Sympatric speciation, the idea that species can arise without a period of geographical isolation, is at the centre of one of the longest running and most contentious debates in evolutionary biology. In The Origin, Darwin presented a verbal model of speciation that did not require geographical isolation. Since that time, support for sympatric models of speciation has fluctuated markedly. Today, it is generally accepted that sympatric speciation is theoretically possible, but there are few empirical studies that show it has occurred in nature. Here, I present evidence that a morphologically diverse group of land snails from Rarotonga represent a previously unrecognised evolutionary radiation, and that this radiation very likely arose by sympatric speciation.In particular, I use mitochondrial and nuclear DNA sequences and morphological characters to delimit seven species of Lamprocystis land snail, five of which were not previously recognised. A phylogenetic analysis incorporating these species and their relatives from across the Pacific establishes that this radiation arose within Rarotonga and a phylogeographic study of two of the Rarotongan species shows it is unlikely that geographic isolation within the island could explain their origin. Taken together, these studies make a strong, albeit circumstantial, case that the Rarotongan Lamprocystis radiation arose by sympatric speciation. However, the forces that drove the origin of these species remain unclear. Theoretical models that predict the possibility of sympatric speciation all include strong ecological competition between species as they begin to form. Ecological data collected for these species shows little evidence for such competition.Several results produced here are of interest over and above their use in determining whether the Rarotongan Lamprocystis arose in sympatry. The delimitation of a new Pacific land-snail radiation is an important result in and of itself, as snails in the Pacific have contributed greatly to our understanding of evolution, but most previously studied radiations in this region are extinct or at risk of extinction. The particular approach I have taken to delimiting the species that make up this radiation should provide a model for the rapid and accurate species delimitation from molecular and morphological evidence. Finally, several results of taxonomic or conservation importance are discussed.