Evidence for Mesozoic and early Cenozoic glaciations is ambiguous at best. Most of the purported evidence including faunal changes, glendonites and stable isotope paleothermometry imply cold conditions but do not explicitly require ice-sheets. In the absence of established indicators of glaciation, such as striated pavements, geomorphic features and tillites, only erratic bearing deposits, if interpreted as ice-rafted dropstones, provide direct evidence of large-scale terrestrial ice. It is shown here that such erratics can be equally well explained by non-glacial processes such as organic-rafting. Erratics in the Chalk of southern England and the middle Cretaceous of Australia are representative of the problems of “glacial” indicators in the Mesozoic.

The interpretation of high-frequency, large-magnitude, eustatic sea-level oscillations requires the existence of large ice sheets throughout much of the Mesozoic and Cenozoic. Although a temporal and spatial relationship between erratic-bearing deposits and transgressions is noted, the lack of demonstrable evidence for glaciation suggests that the amplitude of interpreted sea-level changes may have been exaggerated. We suggest that during non-glacial times the magnitude of any glacio-eustatic signal would have been no greater than 10-20 m and generally much less. Such oscillations would have been swamped by intra-basinal effects. We therefore suggest that global correlation of derived “3rd-order” or equivalent eustatic curves is untenable. We further suggest that beach front erosion explain the origin of erratic-bearing deposits than continental ice-sheets