Did early Homo migrate "out of" or "in to" Africa?

T he origin of our own genus remains frustratingly unclear. Although many of my colleagues are agreed regarding the “what” with respect to Homo, there is no consensus as to the “how” and “when” questions. Until relatively recently, most paleoanthropologists (including the writer) assumed Africa was the answer to the “where” question, but in a little more than a decade discoveries at two sites beyond Africa, one at Dmanisi in Georgia and the other at Liang Bua on the island of Flores, have called this assumption into question. The results of recent excavations at Dmanisi reported in PNAS (1), which suggest that hominins visited that site on several occasions between ca. 1.85 and ca. 1.77 Ma, together with recent reassessments of the affinities of Homo habilis, are further reasons for questioning the assumption that Homo originated in Africa. The site of Dmanisi, which is 34 miles southwest of Tbilisi, is situated on a promontory at the confluence of two rivers, the Masavera and the Pinasaouri. Since the 1930s the main foci of excavations have been its Bronze Age and medieval archeology, but between 1983 and 1987 excavations in part of the medieval village resulted in the recovery of early Pleistocene fossils, and the first of many well-preserved hominin fossils, the D211 mandible, was recovered in 1991. The early Pleistocene sediments at Dmanisi, which are dominated by primary and locally reworked ashfalls, are divided into two major units: stratum A (with subunits A1–A4), which conformably overlies the Masavera Basalt, and stratum B (with subunits B1–B5), which overlies stratum A and is separated from it by a minor erosional disconformity (2).