A right bank tributary of the Niger River, the Mekrou, has formed a strongly incised river bed within clayey to sandy alluvium, locally interrupted by thin layers of gravel and coarser deposits. The alluvia give evidence to different climatic conditions: finer material was accumulated during flooding within a humid period, whereas the arid times seem to be reflected by coarser sediments. The cyclic facies change of sedimentation gives evidence for a repeated shilft in climate and hydrologic conditions, assuming that the alluvia originate in the upper Pleistocene. Some human artefacts are associated to the different gravel layers (subjacent bed = Palaeolithic, intermediate bed = middle Palaeolithic, overlying bed = ‘recent’ Palaeolithic, Neolithic, and iron Metallurgy). The absence of terraces, the occurrence of sandy sediments on the border of the river bed indicates to active morphodynamic processes; some angularly shaped meanders give evidence for a rapid change of drainage and leads to the hypothesis of a modified flow-off by the river’s recent capture.
In southwestern Niger, near Niamey, several thousand singleuse bloomery furnaces have been mapped and identified. The archaeological study of approximately 30 furnaces and their slag reveals the existence of four methods for iron smelting: three types of pit furnace and one slag-tapping type. The slag pit furnaces are clearly differentiated by the form and volume of their pits. All slag-tapping furnaces drain off slag through small openings. The slag is tapped either vertically or laterally. According to radiocarbon dates, the smelting activity developed in the 2nd century AD and intensified through to the 14th century. It continued to evolve until the middle of the 20th century. The low intensity of iron production for these furnaces indicates the products were intended mainly for the local market.