The frequent reassortment of angiosperms to form new communities, a dominant feature of the Quaternary record, has seldom been recognized in earlier periods. In fact, analysis of fossil floras in North America suggests that Tertiary plant communities were relatively stable over long periods of time (for example, Hickey, 1977; MacGinitie, 1969; Wolfe, 1975). This stability suggests that either the record has not been studied in sufficient detail or that the factors controlling plant and environment interactions were less variable than in the Quaternary. If the latter hypothesis is true, the botanical record of the last 1.8x 10^6 yr may be an atypical model on which to base our understanding of community evolution . To assess the importance of short-term variation in Tertiary plant communities, detailed botanical analyses of continuous stratigraphic sections are needed. In particular, a pollen record through a lacustrine sequence is critical for recognizing rapid, floristic or vegetational changes on a regional scale. Thus far, such recognition has been difficult because our understanding of floristic development is based primarily on plant megafossils. While such material provides information on the riparian flora adjacent to a depositional site, it reveals little about regional patterns.
V.—On the Beds Containing the Gelinden Flora