The brittle stars are something of the poor cousin among echinoderm classes. Their tendency to relatively small size and cryptic habitats means that they do not attract the level of attention given to starfish, sea urchins or sea cucumbers. Despite this, they are perhaps the most diverse of the living echinoderm classes, with more recognised species around today than any other.
It should therefore come as no surprise that the internal classification of brittle stars remains decidedly up in the air. The basic framework of the surrent system was established over a hundred years ago by Matsumoto (1915) and changes to this arrangement since have been fairly cosmetic. However, a significant challenge to Matsumoto's system has been arisen following the input of molecular data to the mix: many of Matsumoto's higher groupings have not been supported by moleculat analyses. Perhaps the nail in the Matsumoto system's coffin has come from a recent publication by Thuy & Stöhr (2016) who found that a formal analysis of morphological data also failed to support the pre-existing classification. At this point in time, we know that a new classification of brittle stars is needed but we don't yet know what form it will take.
Perhaps one of Matsumoto's groupings that will survive the transition is the Gnathophiurina. Notable features of this group include a ball-and-socket articulation between the radial shields (large plates that sit on the aboral side of the central body on either side of the insertion of each arm) and the genital plates (sitting below and alongside the radial shields), with the socket in the radial shield and the ball on the genital plate. The genital plates are also firmly fixed to the basal vertebra of each arm. I haven't been able to find what the functional significance of this arrangement is, such as whether it renders the body more flexible that in other groups where the radial-genital plate articulation is more fixed. At least one of the families of Gnathophiurina, the Amphiuridae, includes species that commonly live in burrows with the tips of their arms extended into the water column, using their tube feet to capture food particles (Stöhr et al. 2012). In contrast, some Ophiotrichidae are epizoic, living entwined around black corals and the like. The Gnathophiurina as a whole seem to be most diverse in relatively shallow waters.
Matsumoto's (1915) original concept of the Gnathophiurida included species that are now classified into four families, the Amphiuridae, Ophiotrichidae, Amphilepididae and Ophiactidae, and recent analyses have returned results not inconsistent with this association. In Thuy & Stöhr's (2016) morphological analysis, Gnathophiurina species all belong to, and make up the bulk of, their clade IIIc. In the molecular analysis presented by Hunter et al. (2016), the families belong to two separate clades but the branch separating them is very weakly supported. Further research is needed, of course, but it may turn out that Matsumoto was on to something when he focused on that ball-and-socket joint.
Hunter, R. L., L. M. Brown, C. A. Hill, Z. A. Kroeger & S. E. Rose. 2016. Additional insights into phylogenetic relationships of the Class Ophiuroidea (Echinodermata) from rRNA gene sequences. Journal of Zoological Systematics and Evolutionary Research 54 (4): 269–275.
Matsumoto, H. 1915. A new classification of the Ophiuroidea: with descriptions of new genera and species. Proceedings of the Academy of Natural Sciences of Philadelphia 67 (1): 43–92.
Stöhr, S. T. D. O'Hara & B. Thuy. 2012. Global diversity of brittle stars (Echinodermata: Ophiuroidea). PLoS One 7 (3): e31940.
Thuy, B., & S. Stöhr. 2016. A new morphological phylogeny of the Ophiuroidea (Echinodermata) accords with molecular evidence and renders microfossils accessible for cladistics. PLoS One 11 (5): e0156140.