An empirical test of the relative frequency of bipolar reduction in Beds VI,V, and III at Mumba Rockshelter,Tanzania: implications for the East African Middle to Late Stone Age transition

Mumba Rockshelter, Tanzania, is the only East African site spanning a continuous record of more than 100,000 years, including the Middle Stone Age (MSA) to Late Stone Age (LSA) Transition. Rather than examine the presence or absence of traditional markers of “behavioral modernity”, we have endeavored here to respond to the calls of Shea (2011) and Basell (2008) by examining the proportional similarity of a trait present throughout the MSA–LSA sequence at Mumba Rockshelter: the bipolar technique of lithic reduction. We use a quantified, experimentally derived proxy to track the relative amount of bipolar percussion among Mumba's Beds VI, V, and III, namely, waste shatter. Our examination of Mehlman's previously unanalyzed collections from Mumba demonstrates that in terms of bipolar production Bed V is statistically indistinguishable from the Late Stone Age (LSA) Bed III, but significantly different from the Middle Stone Age (MSA) Bed VI. Given that Bed V dates to 56.9 ± 4.8 − 49.1 ± 4.3 ka cal BP, this result is consistent with other evidence that the origins of the LSA in East Africa began well before 40 kya, and that Mumba Bed V represents one of the earliest manifestations of the LSA in East Africa. We conclude with a discussion of factors that may have influenced the increased dependence of bipolar production at the site.     

End‐Member Mixing Analysis (EMMA) Applied to Sediment Grain Size Distributions to Characterize Formational Processes of the Main Excavation Block,Unit 2, of the Rimrock Draw Rockshelter (35HA3855), Harney Basin,Eastern Oregon (USA)

End‐member mixing analysis (EMMA) was conducted on the grain size distributions (GSD) of 13 sediment samples collected from the main excavation block, Unit 2, of the Rimrock Draw Rockshelter, a deeply stratified, multi‐component Palaeoindian rockshelter in the Harney Basin of eastern Oregon, USA. EMMA confirmed the three stratigraphic units (SU) observed in the field and identified three sedimentary end‐members (EM) that account for 95.52% of Unit 2 site formation. EM 1 comprises 70.72% of the total measured grain‐size variation and represents fluvial deposition, correlating with the bottom stratigraphic unit, SU 3. EM 2 comprises an additional 17.74% of the total measured grain‐size variation and represents fluvially reworked aeolian deposition, correlating with the upper stratigraphic unit, SU 1. EM 3 comprises the remaining 7.06% of the total measured grain‐size variation and represents possibly a two‐component aeolian deposition of suspension and short‐term saltation, correlating with the middle unit, SU 2. The results presented demonstrate the ability of EMMA as a technique to (1) unbiasedly distinguish between the SUs identified in the field using only the measured grain size distributions, (2) provide genetically meaningful and quantitative grain‐size end‐members to identify multiple major depositional processes of site formation, and (3) provide a geologically meaningful interpretation of site formation that is capable of facilitating the formulation of more focused hypotheses regarding human and environment interactions at the site.