Abstract: This article reviews briefly current data and ideas on back-arc basins and suggests that such basins in the Western Pacific might indeed owe their origin to back-arc spreading as postulated by D. Karig(1971).Those basins commonly have an oceanic curst, recognizable magnetic lineations, and high heat-flow. However, not all back-arc basins are so characterized. The Caribbean basins, for example, have neither magnetic lineation nor apparent crustal ages as predieted by a spreading model(Edgar, Saunders, et al., 1973).Recent multi-channel seismic data suggest that the previous deep-sea drilling there has only reached mid-Cretaceous basalt extruded during mid-plate volcanism;the drilling has failed to sample the true oceanic basement(e.g., L. Montadert, personal communication).Current speculations regard the Caribbean as a fragment of the Pacific curst that has incidentally been trapped behind the Caribbean Arc,(e.g., E. L. Winterer, personal communication);the geological history of the Caribbean crust might be comparable to that of the Nauro Basin of the South Pacific(Larsen, Schlanger, et al., 1980).