The vegetation of Campbell Island and its offshore islets was sampled
quantitatively at 140 sites. Data from the 134 sites with more than on
e vascular plant species were subjected to multivariate analysis. Out
of a total of 140 indigenous and widespread adventive species known fr
om the island group, 124 vascular species were recorded; 85 non-vascul
ar cryptogams or species aggregates play a major role in the vegetatio
n. Up to 19 factors of the physical environment were recorded or deriv
ed for each site. Agglomerative cluster analysis of the vegetation dat
a was used to identify 21 plant communities. These (together with cryp
togam associations) include: maritime crusts, turfs, megaherbfields, t
ussock grasslands, and shrublands; mid-elevation swamps, flushes, bogs
, tussock grasslands, shrublands, dwarf forests, and induced meadows;
and upland tundra-like tussock grasslands, tall and short turf-herbfie
lds, bogs, flushes, rock-ledge herbfields, and fellfields. Axis 1 of t
he DCA ordination is largely a soil gradient related to the eutrophyin
g impact of marine spray, sea mammals and birds, and nutrient flushing
. Axis 2 is an altitudinal (or thermal) gradient. Axis 3 is related to
soil reaction and to different kinds of animal influence on vegetatio
n stature and species richness, and Axis 4 also appears to have fertil
ity and animal associations. Autecological interpretation of the data
demonstrates clear niche segregation of congeneric species and ;specie
s with similar growth forms. The notable megaherbs and giant tussocks
may be an adaptation to harvesting nutrients from the aerosol precipit
ate. Heat harvesting in the cool, cloudy, wet, and windy climate may a
lso be implicated. The history of farming and natural disturbances has
resulted in a complex mosaic of vegetation-soil systems of varying ma
turity. Their putative dynamic interrelationships are depicted in term
s of impacts of burning, grazing, marine animals and climate change an
d subsequent recovery or primary and secondary succession.