Tropical Forest Ecology: Sterile or Virgin for Theoreticians? 133mature forest tree species was driven by a
trade-off between colonizing ability and compet-
itive superiority. True, to survive, pioneer species
must convey seeds to gaps soon after they open, or
have them waiting in the soil. Most gaps, however,
are already occupied by seedlings and saplings
of superior competitors: pioneers must outgrow
them and reproduce before being crowded out.
Therefore, the principal factor allowing pioneers
to coexist with superior competitors is not a
competition–colonization trade-off, but the trade-
off between growing fast in bright light and
surviving in shad e(Brokaw 1987, King 1994,
Kitajima 1994). This trade-off affects all aspects
of tree life. More light-demanding species tend to
have more of their seedlings in gaps, higher mor-
tality rates in their seedlings and saplings, lower
sapling wood density, higher growth rates, and
fewer saplings per reproductive adult than more
shade-tolerant counterparts (Wrightet al. 2003).
Sun leaves have high photosynthetic capacity,
which inflates respiratory costs but permits abun-
dant photosynthesis in bright light; shade leaves
with littl eopportunity for rapid photosynth esis
reduce respiratory costs by maintaining low pho-
tosynthetic capacity.Trees with multilayer crowns
designed to spread light over as much leaf surface
as possible grow faster in bright light, whereas
“monolayer” crowns designed to concentrate as
much light as possible on a single layer of leaves
allows herbs and saplings to survive better in
shade (Horn 1971). Lower investment in anti-
herbivore defense allows pioneers,Cecropia,to
maintain a dry matter production per unit leaf
area double that of shade-tolerant competitors,
but they need more light than shade-tolerants to
be able to replace their short-lived leaves before
herbivores eat them (King 1994). Sacrificing dura-
bility by making wood with density only 25% of
th efor est-wid eav erag etranslat esCecropia’s dry
matter production into height growth no shade-
tolerant competitor can match (King 1994).
Obstacles to theory: pest pressure and tree
diversity
Effects on tree diversity of the trade-offs plants
face in resisting different pests and pathogens (and
the corresponding trade-offs pests and pathogensface in attacking different plants) are of central
interest to theorists. If “the jack of all trades
is master of none” (MacArthur 1961), special-
ist pests and pathogens should inflict the most
damage. Even in the tropics, generalist cater-
pillars of some species, such asHylesia lineata
(Saturniidae) can defoliate whole trees (Janzen
1984). Nonetheless, tropical plants suffer most
from species- or genus-specific pests (Janzen 1988,
Novotnyet al. 2002). At all latitudes, diverse
forests and tree plantations suffer less from herbi-
vores, especially relatively specialized herbivores,
than single-species stands. Moreover, a particular
species is less damaged by specialist pests in plots
where it is rarer (Jactel and Brockerhoff 2007).
P est pr essur eis mor eint ens ein th etropics, wh er e
no winter knocks back pest populations (Janzen
1970), and where caterpillars, the most dam-
aging insect pests, are more specialized (Scriber
1973, Dyeret al. 2007). Young tropical leaves
are therefore far more rapidly eaten, despite being
far more poisonous, than temperate-zone dicot
counterparts. Gillett (1962), Janzen (1970) and
Connell (1971) therefore proposed that there are
so many kinds of tropical plants because spe-
cialized pests keep each species rare enough to
make room for many others. This idea unifies a
great variety of data. But is it true? This ques-
tion still arouses vigorous argument (Leighet al.
2004, Leigh 2007). Can mathematical theory
help resolve it?
A mathematical theory of pest pressure and
tree diversity must be based on two propositions:
1 If a tree species is rare enough, the abundance
of consumers specialist upon it declines, whereas
if these consumers are abundant enough, the
density of their host trees declines.
2 A forest’s trees are also limited by light and by
suitabl espac ein which to grow.
To see how pest pressure might influence tree
diversity, consider a community ofntree species,
each with its own species of specialized pest. Let
the biomass per unit area at timetof tree species
iand its specialist consumer beNi(t)andCi(t),
respectively. Let the total density of treesN(t)
beN 1 (t)+N 2 (t)+N 3 (t)+···+Nn(t), where
1 ≤i≤n.Letrib eth ep er capita rat eof incr eas e
of tree speciesiwhen all trees are rare, let con-
sumer speciesidiminish th epopulation growth