This study attempts to reconstruct the vegetational and climatic history of the last 300 000 years for the Taranaki-Wanganui district, North Island, New Lignites in the cover beds of a wellpreserved sequence of marine terraces in South Wanganui Basin provide excellent material for pollen analysis from sequences that are chronologically constrained by the ages of the marine terraces and by the loessand tephrostratigraphy of their cover beds. Anthropogenic influences in New Zealand affect only the last c. 1000 years. Modern pollen samples from a variety of vegetation types from central-western North Island, together with previous modern pollen studies, indicate the palynological representation of taxa and allow the characterisation of the major vegetation types, although these data provide only a limited picture of the source vegetation. This information is used for helping to interpret the fossil pollen spectra. Two sites from the Waverley district, near Wanganui, which together cover the mid and late Holocene, provide an analogue of vegetation assemblages and their sequence of development in the area, for use in interpreting the Pleistocene pollen sequences. The vegetation history from these sites conforms to the known Holocene pattern from sites elsewhere. Polynesian deforestation at Lake Waiau Swamp is estimated to have occurred some time between 685 CAL B.P. and 210 years ago. Five sites from the Pleistocene marine terrace cover beds are pollen analysed in detail. The pollen assemblage zones are correlated with the established deep-sea oxygen isotope record. Combined with previous pollen studies from the terrace cover beds, the results provide a relatively continuous, detailed vegetation history dating from c. 400 ka ago to the present, for the fIrst time in New This record is closely linked to the marine terrace, loess-, and tephrostratigraphy of this region. Because the record is mostly from sequential rather than coeval sites, further work is required to corroborate certain details. The vegetation on the lowlands of Taranaki-Wanganui alternated from grass-shrubland communities during glacial periods, to podocarp-hardwood forest, dominated in the pollen assemblages by Dacrydium cupressinum, during the interglacial times. No particular period is uniquely characterised by its pollen assemblages. In interglacial pollen assemblages, generally some of the hardwood forest elements are severely underrepresented (e.g., Beilschmiedia tawa). The warmest, wettest, and mildest (with rare frosts and droughts) climates during interglacials are suggested by pollen assemblages matching those of the early to mid Holocene, with common Ascarina lucida and Dodonaea viscosa pollen in nearcoastal sites. Interstadial and stadial pollen assemblages are more diffIcult to characterise. In general, interstadial assemblages seem to be dominated by Prumnopitys taxifolia pollen, and· stadial assemblages seem to be dominated by Nothofagus fusca type or Libocedrus pollen, together with common shrub and grasS pollen. The vegetation for Some of the climatic stages, such as the Last and Penultimate Interglacials sensu stricto, are clearly demonstrated for the first time in New N. fusca type, N. menziesii, Libocedrus, Halocarpus, Phyllocladus, Empodisma minus, and Gleichenia are some taxa for which geographic range expansion. can be demonstrated when climates were cooler than present However, the vegetation history of the area generally agrees with the hypothesis that the flora of the Taranaki-W~ganui district has remained essentially the same through the Q\1aternary, although a few records of taxa well outside their present ranges require special explanation. Glacial period pollenrecords show that, while forest was probably largely excluded from the lowlands, it remained in the region during the coldest climatic periods and probably occupied sheltered areas. Broadly synchronous records from separate sites suggest that regional vegetational differences existed, probably due to local site and climate differences, analogous to the present pattern. Occasional natural fIres appear to have occurred throughout the periods studied, but there is no evidence that deforestation due to severe fIres occurred during the late Pleistocene. Palynology and radiocarbon dating of the Ohawe Beach Site, near Hawera, demonstrate that the marine terraces in this region have previously been mapped incorrectly, but in general the marine terrace mapping and chronology of B. J. Pillans (1983:. Upper Quaternary marine terrace chronology and deformation, south Taranaki, New Zealand. Geology 11: 292-297;) is upheld. Radiocarbon dating of separate organic fractions from degraded fossil wood samples at Ohawe Beach show that chemical contamination by young radiocarbon had occurred· and that the real age is beyond the radiocarbon range. This serves as· a warning against simple interpretation of radiOCarbon ages on wood from cover beds in the area. Pollen analyses also suggest that the Fordell Ash is younger than the Mt Curl Tephra, as opposed to previous interpretations . The extinct palynomorphs, Acacia· type andPeromonolites problematicus Couper, are identifIed from the cover beQ sediments, and their time ranges in New Zealand extended until c .. 11 0 ka ago and 90 ka ago, respectively. The implications of these fIndings are discussed. .