The origins and evolutionary history of the New Zealand flora has been the subject of much debate. The recent description of Cyathodophyllum novaezelandieae from early Miocene sediments in New Zealand provides possible evidence for the antiquity of the fleshy fruited epacrids (tribe Styphelieae, Ericaceae) in New Zealand. Yet the extant species in this tribe are thought to be very closely related to or conspecific with Australian taxa, suggesting recent trans-Tasman origins. In order to investigate the origins and evolution of the extant New Zealand Styphelieae we produced molecular phylogenetic trees based on sequences of three plastid regions that include representatives of all the genera of the tribe and eight of the ten New Zealand species. We estimated the range of minimum ages of the New Zealand lineages with Bayesian relaxed-clock analyses using different calibration methods and relative dating. We found strong support for each of the eight extant species of New Zealand Styphelieae being a distinct lineage that is nested within an Australian clade. In all except one case the sister is from Tasmania and/or the east coast of mainland Australia; for Acrothamnus colensoi the sister is in New Guinea. Estimated dates indicate that all of the New Zealand lineages diverged from their non-New Zealand sisters within the last 7 Ma. Time discontinuity between the fossil C. novae-zelandiae (20-23 Ma) and I he origins of the extant New Zealand lineages (none older than 5 Ma) indicates that the fossil and extant Styphelieae in New Zealand are not related. The relative dating analysis showed that to accept this relationship, it would be necessary to accept that the Styphelieae arose in the early-mid Mesozoic (210-120 Ma), which is starkly at odds with multiple lines of evidence on the age of Ericales and indeed the angiosperms. Therefore, our results do not support the hypothesis that Styphelieae have been continuously present in New Zealand since the early Miocene. Instead they suggest a historical biogeographical scenario in which the lineage to which C. novae-zelandiae belongs went extinct in New Zealand, and the extant New Zealand Styphelieae are derived from Australian lineages that recolonised (presumably by long distance dispersal) no earlier than the late Miocene to Pliocene.