The question of what caused the late Quaternary extinction of the Rancholabrean fauna in the California glacial refugium is examined. In the absence of significant climatic and other natural environmental change, man was probably the principal agent of extinction in California. This extinction event marked the beginning of a long history of environmental impact in the area by humans.

The present Mediterranean climate of coastal California is unique in North America and reflects the interaction of several important synoptic controls, principally the North Pacific semipermanent anticyclone, and to a lesser extent the Aleutian low-pressure system and the cool California oceanic current. These synoptic climatic controls, key parts of the global air-sea circulation, were probably operative throughout late Quaternary time as shown by paleoecologic evidence. The thick accumulations of sediments in basins of offshore California indicate that while variable sedimentation regimes reflect changing climatic and oceanographic conditions, the Quaternary climate was probably semiarid as now, even during glacial maxima. Late Quaternary coastal dunes preserve former wind directions and show that prevailing late Quaternary winds were directionally equivalent to modern winds, which are controlled by the North Pacific anticyclone and by interactions between the North Pacific high and the interior basin low. These sand dunes contain buried, datable, carbonate-rich soils. Precipitation then, like the present rainfall regime, was not enough to leach the carbonates from the soils. Charcoal in buried dunes and soils shows that fire was environmentally important throughout the Quaternary, just as it is today. Fossil plants indicate that sclerophyllous vegetation and forest stands of conifers, adapted to a Mediterranean climate, were widely distributed during late Quaternary time. Fossil pollen in the Sierra Nevada indicates the influence of the North Pacific high. The historical precipitation record overlaps a late Holocene tree-ring record permitting extrapolation of the precipitation curve back nearly 600 years. Well-defined wet and dry trends in the precipitation pattern characterized this time span, and provide a possible analog to the earlier Holocene and Pleistocene precipitation regime. The paleoecologic record shows that the late Quaternary climate of coastal California was characterized by regimes similar to those prevailing today. The persistence of a Mediterranean climate in California during the last glaciation contrasts with dramatic climatic changes experienced in glaciated parts of North America. California thus was an Ice Age refugium for animals and cold-sensitive plants.

In the Camp Century core, the difference in oxygen isotope ratio between Wisconsin and Holocene ice seems too large to be purely a climatic effect. A more likely interpretation is that the Wisconsin ice originated at an elevation of at least 500 m above the present station. Total gas content measurements on the core suggest that the elevation difference was about 1300 m. These results are inconsistent with the present ice flow pattern. Three hypotheses are considered: (1) The Wisconsin ice originated near the crest of a high ridge connecting the Greenland ice sheet with an ice sheet on Ellesmere Island. (2) The Wisconsin flow pattern was similar to the present one but ice was much thicker and the ice margin considerably in advance of its present position. (3) The Wisconsin ice originated near the main Greenland ice divide whereas the Holocene ice originates within 50 km of the station. Glacial-geological data are sparse but do not appear to support the first hypothesis, while the uplift data have been interpreted in two widely different ways. The second hypothesis might explain the oxygen isotope values but not the gas content measurements. The third hypothesis is thus considered the most likely one. Differences between Wisconsin and Holocene flow patterns might result from changes in positions of the ice margins as a consequence of lowered sea level in the Wisconsin.

Pollen contained in 22 fossil packrat middens from the Sonoran Desert provides a complementary, but differing, view of the paleoenvironment from that derived by analysis of the associated plant macrofossils. The regional component of the pollen data is in sharp contrast to the locally oriented macrofossils. A total of 84 macrofossil taxa and 47 pollen taxa were identified; only 18 taxa were common to both. The low Index of Similarity, 0.4, indicates that the two sources of fossil information are providing different sets of paleobotanical data. When combined with plant macrofossils and good radiocarbon dating control, the pollen spectra derived from fossil middens are compatable with other paleoenvironmental sequences.

A deep-sea core over 16 m long from the crestal area of the Mediterranean Ridge has been investigated with different techniques, including quantitative micropaleontology, stable isotopes (measured on the epipelagic species Globigerinoides ruber and on the mesopelagic species Globorotalia inflata), and clay mineralogy. The resulting record of climatic fluctuations can be cross correlated to other Mediterranean cores by means of isochronous lithologies (tephra layers and sapropels). The climatic record of the Mediterranean is similar in character, phase, and chronology to the records investigated in the equatorial Pacific and in the Caribbean. Isotope stages 1 to 17 have been recognized. Cyclically repeated stagnant cycles resulting in sapropel deposition complicate both the isotopic and the faunal signal. The isotopic investigations reveal that the temperature change in the surface layers of the eastern Mediterranean was no greater than 8°C in the late “glacial” Pleistocene. The chronostratigraphic and biostratigraphic interpretation of Core KS09 indicate that the mean sedimentation rate was 2.4 cm/1000 years, a value very close to the 2.5 cm/1000 years calculated for the entire Quaternary section at DSDP Site 125, also located in the crestal area of the Mediterranean Ridge in the Ionian Basin. The base of KS09 is likely to be very close to the Brunhes/Matuyama boundary dated at 0.7 my.