The lack of paleontological data concerning some important Alpine sedimentary formations makes more uncertain the paleogeographic and kinematic reconstructions. The Sion-Courmayeur zone is a complex unit of imbricated and folded sediments derived paleogeographically from the Valais domain. Its metamorphosed flysch-type sedimentary content is poorly dated. Along the Sion-Brig segment, this zone only comprises the internal Roignais and Pierre Avoi units. The first one shows a monotonous thin bedded calcareous flysch (St-Christophe beds) lying on the internal Helvetic Series. The overlying Pierre Avoi unit comprises at its base a thin level of black shales bearing polymetallic nodules and is followed by a 500-700m thick accumulation of turbidites, debris flow deposits and olistostromes. While the calcareous elastic material predominates at the base of this coarse-grained detrital succession, the siliciclastic fraction becomes progressively predominant toward the top. Blocks and megablocks are widespread and derive from a Brianconnais Mesozoic cover as well as from Permo-Carboniferous coal measures, both probably located in the external Zone Houillere. It seems likely that the Pierre Avoi sedimentary chaotic formation was deposited within a subduction trench collecting the coarse detritic material from the crustal accretionary prism where the external Zone Houillere suffered detachment and imbrication. The Roignais and Pierre Avoi metasediments were finally subducted underneath, imbricated, ductily folded and suffered a greenschists metamorphic overprint. Neither the St-Christoph flysch nor the Pierre Avoi unit were dated, although radiolarian debris were found in the latter. They were generally considered as Cretaceous in age according to the presumed early Cretaceous age of the pre-flysch beds. Thin sections of samples from the interbedded black or grey schists within both St-Christophe and Pierre Avoi turbidites all yielded radiolarian debris preserved in thin radiolaria-rich laminae. Unfortunately they did not resist the fluorhydric attack. One sample from the middle part of the Pierre Avoi unit, however, yielded free tests of silicified planktonic foraminifers. This microfossil association comprises reworked species ranging in age from Albian-Aptian to Paleogene. The youngest species, autochtonous or reworked, range from late middle Eocene to early Oligocene (?). According to the sedimentological proximal depositional context, this datation probably applies to the entire Pierre Avoi unit, but does not clarify the age of the underlying St-Christophe flysch, which should be in tectonic contact with the previous unit. Nevertheless, these new micropaleontological data constrain the timing of subduction of the European plate under the Brianconnais basement units accreted within the crustal prism in front of the Adria backstop. Trench deposits such as the Pierre Avoi melange (Valais domain, late middle Eocene-early Oligocene ?), the Ultrahelvetic supra-flysch melange (late Eocene-early Oligocene) and the Helvetic supra-flysch melange (early Oligocene) document the forward propagation of the Paleogene subduction front. These new data also set the propagation of the Piemont accretionary prism (mainly the Prealps units): this pile of nappes had not overthrusted the most external part of the Zone Houillere at late middle Eocene-early Oligocene (?) time. At this time the Brianconnais Zone Houillere was not buried but furnished the Valais trench basin with coarse detritic material.
