T H E CONCEPT OF CENTRAL EUROPE is understood here to cover the geographical centre of the European continent (i.e. the territory between the Elbe, Bug and Neman rivers, that is, eastern Germany, Poland, Bohemia, Slovakia and Lithuania), formerly treated in much of the English-speaking world as 'Eastern Europe'. In the past six years, however, this area has been moving closer to the West. This paper shall concentrate on the region north of the Carpathian mountains, particularly the Vistula river-basin and Scandinavia (without Norway), in other words the territory round the Baltic Sea.The period covered by this paper is that following the Marcomannic Wars (AD 167-80), a time of radical changes in the relations between the Romans and barbarians, that lasted until the end of the Roman Empire in the west. In terms of relative chronology the period covers the whole late-Roman period, that is phases C and D. 2 There is not enough space here to present the details of a complicated chronological system, which varies even in the region under discussion. Relative chronology, based particularly on burial material, is also different for male and female graves (that is, based on weapons or ornaments respectively) even within the same culture. The absolute chronology of the phases is a separate problem. The only tools available at present for its solution are datable Roman imports which in effect means Samian pottery and coins.Across almost the whole territory under discussion one can identify phase B2/C1. This is the period when finds in early-and late-Roman styles occur together in the same assemblages. This phase can probably be dated to the period of the Marcomannic Wars; but in my opinion ornaments in early-Roman style continue to occur in female graves until the turn of the second and third centuries AD. The problem of phase Cia and Cib, particularly the question of the separate existence of phase Cia (different from B2/C1) and the absolute dating of phase Cib is a subject for separate discussion. Ilkjser, in a study of finds from bogs (which play a role quite different from grave-goods), has proposed recently that in Jutland phase Cib begins at the turn of the second-third century AD. 3 Phase C2 is connected with the interregional Leuna-Hassleben-Zakrzow (Sacrau) horizon of chieftain graves, which we can date to the second half of the third century. Phase C3-D begins probably in the early fourth century and lasts until c. the mid-fifth century AD. It includes the interregional horizon of finds in the Sosdala-Untersiebenbrunn style, which date from the last quarter of the fourth century onwards.
