Many Cambrian fossils have at times been called ‘corals’. The most coral-like of these are small, cup-shaped, mostly solitary organisms with septa. Some have an operculum over the calice opening. In total, seven orders of Paleozoic corals may be recognised, of which the Tabulata and Rugosa are by far the most important. Unlike the Scleractinia, these two groups have left good fossil records as their skeletons are calcitic and thus more stable than the aragonite skeletons of Scleractinia.
Tabulata are much less variable than rugose or scleractinian corals. They are all colonial and consist of slender tube-like corallites 1-3 mm diameter, crossed internally by transverse partitions, the tabulae. Colonies are typically incrusting, flat or massive, but may be branching. Individual corallites may be in contact or widely separated. Each corallite has a theca (enveloping sheath) and groups of corallites are enclosed in a sheath-like epitheca. The corallites may be connected by fine tubules forming a three-dimensional structure. These tubules may pierce the thecae as mural pores. Where corallites are in close contact, tabulae are usually found. Where corallites are separated, external horizontal plates (dissepiments) also occur. Radially arranged spine-like septa, sometimes forming vertical structures, may be present.
The morphology of the Tabulata does not indicate whether they were zooxanthellate or not. However, as they were colonial, had small corallites, and had growth rates comparable to those of zooxanthellate Scleractinia (up to 18mm per year) it seems probable that they were.
The Tabulata are first seen as a distinct group in the Early Ordovician and thus pre-date the Rugosa. There is no evidence of common ancestry. By Middle Ordovician there was a radiation in Tabulata diversity with six families recognized. These are mostly associated with soft substrates rather than carbonate platforms of high energy environments. Tabulata became significant frame builders during the Silurian but were relatively unimportant during intervals of maximum reef development in the Devonian. They appear to have not re-gained diversity during the Carboniferous or Permian, although they were periodically abundant.
Rugosa have a much greater resemblance to Scleractinia than to the Tabulata. Mature colonies have an array of growth forms comparable to Scleractinia (see below) although the majority of taxa are solitary or form colonies where individual corallites are large and dominant. Rugosa have basic structural components at least superficially similar to Scleractinia, the main difference being in the arrangement of the septa. Tabulae analogous to those of Tabulata are usually very abundant and so is a complex array of dissepiments which appear analogous to those of Scleractinia.
The earliest Rugosa were solitary and of simple construction. Complex colonies had evolved by Late Ordovician and thereafter the few evolutionary trends that are seen are mostly in skeletal detail and most have reversed one or more times. Differences in the pattern of insertion of septa make it unlikely that the Scleractinia arose from the Rugosa although this view is disputed.