Abstract The examination of fossil assemblages of Coleoptera (beetles) in North America is a relatively new but rapidly growing research area. This chapter is an attempt to use suites of fossil Coleoptera to establish the nature and timing of different paleoenvironments at the margins of the Laurentide Ice Sheet as it advanced and retreated from the Great Lakes region. Some of the work is synthesized from previous publications, but some important data are from unpublished sites. Before the various sites illustrated in Figure 1 are discussed, certain assumptions must be restated that are implicit in the study of Quaternary insect assemblages. The first of these is that beetle species can be identified by characters that are not always used by neotaxonomists. The latter frequently describe species by using portions of the exoskeleton that are destroyed or scattered following the death of the organism. In many cases a specific determination cannot be made with the available fossil components, but even narrowing the identification to a species group often is sufficient to provide valuable ecological data. The second assumption is that species have not evolved during the Pleistocene. Early in the century a general belief prevailed that intense speciation occurred during periods of glaciation. This was one of the greatest barriers to the development of paleoentomology, and it was not until the work of Henriksen (1933), Lindroth (1948), and Coope (1959, 1968; Coope and others, 1961) that this belief was shown to be fallacious. Beetles recovered from Quaternary deposits can be matched precisely with modern species, although they may occur currently thousands of kilometers away from the fossil locality. Examples of major late Pleistocene geographic shifts are demonstrated in papers by Coope (1973) and Hammond and others (1979). Some general statements can be added to the two basic tenets stated above. Paleoenvironments or paleoclimates are not established by use of individual species, but rather of species assemblages. Inherent is the assumption that we know the full modern distribution of a species (or at least the range limits), and that adequate ecological data are available. We assume that the ecology of the fossil species is the same as that of the modem species, and that little or no physiological change has occurred during this period. Publications commenting on some of the points mentioned above include two excellent summaries by Coope (1970, 1979). North American reviews are provided by Ashworth (1979), Matthews (1977, 1980), Morgan and Morgan (1980a, 1980b), and Morgan and others (1983a).