In this paper a general account is given of a peculiar facies in the Lower Carboniferous of Northern England, the origin of which has given rise to much controversy. This facies is most prominently displayed in the limestone hills named "reef knolls" by Tiddeman. The evidence is reviewed and the conclusion reached that these features are in fact reef-like in origin and are not merely unconsolidated banks. The reefs occur near the margins of sedimentary basins, and some form a more or less continuous rim between shelf and basin. The "marginal" reefs are characteristic of the upper Dinantian. Another type consists of mound-like discrete masses which are usually found within the basins, though small examples occur in the shelf areas. This type is dominant in the basin areas nd is characterized by the presence of original dips on the flanks of the knolls. A difficulty in ascribing these structures to reefs is that in much of the limestone the fossils have been destroyed and the nature of the reef-builders is in doubt. The reefs, however, do contain a rich and varied fauna, particularly of brachiopods and crinoids. Bryozoans are abundant locally and it is argued that these creatures were prominent in building the framework of the reef by acting as sediment binders. They commonly constitute the nucleus of a form of fibrous calcite which is widespread in the reef limestone and peculiar to it. Calcareous algae may have been important reef-builders. They are rarely preserved, but it has been suggested (Black, 1954) that the unbedded calcite mudstone characteristi of the reef limestone might be recrystallized algal skeletons.