phyte colonization than does bare bark by "roughen-
ing" the surface of the stem and providing more loca-
tions for colonization by epiphyte propagules and
retention of water and nutrient. As the epiphyte com-
munity develops, the host tree may derive more
leachates from the epiphytes and put forth more AAR,
which would in turn create a better substrate for fur-
ther epiphyte colonization.
Acknowledgments I thank Israel Mendez for his help
in the field. I acknowledge the Tropical Science Cen-
ter and the MCFP. Fieldwork was completed while I
was a graduate student at the College of Forest Re-
sources, University of Washington, and received sup-
port from the Minority Affairs Office and the Bloedel
Fellowship. I thank Charles Grier, members of my
graduate committee, and David Benzing. This was
funded by a Man and the Biosphere research grant to
C. Grier. Data analysis was partially supported with
a grant from the National Science Foundation (DEB
90-18006), administered through The Evergreen State
College.FINE UTTER DYNAMICS WITHIN THE TREE CANOPY
OF A TROPICAL CLOUD FOREST
•Nalini M. Nadkarni&TeriJ.Matelsonotential sources of nutrients for epiphytes in-
clude an array of both autochthonous (within-
ecosystem) and allochthonous (outside of eco-
system) sources (Table 9.24). It is important to distin-
guish between these two source types in forests where
nutrient availability may limit productivity of the eco-
system (Grubb 1977, Vitousek 1984). If epiphytes were
to obtain all of their nutrients from autochthonous
sources, they would simply be diverting nutrients
from the tree-ground flux pathway; they would not
increase the total pool but merely change the form or
compartment in which nutrients were stored. Alter-
natively, if they were to obtain nutrients from allo-
chthonous sources, they would potentially increase
the total nutrient input to the ecosystem in addition
to altering the form and location of these nutrients.
Because epiphytic plants have no direct vascular
connection to the bank of nutrients in the forest floor,
they must rely on morphological and physiological
attributes such as litter-impounding pools, foliar tri-
chomes, insectivory, myrmecochory, and poikilohy-
dric foliage to acquire and conserve nutrients in an
environment that may deliver nutrients only sporadi-
cally and in dilute concentrations (Benzing and See-
man 1978, Nadkarni 1981). Some specialized epi-Table 9.24. Potential sources of nutrient input to
epiphyte communities in forest ecosystems.
Autochthonous sources Allochthonous sourcesphytes gain nutrients exclusively from precipitation
and dry deposition (Benzing 1983), but many epi-
phytes are unable to obtain nutrients from atmo-
spheric sources. These epiphytes, such as woody
shrubs in the Ericaceae, have well-developed root
systems that penetrate the dead organic matter and
"crown humus" (Jenik 1973) that accumulates as mats
on upper branch surfaces. We quantified the nutrient
dynamics of one potential autochthonous source of
nutrients to the epiphyte community, the abscised
plant litter that is intercepted within the upper tree
canopy by inner branches and their epiphytes, in the
leeward cloud forest area of the MCFP (Nadkarni and
Matelson 1991).
Fine litter dynamics within the canopy may be criti-
cal for epiphyte productivity and may differ from lit-
ter dynamics on the forest floor for three reasons. First,
canopy litter may be ephemeral, as it can be removed
from branches by within-canopy disturbances such as
wind, rain, and arboreal animal activities. In contrast
to fallen leaves on the forest floor, which can shift their
position on the ground with only minor consequences
for plants rooted in soil (Orndorff and Lang 1981), leaf
movement in the canopy may remove a potentially
substantial contribution to epiphyte nutrition. Second,
leaf litter in the canopy may be deposited in smaller
amounts than is leaf litter in the forest floor due to lack
of input from subcanopy and understory vegetation.
Third, decomposition rates of litter deposited and re-
tained within the canopy may differ from litter on the
forest floor due to microclimate and substrate differ-
ences between the canopy and forest floor, as well as
differences in community structure and density of
macroinvertebrate detritivores and microbial decom-
posers (Nadkarni and Longino 1990).341 Ecosystem Ecology and Forest DynamicsSoil-rooted phytomass Atmospheric
Intercepted litterfall wet deposition
Bark decomposition dry deposition
Leachate of live foliage gaseous input (including
Animal defecation and death nitrogen-fixation)P