Agetolites is a problematic Late Ordovician genus possessing traits of both tabulate and rugose corals. The presence of numerous mural pores has often been considered to indicate a relation to tabulates, although an affinity to rugosans has also been proposed, based mainly on well-developed septa that alternate in length. To further consider the taxonomic position of Agetolites, growth characteristics of coralla representing three species from the Xiazhen Formation in South China are documented and assessed, focusing on modes of corallite increase. Three major modes of increase are recognized. By far the most common mode involves the development of an offset from a connective mural pore, without a clear relationship to a particular parent corallite. This mode of increase is usually associated with corner pores, but in one case occurs at a wall pore. The lateral mode of increase, which is relatively uncommon, is a typical feature in corallites along the boundary of intergrowths with stromatoporoids. The axial mode of increase is rare, occurring during rejuvenation of a damaged corallite or during regeneration following termination of a corallite. The mode of corallite increase that is characteristic of Agetolites, involving a connective mural pore and occurring without evidence of a particular parent, supports the interpretation that this genus is not a rugosan or a typical favositid tabulate. Mural pores are unknown in rugosans, and offsets arise from distinct parent corallites in favositids. The Ordovician genus Lichenaria, considered a representative of the most primitive stock of tabulate corals, shows the closest similarities with types of increase in Agetolites. Certain aspects of lateral and axial increase in Agetolites are comparable to features in a few more genera of Ordovician tabulates, further supporting a tabulate affinity. The phylogenetic relation of Agetolites to those and other tabulate genera, however, remains unresolved.