplift rates along tectonically active coastlines. Unfortunately, dating of marine terraces commonly proves frustrating because few terraces contain solitary corals necessary for U-series dating. This study examines the utility of oxygen isotope stratigraphy to constrain the ages of marine terraces in southern California. where a late Quaternary delta (super 18) O sequence (ODP site 893) is available as a standard for comparison. Correlation of marine terraces based on oxygen isotope stratigraphy relies on the assumption that marine terrace mollusks of different ages preserve distinct isotopic signatures characteristic of the ocean water in which they formed. Evidence from modern (living) samples indicates that this is the case, so oxygen isotope stratigraphy provides a powerful tool for helping to constrain ages of marine terraces. In this study, the oxygen isotopic signatures of samples of Olivella biplicata from undated terraces are compared to those of samples from U-series dated "calibration points" from oxygen isotope Stage 3, substage 5a, and substage 5e. The Isla Vista terrace, in Goleta, is U-series dated at 47 ka + or -500 yr (late Stage 3) (Gurrola et al., 1996), and the Santa Barbara City College terrace, in downtown Santa Barbara, is U-series dated at 70+ or -2 ka (substage 5a) (Gurrola, et al., 1997). These terraces preserve isotopic signatures of 0.58 per mil and 0.09 per mil, respectively. This difference in isotopic signatures provides the basis for correlation of undated terraces to dated sites in the Santa Barbara and Ventura areas. Terraces dated at substage 5e (Eemian) reveal surprising results, however, Isotopic signatures of 0.25 per mil, 0.53 per mil and 0.45 per mil from the Nestor terrace in San Diego and the Forney Cove and Near Point terraces on Santa Cruz Island indicate surprisingly cold conditions, in direct contrast with the bulk of evidence that points to Eemian ocean temperatures at least equal to or warmer than today's ocean. XRD analysis shows that the shells are free of diagenetic material, and the isotopic data are sound. Furthermore, data from Ocean Drilling Project Hole 893A, in the Santa Barbara Basin, clearly show Eemian ocean temperatures as warm or warmer than present day temperatures, implying that the terraces sampled are not Eemian in age. Values from the substage 5e shells represent significantly cooler temperatures than those recorded by modern shells from the mainland and Channel Islands, as well as those implied by shells from a warm assemblage at Second and Beacon Streets in San Pedro. This evidence indicates that the U-series dates on these terraces may be inaccurate, or misinterpreted.