Spatial Variation in Tree Species Composition 25finer scale, the one of most interest for land-use
planners and conservationists, it is likely that the
simple correlative model predicted by Francis and
Currie (2003) will fail.
Niche-assembly and dispersal-based theories
highlight different constraints on the distribu-
tion of plant species. The niche-assembly the-
ory emphasizes the physiological constraints on
plants, constraints that cannot be ignored in the
study of plant distributional patterns. Many of
the most common tree species of the Amazon
forest never occur in swampy areas, or in areas
with intense dry seasons. Species in the Melas-
tomataceae and Vochysiaceae are tolerant to soils
rich in free aluminum (Al^3 +), while most other
tropical tree families are not (Jansenet al. 2002);
some species are capable of ectomycorrhizal asso-
ciations which puts them at an advantage in
phosphorus-limited environments. Moreover, it is
likely that distance will not limit the spread of
plants with small propagules; the minute spores
of ferns are known to be dispersed great distances,
thereby makin gthe fern community effectively
panmictic on large scales (Wolf et al. 2001).
On the other hand, history should also be an
important predictor of the distribution of many
plant species. The Central American and South
American tree floras remain clearly differentiated
3 million years after the closure of the Panama
land-bridge (Gentry 1982, Dicket al. 2005). A
look at almost any distribution map of congeneric
species in the Flora Neotropica shows that species
with similar functional features and broad envi-
ronmental niches occur in clearly distinct and
restricted areas (consider, e.g., the case of the
neotropical palms in Hendersonet al. 1995),
despite the fact that they have had millions of
years to spread across the continent. These few
examples show that spatial species mixin gat
the regional scale is not a particularly efficient
mechanism.Thus, niche-assembly theories can be
criticized because they tend to minimize the role of
dispersal limitation, while the dispersal-assembly
theories can be criticized for ignoring species’
physiological peculiarities.
The debate over the validity of the neutral
theoryisnowbehindus.Thefundamentalingredi-
ents of the neutral theory, namely the crucial role
of unpredictability and of dispersal limitation – in
short, of historical factors – can hardly be debated.
The signature of this effect, the space dependence
of floristic diversity, has been demonstrated in var-
ious studies in different sites and with different
plant groups. The fundamental premise of classi-
calcommunityecology,theinfluenceof theabiotic
environment on floristic turnover, has also been
confirmed in all these studies. Future unifying
theories should take into account both historical
fluctuations and environmental determinism. Yet
much remains to be done to achieve this goal,
simply because we still know so little about how
plants respond to their abiotic environment. It is
known that some chemical elements of the soil are
necessary while others are toxic, and that trade-
offsexistamongphysiologicalfunctions,including
nutrient use efficiency (Reichet al. 1999), but the
balance for each species, or across species lineages,
remains poorly documented. It may still be useful
to explore the two theories independently, as long
as it is understood that they represent only part
of the larger picture (Ricklefs 2004). One must
not forget that most theories in ecology have an
illustrative purpose. In May’s (1973) words, eco-
logical models “are at best caricatures of reality,
and thus have both the truth and the falsity of
caricatures.”ACKNOWLEDGMENTS
I thank Guillem Chust, Jim Dalling, Nicolas
Mouquet, and Oliver Phillips for useful correspon-
dence, and Christophe Thébaud, Stefan Schnitzer,
Kalle Ruokolainen, and an anonymous reviewer
for pointin gout a number of deficiencies in previ-
ous versions. Finally, I thank S Schnitzer for offer-
in gan opportunity to update this work in January
2008, some 2 1/2 yrs after the manuscript was
first written.REFERENCES
Ashton, P.S. (1964) Ecological studies in the mixed dipte-
rocarp forests of Brunei State.Oxford Forestry Memoirs
25, 1–75.
Ashton, P.S. (1976) Mixed dipterocarp forest and its
variation with habitat in the Malayan lowlands: a
re-evaluation at Pasoh.Malayan Forester39, 56–72.