Pioneering the use of radiolarians for terrane and collage analysis in the 70ies, D.L. Jones greatly contributed to establish this microfossil group as a familiar component of the Cordilleran tool box. Since then, a wide number of Paleozoic-Mesozoic radiolarian localities have been complementing initial data from ammonites, corals, foraminifers, and conodonts. More specifically, radiolarian oceanic signature opened up a window for the study of Cordilleran sutures, and what they represent for structural analysis and paleogeographic reconstructions. Due to slow sedimentation rates and tectonic underplating, surfaces of chert depocenters may however have been underestimated, as is revealed by the unsuspectedly long age ranges of Cordilleran radiolarites. An overview of Ordovician-Cretaceous chert terranes age ranges from North America shows that radiolarian deposition commonly reach 100 to ~200 m.y. within units broadly arranged into a westward-younging pattern. Parallel data are now being established on the other side of the Pacific ocean, for instance in Siberia and East Asia where radiolarian studies have grown significantly over the last decade and complement initial North American results. Still, the relatively weak provincial signature of pelagic fauna such as radiolarians hinders paleogeographic reconstructions. Evaluating longitudinal displacements has been based on benthic / nectobenthic faunal assemblages found in carbonate successions, but the notion of being 'Tethyan' has enthusiastically been used as an adaptable proxy for either low latitude, distant longitude, or both. Probabilistic models based on faunal biogeography (i.e. Belaski and Runnegar 1994) were unfortunately not developed further, and we still lack quantitative, reliable constraints on terrane paleolongitudes. After 30 years of exploration, what do radiolarian-bearing oceanic terranes tell us about the geodynamic evolution of Panthalassa, not only from a North-American point of view, but on a global circum-Pacific perspective ?