We reconstructed the mitochondrial phylogeny of the species of the brittle star genus Ophioderma, using sequences of the Cytochrome Oxidase I gene (COI) to address four questions: (i) Are the species of Ophioderma described on morphological evidence reflected in mitochondrial genealogy? (ii) Which species separated from which? (iii) When did speciation events occur? (iv) What is the rate of COI evolution in ophiuroids? We found that most of the 22 described species we sampled coincide with monophyletic clusters of COI sequences, but there are exceptions, most notably in the eastern Pacific, in which three undescribed species were indicated. The COI phylogeny lacks resolution in the deeper nodes, but it does show that there are four species pairs, the members of which are found on either side of the central American Isthmus. Two pairs with a genetic distance of ~ 4% between Atlantic and Pacific members were probably split during the final stages of Isthmus completion roughly 3 million years ago. The rate of divergence provided by these pairs allowed the calibration of a relaxed molecular clock. Estimated dates of divergence indicate that the lineages leading to extant species coalesce at times much older than congeneric species in other classes of echinoderms, suggesting that low extinction rates may be one of the reasons that ophiuroids are species-rich. The mean rate of COI substitution in Ophioderma is three times slower than that of echinoids. Conclusions of previous mitochondrial DNA studies of ophiuroids that relied on echinoid calibrations to determine divergence times need to be revised.