A bipolar disjunction is an extreme, yet common, biogeographic pattern in non-vascular plants, yet its underlying mechanisms (vicariance or long-distance dispersal), origin and timing remain poorly understood. Here, combining a large-scale population dataset and multiple dating analyses, we examine the biogeography of four bipolar Polytrichales mosses, common to the Holarctic (temperate and polar Northern Hemisphere regions) and the Antarctic region (Antarctic, sub-Antarctic, southern South America) and other Southern Hemisphere (SH) regions. Our data reveal contrasting patterns, for three species were of Holarctic origin, with subsequent dispersal to the SH, while one, currently a particularly common species in the Holarctic (Polytrichum juniperinum), diversified in the Antarctic region and from here colonized both the Holarctic and other SH regions. Our findings suggest long-distance dispersal as the driver of bipolar disjunctions. We find such inter-hemispheric dispersals are rare, occurring on multi-million-year timescales. High-altitude tropical populations did not act as trans-equatorial 'stepping-stones', but rather were derived from later dispersal events. All arrivals to the Antarctic region occurred well before the Last Glacial Maximum and previous glaciations, suggesting that, despite the harsh climate during these past glacial maxima, plants have had a much longer presence in this southern region than previously thought.