ern China (Heller and Liu, 1982); and pollen sequences in the Netherlands, northern Italy, and Macedonia (van der Hammen et al., 1971). Prominent features in these records include a large initial step toward glacial-like climates at or near 2.4 Ma, cycles of modest size and relatively high frequency until near the Jaramillo magnetic reversal (~0.9 Ma), and larger cycles displaying a 100,000-yr periodicity during the late Pleistocene.
Regions located well away from the Northern Hemisphere ice sheets might be expected to show climatic variations during the Late Pliocene and Pleistocene in response to other factors. Within middle and high latitudes of the Northern Hemisphere, however, the relatively continuous records available all appear to fluctuate sympathetically with the ice sheets. Some discrepancies may occur in the 2.3 to 0.9 Ma interval, where the isotopic record appears to detect more cycles than the loess or pollen sequences, but this may reflect different thresholds of sensitivity in the detection of small-amplitude 41,000 yr cycles.
Less continuous or well-dated stratigraphic sequences representing shallow marine and terrestrial environments yield fossils that document important climatic events. For example, the record of vegetational change along the northwestern margin of the Mediterranean at about 2.3 Ma points to a shift to drier conditions. Forests shrank and steppe vegetation, including sagebrush and weedy plants, expanded. Generally contemporaneous changes in the composition of some European floras reflect cooling (Suc, 1984). At about this time (the Praetiglian interval), what remained of the subtropical Malayan element disappeared from the flora of northwestern Europe (van der Hammen et al., 1971; also see Figure 7.1). In Africa, palynofloras reveal sweeping climatic change at about 2.5 Ma., although the precise timing remains to be determined. In the Ethiopian uplands, climates became cooler and drier than today, whereas they had previously been warmer and moister (Bonnefille, 1985). Apparently this pulse of climatic change was soon partly reversed. The fact that even today, during a glacial minimum, Africa is cooler and drier than it was prior to 2.5 Ma may, however, indicate that this region has experienced a fundamental climatic change unrelated to glacial maxima.
In the Western Atlantic region and the Americas, climatic changes are less well documented, but in the high plain of Bogota, Columbia, fossil pollen records indicate cooling at about 2.5 Ma (van der Hammen, 1985). Cores obtained by drilling in the Bahamas document dramatic change a bit earlier. Here, magnetostratigraphy has provided a detailed chronology for a relatively complete stratigraphic sequence (McNeill et al., 1988). Fossils in the core reveal that sea-level dropped briefly at about 2.65 Ma and San Salvador ceased to grow as a coral atoll. The fact that the biotic crisis in the Bahamas occurred as early as 2.65 Ma reflects the fact that a transition to the ice age was a complex event that spanned more than 200,000 yr (Figure 7.2). The precise chronology of climatic change remains to be established for many geographic areas, but the magnetic reversal separating the Gauss and Matuyama Chrons at 2.5 Ma should prove useful here.
The Pliocene climatic changes that affected terrestrial biotas of the North Atlantic region included a general cooling of climates, with increased seasonality in many areas and widespread aridification on the land. Among the results were the migration of many species to favorable habitats; the extinction of species unable to escape intolerable new conditions through migration; and the origin of new species adapted to the new conditions.
Some of the biotic consequences of the climatic changes, in fact, represent key evidence documenting the changes. Foremost of these were previously described transformations of terrestrial floras, which we review in discussing the general topic of biotic change.
Fossil pollen have provided most of our information about Pliocene terrestrial floras of the North Atlantic region. Although the evidence assembled to date is patchy, it indicates that drying of climates was at least as significant as thermal change. Often it was an accentuation of the dry season that had the greatest impact on floras. In general, grasslands expanded at the expense of forests, which require moist conditions. Climatic changes had their primary effects on terrestrial mammals indirectly, through their influence on vegetation.
In some areas, orographic effects of regional tectonic activity had a more pronounced impact on biotas than did global climatic change. Africa, however, experienced aridification so pervasive that it could not have resulted from small-scale tectonics. Fossil pollen reveals that savannas spread at the expense of forests, not only throughout northern Africa, where the trade winds prevail, but also in Kenya (Bonnefille, 1976). The dramatic change that occurred at about 2.5 Ma entailed both cooling and drying of climates. While a slight cooling probably had some effect on African floras, aridification was the most influential climatic change. Low rainfall is the primary factor that separates grasslands from forested areas today.
The onset of the ice age had a greater impact on mammals in Africa than in Europe or North America. Presum-