The morphological identification and proper delineation of subterranean fauna species has challenged taxonomists for decades. Here, we investigate the suitability of quantitative methods of recognizing species in a subterranean radiation of short-tailed whipscorpions (Arachnida: Schizomida: Hubbardiidae) in a small part of the Hamersley Range in the Pilbara region of Western Australia. A molecular phylogeny was used to test six methods of species delineation, each founded on different assumptions. By comparing the results against existing, well-described schizomids in the Pilbara, we suggest that the bGMYC (Bayesian General Mixed Yule-Coalescent) and, to a lesser degree, ABGD (Automatic Barcode Gap Discovery) models provide a framework on which to base further investigations. Neither method appears likely to reliably delineate subterranean species in habitats where there is considerable genetic structuring and small numbers of animals. Nevertheless, the models infer an impressive diversity of up to 50 subterranean schizomid species across the Pilbara as a whole, and up to 15 species in the study area. We also show that both the timing and the mode of speciation is consistent with a model of independent movements into the subterranean matrix after the Middle Miocene, followed by subsequent in situ diversification of the Hamersley fauna throughout the Pliocene and Early Pleistocene when arid steppe habitats first emerged in north-western Australia. Molecular dating indicates that this diversification is comparably recent and ongoing. While the drivers of continued speciation are unclear, the variable ranges of species and the sympatric (or micro-allopatric) occurrence of several species suggests dispersal is poor, which fits with the high level of genetic structuring observed.