Abstract Rivers and river-borne deposits have always been a major attraction for hominins as an important source of sustenance and settlements. Hence, fluvial deposits have long been an important source of evidence for early human occupation throughout the Old World. Apart from being an important palaeoclimatic marker, fluvial sequences have provided archaeologists with frameworks for correlation, along with Palaeolithic markers discovered within them. Moreover, given the influx of sediments eroded and deposited by Indian rivers, these could have acted as a centre of hominin activities. Palaeolithic research in India has been concentrated around some of its major river valleys, which have yielded a rich record of hominin occupation. So far, 305 Palaeolithic sites have been reported from a gravel context throughout the country, yielding Lower to Upper Palaeolithic, Mesolithic and Neolithic evidence. However, most of the derived evidence is secondary deposits and stands contested based on its contextuality. Nevertheless, its importance as a source of information about hominin activity cannot be underestimated. This review presents a provisional synthesis of all of the Indian Palaeolithic sites reported from gravel contexts, thereby presenting scope for future multidisciplinary research at these localities.

ABSTRACT The ~11-km-wide, Paleoproterozoic Dhala impact structure in north-central India comprises voluminous exposures of impact melt breccia. These outcrops are discontinuously spread over a length of ~6 km in a semicircular pattern along the northern, inner limit of the monomict breccia ring around the central elevated area. This study of the magnetic fabrics of impact breccias and target rocks from the Dhala impact structure identified a weak preferred magnetic orientation for pre-impact crystalline target rocks. The pre- and synimpact rocks from Dhala have magnetite and ilmenite as common magnetic phases. The distributions of magnetic vectors are random for most impact melt breccia samples, but some do indicate a preferred orientation. Our anisotropy of magnetic susceptibility (AMS) data demonstrate that the shape of susceptibility ellipsoids for the target rocks varies from prolate to oblate, and most impact melt breccia samples display both shapes, with a slight bias toward the oblate geometry. The average value for the corrected degree of anisotropy of impact melt rock (P′ = 1.009) is lower than that for the target rocks (P′ = 1.091). The present study also shows that both impact melt breccia and target rock samples of the Dhala structure have undergone minor postimpact alteration, and have similar compositions in terms of magnetic phases and high viscosity. Fine-grained iron oxide or hydroxide is the main alteration phase in impact melt rocks. Impact melt rocks gave a narrow range of mean magnetic susceptibility (K m ) and P′ values, in contrast to the target rock samples, which gave K m = 0.05–12.9 × 10 −3 standard international units (SI) and P′ = 1.036–1.283. This suggests similar viscosity of the source magma, and limited difference in the degrees of recorded deformation. Between Pagra and Maniar villages, the K m value of impact melt breccias gradually decreases in a clockwise direction, with a maximum value observed near Pagra (K m = 1.67 × 10 −3 SI). The poor grouping of magnetic fabrics for most impact melt rock samples implies local turbulence in rapidly cooled impact melt at the front of the melt flow immediately after the impact. The mean K 1 for most impact melt samples suggests subhorizontal (<5°) flow in various directions. The average value of K m for the target rocks (4.41 × 10 −3 SI) is much higher compared to the value for melt breccias (1.09 × 10 −3 SI). The results of this study suggest that the melt breccias were likely part of a sheet-like body of sizeable extent. Our magnetic fabric data are also supported by earlier core drilling information from ~70 locations, with coring depths reaching to −500 m. Our extensive field observations combined with available widespread subsurface data imply that the impact melt sheet could have covered as much as 12 km 2 in the Dhala structure, with an estimated minimum melt volume of ~2.4 km 3 .