In his monograph, Mortensen (1948) simply placed all scutelliforms into a single family, the Scutellidae. Durham (1955) began the process of classifying this large clade and set up a number of familes. However, he provided no hierarchical structure for the 10 families that he distinguished and no formal classification has been proposed subsequently. The most thorough treatment of this group remains unpublished (Mooi, unpublished Ph. D. thesis).
Taking the living members first, there appear to be two major extant groups, the Echinarachnidae, characterized by having food grooves that form a primary trunk over the first few abulacral plates and whose water vascular system is confined to ambulacral plates, and the remaining forms, which have the food grooves bifurcating at the end of the basicoronal circlet and which have tube feet and food grooves that expanded across the interambulacral zones. These sister clades are here recognized at superfamily level, as Echinarachnidea and Scutellidea respectively.
The Scutellidea can be further subdivided into a number of families of which the Mellitidae and Astriclypeidae are sister taxa, united by having a well-developed microcanal system internally, spines and tubercles on the oral surface clearly differentiated into food-gathering and locomotory areas, and the periproct opening in the first post-basicoronal interambulacral plate. Mellitidae are the more derived as they have an anal lunule. Although both clades possess ambulacral notches or lunules, these are actually constructed in very different ways, as noted by Seilacher (1979) and are presumably homoplasious.
The extant sister group to the Mellitidae plus Astriclypeidae are the Dendrasteridae, a group of sand dollars lacking lunules and with a less well-developed system of internal buttressing with V-shaped interradial channels and a more open, stellate mesh in ambulacral zones. Dendrasteridae have their periproct broadly marginal and opening in plates posterior to the first post-basicoronal plate. Major branches of the food groove extend into the centre of interambulacral columns towards the periphery.
The placement of fossil taxa remains problematic, but the Protoscutellidae, with their simple food groove that bifurcates only after the second or third pair of ambulacral plates, probably lie between the Echinarachnidea and Scutellidea. They cannot, however, be direct ancestors to Scutellidea, since they have five gonopores, a derived condition. Iheringiella is somewhat unusual in having food grooves that bifurcate at the end of the second pair of ambulacrals. The branches from the primary trunk form slightly raised tracts on the test rather than grooves, as in Echinarachniidae.
The Scutellidae as recognized by Durham (1955) contains a mixture of extinct primitive forms, but Scutella and Parascutella have the internal buttressing and food-groove arrangement of the Dendrasteridae, but retain the primitive condition of having continuous interambulacral zones. Abertella and the very closely related Parmulechinus, form a sister group to the clade (Astriclypeidae+Mellitidae), having the dense internal meshwork of that clade, but with the periproct opening in a more posterior interambulacral plate and no strong differentiation of oral tuberculation.
The monophorasterids represent another entirely extinct clade. These have an anal lunule and are usually considered as sister group to the mellitids. Furthermore, like mellitids, their periproct lies associated with the first pair of post-basicoronal interambulacral plates. However, against this is the fact that they have extremely narrow interambulacral zones and neither accessory pores nor food grooves extend across into interambulacral zones.
Scutaster has festooned lunules, like Mellitidae, but lacks an anal lunule and has very narrow interambulacral zones. Until it is redescribed from better material, it is left as incertae sedis within the Scutellidea.
Neotenous scutelliforms are a major problem, being difficult to distinguish from small laganiforms. Mooi (1990) and Mooi & Chen (1996) have done a great deal to clarify the position of some of these small forms. In particular, the presence of a diverticular weight-belt provides an excellent character for separating small neotenous scutelliforms from laganiforms.
The classification adopted here is as follows (*= extinct):-
Infraorder Scutelliformes Haekel, 1896
Superfamily Echinarachniidea Lambert, 1914
- Family Echinarachniidae Lambert, 1914
- *Family Protoscutellidae Durham, 1955
- *Genus IheringiellaBerg, 1898
Superfamily Scutellidea Gray, 1825
- Family Scutellidae Gray, 1825
- Family Astriclypeidae Stefanin, 1911
- *Family Monophorasteridae Lahille, 1896
- Family Mellitidae Stefanini, 1911
A checklist and detailed key to the genera of Scutelliformes can be found in Mooi (1989).