Antarctica is a continent locked in ice, with almost 99.7% of current terrain covered by permanent ice and snow, and clear evidence that, as recently as the Last Glacial Maximum (LGM), ice sheets were both thicker and much more extensive than they are now. Ice sheet modelling of both the LGM and estimated previous ice maxima across the continent give broad support to the concept that most if not all currently ice-free ground would have been overridden during previous glaciations. This has given rise to a widely held perception that all Mesozoic (pre-glacial) terrestrial life of Antarctica was wiped out by successive and deepening glacial events. The implicit conclusion of such destruction is that most, possibly all, contemporary terrestrial life has colonised the continent during subsequent periods of glacial retreat. However, several recently emerged and complementary strands of biological and geological research cannot be reconciled comfortably with the current reconstruction of Antarctic glacial history, and therefore provide a fundamental challenge to the existing paradigms. Here, we summarise and synthesise evidence across these lines of research. The emerging fundamental insights corroborate substantial elements of the contemporary Antarctic terrestrial biota being continuously isolated in situ on a multi-million year, even pre-Gondwana break-up timescale. This new and complex terrestrial Antarctic biogeography parallels recent work suggesting greater regionalisation and evolutionary isolation than previously suspected in the circum-Antarctic marine fauna. These findings both require the adoption of a new biological paradigm within Antarctica and challenge current understanding of Antarctic glacial history. This has major implications for our understanding of the key role of Antarctica in the Earth System.