The author examines the competing hypotheses that have been advanced to explain the origin and migration routes of the Finns, including the suggestion that they formerly inhabited parts of the North European plain immediately to the south of Scandinavia during the latter part of the Wurm glaciation. As the ice retreated northwards at the end of the Fourth or Wurm glaciation, the ancestors of the Finns similarly moved into Finland from a more easterly part of the North European plain, along the eastern rim of the expanding lake which later became the Baltic Sea. Key Words: Finns, Saami, Finno-Ugric languages, Indo-European languages, Upper Paleolithic, Mesolithic, Scandinavia, Baltic Sea, North European archeology, Wurm glaciation. In 1980, 1 included a model for the spread of Mesolithic hunters into deglaciating northern Europe at the end of the Ice Age as a chapter of my PhD thesis. A slightly modified version was later published in an article, A model for the early settlement of Finland (1987). The latter was intended as a provocation to prompt discussion, and that it did (Dolukhanov 1989; Leskinen 1989; Welinder 1989; Nunez 1989); though perhaps not enough. Subsequently, I have updated the subject in lectures and papers on various occasions (e.g. Nunez 1990, 1995; Nunez and Taavitsainen 1992). The present article is, as the heading suggests, a reassessment of this 17 year-old model in the light of old and new information. On the Dates Used There is no longer an excuse not to use calibrated radiocarbon dates in Finnish archeology now that the tree-ring calibration curve extends far back enough to encompass the whole of Finnish prehistory. This work uses tree-ring calibration for conventional radiocarbon dates back to 10,000 bp (ca. 9200 cal BC) and the calibration based inferred atmospheric spline of coral data for conventional dates within 10,000-18,800 bp (ca. 9200-22,000 cal BC). Since this may be confusing to researchers used to other dating systems, the calibrated values for dates between 7500 and 22,000 cal BC will be followed by the corresponding conventional dates labeled bp. To avoid an overlap between calibrated dates and those that fall outside the calibration range, conventional dates beyond 18,800 bp will be adjusted by adding 3,000 years and labelled BC. Thus 22,000 bp will become 23,000 BC. The 3,000 years roughly corresponds to the difference between the last calibratable conventional date and its calibrated value: 18,899 bp and 21,950 cal BC. All calibrated dates used here have been obtained with Stuiver's and Raimer's (1993) program. As was the case in the original model (Nunez 1987), the central calibrated values have been rounded off to the nearest 100 years. The Original Model The original model is rather general, both because when it was created - and even now - our knowledge of the vast territories involved was very patchy and because simple general models are best suited to explain long-term processes over large areas. Like T.H. White's hawk, I tried to look at late glacial Europe from such a height that details became blurred and the general underlying processes discernible. The model was based mainly on the extant paleoenvironmental and archeological data and assumed the following: 1. During the glacial maximum, ca. 23,000 BC - 19,500 cal BC (ca. 22,000-18,000 bp) human groups had successfully adapted to life in the inhabitable zone in the periphery of the ice sheet - the so-called marginal zone. 2. It took at least four millennia for the ice border to retreat from its maximum position across northern Europe to southern Finland. 3. The territories uncovered by the gradually receding ice sheet logically would have become occupied by those human groups living in adjacent areas of the marginal zone. 4. These groups spread metachronically (time-transgressively) with the gradually retreating ice border, reaching southern Finland by 8000 cal BC (ca. …