A little-explored line of evidence for the antiquity and nature of early vegetation on land is the soils in which they grew. Vegetation is one of a number of factors known to play an important role in forming modern soils and soil features. As in studies of the role of organisms in modern soil formation, what are needed are fossil soils that supported different ancient ecosystems, but for which other soil-forming factors, such as palaeoclimate, palaeogeographical setting, parent materials and duration of formation, were closely comparable. This preliminary petrographic and chemical study compares four palaeosols; three are from the states of Pennsylvania and New York, U.S.A., and the fourth from the Potwar Plateau region of northern Pakistan. All appear to have formed in a subhumid, seasonally dry, subtropical climate, on the alluvial outwash of major mountain ranges, largely formed of sedimentary and metamorphic rocks, and over periods of only a few thousand years. These palaeosols are here named the Potters Mills clay (late Ordovician), Lehigh Gap clay (late Silurian), Peas Eddy clay (late Devonian) and Lal clay (late Miocene). Successively younger palaeosols show increasing degree of weathering, more clayey texture and better soil structure. Deep burrows are abundant in late Ordovician palaeosols, and are evidence of sizeable (3-16 mm diameter) soil animals. Bioturbation in the surface of the late Silurian palaeosol may have been produced by animals or vascular land plants. Large root traces and remains of leaf litter are indications that the late Devonian palaeosol supported a low diversity, streamside gallery forest. Weak redistribution of iron in this palaeosol may have been produced by phenolic and other herbivore suppressant toxins from these early trees. The late Miocene palaeosol is extensively bioturbated, presumably by termites and other creatures. Judging from its root traces and associated sediments and fossils, it supported gallery forest in a region of grassy savanna groveland. These early results encourage the belief that fossil soils may provide useful evidence for the nature of early ecosystems on land, not only complementary to that of early terrestrial fossils, but also in sedimentary sequences too oxidized and acidic to allow preservation of fossil plants and animals.