The Stavelot-Venn Massif (SVM) is the major Cambrian-Ordovician inlier of the Ardenne Allochthon (Southern Belgium). The SVM belongs to the Avalonian microcontinent and was affected by the Caledonian and Variscan orogenies. The Ardenne Allochthon constitutes the northwestern part of the Rheno-Hercynian domain, just south of the Variscan northern front. By contrast, the comparable Avalonian Cambro-Silurian Brabant Massif (northern Belgium) is located just to the north of this front and is not affected by the Variscan orogeny. In this study, we compiled the SVM available but dispersed data concerning the the Cambrian-Ordovician series (field, geochemistry, isotopes). In addition, we acquired new data especially LA-ICP-MS detrital and magmatic zircon ages and Nd isotopes. The comparison with the Brabant Massif, which also belongs to Avalonia and whose geological history is well known, is particurlarly enlightening. SVM Cambrian and lowermost Ordovician depositional environments are similar to the Brabant Massif but sediment thickness is significantly lower (2000 m vs 9000 m). During the remaining Ordovician and Silurian, after the opening of the Rheic Ocean, Brabant and SVM behaved differently, pointing to the existence of two different basements, whereby the basement underneath the SVM acted more rigidly (metacratonic). The SVM environment matches a rift shoulder while the Brabant was located in the rift itself. Our comprehensive study of detrital zircon ages from Cambrian-Ordovician sediments establishes the relative contributions over time of three composite sources: the West African craton, the Western Amazonian craton and the Pan-African orogen. This allowed us to determine a fine record of the tectonic events, distant or local, at the origin of the supply variations from these major sources. The detrital zircon age pattern of the Pridoli conglomerate (Ardennian unconformity) deposited after a sedimentation hiatus of 45 Myrs, is very distinctive. It shows a continuous record between 467 Ma and 420 Ma (97% zircons) that we relate to the activity of a large igneous province (LIP). The latter would be located along the south-eastern boundary of Avalonia, denying the existence of a Silurian island arc as previously proposed. The geochemical evidence indicates that most magmatic rocks have a crustal origin in relation to melting in an intracontinental setting due to stress applied at the Avalonia plate margin. Finally, we propose a geodynamic model in which the rigid basement of the Rheno-Hercynian domain originated from the tearing of a metacratonic fragment of the West African craton, which left a large scar on its western margin (Mauritania/Senegal). This metacratonic fragment includes only the Silurian Brabant western foreland and extends to England, including the Midlands microcraton.