Aspects of succession of subalpine vegetation on Mt Taranaki were examined in a vegetation chronosequence, initiated by four landslides on the same sideslope in 1333-1343 A.D., 1625 A.D., 1811-1817 A.D. and 1968-1971 A.D .. In three stands Libocedrus bidwillii (kaikawaka) subalpine low forest was displayed at different stages of development while the fourth stand without kaikawaka comprised mainly low herbaceous vegetation. Core samples were collected from the three kaikawaka populations to establish the crossdating pattern. describe anomalous ring features, establish that the four stands formed a vegetation chronosequence and to examine the age-size relationships of this species. The crossdating pattern found in the present study confirms that reported for kaikawaka at two other sites on Mt Taranaki. The absence of the 1934-1938 signature in cores from the youngest stand suggests the climatic signal is being suppressed in younger trees. Various types of anomalous rings were described for kaikawaka and -frequencies per core given. A number of features not previously reported for this species were recognised: 1) resin bands as a type of false ring 2) the presence of-damaged rings, their synchronous occurrence within a tree, in different trees and with false rings 3) two types of partially absent rings 4) a highest frequency of absent rings for any single radius in datable cores of 3.4 % and 5) the phenomenon of displaced marker rings. A linear relationship between tree age at 0 cm downslope and DBH of kaikawaka was found to be significant at the 0.1% level but as in other studies a wide variation of age within any one size class suggested that ages predicted from DBH data may be misleading. Floristic composition of the four stands was recorded on a stratum basis and analysed by lists and through tables of species in common, species unique, total number and identity of species present and Jaccard's and Sorensen's similarity indices. These data showed that the most diverse stratum in all four stands was the ground cover and the major between-stand differences in floristic composition occurred in this stratum. The youngest stand was the most distinctive floristically with twenty-seven species of vascular plants exclusive to this stand. Few (<8) species were found exclusive to each of the 3 older stands. However, 11 species were exclusive to the two older stands and 25 species occurred in all 4 stands. These observations can be interpreted using a combination of the relay-floristic and initial-floristic-composition models. Some early-successional species persist for a period less than 170 years, others invade throughout the period of the chronosequence and another group are part of the initial flora yet persist throughout. Kaikawaka was invading and regenerating in existing vegetation at the site (three older stands) so can participate in secondary succession. The main species in each stratum were ranked in order of decreasing cover using a semiquantitative estimate method or a point-estimate transect method. The species replacement sequence in the top layer of the four stands was described and follows a trend with time of physiognomic dominance by species with larger size at maturity. Life-form composition of the four stands (number and percentage of species per life-form category) was compared on a stratum basis through tables and lists. This analysis showed the three older stands were similar in life-form composition except for the presence of hemi-epiphytes exclusive to the oldest stand. The youngest stand was distinctive in life-form composition with a relatively large number of composite herb species present. Kaikawaka population structures were examined and compared using DBH size-class frequency distributions which also showed size-class mortality patterns. These distributions in the three older stands were dissimilar and indicated waves or pulses of kaikawaka regeneration have occurred rather than continuous recruitment. This intermittent mode of regeneration was linked to disturbance of various kinds over a range of scales rather than to climatic change. Total basal area, mean basal area per tree and stem density were also calculated for each population and found to increase with time in the case of the former two parameters. The cohort-senescence dieback theory and European forest life-cycle theory appear applicable to the study area and offer hypotheses to explain the widespread canopy dieback of kaikawaka found on Mt Taranaki. Whereas a full range of forest life-cycle phases were represented in the small area (<5ha) of the study site, elsewhere on a debris-fan only the dieback phase was evident over a much larger area (>200ha). This was interpreted as reflecting the dependence of disturbance regimes on topography, substrate and vegetation with different disturbance regimes operating in the two areas.