370 Chris J. Peterson and Walter P. Carson
interpretation of initial floristics, all or the vast
majority of species that will eventually become
prominent later in succession are present at the
outset. Species that dominate early do so because
of rapid growth rates or because a large supply of
propagules were present in the first year of suc-
cession, or both (Table 22.1). Late successional
species that were present at the outset become
dominant thereafter because they develop more
slowly and ultimately outlive the early occupants.
Thus the primary determinant of species turnover
is differences in life-history characteristics, espe-
cially growth rate and lifespan. The broad trends
in growth rate and lifespan observed in early tropi-
cal secondary succession (from early successional
grasses, herbs, and shrubs, to later successional
short-livedpioneertrees,andfinallytolonger-lived
pioneer trees) appear to be consistent with this
model (Guariguata and Ostertag 2001), but not
if late successional and primary tree species are
considered. Still, Finegan (1996) maintained that
alllong-livedpioneertrees,aswellasmanymature
forest trees, typically establish in the first few years
of succession, thus offerin gsupport for the ini-
tial floristics model in post-agricultural habitats.
Wilsonet al. (1992) pointed out that researchers
are not consistent in their use of the IFC model,
and have actually used two variants. The first,
described above, is closer to Egler’s original con-
cept and the second is called pre-emptive initial
floristics. In this latter variant, the species that col-
onize a new site prevent the establishment of later
successionalspecies,untilaftertheinitialcolonists
die. However, this expanded definition of IFC is
not consistent with Egler’s (1954) original model.
We argue that the defining characteristic of IFC is
the presence of many of the primary forest or late
successional species durin gthe first few years of
succession.
Nucleation
Yarranton and Morrison (1974) developed a
facilitation-based nucleation model for stressful
habitats that have low resource availability; these
habitats are typically inimical to colonization.This
model is a spatial extension of the Clementsian
relay floristics model, whereby early colonizers,
typically woody species, facilitate and promote
the colonization of additional woody species. It
was originally applied to succession on temper-
ate dunes. Under the nucleation model, a few
successful early colonists ameliorate harsh local
conditions (e.g., enhancing local nutrient and
water status via shadin gand litter), thereby facil-
itatin gthe establishment and survival of later
arrivals in the immediate proximity of the early
residents (Table 22.1). The early residents also
serve as foci for enhanced localized seed disper-
sal and accumulation (e.g., via serving as perches
for birds). Spatially, the model predicts that suc-
cession will proceed outward from some number
of recruitment foci of early colonists, which grow
and coalesce as additional late arrivals establish in
the vicinity of the early colonists. Thus the nucle-
ation model is spatially explicit. In its most general
form it makes no particular predictions about
future species composition though it could easily
be refined to do so based upon differential species
responses to stressful habitats and colonization
ability and dispersal mode. The nucleation model
differs substantially from all of the other models in
explicitly predicting high levels of neighborhood-
scale heterogeneity in vegetation structure and
composition durin gsecondary succession. Under
this model, as under relay floristics, the critical dif-
ferences amon gspecies are in their tolerance of
stressful conditions that occur in recently aban-
doned habitats. Here, pioneer species are defined
by their ability to both arrive in these habitats
and more importantly cope with stressful condi-
tions (e.g., low resource availability; Table 22.1).
Animal effects were given only passing mention as
dispersers that brin glate successional propa gules
into the “zone of influence” of early successional
nuclei.Gradient in timePickett (1976) and Drury and Nisbet (1973)
developed explanations for species turnover dur-
in gsuccession that focused on the contrastin g
life-history traits (particularly differential growth
rates, dispersal ability, lifespan, etc.) of species that
dominated early versus late durin gsuccession.
Both perspectives viewed secondary succession as
a gradient in time whereby temporally chang-
in gbiotic and abiotic conditions favored species