Austral Ecology (2003) 28, 666673

Subtropical montane tree litter decomposition: Links with

secondary forest types and species shade tolerance
DIEGO E. GURVICH,1* TOMS A. EASDALE2 AND NATALIA PREZ-HARGUINDEGUY1
1
Instituto Multidisciplinario de Biologa Vegetal (FCEFyN, UNCCONICET) CC 495, 5000,
Crdoba, Argentina (Email: dgurvich@com.uncor.edu) and 2Laboratorio de Investigaciones
Ecolgicas de las Yungas (UNT), B. Universitario, Tucumn, Argentina

Abstract Litter decomposition plays an important role in secondary forest recovery in the tropics. In this study
we assessed the decomposition rates of tree litter in species from different secondary forest types and with different
shade tolerances. The three secondary forest types analysed are related to the effects of different previous land use
intensities. The typical forest type (TYP) is related to low land use intensity, Alnus acuminata-dominated forest
(ALN) type is related to medium land-use intensity and Amomyrtella gili-dominated forest type (AMO) is related
to high land use intensity. The effect of shade tolerance was assessed using maximum height of each species as an
indicator of its light requirements. Associations with leaf functional traits such as specific leaf area (SLA), and tensile
strength (LTS) were also assessed. We found that leaves of species from the TYP forest type decompose faster than
those of the ALN and AMO forest types. These changes were consistent with differences in the SLA of the species,
which was higher in the TYP forest type than in the ALN and AMO forest types. SLA, LTS and decomposition
were not significantly correlated with tree maximum height. Our results show that the secondary forest types, which
are related to land use intensities prior to abandonment have an important influence on litter decomposition. This
implies potential long-term effects on soil properties and species composition.

Key words: decomposition, leaf tensile strength, litter quality, maximum height, secondary forests, shade tolerance,
specific leaf area.

INTRODUCTION highly degraded sites where organic matter is low

(Weaver et al. 1987) generally have a lower leaf quality
Anthropogenic secondary forests cover large and and lower decomposition rates, which may in turn
growing areas of land in the tropics (Brown & Lugo retard forest recovery (Hobbie 1992; van der Putten
1990). Soil processes are key factors that regulate forest 1997).
recovery following the cessation of logging, agriculture Light incidence is an important selective force on
or grazing (Zou et al. 1995). Among these processes, species strategies in forest ecosystems (Lawton 1984;
litter decomposition plays a very important role Kikuzawa 1995; Thomas 1996; Barker et al. 1997;
through the strong control that litter quality exerts on Poorter 2001). The marked light deficiency under
soil organic matter formation and nutrient availability forest canopies forces understorey species to have a
(Anderson & Flanagan 1989; Scott & Binkley 1997; high lamina area per unit leaf weight (later referred
Arunachalam et al. 1998; Thomas & Prescott 2000). as specific leaf area or SLA) in order to enhance their
It is to be expected that differences in secondary light interception (Smith et al. 1998; Meziane &
forest species composition, caused by different land Shipley 1999; Poorter 2001). Surprisingly, no study
uses and intensities, will produce differences in leaf has assessed the relationship between shade tolerance
quality and decomposition rates (de Mesquita et al. and litter decomposition although it is well known that
1998), which in turn affect the sites soil attributes SLA is linked to decomposition (Cornelissen &
(Garca-Montiel & Binkley 1998). Land use or Thompson 1997; Cornelissen et al. 1999). In this
intensity could then influence forest recovery not only context, it is expected that canopy species, which
through a direct effect on soil quality but also through usually have lower SLA values than understorey
long-term effects on soil processes brought about by species, will also have lower decomposition rates than
the decomposition properties of the species that understorey species.
colonize the site. In other words, trees growing in Three main secondary forest types are present in
montane forests around Los Toldos Valley (northwest
Argentina). These forest types differ markedly in tree
*Corresponding author. species composition and relative abundances (Table 1),
Accepted for publication June 2003. which relates to different land use intensities on the
 L I T T E R D E C O M P O S I T I O N F R O M S U B T R O P I C A L M O N TA N E T R E E S 667

three sites prior to abandonment (Easdale 1999). The METHODS

aim of the present study was to investigate tree litter
decomposition patterns in relation to secondary forest
types and species position along a light gradient. Our Study area
hypothesis was that both factors would affect
decomposition rates and that this variation would be Plant material was collected at Los Toldos Valley, Salta
associated with functional traits such as SLA and leaf Province, northwest Argentina (2217S, 6442W).
tensile strength (LTS). We expected species from the The area is located at 1500 m a.s.l. and the vegetation
secondary forest types that had experienced a higher of the area corresponds to the upper montane forest
land use intensity prior the abandonment would have belt of the Yungas biogeographical province, Amazo-
low SLA and high LTS, and thus low decomposition nian domain (Cabrera 1976). Mean annual temper-
rates. Additionally, we expected shorter and more ature is 11.7C and frosts are common during winters.
shade tolerant species to have higher SLA, lower LTS, Mean annual precipitation is 1300 mm, with dry
and therefore higher decomposition rates. winters ameliorated by dew (Hunzinger 1995).

Table 1. List of trees species, families, secondary forest types, litter mass loss after 15 weeks of decomposition, and relative
abundance of the species in each forest type

Species Family Litter mass loss (%) Relative abundance (%)

ALN
Alnus acuminata AMO, TYP. Betulaceae 19.02 54.01
Myrsine coriacea AMO, TYP. Myrsinaceae 8.83 16.33
AMO
Amomyrtella gili ALN,TYP. Myrtaceae 5.26 41.67
Clethra scabra ALN, TYP. Clethraceae 13.57 2.09
Ilex argentina TYP Aquifoliaceae 16.79 2.64
Maytenus cuezzoi TYP Celastraceae 17.13 0.32
Myrica pubescens ALN, TYP Myricaceae 7.22 9.70
Roupala meisneri TYP Proteaceae 8.50 0.49
Ternstroemia congestiflora Theaceae 16.71 1.16
TYP
Baccharis latifolia AMO, ALN Asteraceae 20.24 0.54
Berberis jobii ALN Berberidaceae 17.10 0.31
Celtis iguanaea Celtidaceae 53.91 0.09
Cinnamomum porphyrium AMO Lauraceae 14.31 1.87
Crinodendron tucumanum Elaeocarpaceae 35.84 0.10
Duranta serratifolia AMO Verbenaceae 12.61 3.12
Escallonia sp. AMO Saxifragaceae 21.19 0.95
Eugenia hyemalis Myrtaceae 15.46 0.20
Eupatorium saltense AMO, ALN Asteraceae 49.54 1.16
Fagara coco AMO, ALN Rutaceae 52.24 2.68
Myrcianthes mato AMO Myrtaceae 16.45 5.15
Parapiptadenia excelsa AMO, ALN Fabaceae 12.20 3.29
Podocarpus parlatorei AMO, ALN Podocarpaceae 16.59 30.14
Randia armata Rubiaceae 62.14 0.09
Rhamnus sphaerosperma Rhamnaceae 53.84 0.26
Sambucus nigra Caprifoliaceae 42.74 0.34
Schinus meyeri AMO, ALN Anacardiaceae 20.18 7.45
Scutia buxifolia Rhamnaceae 14.75 0.49
Sebastiania anisandra Euphorbiaceae 30.25 0.22
Solanum symmetricum AMO, ALN Solanaceae 17.94 0.08
Tabebuia lapacho ALN Bignoniaceae 26.22 0.51
Terminalia triflora AMO Combretaceae 25.17 1.58
Vassobia breviflora Solanaceae 50.73 1.48
Viburnum seemenii AMO, ALN Caprifoliaceae 26.87 3.08
Xylosma pubescens AMO Flacourtiaceae 13.79 1.78

ALN, Alnus acuminata-dominated secondary forest; AMO, Amomyrtella gili-dominated secondary forest; TYP, typical sec-
ondary forest. Superscripts after species names indicate the forest type where the species can also be found but with a very low
abundance.
668 D. E. GURVICH ET AL.

Species selection and leaf collection precipitation during the experiment were typical for
this area (de Fina 1992).
Thirty-four tree species (33 angiosperms and one This study was not intended to simulate the in situ
gymnosperm) from 29 plant families were selected decomposition of each species in the different forest
based on their abundance in the three secondary forest types but was rather designed to compare decompo-
types (Table 1) (Easdale 1999). These species together sition rates among species of different leaf qualities
comprise more than 90% of the vegetation coverage. and distribution, under a standard environment
From January to April 2000 we collected fresh leaf litter (Cornelissen 1996; Prez-Harguindeguy et al. 2000).
from at least four randomly selected mature individuals After 15 weeks incubation, adhering soil, soil fauna,
of each species. We also collected living leaves from six and other extraneous material were removed from the
individuals of each species for measurements of remaining leaf litter by swiftly rinsing with water. Litter
functional traits. These leaves were stored in poly- samples were dried for 48 h at 60C, then weighed.
ethylene bags at 4C for 48 h before processing. Decomposition rate was defined as the percentage of
Species nomenclature follows Zuloaga and Morrone weight loss after 15 weeks of burial.
(1996a,b).

Secondary forest types

Litter preparation
Forests around the Los Toldos Valley are arranged in a
We used the litter-bag technique to asses litter mosaic of primary and various secondary forest stands
decomposition. Litter samples were sorted and cleaned of different ages and species composition, which
following Cornelissen (1996). Six samples per species originated after activities by local people such as
of 1.0 0.1 g litter were oven-dried at 60C for 48 h, extensive grazing and slash and burn agriculture
and then sealed into tube-shaped nylon bags of 0.3 mm (Brown 1995; Reboratti 1996). The patches of second-
mesh size. This mesh size does not allow the inverte- ary forest can be grouped into three types on the basis
brate mesofauna to contribute to the decomposition of their relative species composition. These forest types
process, and may also affect some of the environmental are associated with different previous land use inten-
conditions experienced by the decomposing litter sities, but not with current land-use type (e.g. grazing
(Anderson 1975; Louisier & Parkinson 1976). These or agriculture). Typical secondary forest (TYP) is
effects may be important as far as relative weight losses associated with the lowest land use intensity; Alnus
of individual species are concerned. However, their acuminata-dominated forest (ALN) is associated with
effect is considered small compared with that produced low or medium land use intensity; and Ammomirtella
by bacteria, protozoa and fungi, and would not guilli-dominated forest (AMO) is associated with pre-
significantly alter the ranking of species with regard to vious high land-use intensities where superficial soil
weight loss (Cornelissen 1996; Prez-Harguindeguy layers have been almost completely lost (Easdale
et al. 2000), which was the objective of our study. 1999). The three forest types are discrete units in terms
of relative abundance of species, with some species
occurring in more than one unit (Table 1). Dendro-
Decomposition treatments chronological forest age estimations, which range from
14 to 81 years, indicate that there is no clear suc-
All litter samples were remoistened and buried simul- cessional relationship among the different secondary
taneously in a purpose-built decomposition bed at forest types (Easdale 1999).
Reserva Experimental Horco Molle (Universidad
Nacional de Tucumn, Yerba Buena, Argentina) on 21
May 2000. The top 20 cm of soil in a 4 m  4 m bed Shade tolerance
was removed the day before burial. Samples were
randomly placed and buried 10 cm below ground with To avoid possible noise produced by mixing species
mixed litter leaf-mould at different stages of decay and associated with different land use intensities, compari-
soil from an adjacent area in order to homogenize sons of decomposition rates among plants differing in
physical conditions, reduce the effect of the unpre- shade tolerance were performed only with the 25 TYP
dictable environment close to the surface, and avoid forest species. For these analyses, we used the maxi-
damage by birds and mammals (Prez-Harguindeguy mum height of each species as an indicator of species
et al. 2000). The decomposition bed was surrounded position in the light gradient (Thomas 1996) and
by a metal mesh to exclude small mammals and birds. assumed that shorter species had a higher shade
Six replicates per species were used. Samples were tolerance. Maximum height was obtained both from
buried for 15 weeks under natural autumnwinter previously published studies (Legname 1982; Killeen
temperature and rainfall conditions. Temperature and et al. 1993) and from direct field measurements.
 L I T T E R D E C O M P O S I T I O N F R O M S U B T R O P I C A L M O N TA N E T R E E S 669

Leaf attributes ments were expressed as the force (newtons), needed

to tear a leaf sample of known width (mm). The tensile
We also measured leaf attributes known for their associ- strength of living leaves was used to represent that of
ation with plant strategies: LTS and SLA (Bazzaz et al. leaves in the litter, because lignin-rich structures that
1987; Coley & Barone 1996; Daz & Cabido 1997). provide toughness remain intact during senescence
These attributes of living leaves influence the inter- (Prez-Harguindeguy et al. 2000).
specific variability in decomposition rates of the leaf The SLA was calculated as mm2 leaf area per mg dry
litter (Berendse 1994; Grime et al. 1996; Cornelissen & mass. Leaves were arranged between a white paper
Thompson 1997; Wardle et al. 1998). Six replicates per sheet and a sheet of glass and scanned. The average leaf
species taken from different individuals were used in area of each replicate was calculated by using Opti-
all cases. metrics Software. Compound leaves were treated as a
Measurements of LTS (resistance to tearing) were whole, without separating leaflets. Leaves were weighed
taken as an index of leaf physical quality by using the after being dried at 70C until a constant mass was
method of Hendry and Grime (1993). These measure- achieved.

Fig. 1. Relationship between (a) specific leaf area and (b) leaf tensile strength with litter mass loss in 34 tree species. r, Spearman
rank correlation coefficient. Podocarpus parlatorei was excluded from the leaf tensile strength analyses.

Fig. 2. Differences among tree species in three secondary forest types in (a) decomposition, and (b) specific leaf area. The box-
plot shows the distribution of the values according to the medians (central line), the mean (dots), the 25 and 75% quartiles (box)
and the ranges (whiskers). Different letters over the bars indicate significant differences among the three forest types (P  0.05;
KruskallWallis and multiple comparison test). ALN, Alnus acuminata-dominated secondary forest; AMO, Amomyrtella gili-
dominated secondary forest; TYP, typical secondary forest. Superscripts after species names indicate the forest type where the
species can also be found but with a very low abundance.
670 D. E. GURVICH ET AL.

Data analysis Of all species, we found significant correlations

between decomposition rate and both SLA and LTS
Because the data distribution was not normal (Fig. 1). Podocarpus parlatorei was excluded from the
(Hollander & Wolfe 1972), the Spearman rank correla- LTS analysis because of its very tough leaves (5.94
tion was used to test the relationships between percent- versus 1.60.3 N mm1 in the other species). SLA
age mass loss, maximum tree height and leaf attributes. was also strongly correlated with LTS (r = 0.69,
The KruskallWallis test (InfoStat 2002) and multiple P < 0.000).
comparisons test (Marascuilo & McSweeney 1977)
were used for comparisons among forest types.
Decomposition patterns and secondary forest
types
RESULTS
When species were grouped according to their distri-
bution in different forest types, we found two groups of
Decomposition patterns and leaf traits species characterized by different decomposition rates
(Fig. 2a). Species typical of AMO and ALN forests had
After incubation in litter bags for 15 weeks, the leaf the slowest decomposition rates, whereas species from
litter samples of the 34 sampled species showed a wide TYP forests had faster decomposition rates. This result
range of decomposition rates (Table 1). Randia armata, was consistent with the SLA pattern among the forests
an understorey tree, decomposed the fastest, with a (Fig. 2b), which revealed significantly lower values for
62% loss of mass, whereas Amomyrtella guilli, Myrica ALN and AMO forest types than for the TYP forest
pubescens, Roupala meisneri and Myrsine coriacea lost type. We did not find significant differences in LTS
the least mass, ranging from 5 to 9%. among the forest types.

Fig. 3. Relationships between (a) tree litter mass loss, (b) leaf tensile strength, and (c) specific leaf area with maximum tree
height. This analysis was carried out only for the 24 species from the TYP forest type. r, Spearman rank correlation coefficient.
Podocarpus parlatorei was excluded for the leaf tensile strength analyses.
 L I T T E R D E C O M P O S I T I O N F R O M S U B T R O P I C A L M O N TA N E T R E E S 671

Decomposition patterns and shade tolerance Decomposition and shade tolerance

Maximum tree height was not correlated with leaf Although Poorter (2001) found that SLA increased
decomposition rate, LTS or SLA (Fig. 3ac). Thus with shade tolerance, our study did not find a signifi-
there is no evidence from this analysis that shade cant correlation between maximum tree height and
tolerance affects any of these characteristics of leaves. SLA, nor could we find a significant correlation
between decomposition and maximum tree height.
Tree maximum height is probably not precise
enough to estimate shade tolerance and hence was not
DISCUSSION
linked with other leaf traits. An alternative measure-
ment to assess shade tolerance is asymptotic height
Decomposition and leaf traits (see Thomas 1996). Additionally, the measurement of
photosynthetic active radiation available at each species
Our results support claims that plant functional traits canopy level is a more quantitative predictor of shade
have an influence on decomposition rates (Grime tolerance.
1979; Chapin 1980; Coley 1980; Herms & Mattson To conclude, our results show that SLA and
1992). In agreement with the results of Cornelissen and decomposition are both high in the TYP forest type.
Thompson (1997) for temperate forests and grassland The secondary forest types, which can be associated
species, SLA was correlated with decomposition rate. with different land use intensities, would therefore have
LTS was also associated with decomposition, in a strong effect on litter decomposition through its
accordance with patterns found by Cornelissen et al. effect on species composition. These differences in
(1999) and Prez-Harguindeguy et al. (2000) across a litter decomposition could have long-term effects on
wide range of species and functional types. forest dynamics, affecting their recovery in terms of soil
properties and future species composition. Contrary to
our expectations, tree maximum height, an index of
Decomposition and the secondary forest types species shade tolerance, was not associated either with
leaf traits or decomposition.
When species were grouped according to the second-
ary forest types they were associated with, we found
differences in decomposition. AMO and ALN forest
types, which developed after high or medium land use ACKNOWLEDGEMENTS
intensities, were characterized by species with low
decomposition rates, whereas the TYP forest type, Research was supported by the Pro-Yungas Found-
which originated after abandonment following lower ation and CONICET. We are grateful to C. Martini,
land use intensities, mostly comprised species with S. Quinzio, A. Cosacov, G. Quero, M. Paolorossi,
relatively fast decomposition rates. L. Heil, and A. Sabella for field assistance. An early
Land use type and intensity are known to have version of this manuscript was substantially improved
long-term effects on soil characteristics and processes by the comments of A. Cingolani, S. Daz, M. Zak and
(Compton et al. 1998), and on vegetation dynamics P. Tecco. The suggestions of S. Halloy, A. Mark,
(Zou et al. 1995). In the short term, land use controls M. Bull and an anonymous referee improved the
tree establishment through an influence on the initial manuscript substantially. We thank D. Abal-Sols for
conditions of the site (Miller 1984). In our study, drawing the figures.
species such as Amomyrtella guilli and Myrica pubescens,
typical of the more degraded stands (ALN and AMO
forests), would be likely to establish under more limit-
REFERENCES
ing conditions through their slow growth and stress
tolerance. More mesic, fast-growing species, such as
Anderson J. M. (1975) Succession, diversity and trophic relation-
Randia armata and Celtis iguanaea, are more frequent ships of some soil animals in decomposing leaf litter. J. Anim.
in the less degraded sites (TYP forests). ALN and Ecol. 44, 47595.
AMO forests are characterized by species with slow Anderson J. M. & Flanagan P. W. (1989) Biological processes
decomposition rates, which would in turn lead to regulating organic matter dynamics in tropical soils. In:
slower nutrient release and availability in the soil. Dynamics of Soil Organic Matter in Tropical Ecosystems (eds
Therefore, the initial differences in tree composition D. C. Coleman, J. M. Oades & G. Uehara) pp. 97123.
University of Hawaii Press, Honolulu.
between the more degraded and the less degraded sites
Arunachalam A., Maithani K., Pandey H. N. & Tripathi R. S.
could lead to long-term effects on forest dynamics (1998) Leaf litter decomposition and nutrient mineralization
mediated by treesoil feedback systems (Hobbie 1992; patterns in regrowing stands of a humid subtropical forest
Compton et al. 1998). after tree cutting. For. Ecol. Manage. 109, 15161.
672 D. E. GURVICH ET AL.

Barker M. G., Press M. C. & Brown N. D. (1997) Photosynthetic Herms D. A. & Mattson W. J. (1992) The dilemma of plants: To
characteristics of dipterocarp seedlings in three tropical rain grow or defend. Quart. Rev. Biol. 67, 283325.
forest light environments: A basis for niche partitioning? Hobbie S. E. (1992) Effects of plant species on nutrient cycling.
Oecologia 112, 45363. Trends Ecol. Evol. 7, 3369.
Bazzaz F. A., Chiariello N. R., Coley P. D. & Pitelka L. F. (1987) Hollander M. & Wolfe D. A. (1972) Non-Parametric Statistical
Allocating resources to reproduction and defense. Bioscience Methods. John Wiley and Sons, New York.
37, 5867. Hunzinger H. (1995) La precipitacin horizontal: Su impor-
Berendse F. (1994) Litter decomposability: A neglected com- tancia para el bosque y a nivel de cuencas en la Sierra de San
ponent of plant fitness. J. Ecol. 82, 18790. Javier, Tucumn, Argentina. In: Investigacin, Conservacin y
Brown A. D. (1995) Las selvas de montaa del noroeste de Desarrollo en Selvas Subtropicales de Montaa (eds A. D.
Argentina: Problemas ambientales e importancia de su con- Brown & H. R. Grau) pp. 538. Proyecto de Desarrollo
servacin. In. Investigacin, Conservacin y Desarrollo en Agroforestal LIEY, Tucumn.
Selvas Subtropicales de Montaa (eds A. D. Brown & H. R. InfoStat (2002) Infostat Versin 1.1. Grupo InfoStat, FCA,
Grau) pp. 918. Proyecto de Desarrollo Agroforestal Universidad Nacional de Crdoba, Crdoba.
LIEY, Tucumn. Kikuzawa K. (1995) Leaf phenology as an optimal strategy for
Brown S. & Lugo A. E. (1990) Tropical secondary forests. carbon gain in plants. Can. J. Bot. 73, 15863.
J. Trop. Ecol. 6, 132. Killeen T. J., Garca E. & Beck S. G. (1993) Gua de rboles de
Cabrera A. L. (1976) Regiones Fitogeogrficas Argentinas. Editorial Bolivia. Herbario Nacional de Bolivia & Missouri Botanical
ACME SACI, Buenos Aires. Garden, Saint Louis.
Chapin F. S. III (1980) The mineral nutrition of wild plants. Lawton R. O. (1984) Ecological constraints on wood density in
Annu. Rev. Ecol. Syst. 2, 23360. a tropical montane rain forest. Am. J. Bot. 71, 2617.
Coley P. D. (1980) Effects of leaf age and plant life history Legname P. R. (1982) rboles Indgenas del Noroeste Argentino
patterns on herbivory. Nature 283, 5456. (Salta, Jujuy, Santiago del Estero y Catamarca). Opera
Coley P. D. & Barone J. A. (1996) Herbivory and plant Lilloana XXIV, Tucumn.
defenses in tropical forests. Ann. Rev. Ecol. Syst. 27, Louisier J. D. & Parkinson D. (1976) Litter decomposition in a
30535. cool temperate deciduous forest. Can. J. Bot. 54, 41936.
Compton J. E., Boone R. D., Motzkin G. & Foster D. R. (1998) Marascuilo L. & McSweeney M. (1977) Nonparametric and
Soil carbon and nitrogen in a pineoak sand plain in central Distribution-Free Methods for the Social Sciences. Wadsworth
Massachusetts. Oecologia 116, 53642. Publishing, City.
Cornelissen J. H. C. (1996) An experimental comparison of leaf Meziane D. & Shipley B. (1999) Interacting determinants of
decomposition rates in a wide range of temperate plant specific leaf area in 22 herbaceous species: Effects of
species and types. J. Ecol. 84, 57382. irradiance and nutrient availability. Plant Cell Environ. 22,
Cornelissen J. H. C., Prez-Harguindeguy N., Daz S. et al. 44759.
(1999) Leaf structure and defence control litter decompo- Miller H. G. (1984) Dynamics of nutrient cycling in plantation
sition rate across species and life forms in regional floras of ecosystems. In: Nutrition of Plantation Forests (eds G. D.
two continents. New Phytol. 143, 191200. Bowen & E. K. S. Nambiar) pp. 5378. Academic Press,
Cornelissen J. H. C. & Thompson K. (1997) Functional leaf New York.
attributes predict litter decomposition rate in herbaceous Prez-Harguindeguy N., Daz S., Cornelissen J. H. C., Vendra-
plants. New Phytol. 135, 10914. mini F., Cabido M. & Castellanos A. (2000) Chemistry and
de Fina A. L. (1992) Aptitud Agroclimtica de la Repblica toughness predict leaf litter decomposition rates over a wide
Argentina. Academia Nacional de Agronoma y Veterinaria, spectrum of functional types and taxa in central Argentina.
Buenos Aires. Plant Soil 218, 2130.
de Mesquita R. C. G., Workman S. W. & Neely C. L. (1998) Poorter L. (2001) Light-dependent changes in biomass allocation
Slow litter decomposition in a Cecropia-dominated and their importance for growth of rain forest tree species.
secondary forest of Central Amazonia. Soil Biol. Biochem. Funct. Ecol. 15, 11323.
30, 16775. Reboratti C. (1996) Sociedad, Ambiente y Desarrollo en la Alta
Daz S. & Cabido M. (1997) Plant functional types and eco- Cuenca del Ro Bermejo. Institute de Geografa, Facultad de
system function in response to global change: A multiscale Filosofa y Letras, UBA, Buenos Aires.
approach. J. Veg. Sci. 8, 46374. Scott N. A. & Binkley D. (1997) Foliage litter quality and annual
Easdale T. (1999) Relacin de Disturbios y Factores Ambientales con net N mineralization: Comparison across North American
la Diversidad, Composicin y Estructura de Comunidades forest sites. Oecologia 111, 1519.
Leosas en el Valle de Los Toldos,Yungas Argentinas. Final report Smith W. K., Bell D. T. & Shepherd K. A. (1998) Associations
FA96, World Wildlife Fund. between leaf structure, orientation, and sunlight exposure
Garca-Montiel D. C. & Binkley D. (1998) Effect of Eucalyptus in five Western Australian communities. Am. J. Bot. 85,
saligna and Albizia falcataria on soil processes and nitrogen 5663.
supply in Hawaii. Oecologia 113, 54756. Thomas S. C. (1996) Asymptotic height as a predictor of growth
Grime J. P. (1979) Plant Strategies and Vegetation Processes. John and allometric characteristics in Malaysian rain forest trees.
Wiley and Sons, Chichester. Am. J. Bot. 83, 55666.
Grime J. P., Cornelissen J. H. C., Thompson K. & Hodgson J. G. Thomas K. D. & Prescott C. E. (2000) Nitrogen availability in
(1996) Evidence of a causal connection between anti- forest floors of three species on the same site: The role of
herbivore defence and the decomposition rate of leaves. litter quality. Can. J. For. Res. 30, 1698706.
Oikos 77, 48994. van der Putten W. (1997) Plant-soil feedback as a selective force.
Hendry G. A. F. & Grime J. P. (1993) Methods in Comparative Trends Ecol. Evol. 12, 16970.
Plant Ecology: A Laboratory Manual. Chapman & Hall, Wardle D. A., Barker G. M., Bonner K. I. & Nicholson K. S.
London. (1998) Can comparative approaches based on plant eco-
 L I T T E R D E C O M P O S I T I O N F R O M S U B T R O P I C A L M O N TA N E T R E E S 673

physiological traits predict the nature of biotic interactions sition and litter dynamics of a tropical rain forest in Puerto
and individual plant species effects in ecosystems? J. Ecol. Rico. For. Ecol. Manage. 78, 14757.
86, 40520. Zuloaga F. O. & Morrone O. (1996a) Catlogo de las Plantas
Weaver P. L., Birdsey R. A. & Lugo A. E. (1987) Soil organic Vasculares de la Repblica Argentina. Monogr. Syst Bot. Mo.
matter in secondary forests of Puerto Rico. Biotropica 19, Bot. Gard. 60, 1323.
1723. Zuloaga F. O. & Morrone O. (1996b) Catlogo de las Plantas
Zou X., Zucca C. P., Waide R. B. & McDowell W. H. (1995) Vasculares de la Repblica Argentina. Monogr. Syst Bot. Mo.
Long-term influence of deforestation on tree species compo- Bot. Gard. 74, 11269.