10/1/2004

The Botanical Review

By Randall W. Myster

   I. Abstract/Resumen 

  II. Introduction 

 III. Clearing, Farming, and Abandonment Practices in the Neotropics 

  IV. The Post-Agricultural Environment and Disturbance Regime 

   V. Patterns of Tree Invasion, Establishment, and Growth 

  VI. Processes Affecting Tree Patterns 

      A. Crop Signatures 

      B. Seed Processes 

      C. Seedling Processes 

      D. Tree Processes 

 VII. Synthesis 

VIII. Restoration, Management, and Conclusions 

  IX. Acknowledgments 

   X. Literature Cited 

I. Abstract

Because clearing forest for agriculture is the most common disturbance in the Neotropics, studies of post-agricultural recovery need to be conducted, both to understand rain-forest function and structure and to address important social issues such as deforestation, restoration, and sustainability. To assist that effort, the clearing, planting, cultivation, harvesting, and abandonment practices for common crops in the Neotropics, the post-agricultural environment and disturbance regime, and the recovery mechanisms are reviewed for their influence on the succession that follows abandonment. An important focus is on the four historical effects, or signatures, of crops: crop persistence, crop root exudate persistence, persistence of associated species, and indirect effects of herbicide and fertilization regimes. In addition, the effects of cattle introduced after cropping, such as hummock creation, soil compaction, and dung deposition, are discussed. Further, permanent plot tree data from Puerto Rico and Ecuador are summarized to guide an understanding of how trees invade and replace themselves. Finally, tabular summaries of completed Neotropical seed/seedling/sapling field experiments are used, both to examine what is known about the mechanisms of old-field succession in the Neotropics and to suggest research directions and successful restoration strategies. A species-specific and field type-specific investigation of tree-replacement mechanisms in the future is suggested, leading to replacement modeling using data from permanent plots.

II. Introduction

The study of post-agricultural (i.e., old-field) succession has helped ecologists understand the processes that structure plant communities (Strong et al., 1984) and the role of history and initial condition in community development (Myster & Pickett, 1990b, 1994). In addition, old-field research has led to important ecological theories, such as the initial floristic composition hypothesis (Egler, 1954; Paine, 1977) and the resource ratio hypothesis (Tilman, 1988). Using permanent plots established at abandonment as the backbone of old-field research (e.g., the Buell-Small succession study in New Jersey: Buell et al., 1971; Myster & Pickett, 1990a), scientists have been successful in finding both the pattern of response over time after abandonment from crops (e.g., plant cover: Myster & Pickett, 1990b) and many of the mechanisms determining that pattern (see Myster, 1993; Bazzaz, 1996).

Unfortunately, a disproportionate number of these old-field and pasture studies (see Castro et al., 1986; Hill et al., 1992) have taken place in the temperate-forest regions of the United States (e.g., Oosting, 1942; Bazzaz, 1968; Buell et al., 1971; Pickett, 1982; Tilman, 1988; Myster, 1993) and Europe (Grime, 1977; Miles, 1979), even though recovery of agricultural areas is critical to ecosystems all over the world (Borhidi, 1988; Matson et al., 1997). This is particularly the case in the Neotropics, where the conversion of tropical forest to agriculture and then, usually, to pasture is the most common kind of disturbance (Buschbacher, 1986; Uhl et al., 1988; Fearnside, 1993; Thomlinson et al., 1996). Not surprisingly, recovery of these areas after agriculture (Donfack et al., 1995; Fernandes & Sanford, 1995; Quintana-Ascencio et al., 1996) addresses important Neotropical issues, such as: deforestation and forest regeneration (Lanty, 1981; Grainger, 1988; Brown & Lugo, 1990a; Myers, 1991; Singh, 1993; Skole & Tucker, 1993); forest ecosystem restoration (Brown & Lugo, 1990b; Lugo, 1991); sustainability of agriculture (Brown & Lugo, 1990a; Serrao & Toledo, 1990); maintenance of biodiversity (Brown & Lugo, 1990a; Aide & Cavelier, 1994); and impacts of global climate change on forest dynamics (Padoch & Vayda, 1983; Hobble, 1992; Mabberley, 1992; Keller et al., 1993; Keller & Reiners, 1994). In addition, these areas may help create a buffer between primary forest and more intensely human-impacted areas (Brown & Lugo, 1990a).

Studies of post-agricultural recovery may have been neglected because of an implicit assumption that all disturbances have the same effect--opening canopies in the forest--so only studies of tree-fall gap dynamics (Ewel, 1980; Gomez-Pompa & Vazquez-Yanes, 1981; Denslow, 1987) or "secondary" forests (Guariguata & Ostertag, 2001) are required. However, research in Neotropical forest-succession areas other than treefall gaps, such as landslides (Myster & Fernandez, 1995; Myster, 1997; Myster & Walker, 1997), abandoned plantations (Lugo, 1991; Myster, 2003a), and old fields / pastures (Purata, 1986; Uhl et al., 1988; Donfack et al., 1995; Fernandez & Sanford, 1995; Nepstad et al., 1996, Quintana-Ascencio et al., 1996; Rivera et al., 2000; Chinea, 2002; Ferguson et al., 2003; Myster, 2003a, 2003b, 2003c), has shown that both the abiotic environment and the biotic recovery mechanisms vary widely between disturbances. In particular, post-agricultural areas are likely to be starkly different from treefall gaps because they are larger in area than are treefall gaps and, therefore, have a greater effect on dispersing animals (Janzen, 1990), they remove most of the intact aboveground vegetation, they have exotic species grown in them at high density for an extended period of time (the duration of cultivation), they may undergo burning and recurrent weeding, and they can have lasting historical effects of past crop use after abandonment (crop signature: Myster & Pickett, 1990b, 1994).

For scientific, social, political, and ecological reasons, then, research is needed in Neotropical old fields. To help guide this research effort, important issues concerning post-agricultural recovery in the Neotropics, such as common clearing, planting, cultivation, harvesting, and abandonment practices, are examined, along with the post-agricultural environment and disturbance regime and the successional mechanisms that control tree replacements. This review goes farther than more general secondary-forest succession reviews that have been published in the past decade (e.g., Finegan, 1996; Holl & Kapelle, 1999; Aide, 2000; Guariguata & Ostertag, 2001) by focusing directly on Neotropical old fields and including data on successional pattern and mechanisms, working definitions of old-field function and structure, comparisons with disturbances of both greater and lesser severity for better synthesis, no false dichotomies such as primary vs. secondary forest, natural vs. manmade, pioneer vs. climax species, new growth vs. old growth, and a comprehensive framework of tree-tree replacement mechanisms that all old-field studies can feed into. Plantations will not be discussed directly; good reviews of their dynamics include Lugo (1991) and Guariguata et al. (1995).

The focus is on the common crops of maize (Zea mays), sugarcane (Saccharum officinarum), banana (Musa spp.), coffee (Coffea spp.) cassava (Manihot esculenta), beans (Phaseolus spp.), and rice (Oryza spp.) and on pastures usually established after cropping. How these crops and other aspects of agricultural practices may affect the tree succession that follows abandonment is a major focus. Furthermore, I argue that pastures are a kind of old-field succession because they are usually converted from depleted agricultural fields when cows are allowed to enter. Finally, extensive data tables from completed mechanistic studies show key successional mechanisms and important variation in how they produce tree replacements.

III. Clearing, Farming, and Abandonment Practices in the Neotropics

Archaeologists have evidence that maize was grown in the Amazon Basin some 6000 years ago. All crops were first propagated vegetatively and then by seed (Mabberley, 1992; Gomez-Pompa & Bainbridge, 1997). Since the invasion of the Spanish and the subsequent genocide of the native "Indians," additional techniques such as the use of plowing animals, exotic cultivars, and mechanization have been practiced.

Currently, two general forest-clearing techniques are found in the Neotropics, both of which increase light quantity and contact of rainfall with the soil but eventually decrease soil fertility. The first is mechanical clearing (e.g., use of a bulldozer), with the removal of the woody biomass. It is the most severe method practiced in the Neotropics and often leads to changes in soil physical properties, but a similar manual clearing technique can reduce these effects (Purata, 1986). The second forest-clearing technique is shifting agriculture (slash and burn: Uhl, 1987; Tinker et al., 1996). It is a common clearing method that existed long before mechanical clearing, but it is being replaced by more technologically advanced farming in many areas. This method usually involves the cutting down to the ground, or slashing, of the natural vegetation and then burning it, a rotation of fields rather than crops, and long fallow periods to help maintain soil fertility by allowing vegetation to regenerate (Grigg, 1978). Although burning can produce a short pulse of some nutrients, like phosphorus, it also volatilizes both organic matter and nitrogen and raises soil temperature. In the long term this kind of farming produces nutrient-poor soil and reduces the soil seed bank. Consequences of these clearing and plowing methods include carbon loss, combination of the soil organic and mineral horizons, and more soil aeration influencing microbial activities such as nutrient release and mineralization (Bazzaz, 1996).

Table I highlights farming practices for common crops after clearing, showing the great variety of both techniques and use of natural processes to maximize production. In addition to varying fallow and crop periods (shorter fallow periods are often used when the cost of land clearing is low), farming practices can include growing a mix of crops and using a variety of both traditional "low-input" (e.g., seed planting with digging sticks called "dibbles") and modem "high-input" (e.g., mechanical clearing with application of fertilizer and water) planting methods (Scatena et al., 1996). Intercropping is another practice used for some crops; in this method additional crops are planted between rows of existing crop. The decision as to when to rotate fields and for how long to leave them fallow includes consideration of weed (Qiang & Hu, 1999) and insect numbers, levels of both soil fertility and productivity, labor availability, and dietary needs (Staver, 1989).

Further, there is a common synergism among agricultural uses in the Neotropics because most often an area is cropped first and then, with declining productivity, becomes pasture as livestock grazing leads to domination by grasses (Scatena et al., 1996). Grass species can be either seeded into the pasture (e.g., Setaria sphacelata in Ecuador: Myster & Sarmiento, 1998) or colonize naturally (Panicum maximum, Brachiaria humidicola, Paspalum spp., and Hyparrhenia rufa in Amazonia: Nepstad et al., 1996). Because commercial fertilizer is expensive, many of these fields are first converted to pasture by letting in cattle and become truly abandoned only years later, when cattle are excluded.

IV. The Post-Agricultural Environment and Disturbance Regime

The Neotropics consist of the American landmass and associated islands lying between the Tropic of Cancer (23.5 [degrees] N) and the Tropic of Capricorn (23.5[degrees] S; see Figure 1 for locations of major research sites). Between these latitudes the sun has its least declination north and south, greatly reducing light and temperature seasonally and allowing year-round plant growth. Within the Neotropics temperatures can attain 38 [degrees]C but have a usual range between 17[degrees]C and 27[degrees]C, and annual precipitation can reach 6m with seasonal wet and dry cycles (Mabberley, 1992). However, rainfall is influenced by El Nino southern oscillations that occur every 3-7 years and can cause floods, droughts, and fires (Sanford et al., 1985).

[FIGURE 1 OMITTED]

Many Neotropical soils are old, highly weathered, and leached oxisols and ultisols, well drained, with a thin litter layer (Richter & Mammer, 1991; Mabberley, 1992; Finegan, 1996) and low organic matter content. Also common in the Neotropics are andisols derived from volcanic activity, which are very fertile and support intensive agriculture (Vitousek & Sanford, 1986). Many Neotropical soils have a higher clay content than do temperate soils and are low in phosphorus. Consequently, plant growth in the Neotropics may be more limited more by phosphorus than by other nutrients (Vitousek, 1984).

Biomes in the Neotropics that could be cleared for cultivation include deserts, grasslands, and forests (some forests are evergreen moist; others are seasonally dry). However, forests are most often used for cultivation. These forests have high plant diversity where mycorrhizae are common, effectively helping plants forage for phosphorus and other soil nutrients (Lugo & Lowe, 1995). Deserts are typically nonarable, tropical grasslands usually have periodic droughts, and large areas are often under water, making cultivation much more difficult (Mabberley, 1992).

At the moment of abandonment the abiotic environment is affected by the lack of woody vegetation, even though a few large trees and shrubs, which may have been too difficult to remove originally, may remain. Therefore, these areas will have high light quantity and quality and substantial rainfall contact with the soil. However, there may be low soil fertility and losses of fine roots (Da Oliveira Carvalheiro & Nepstad, 1996), mycorrhizae, and soil horizon/structure (Purata, 1986) resulting from the mechanical clearing of the area for crops. If shifting agriculture (slash and bum: Uhl, 1987) was used there may also have been a short pulse of some nutrients, like phosphorus, but with a volatilization of both organic matter and nitrogen (Montagnini & Sancho, 1994) and with an increase in soil temperature (Grigg, 1978). Such loss of soil organic matter can also lead to reduced soil fertility. These slash-and-burn methods produce, in general, nutrient-poor soils with carbon loss and a poor soil seed bank. In addition, slash and burn may lead to a mixing of the soil organic and mineral horizons and increased soil aeration/oxidation, which can influence microbial activities affecting nutrient release and mineralization (Bazzaz, 1996). Finally, application of fertilizer, herbicide, or water may have other consequences (see Scatena et al., 1996) following cropping.

Air temperature, vapor pressure deficit, and photon flux density have been shown to be much higher in post-agricultural areas than in the adjacent rain forest (Holl, 1999). The range of light intensity at ground level is 121-223 micromoles per [cm.sup.2] per second of photosynthetic active radiation [PAR] for a 15-second average (Myster, 2003a), with higher average air and soil temperature but lower average relative humidity than in forest (Uhl et al., 1988).

Nitrogen may be a key soil nutrient (Neill et al., 1995, 1997; Piccolo et al., 1997) and has been found to range between 0.027 and 0.080 in dry soil in kg/[m.sup.2] to 9-cm depth (Myster, 2004). Rhodes et al. (1998) and Rhoades and Coleman (1999) found less than 20% of nitrogen in old-field soils compared with nearby forest soils, but under Inga spp. trees soil N[O.sub.3] was four times higher, nitrification was five times higher, and soil organic matter was 7% higher compared with open microsites. However, these soils are generally wetter, with higher bulk densities (Rhodes et al., 1998). Ellingson et al. (2000) found a dramatic but short-term increase in inorganic nitrogen (up to 57 kg per ha) and in soil pH (as much as 2.3 pH units) after slash and burn in a Mexican pasture. Further, Reiners et al. (1994) report that pastures show a decrease in acidity and an increase in some base exchange properties, an increase in bulk density with its decrease in porosity, higher concentrations of N[H.sup.4], lower concentrations of N[O.sup.-3], lower rates of nitrogen mineralization, and lower rates of nitrification showing mixed soil fertility signals, higher cation concentration, and higher pH than forest soils.

There is a depletion of potassium in pasture soils but an increase in percentage of phosphorus retention compared with forest soils (Holl, 1999). Holl (1998a) also found clear patch differences in potassium, and Myster (2004) found that potassium ranged between .022 and .059 in pastures and fields. Phosphorus ranged between .002 and .005 in pastures and fields as well (Myster 2004), but high earthworm densities in pastures compared with the mature forest (Zou & Gonzalez, 1997, Leon & Zou, 2004) may help increase phosphorus availability (Aide & Cavelier, 1994). Soil carbon, an important control of nutrient availability (Brown & Lugo, 1990b; Desjardins et al., 1994; Veldkamp, 1994; Trumbore et al., 1995; Neill et al., 1996; Fearside & Barbosa, 1998), was reduced after 15 years of succession, but pasture soils showed a 26% increase in soil bulk density compared with forest soils (Rhoades et al., 1998).

All studies point to old-field succession as a patch dynamic process where shrub patches have higher levels of soil and litter nutrients than do grass patches (Vieira et al., 1994). As time passes, light quantity at ground level should decrease, but nutrients, soil organic matter, and litter should increase (Webb et al., 1972; Opler et al., 1977). Due to year-round vegetative growth and other favorable climatic features, responses and rate of recovery for pastures may be faster in the Neotropics (especially in areas that were once evergreen moist rain forest) than in the temperate zone, with nutrient cycling and water retention recovering more quickly than diversity and species composition (D. Schaefer, pets. comm.).

In general, a disturbance regime is characterized by its severity, size, spatial location, and frequency (Myster & Pickett, 1990a; Myster, 2001). The recovery that occurs after agriculture can also be conceptualized this way. First, old-field disturbance is of relatively moderate severity because, although most aboveground vegetation is cleared (or burned) and removed (Myster, 2003d), fields keep both an intact soil profile and, if not burned, a humus layer. Landslides (Myster & Fernandez, 1995; Myster, 1997; Myster & Walker, 1997) would be an example of a more severe disturbance because soil layers are lost. Second, plots used for agriculture in the Neotropics are often no more than a few to tens of hectares in size, with larger combined areas reserved for use as pastures after crop productivity decreases (Uhl et al., 1988; Scatena et al., 1996). Third, many crops are specific to certain spatial locations and elevations in Neotropical landscapes (Patzelt, 1996). For example, coffee (Coffea spp.) is often grown in mountainous areas between 1000 m and 2000 m, with banana/plantain (Musa spp.), maize (Zea mays), and sugarcane (Saccharum officinarum, Purata, 1986; Scatena et al., 1996) grown in the wetter valleys between 0 m and 1000 m. And finally, the frequency of old-field disturbance includes the number of times a field has been cultivated, the length of each cropping period (e.g., duration, which for pastures is the number of years cattle are present), its fallow time, and the order of crop rotation. Furthermore, fields are usually fallow longer than they are in crop, are reused with different crops and fallow periods until abandoned, are used for 5-10 years (shorter than temperate fields) until productivity declines (Uhl et al., 1988), and are not plowed when abandoned (Scatena et al., 1996).

The disturbance regime should also include aspects of cattle usage. Use of the land as pasture often begins when cattle (and sometimes horses) are let into cleared areas that were recently used for cultivation and ends when cattle are excluded (Humphreys, 1978; Whiteman, 1980). The longer cattle are present, the more pronounced are the effects (Hecht, 1993), such as trampling of vegetation, creation of hummocks, deposition of dung, and compaction of the soil, which increases soil bulk density and leads to resistance to penetration, to the formation of sesquioxides, and to the retardation of soil nitrogen transformations due to waterlogged conditions (Rhoades et al., 1998). Finally, because some of these areas have seasonal dry periods, recurrent fires are a possibility, in which case grasses, with their soil-level meristems, would be favored.

V. Patterns of Tree Invasion, Establishment, and Growth

Succession studies both in the temperate zone and the Tropics suggest that research should focus primarily on the dynamics of the functionally and structurally dominant species that were there before the disturbance and that will subsequently reinvade, reestablish, and regrow. Because most agricultural areas in the Neotropics were once forested, this means the invasion, establishment, and growth of trees (McDonnell & Stiles, 1983; Myster, 1993; Nepstad et al., 1996). A critical part of that investigation must be permanent vegetation plots where individual trees are marked and followed through time, a technique that has proved invaluable in temperate old-field studies. Unfortunately, there have been only two such Neotropical studies, with recurrent tree sampling at the same time intervals in plots established at abandonment (in Puerto Rico: Myster, 2003b; in Ecuador: Author, unpub, data--which has limited what can be said about how these areas change over time.

However, based on temperate old-field studies (for reviews, see Myster, 1993; Bazzaz, 1996) and Neotropical succession studies in gaps (e.g., Denslow, 1987; Everham et al., 1996), in landslides (Myster & Fernandez, 1995; Myster, 1997; Myster & Walker, 1997), and in abandoned crop fields (Purata, 1986; Donfack et al., 1995; Fernandes & Sanford, 1995; Quintana-Ascencio et al., 1996: Myster, 2003a, 2003b, 2003c), it is expected that after agriculture there would be increases in species richness, biomass, density, height, basal area, vertical structure, and productivity over time (Author, unpub. data). In addition, invading species have greater mycorrhizal association, larger seeds size, and more shade tolerance (Bazzaz & Pickett, 1980; Finegan, 1996).

Pasture plot data sampled in Puerto Rico show that Neotropical old-field trees enter, peak, and decline in an individualistic pattern, present for long periods with overlapping population curves (Myster, 2003b). On Neotropical islands, trees common in other seres (e.g., Cecropia spp., Schefflera spp.: Myster & Walker, 1997) may be absent after agriculture, replaced by trees such as Syzygium jambos, Calophyllum calaba, Clidemia strigilliosa, Miconia prasina, M. impetiolaris, Tabebuia heterophylla, and Piper hispidum (Myster, 2003b). However, chronosequences both in Puerto Rico (Aide et al., 1995, 2000) and in Brazil (Uhl et al., 1988) suggest that this may be only temporary. Common successional species and genera, however, such as Cecropia monostachya, Acalypha spp., Swartzia spp., and Psidium spp., have been found since abandonment on the Neotropical mainland (Author, unpub, data). In all cases the type of past crop (e.g., coffee, banana, sugarcane, pastures) in a field leads to individual plant responses, arguing strongly for fields to be investigated separately and for old-field succession not to be viewed as a single process. For example, in recovering coffee plantations, unique residual coffee (Coffea arabica) trees dominate, as do Guarea guidonia, Andira inermis, Homalium racemosum, and Casearia decandra (Myster, 2003b). As more data come in from the permanent plots, we may see more tree species in the Melastomataceae and Rubiaceae and such genera as Alchornea, Cordia, Inga, Trema, and Ceiba, as found in older Neotropical landslide permanent plots (Myster & Walker, 1997).

Permanent plots on abandoned pastures also show life-form changes such as a domination by grass (e.g., Panicum sp., Axonopus compressus) that peaks in cover in year three and then declines, fern (Nephrolepsis sp., Thelpteris deltoidea) and herbaceous (e.g., Commelina sp., Desmodium adscendens. Clidemia hirta) cover that also peaks in year three but at lower cover levels than grass, woody vines (e.g., Ipomea sp.) that continue to increase in cover over the first five years, and trees and shrubs that enter, peak, and decline in an individualistic pattern. Permanent plots in sugarcane fields show similar trends of grass dominating early (mostly sugarcane itself, Saccharum officinarum) and then declining to 50% cover levels, with Panicum spp. and Brachiaria subquadripara also present, ferns (e.g., Nephrolepsis sp., Thelypteris deltoidea) and forbs (e.g., Desmodium sp., Bidens pilosa, Lantana camara) as a small part of the flora that begins to decline immediately with cover of woody species increasing steadily, trees showing this gradual increase with members of the family Melastomataceae most common, and P. aduncum more common here than in other fields. In the plots located in recovering banana fields, grass and Musa sp. dominate early and then decline, but to greater cover levels than in the sugarcane. Ferns attain greater cover levels in banana than in sugarcane but forbs are a small part of the flora. Also, woody species increase, but at smaller cover levels than in sugarcane. There were many of the same tree species in banana fields as in sugarcane fields, but with additional species (e.g., C. monostachyma) present. Finally, in the seeded pastures, grass dominated almost completely (mainly the planted grass Setaria sphacelata), with a few woody plants showing in year five. Fern and forb cover was small, and pastures had few trees, but the species they did have were common to the sugarcane and banana fields.

Community parameters in pasture plots show after five years: two or three strata or horizontal levels that developed quickly; total stems at 200 per 250 [m.sup.2] plot (compared with 48-152 stems per 100 [m.sup.2] plot in Brazil; Uhl et al., 1988); tree richness constant at 20 species (also, 20 in Brazil but in a 100[m.sup.2] plot; Uhl et al., 1988); tree-stem evenness approximately 0.7; productivity peaking at year two at 0.5 kg/[m.sup.2]/yr (10 tons per ha per yr in Brazilian pastures; Uhl et al., 1988) and then declining slightly; turnover (for the formula, see Myster & Pickett, 1994) decreasing over the first five years to 50%; and mean height at 250 cm and total basal area of stems at 1000 [cm.sup.2]. For the sugarcane fields, total cover, an indicator of stratification and developing structure, was greater compared with the other fields, and banana greater than pasture, starting at 200% cover and gradually declining with time. Species richness was also greater in the sugarcane fields compared with the other fields, and banana had more species than did pasture. Total stems were slightly more in the banana fields compared with the sugarcane fields, and pastures had few stems. Compared with banana, mean stem height was also greater in the sugarcane and increased through time, while pasture stems remained small. Basal area stabilized at 500 [cm.sup.2] for sugarcane and banana fields (Buschbacher et al., 1988).

There is a critical need for more long-term plots after agriculture in the Neotropics in order to capture the complete range of successional responses. Puerto Rican and Ecuadorean study plots begin an exploration of the pattern of post-agricultural succession in the Neotropics, but more plots are needed (best at the other active research sites; Fig. 1) in order to examine whether these tree patterns continue in other old fields and pastures. Ideally, these plots should be started at abandonment, researchers should tag individual tree stems, and include replication of fields with the same crop history.

VI. Processes Affecting Tree Patterns

Critical in old-field investigation are the mechanisms that control individual tree replacements (Myster & Pickett, 1992a; Myster, 2001), because mechanisms work on individual plants, not on whole species. Individual replacements may lead to the temporal change in species composition called "succession," if all individuals of a species are replaced and individuals of new species enter the sere. These compositional changes may then define much of old-field patch dynamics. The resulting plant-based functions, such as productivity and decomposition, within those patches are dominant in defining ecosystems and landscapes (Myster, 2002a). Finally, replacement mechanisms have many sources of variation in the way they operate where neighborhood plants work to either increase or decrease the action of mechanisms. For example, in the case of dispersal, plants next to the plant being replaced may either attract dispersal agents (McDonnell & Stiles, 1983; Holl, 1998b) or repel them, as when mammals prefer certain patches to others.

No plant lives forever. Every plant will eventually be replaced, usually by one with a longer life span and a larger mature size. However, when plants become large enough to form a canopy, they enter a thinning phase in which, even though plants continue to die, they are not replaced by other individuals. Instead, living plants nearby adjust their growth to fill in the gap left by the dead plant. At some point plants reach a critical size where their death can no longer be compensated for by this "neighborhood growth," and they are then replaced by faster-growing, smaller plants. This is the fundamental cycle of terrestrial vegetation and continues indefinitely, unless underlying environmental gradients change or the species pool changes through loss or gain of species (Myster, 2001). Because the world's terrestrial vegetation is constantly undergoing replacement processes, plan-plant replacements are one of the most fundamental ecological processes on earth and a key element in vegetation dynamics, the mechanisms of which are most critical at the seed and seedling stages of plant development (the regeneration niche; Grubb, 1977).

A. CROP SIGNATURES

One important way in which agriculture differs from other kinds of Neotropical disturbance (e.g., landslides, hurricanes, tree-fall gaps) is the potential lingering effect of previous crops after abandonment (Myster & Pickett, 1990b). These signatures may occur in at least four ways. First, crops with extensive root systems and large root-to-shoot ratios can persist for long periods of time, especially if fields are not plowed before abandonment (e.g., grasses grown for hay [Daclylis glomerata] persist in the temperate zone; Myster & Pickett, 1990b, 1994). Second, crop root exudates may persist and affect litter chemistry, fungus, soil microbes, decomposition, and seed predators (Bazzaz, 1996). Third, species that were associated with the crop may persist (e.g., Inga spp. and Guarea guidonia [planted for shade in coffee plantations; Barros et al., 1995], Citrus spp. [planted fruit trees; Garcia-Montiel & Scatena, 1994], or unwanted weeds [Vengris, 1953]) and compete well with the native plants (Myster & Pickett, 1992a) or affect succession by altering soil-nutrient availability (Inga spp. can fix nitrogen: Myster, 2003a; Myster & Walker, 1997). And fourth, fertilization and herbicide practices may produce both residual abiotic (e.g., higher than normal levels of phosphorus or nitrogen) and biotic effects (e.g., lower abundances and loss of some microbe and insect species).

Because pastures are usually not plowed when abandoned (Scatena et al., 1996), grasses persist after abandonment and may have a host of long-term effects or signatures, as has been seen in temperate old fields (reviewed in Myster, 1993). These effects include reduction of annual and biennial cover, exclusion of woody species, and alteration of whole successional pathways (Myster & Pickett, 1990a, 1990b) and have been seen in Neotropical pasture plots (Myster, 2003b). While pastures may be viewed as crops of grass (e.g., Setaria sphacelata in Ecuador [Myster & Sarmiento, 1998], and Panicum maximum and Brachiaria spp. in Brazil [Uhl et al., 1988]) that are later abandoned when livestock are excluded, there are at least two fundamental differences between pastures and old fields: the grass may not be planted (as crops are) but may result from natural colonization (e.g., Panicum spp. and Brachiaria subquadripara in Puerto Rico [Myster, 2003b] and Paspalum spp. and ttyparrhenia rufa in Amazonia [Nepstad et al., 1996]); and livestock can alter pastures dramatically by depositing dung, trampling vegetation, creating hummocks, and compacting the soil, thereby increasing soil bulk density and resistance to water penetration.

B. SEED PROCESSES

Completed field studies (Table II) show how the amount or strength of tree replacement mechanisms at the seed stage vary among microsites, species, and field types. Such data exploring variation in the working of individual mechanisms can begin to weight different mechanisms against each other to help decide which ones may be most important in determining a particular tree replacement. Such replacements are bound to be due to the action of multiple mechanisms, as expressed in Table II. Other benefits include the identification of gaps in our knowledge of these mechanisms and, ultimately, how mechanisms might be incorporated in future probabilistic replacement models. It is at first clear that both the seed bank (Garwood, 1989) and seed pathogens need to be investigated more and at other sites (see Figure 1). There is some evidence that tree seeds are rare in the seed bank but that grasses and herbs are common (Uhl, 1987; Nepstad et al., 1996). Variation is wide among sites for number of seeds dispersed, suggesting that more detail is needed, especially for microsites and species (Table II). Data do show that wind dispersal is more common than animal dispersal (Guevara & Laborde, 1993; Cardoso et al., 1996), probably because animals wish to avoid open areas, and that areas next to forests and under woody patches are favored (Gorchov et al., 1993; Galindo-Gonzalez et al., 2000). Further exploration is also needed for germination, which has been shown to vary greatly among species and microsites (Aide & Cavelier, 1994; Holl, 1999; Myster, 2004).

Seed-predation studies are most numerous and well designed among the tree-replacement mechanism studies. Results show heavy seed losses everywhere. Because seed predation is so important, issues, such as density of predators in these areas, their behavior, and the availability and structure of cover where they hide, are probably very important. For example, predators seem to prefer the thick grass of pastures to the relative openness of a coffee plantation (Myster, 2003a). Although pathogenic diseases may not be as important as predation and other mechanisms, they may be key in specific field types such as sugarcane plantations, with their mats of decaying cane biomass. Reviewed results are comparable in strength and variation to these other Neotropical succession studies that examine many of the same mechanisms (Purata, 1986; Rico-Gray & Garcia-Franco, 1992; Donfack et al., 1995; Fernandes & Sanford, 1995; Garcia, 1995; Everham et al., 1996; Quintana-Ascencio et al., 1996; Myster, 1997; Medellin & Gaona, 1999).

C. SEEDLING PROCESSES

Completed field studies (Table III) show how the amount or strength of tree-replacement mechanisms at the seedling/sapling stage vary among microsites and field types. Such data explore the same variation in the working of individual mechanisms as those for seeds in Table II and also help identify gaps in our knowledge. Just as with seed pathogens, disease at the seedling/sapling stage needs to be investigated more and at other sites (see Figure 1). Although significant areas of leaf tissue can be lost to herbivory, which may even cause some seedling mortality (Holl & Quiros-Nietzen, 1999), competition seems to be the major mechanism working at this stage (Table III). However, herbivory and pathogenic disease may be important in terms of their effect on competition, even by altering competitive outcome among trees. Future field experiments should examine these interactions, along with resource availability. For example, persistence of past crops like coffee trees or large-leaved banana may increase competition for light, while pasture grasses may increase competition for water and soil nutrients. Competition for nitrogen and its patchy availability may also be key, especially where nitrogen-fixing trees like Inga spp. have been planted (coffee plantations; Myster, 2003a).

D. TREE PROCESSES

Even though studies on the seed/seedlings stages are most important, studies with larger individuals are needed for a complete understanding of old-field succession. Permanent plots in place now should help fill in those information gaps in the future, such as the pattern of succession. In addition, analysis of those data can implicate mechanisms among trees. However, field experiments need to include later stages. Those experiments can be physically demanding, but several manipulations can be performed; for example, removing species in field plots and manipulating herbivory and pathogenic levels. Also, patch effects often translate into effects of trees.

Because these areas were once rain forests, the dynamics of the old-field trees is a major concern. Indeed, the most obvious succession pattern may be the slow rate of invasion by trees into many of these areas. Tree seedlings and saplings may overcome these "barriers" and invade successfully in several ways: after incomplete clearing of woody vegetation (Nepstad et al., 1994, 1996), leading to tree sprouting from coppiced tree trunks or root stocks (Rico-Gray, 1991); as original trees remain and function as "nurse" trees facilitating seed dispersal and other regeneration mechanisms (McDonnell & Stiles, 1983; Myster, 1993; Holl, 1998b); or forest-edge shrubs spreading into the field asexually (e.g., Miconia spp. in Puerto Rican fields) and thereby allowing trees to invade quickly from the forest edges, independent of reproduction from dispersed seed or from the weak seed bank (Garcia, 1995). These early asexual woody plants can then facilitate later tree invasion by overtopping and killing the grass or other remaining crops. Some of these dynamics may be size-dependent because of post-abandonment human firewood collection of "useful"-sized trees or browsing of small trees by mammals, leading to an extended or arrested mid-succession old-field phase.

VII. Synthesis

Based on what we know now about regeneration mechanisms, tree succession in the Neotropics seems to be controlled by: competition with trees, shrub and large herbs for light (Holl, 1998a) and with grasses for water and soil nutrients, due in part to the high grass root mass per volume of soil (Gerhardt & Fredriksson, 1995; Holl, 1998a); drought stress (Gerhardt, 1993); high loss of seeds to predators (and at various spatial scales; Myster, 2003a); a high loss of seedlings to competition (Myster, 2004) or herbivory (Reader, 1992; Holl & Quiros-Nietzen, 1999); and a lack of seed input from the surrounding area or from the soil (Gomez-Pompa & Vazquez-Yanes, 1981; Hammond, 1995; Nepstad et al., 1996; da Silva et al., 1996; Holl, 1998b, 1999; Sarmiento, 1997; Myster, 2003a, 2004).

Taken together, plot data and experimental results suggest that successional development in these fields follows this timeline: early succession is dominated by the continuation of the past crop; after that, the edge effect of the adjacent rain forest dominates (Gorchov et al., 1993); then shrubs and woody vines invade, creating patch structure; and, finally, trees invade in microsites created by the shrubs and grow to form a canopy (for a similar model of temperate old fields, see Myster, 1993). Shrubs and their dynamics play a key role in Neotropical old-field succession; that is, the species and the cover or biomass of shrubs in a pasture largely determine its environmental conditions and regeneration mechanisms (Zahawi & Augspurger, 1999). The ability of many shrubs to attract seeds and shade out grasses, thereby facilitating tree invasion, and to spread asexually (e.g., Miconia spp.) leads to their domination in defining pasture dynamics. Consequently, aboveground competition dominates over belowground (Holl, 1998a). Such positive woody effects can also be seen near downed trees or logs (Peterson & Haines, 2000; Slocum, 2000). However, grass/herb patches may reduce drought stress in some cases (Aide & Cavelier, 1995). This was also true in New Jersey old fields, where the grass species was Dactylis and the shrubs included Rhus glabra and Rosa multiflora (Myster & Pickett, 1992a). However, it should be true everywhere that the inhibiting presence of grasses in pastures leads to a slow recovery (Aide et al., 1995, 1996; Myster, 2003b) compared with other post-agricultural areas.

As the length of time an area is in crop, or its duration, increases, the length of successional or recovery time may also lengthen (Purata, 1986; Waller et al., 1986). For example, a short cropping period can lead to large contributions from the seed bank, and trees may regenerate quickly from stump or root sprouts (Uhl et al., 1988; Rico-Gray, 1991). As the cropping period increases, decay of rootstocks and tree stumps leads to grass invasion and prolonged tree invasion (Nepstad et al., 1996). However, old fields may overcome this to a degree through regeneration from clonal shrubs such as Miconia spp. and the creation of suitable microsites by these shrubs for the advancing trees (Vieira et al., 1994; Myster, 2003b).

In general, these studies and others in Neotropical fields confirm many of the expectations from temperate studies (see Myster, 1993) where major successional mechanisms seem to be seed rain (Myster & Pickett, 1992b), seed predation, and seedling herbivory/competition. The recovering grass fields of New Jersey showed key mechanisms of seed predation and competition, but also seedling predation, frost heaving, and damage from falling stems and branches (Myster, 1993). In comparison, Neotropical permanent plots show similar life-form changes, such as: domination by grass (e.g., Panicum sp., Axonopus compressus) that peaks in year three and then declines (compared with eight-year grass dominance in New Jersey old fields grown in Dactylis glomerata, see references in Myster, 1993); fern (Nephrolepsis sp., Thelpteris deltoidea) and herbaceous (e.g., Commelina sp., Desmodium adscendens, Clidemia hirta) cover that also peaks in year three but at lower cover levels than grass; woody vines (e.g., Ipomea sp.) that continue to increase in cover over the first five years; and trees and shrubs that enter, peak, and decline in an individualistic pattern (seen also in the Dactylis spp. fields in New Jersey [Myster, 1993], where tree invasion was suppressed until Dactylis declined).

Finally, for further synthesis we need to compare old-field results and mechanisms with those from a more severe disturbance. During the recovery from landslides, for example, seed rain showed 48 seeds per [m.sup.2] over a similar time interval as studies in Table II and the seed bank was 3 seedlings for a similar soil sample, there was only a 10% loss of seeds to predators with 50% of the seeds lost to pathogens and 25% of them germinating, and 20% tree stem survival over a one-year period and leaf area losses to pathogens in the 1-3% range and to herbivory as much as 34% (Myster & Fernandez, 1995; Myster, 1997, 2002b). This shows a reduced seed rain and seed bank as the severity of the disturbance increases, with more seeds lost to pathogens but fewer seeds lost to predators.

VIII. Restoration, Management, and Conclusions

Restoration ecology attempts to manage natural succession processes toward an end that better satisfies human needs (Moran et al., 1996). In these old-fields, managers may want, for example, to retard succession, to quicken the rate of recovery, perhaps by increasing facilitating mechanisms such as dispersal and decreasing inhibitory mechanisms such as predation and competition, in order to generate rain-forest timber for use as a sustainable resource or to create rain-forest habitat for rare species, to hold succession (Niering & Goodwin, 1974) in a state where agricultural is less costly to maintain, or to alter the natural successional trajectory itself (Myster & Walker, 1997) to match soil conditions for maximum crop production. Mechanisms of tree replacement, as discussed in this review, can be manipulated toward any of these ends, either as inhibitors or facilitators.

Returning land to high productivity in as short a period of time as possible and then maintaining its sustainability are major concerns (Nepstad et al., 1991). In all these cases post-agricultural successional studies are critical for a complete understanding of how these Neotropical areas regenerate and function. At least two restoration methods have had success after other disturbances (for landslides, see Myster & Sarmiento, 1998): the planting of native seedlings and nurse trees (particularly if they can fix nitrogen), and the addition of fertilizer (Guariguata et al., 1995; Rhoades et al., 1998).

A special issue of the journal Restoration Ecology (see Aide, 2000 and papers therein, such as Holl et al., 2000; Peterson & Haines, 2000; Silver et al., 2000; Slocum, 2000; Zimmerman et al., 2000) dealt with pastures. Many of the suggestions there have been made for the restoration of other tropical disturbed areas (e.g., for landslides, see Myster & Sarmiento, 1998) and revolve around finding both suitable microsites for recruitment (Malo & Suarez, 1995; Myster, 2003a) and identifying recovery "barriers" (Nepstad et al., 1990; Aide & Cavelier, 1994). The planting of native seedlings in good microsites is a common approach and should include a few larger trees or shrubs to act as recruitment foci, by attracting bird dispersers (McDonnell & Stiles, 1983; Howe, 1984; Holl, 1998b) and acting as "nurse" trees (e.g., Guevara et al., 1992; Vieira et al., 1994; Rhoades & Coleman, 1999), creating positive tree feedback mechanisms. Another restoration strategy is the addition of nutrients, especially if soil was removed or damaged. One may do both, of course, by adding nitrogen-fixing species such as Inga spp. However, it needs to be kept in mind that old-field succession must be understood scientifically in order for it to be managed well and that, consequently, the successional processes and mechanisms reviewed here should be a key part of any restoration efforts.

An important thrust of future research should then be on both the pattern and the processes of tree invasion and establishment, with a significant focus on crop signatures. This review of old-field and pasture studies in the Neotropics can help that effort by assisting in the construction of hypotheses and predictions about key species-specific and field-type-specific regeneration mechanisms after agriculture. Researchers need to be interested in how mechanisms vary over space and time and how they are affected by the resident tree species and by a field's history, including its crop signature. For those studies showing species variation in mechanisms, summaries such as Tables II and III could then be used to construct replacement probabilities among species in each field type for each specific mechanism. From there, mathematical analysis could lead to predictions of tree-replacement sequences and their comparison with actual replacement sequences found in permanent plot data.

Old-field studies are paramount for an understanding of how Neotropical rain forests regenerate and function. To date we have some understanding of the pattern of their recovery, but it is limited to only two sites, making establishment of permanent plots at other sites (see Figure 1) critical for a more complete examination of pasture succession. In addition, we have a fair amount of information about mechanisms driving those patterns of recovery and tree replacement; however, a primary thrust of future research should also be on the processes of tree invasion and establishment (see Myster, 1993). This would naturally lead to additional modeling, both of individual tree replacements and of tree composition and abundance. With these models, prediction of tree densities in entire pastures is then possible if the number and starting tree abundance in each patch type, as well as the number of patches of each type in the pasture, are known.

Finally, I suggest that the ecophysiology of the dominant species needs to be better investigated (Gerhardt & Fredriksson, 1995) as a possible driver of the replacements (tolerance model: Myster & Pickett, 1992a), focusing on how individual tree species grow in the presence of different resources. Also, environmental conditions need to be sampled in more sites, and investigations need to be carried out in a patchwise manner (for patch differences in regeneration mechanisms, see Vieira et al., 1994). Understanding mechanisms of old-field succession and uncovering gaps in our knowledge may in turn lead to construction of hypotheses and predictions about pasture regeneration, to be tested in field experiments and in comparative pattern analysis using permanent plots. Results show that future research should have a more exact within-patch (Myster, 2003c) and individual tree species focus.

IX. Acknowledgments

I wish to thank O. Mironova, P. Burton, M. Palmer. D. Gorchov, J. Parrotta, S. Gleeson, N. Brokaw, and K. Holl for commenting on a previous version of this article. I also wish to thank G. Reyes for her help in finding reference material and T. Bower for his help with the map. This research was performed tinder grant DEB-9411973 from the National Science Foundation to the Institute for Tropical Ecosystem Studies, University of Puerto Rico, and to the International Institute of Tropical Forestry, as part of the long-term Ecological Research Program in the Luquillo Experimental Forest.

Table I 

Planting, cultivation, and harvesting practices of common crops 

in the Neotropics 

Crop        Planting             Cultivation 

Sugarcane   Hand planted from    Stumps cleared of 

            cuttings at wet      sprouts; rows of 

            season               vines tilled; 

                                 grasses resprout 

Corn        Seeds hand planted   Crop rotation; 

                                 intercropping with 

                                 clover and beans 

Coffee      Grown seedlings      Legumes often 

            planted in deep      provide shade 

            holes 

Bananas     Corms and root       Weak roots, so 

            suckers placed in    weeds hand cut 

            cleared areas at 

            wet season 

Pasture     Cows let in; grass   Fire can reduce 

            seed may be added    woody growth 

Cassava     Hand tilled, then    Grows well on 

            stem cuttings        low-fertility soils; 

            planted in holes     intercropped with 

            with organic         beans 

            matter added; 

            grafting common 

Beans       Planted by stick     Commonly intercropped 

            or hoe; seeds then   with corn and other 

            added but need       crops 

            tilled soil to 

            germinate 

Rice        Flooded areas        Flooding regime 

            tilled; seedlings    must be maintained; 

            then hand planted    some weeding 

Crop        Harvesting           General references 

Sugarcane   Stems hand cut;      Husz, 1972; Fauconnier, 

            15-20 years of       1992 

            yields 

Corn        Cobs harvested by    Schrimpf, 1966; 

            hand                 Rouanet, 1987 

Coffee      Hand picked; trees   Wellman, 1961; 

            last for decades     Cambrony, 1992 

Bananas     Bunches hand cut;    Seelig, 1969; Reynolds, 

            may last 5-20        1981 

            years 

Pasture     Livestock excluded   Humphreys, 1978; 

                                 Whiteman, 1980 

Cassava     Best in cooler       Hall, 1984; Cock, 

            months; variable     1985 

            after 7-18 months 

            of growth 

Beans       Hand picked, then    Duke, 1981; 

            hoed under           Schoonhoven & 

                                 Voysest, 1991 

Rice        All aboveground      Chandler, 1979; 

            biomass harvested,   Lang, 1991 

            then machine 

            picked 

Table II 

Replacement mechanisms in completed seed studies of Neotropic oldfields and pastures. Consult the specific references for details both on methodology and on species and other sources of variation in the workings of each mechanism. 

Mechanism,             Time                  Field 

amount                period     Microsite   type        Reference 

Seed dispersal 

  6-92/[m.sup.2]      1 year     Grass/      Pasture     Vieira et al., 

                                 shrub                   1994 

  23-100/[m.sup.2]    1 year     Grass/      Pasture     Holl, 2002 

                                 shrub 

  190/[m.sup.2]       1 year                 Pasture     Holl, 1999 

  39-4610/[m.sup.2]   1 year                 Pasture     Sarmiento, 

                                                         1997 

  0-8/[m.sup.2]       1 month                Pasture     Aide & 

                                                         Cavelier, 1994 

  10/[m.sup.2]        2 weeks                Pasture     Holl, 1998b 

  2-990/[m.sup.2]     1 year                 Pasture     Nepstad et 

                                                         al., 1996 

  400/[m.sup.2]       3 months   Edge        Oldfields   Myster, 2004 

Seed bank 

  1676                6 months               Pasture     Zahawi & 

                                                         Augspurger, 

                                                         1999 

  10/10g soil         2 months   Edge        Oldfields   Myster, 2004 

Seed predation 

  95% loss            1 month                Oldfields   Notman & 

                                                         Gorchov, 2001 

  90%loss                        Edge        Pasture     Sarmiento, 

                                                         1997 

  4-100% loss         19 days                Pasture     Aide & 

                                                         Cavelier, 1994 

  59-68% loss         24 days    Grass/      Pasture     Holl, 2002 

                                 shrub 

  50-75% loss         14 days                Coffee/     Myster, 2003a 

                                             pasture 

  50-79% loss         14 days    Grass/      Pasture     Myster, 2003c 

                                 shrub 

  63% loss            30 days    Edge        Pasture     Holl & Lulow, 

                                                         1997 

  >80% loss           20 days                Pasture     Nepstad et 

                                                         al., 1990 

  67-90% loss         14 days    Edge        Oldfield,   Myster, 2004 

Seed pathogenic disease 

  20-45% loss         14 days                Coffee/     Myster, 2003a 

                                             pasture 

  5-10% loss          21 days    Edge        Oldfields   Myster, 2004 

Germination 

  0-53%               6 months               Pasture     Holl, 1999 

  77%                                        Pasture     Vieira et al., 

                                                         1994 

  15-35%              14 days                Coffee/     Myster, 2003a 

                                             pasture 

  9-77%               3 months   Grass/      Pasture     Holl, 2002 

                                 shrub 

  5-10%               21 days    Edge        Oldfields   Myster, 2004 

Table III 

Replacement mechanisms in completed seedling/sapling studies of Neotropic oldfields and pastures. Stem disease and stem herbivory are given as percentages of leaf area lost, and competition refers to percentage of stem survival. Consult the specific references for 

details both on methodology and on species and other sources of variation in the workings of each mechanism. 

Process,       Time        Microsite     Field       Reference 

amount         period                    type 

Stem disease 

  3%           6 months    Edge          Oldfields   Myster, 2004 

Stem herbivory 

  35%          16 days                   Pasture     Nepstad et al., 1996 

  <25%                                   Pasture     Gerhardt, 1993    33%                                                Nepstad et al.,                                                       1990    5%           6 months    Edge          Oldfields   Myster, 2004    Stem competition      40%          5 months                              Aide & Cavelier,                                                       1994    95%          1 year      Grass/shrub   Pasture     Holl, 1998a    ><5%          3 year                    Pasture     Gerhardt, 1993    100%                                   Pasture     Vieira et al.,                                                       1994    1-10% loss   18 months   Grass/shrub   Pasture     Holl, 2002    30-50%       6 months    Edge          Oldfields   Myster, 2004  >

X. Literature Cited

Aide, T. M. 2000. Clues for tropical forest restoration. Restoration Ecology 8: 327.

-- & J. Cavelier. 1994. Barriers to lowland tropical forest restoration in the Sierra Nevada de Santa Marta, Colombia. Restoration Ecology 2: 219-229.

--, J. K. Zimmerman, L. Herrera, M. Rosario & M. Serrano. 1995. Forest recovery in abandoned tropical pastures in Puerto Rico. Forest Ecology and Management 77: 77-86.

--, --, M. Rosario M. & H. Marcano. 1996. Forest recovery in abandoned cattle pastures along an elevational gradient in northeastern Puerto Rico. Biotropica 28: 537-548.

--, J. B. Pascarella, L. Rivera & H. Marcano-Vega. 2000. Forest regeneration in a chronosequence of tropical pastures: Implications for restoration ecology. Restoration Ecology 8: 328-338.

Barros, R. S., M. Maestri & A. B. Rena. 1995. Coffee crop ecology. Tropical Ecology 36: 1-19.

Bazzaz, F. A. 1968. Succession on abandoned fields in the Shawnee Hills, southern Illinois. Ecology 49: 924-936.

--. 1996. Plants in changing environments: Linking physiological, population and community ecology. Cambridge Univ. Press, Cambridge.

-- & S. T. A. Pickett. 1980. The physiological ecology of tropical succession: A comparative review. Annual Review of Ecology and Systematics 11: 287-310.

Borhidi, A. 1988. Vegetation dynamics of the savannization process on Cuba. Vegetatio 77:177-183.

Brown, S. & A. E. Lugo. 1990a. Tropical secondary forests. Journal of Tropical Ecology 6: 1-32.

-- & --. 1990b. Effects of forest clearing and succession on the carbon and nitrogen content of soils in Puerto Rico and US Virgin Islands. Plant and Soil 124: 53-64.

Buell, M. F., H. F. Buell, J. A. Small & T. G. Siccama. 1971. Invasion of trees and secondary succession on the New Jersey Piedmont. Bulletin of the Torrey Botanical Club 98: 67-74.

Buschbacher, R. J. 1986. Tropical deforestation and pasture development. Bioscience 36: 2-28.

--, C. Uhl & E. A. S. Serrao. 1988. Abandoned pastures in eastern Amazonia, II. Nutrient stocks in the soil and vegetation. Journal of Ecology 76: 682-699.

Cambrony, H. 1992. Coffee growing. Macmillan, New York.

Cardoso da Silva, J. M., C. Uhl & G. Murray. 1996. Plant succession, landscape management, and the ecology of frugivorous birds in abandoned Amazonian pastures. Conservation Biology 10: 491-503.

Castro, I. A., S. Sterling & E. F. Galiano. 1986. Multispecies pattern-analysis of Mediterranean pastures in 3 stages of ecological succession. Vegetatio 68: 37-42.

Chandler, R. F. 1979. Rice in the Tropics: A guide to the development of national programs. Westview Press, Boulder, CO.

Chinea, J. D. 2002. Tropical forest succession on abandoned farms in the Humacao Municipality of eastern Puerto Rico. Forest Ecology and Management 167: 195-207.

Cock, J. 1985. Cassava, new potential for a neglected crop. Westview Press, London.

Da Oliveira Carvalheiro, K. & D. C. Nepstad. 1996. Deep soil heterogeneity and fine root distribution in forests and pastures of eastern Amazonia. Plant and Soil 182: 279-285.

Da Silva, J. M., C. Uhl & G. Murray. 1996. Plant succession, landscape management and the ecology of frugivorous birds in abandoned Amazonian pastures. Conservation Biology 10: 491-503.

Denslow, J. S. 1987. Tropical rainforest gaps and tree species diversity. Annual Review of Ecology and Systematics 18: 431-452.

Desjardins, T. F., B. Andreux, V. Volkoff & C. C. Cerri. 1994. Organic carbon and C13 contents in soils and soil size-fractions, and their changes due to deforestation and pasture installation in eastern Amazonia. Geoderma 61:103-118.

Donfack, P., C. Floret & R. Pontanier. 1995. Secondary succession in abandoned fields of dry tropical northern Cameroon. Journal of Vegetation Science 6: 499-508.

Duke, J. A. 1981. Handbook of legumes of world economic importance. Plenum Press, New York.

Egler, F. E. 1954. Vegetation science concepts, I. Initial floristic composition--A factor in old-field vegetation development. Vegetatio 4: 412-417.

Ellingson, L. J., J. B. Kauffman, D. L. Cummings, R. L. Sanford & V. J. Jaramillo. 2000. Soil N dynamics associated with deforestation, biomass burning, and pasture conversion in a Mexican tropical dry forest. Forest Ecology and Management 137: 41-51.

Everham, E. M., III, R. W. Myster & E. VanDeGenachte. 1996. Effects of light, moisture, temperature, and litter on the regeneration of five tree species in the tropical montane wet forest of Puerto Rico. American Journal of Botany 83: 1063-1068.

Ewel, J. 1980. Tropical succession: Manifold routes to maturity. Biotropica 12: 2-7.

Fauconnier, R. 1992. Sugarcane. Macmillan, New York.

Fearnside, P. M. 1993. Deforestation in Brazilian Amazonia: The effect of population and land tenure. Ambio 22: 537-545.

-- & R. I. Barbosa. 1998. Soil carbon changes from conversion of forest to pasture in Brazilian Amazonia. Forest Ecology and Management 108:147-166.

Ferguson, B. G., J. Vandermeer, H. Morales & D. M. Griffith. 2003. Post-agricultural succession in El Peten, Guatemala. Conservation Biology 17: 818-828.

Fernandes, D. N. & R. L. Sanford. 1995. Effects of recent land-use practices on soil nutrients and succession under tropical wet forest in Costa Rica. Conservation Biology 9:915-922.

Finegan, B. 1996. Pattern and process in neotropical secondary rain forests: The first 100 years of succession. Trends in Ecology and Evolution 11: 119-124.

Galindo-Gonzalez, J., S. Guevara & V. J. Sosa. 2000. Bat- and bird-generated seed rains at isolated trees in pastures in a tropical rainforest. Conservation Biology 14: 1693-1703.

Garcia, M. A. 1995. Relationships between weed community and soil seed bank in a tropical agro-ecosystem. Agriculture, Ecosystems and Environment 55: 139-146.

Garcia-Montiel, D. C. & F. N. Seatena. 1994. The effect of human activity on the structure and composition of a tropical forest in Puerto Rico. Forest Ecology and Management 63: 57-78.

Garwood, N. C. 1989. Ecology of soil seed banks. Academic Press, New York.

Gerhardt, K. 1993. Tree seedling development in tropical dry abandoned pasture and secondary forest in Costa Rica. Journal of Vegetation Science 4: 95-102.

-- & D. Fredriksson. 1995. Biomass allocation by broad-leaf mahogany seedlings, Swietenia macrophylla (King), in abandoned pasture and secondary dry forest in Guanacaste, Costa Rica. Biotropica 27: 174-182.

Gomez-Pompa, A. & D. A. Bainbridge. 1997. Tropical forestry as if people mattered. Pp. 408-422 in A. E. Lugo & C. Lowe (eds.), Tropical forests: Management and ecology. Ecological Studies, 112. Springer-Verlag, New York.

-- & C. Vazquez-Yanes. 1981. Successional studies of a rain forest in Mexico. Pp. 246-266 in D. C. West, H. H. Shugart & D. B. Botkin (eds.), Forest succession: Concepts and application. Springer-Verlag, New York.

Gorchov, D. L., F. Cornejo, C. Ascorra & M. Jaramillo. 1993. The role of seed dispersal in the natural regeneration of rain forest after strip-cutting in the Peruvian Amazon. Vegetatio 107/108: 339-349.

Grainger, A. 1988. Estimating areas of degraded tropical lands requiring replenishment of forest cover. International Tree Crops Journal 5:31-61.

Grigg, D. B. 1978. Shifting agriculture. Cambridge Univ. Press, Cambridge.

Grime, J. P. 1977. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 111: 1169-1194.

Grubb, P. J. 1977. The maintenance of species-richness in plant communities: The importance of the regeneration niche. Biological Reviews of the Cambridge Philosophical Society 52: 107-145.

Guariguata, M. R. & R. Ostertag. 2001. Neotropical secondary forest succession: Changes in structural and functional characteristics. Forest Ecology and Management 148: 185-206.

--, R. Rheingans & F. Montagnini. 1995. Early woody invasion under tree plantations in Costa Rica: Implications for forest restoration. Restoration Ecology 3: 252-260.

Guevara, S. & J. Laborde. 1993. Monitoring seed dispersal at isolated standing trees in tropical pastures: Consequences for local species availability. Vegetatio 107/108: 319-338.

--, J. Meave de Castillo, P. Moreno-Casasola & P. Laborde. 1992. Floristic composition and vegetation structure under isolated tree in Neotropical pastures. Journal of Vegetation Science 3: 655-664.

Hall, J. R. 1984. A study of the growth and yield of cassava (Manihot esculenta Crantz) when grown as a perennial crop. Master's thesis, Univ. of Queensland.

Hammond, D. S. 1995. Postdispersal seed and seedling mortality of tropical dry forest trees after shifting agriculture, Chiapas, Mexico. Journal of Tropical Ecology 11:295-313.

Hecht, S. B. 1993. The logic of livestock and deforestation in Amazonia. Bioscience 34: 687-695.

Hill, M. O., D. F. Evans & S. A. Bell. 1992. Long-term effects of excluding sheep from hill pastures in North Wales. Journal of Ecology 80:1-13.

Hobbie, S. E. 1992. Effects of plant species on nutrient cycling. Trends in Ecology and Evolution 7: 336-339.

Holl, K. D. 1998a. Effects of above- and below-ground competition of shrubs and grass on Calophyllum brasiliense (Camb.) seedling growth in abandoned tropical pasture. Forest Ecology and Management 109: 187-195.

--. 1998b. Do bird perching structures elevate seed rain and seedling establishment in abandoned tropical pasture? Restoration Ecology 6: 253-261.

--. 1999. Factors limiting tropical rain forest regeneration in abandoned pasture: Seed rain, seed germination, microsite and soil. Biotropica 31: 229-242.

--. 2002. Effect of shrubs on tree seedling establishment in an abandoned tropical pasture. Journal of Ecology 90: 179-187.

-- & M. Kappelle. 1999. Tropical forest recovery and restoration. Trends in Ecology and Evolution 14: 378-379.

-- & M. E. Lulow. 1997. Effects of species, habitat and distance from edge on post-dispersal seed predation in a tropical rainforest. Biotropica 29: 459-468.

-- & E. Quiros-Nietzen. 1999. The effect of rabbit herbivory on reforestation of abandoned pasture in southern Costa Rica. Biological Conservation 87: 391-395.

--, M. E. Loik, E. H. V. Lin & I. A. Samuels. 2000. Tropical montane forest restoration in Costa Rica: Overcoming barriers to dispersal and establishment. Restoration Ecology 8: 339-349.

Howe, H. F. 1984. Implications of seed dispersal by animals for tropical reserve management. Conservation Biology 30: 261-281.

Humphreys, L. R. 1978. Tropical pastures and fodder crops. Longman Group, London.

Husz, G. S. 1972. Sugarcane: Cultivation and fertilization. Ruhr-Stickstoff, Bochum, Germany.

Janzen, D. H. 1990. An abandoned field is not a tree fall gap. Vida Silviculture. Neotropica 2: 64-67.

Keller, M. & W. A. Reiners. 1994. Soil-atmosphere exchange of nitrous oxide, nitric oxide and methane under secondary succession of pasture to forest in the Atlantic lowlands of Costa Rica. Global Bingeochemical Cycles 8: 399-409.

--, E. Veldkamp, A. M. Weitz & W. A. Reiners. 1993. Effect of pasture age on soil trace-gas emissions from a deforested area of Costa Rica. Nature 365: 244-246.

Lang, J. 1991. Feeding a hungry planet: Rice, research and development in Asia and Latin America. Univ. of North Carolina Press, Chapel Hill.

Lanly, J. P. 1981. Sintesis regional. Pp. 1-86 in Los recursos forestales de la America tropical. Food and Agriculture Organization / United Nations Environmental Program, Rome.

Leon, Y. S. & X. Zuo. 2004. Plant influences on native and exotic earthworms during secondary succession in old tropical pastures. Pedobiologia 48: 215-226.

Lugo, A. E. 1991. Comparison of tropical tree plantations with secondary forests of similar age. Ecological Monographs 62: 1-41.

-- & C. Lowe. 1995. Tropical forests: Management and ecology. Springer-Verlag, New York.

Mabberley, D. J. 1992. Tropical rain forest ecology. Chapman and Hall, New York.

Malo, J. E. & F. Suarez. 1995. Establishment of pasture species on cattle dung: The role of endozoochorus seeds. Journal of Vegetation Science 6:169-174.

Matson, P. A., W. J. Patton, A. G. Power & M. J. Swift. 1997. Agricultural intensification and ecosystem properties. Science 277: 504-509.

McDonnell, M. J. & E. W. Stiles. 1983. The structural complexity of old field vegetation and the recruitment of bird-dispersed plant species. Oecologia 56:109-116.

Medellin, R. A. & O. Gaona. 1999. Seed dispersal by bats and birds in forest and disturbed habitats of Chiapas, Mexico. Biotropica 31: 478-485.

Miles, J. 1979. Vegetation dynamics. Chapman and Hall, London.

Montagnini, F. & F. Sancho. 1994. Net nitrogen mineralization in soils under six indigenous tree species, and abandoned pasture and a secondary forest in the Atlantic lowlands of Costa Rica. Plant and Soil 162: 117-124.

Moran, E. F., A. Packer, E. Brondizzio & J. Tucker. 1996. Restoration of vegetation cover in the eastern Amazon. Ecological Economics 18: 41-54.

Myers, N. 1991. Tropical deforestation: The last situation. Bioscience 41,282.

Myster, R. W. 1993. Tree invasion and establishment in old fields at Hutcheson Memorial Forest. Botanical Review 59: 251-272.

--. 1997. Seed predation, disease and germination on landslides in neotropical lower montane wet forest. Journal of Vegetation Science 8: 55-64.

--. 2001. Mechanisms of plant response to gradients and after disturbances. Botanical Review 67: 441-452.

--. 2002a. The use of productivity and decomposition to address functional redundancy in the Neotropics. Community Ecology 3:51-57.

--. 2002b. Foliar pathogen and insect herbivore effects on two landslide tree species in Puerto Rico. Forest Ecology and Management 169:231-242.

--. 2003a. Seed regeneration mechanisms over fine spatial scales on recovering coffee plantation and pasture in Puerto Rico. Plant Ecology 166:199-205.

--. 2003b. Vegetation dynamics of a permanent pasture plot in Puerto Rico. Biotropica 35: 422-428.

--. 2003c. Effects of species, density, patch-type and season on post-dispersal seed predation in a Puerto Rican pasture. Biotropica 35: 542-546.

--. 2003d. Using biomass to model disturbance. Community Ecology 4: 101-105.

--. 2004. Regeneration filters in post-agricultural fields of Puerto Rico and Ecuador. Plant Ecology 172: 199-209.

-- & D. S. Fernandez. 1995. Spatial gradients and patch structure on two Puerto Rican landslides. Biotropica 27: 149-159.

-- & S. T. A. Pickett. 1990a. The Buell Successional Study at the Hutcheson Memorial Forest Center. Permanent Plotter 3: 2-3.

-- & --. 1990b. Initial conditions, history and successional pathways in ten contrasting old fields. American Midland Naturalist 124: 231-238.

-- & --. 1992a. Dynamics of plant associations in ten old fields during 31 years of succession. Journal of Ecology 80: 291-302.

-- & --. 1992b. Effects of palatability and dispersal mode on spatial patterns of trees in old fields. Bulletin of the Torrey Botanical Club 119: 145-151.

-- & --. 1994. A comparison of rate of succession over 18 yr in 10 contrasting old fields. Ecology 75: 387-392.

-- & F. O. Sarmiento. 1998. Seed inputs to microsite patch recovery on two Tropandean landslides in Ecuador. Restoration Ecology 6: 35-43.

-- & L. R. Walker. 1997. Plant successional pathways on Puerto Rican landslides. Journal of Tropical Ecology 13: 165-173.

Neill, C., M. A. Piccolo, P. A. Steudler, J. M. Melillo, B. J. Feigl & C. C. Cerri. 1995. Nitrogen dynamics in soils of forests and active pastures in the western Brazilian Amazon Basin. Soil Biology and Biochemistry 27:1167-1175.

--, B. Fry, J. M. Melillo, P. A. Steudler, F. Moraes & C. C. Cerri. 1996. Forest and pasture-derived carbon contributions to carbon stocks and microbial respiration of tropical pasture soils. Oecologia 107: 113-119.

--, C. Neill, J. M. Melillo, P. A. Steudler, C. C. Cerri, J. E L. Moraes, M. C. Piccolo & M. Brito. 1997. Soil carbon and nitrogen stocks following forest clearing for pasture in the southwestern Brazilian Amazon. Ecological Applications 7: 1216-1225.

Nepstad, D. C., C. Uhl & E. A. S. Serrao. 1990. Surmounting barriers to forest regeneration in abandoned, highly degraded pastures: A case study from Paragominas, Para, Brazil. Pp. 215-229 in A. B. Anderson (ed.), Alternatives to deforestation: Steps toward sustainable use of the Amazon rain forest. Columbia Univ. Press, New York.

--, -- & --. 1991. Recuperation of a degraded Amazonian landscape: Forest recovery and agricultural restoration. Ambio 20: 248-255.

--, C. R. Carvalho, E. A. Davidson, P. Jipp, P. Lefebvre, G. H. Negreiros, H. Silva, T. A. Stone, S. Trumbore & S. Vieira. 1994. The role of deep roots in the hydrologic and carbon cycles of Amazonian forests and pastures. Nature 372: 666-669.

--, C. Uhl, C. A. Perira & J. M. C. De Silva. 1996. A comparative study of tree establishment in abandoned pasture and mature forest of eastern Amazonia. Oikos 76: 25-39.

Niering, W. A. & R. H. Goodwin. 1974. Creation of relatively stable shrub lands with herbicides: Arresting "succession" on rights of way and pastureland. Ecology 55: 784-795.

Notman, E. & D. L. Gorchov. 2001. Variation in post-dispersal seed predation in mature Peruvian lowland tropical forest and fallow agricultural sites. Biotropica 33: 621-636.

Oosting, H. J. 1942. An ecological analysis of the plant communities of Piedmont, North Carolina. American Midland Naturalist 28:1-126.

Opler, P. A., H. G. Baker & G. W. Frankie. 1977. Recovery of tropical lowland forest ecosystems. Pp. 379-419 in J. Cairns Jr., K. L. Dickson & E. E. Herricks (eds.), Recovery and restoration of damaged ecosystems. Univ. of Virginia Press. Charlottesville.

Padoch, C. & A. E Vayda. 1983. Patterns of resource use and human settlement in tropical forest. Pp. 301-313 in F. B. Golley (ed.), Tropical rain forest ecosystems. Elsevier, New York.

Paine, R. T. 1977. Controlled manipulations in the marine intertidal zone and their contributions to ecological theory. Proceedings of the Academy of Natural Sciences 12: 245-270.

Patzelt, E. 1996. Flora del Ecuador. Banco Central del Ecuador, Quito.

Peterson, C. J. & B. L. Haines. 2000. Early successional patterns and potential facilitation of woody plant colonization by rotting logs in premontane Costa Rican pastures. Restoration Ecology 8: 361-369.

Piccolo, M. A., C. Neill & C. C. Cerri. 1997. Net nitrogen mineralization and net nitrification along a tropical forest-to-pasture chronosequence. Plant and Soil 162: 61-70.

Pickett, S. T.A. 1982. Population patterns through twenty years of old field succession. Vegetatio 49: 45-59.

Purata, S. E. 1986. Floristic and structural changes during old-field succession in the Mexican tropics in relation to site history and species availability. Journal of Tropical Ecology 2: 257-276.

Qiang, S. & J. Hu. 1999. Quantitative analysis of weed communities in cotton fields in cotton-growing regions of Jiangsu Province. Acta Ecologia Sinica 19: 810-816.

Quintana-Ascencio, E F., M. Gonzalez-Espinosa, N. Ramirez-Marcial, G. Dominguez-Vazquez & M. Martinez-Ico. 1996. Soil seed banks and regeneration of tropical rain forest from milpa fields at the Selva Lacandona, Chiapas, Mexico. Biotropica 28: 92-209.

Reader, R. J. 1992. Herbivory, competition, plant mortality and reproduction on a topographic gradient in an abandoned pasture. Oikos 65: 414-418.

Reiners, W. A., A. E Bouwman, W. E Parsons & M. Keller. 1994. Tropical rain forest conversion to pasture: Changes in vegetation and soil properties. Ecological Applications 4: 363-377.

Reynolds, P. K. 1981. The banana: Its history, cultivation and place among stable foods. Gordon Press, New York.

Rhoades, C. C. & D. C. Coleman. 1999. Nitrogen mineralization and nitrification following land conversion in montane Ecuador. Soil Biology and Biochemistry 31: 1347-1354.

--, G. E. Eckert & D. C. Coleman. 1998. Effect of pasture trees on soil nitrogen and organic matter: Implications for tropical montane forest restoration. Restoration Ecology 6: 262-270.

Richter, D. D. & L. I. Mammer. 1991. Soil diversity in the tropics. Advances in Ecological Research 21: 315-389.

Rico-Gray, V. 1991. Propagule availability and the regeneration of tropical forests. Trends in Ecology and Evolution 6:143-144.

-- & J. G. Garcia-Franco. 1992. Vegetation and soil seed bank of successional stages in tropical lowland deciduous forest. Journal of Vegetation Science 3:617-624.

Rivera, L. W., J. K. Zimmerman & T. M. Aide. 2000. Forest recovery in abandoned agricultural land in a karst region of the Dominican Republic. Plant Ecology 148: 115-125.

Rouanet, G. 1987. Maize. CTA Macmillan, London.

Sanford, R. L., J. K. Saldarriaga, D. Clark, C. Uhl & R. Herrera. 1985. Amazon rainforest fires. Science 227: 53-55.

Sarmiento, F. O. 1997. Arrested succession in pastures binders regeneration of Tropandean forests and shreds mountain landscapes. Environmental Conservation 24: 14-23.

Scatena, E N., R. T. Walker, A. K. O. Homma, A. J. Deconto, C. A. P. Ferreira, R. A. Carvalho, A. C. P. N. Derocha, A. I. M. Dos Santos & P. M. De Oliveira. 1996. Cropping and fallowing sequences of small farms in the "terra firme" landscape of the Brazilian Amazon: A case study from Santarem, Pardi. Ecology and Economics 18: 29-40.

Schoonhoven, A. & O. Voysest (eds.). 1991. Common beans: Research for crop improvement. Redwood Press, Melksham, England.

Schrimpf, K. 1966. Maize: Cultivation and fertilization. Ruhr-Stickstoff, Bochum, Germany.

Seelig, R. A. 1969. Bananas. United Fresh Fruit and Vegetable Association, Washington, DC.

Serrao, E. S. A. & J. M. Toledo. 1990. The search for sustainability in Amazonian pastures. Pp. 195-214 in A. B. Anderson (ed.), Alternatives to deforestation: Steps toward sustainable use of the Amazon rain forest. Columbia Univ. Press, New York.

Silver, W. L., R. Ostertag & A. E. Lugo. 2000. The potential for carbon sequestration through reforestation of abandoned tropical agricultural and pasture lands. Restoration Ecology 8: 394-407.

Singh, K. D. 1993. The 1990 tropical forest resources assessment. Unasylva 174: 10-19.

Skole, D. & C. Tucker. 1993. Tropical deforestation and habitat fragmentation in the Amazon: Satellite data from 1978 to 1988. Science 260:1905-1910.

Slocum, M. 2000. Logs and fern patches as recruitment sites in a tropical pasture. Restoration Ecology 8: 408-413.

Stayer, C. 1989. Why farmers rotate fields in maize-cassava-plantain bush fallow agriculture in the wet Peruvian Amazon. Human Ecology 17: 401-426.

Strong, D. R., J. H. Lawton & T. R. E. Southwood. 1984. Insects on plants: Community patterns and mechanisms. Blackwell Scientific Publications, Oxford.

Thomlinson J. R., M. L Serrano, T. del M. Lopez, T. M. Aide & J. K. Zimmerman. 1996. Land-use dynamics in a post-agricultural Puerto Rican landscape (1936-1988). Biotropica 28:525-536.

Tilman, D. 1988. Plant strategies and the dynamics and structure of plant communities. Princeton Univ. Press, Princeton, NJ.

Tinker, P. B., J. S. Ingrain & S. Struwe. 1996. Effects of slash-and-burn agriculture and deforestation on climate change. Agriculture, Ecosystems and Environment 58: 13-22.

Trumbore, S. E., E. A. Davidson, P. B. Camargo, D. C. Nepstad & L. A. Martinelli. 1995. Belowground cycling of carbon in forests and pastures of eastern Amazonia. Global Biogeochemical Cycles 9:515-528.

Uhl, C. 1987. Factors controlling succession following slash-and-burn agriculture in Amazonia. Journal of Ecology 75: 377-407.

--, R. Buschbacher & E. A. S. Serrao. 1988. Abandoned pastures in eastern Amazonia, I. Patterns of plant succession. Journal of Ecology 73: 663-681.

Veldkamp, E. 1994. Organic carbon turnover in three tropical soils under pasture after deforestation. Soil Society of America Journal 58:175-180.

Vengris, J. 1953. Weed populations as related to certain cultivated crops in the Connecticut River Valley, Mass. Weeds 2: 125-134.

Vieira, I. C. G., C. Uhl & D. Nepstad. 1994. The role of the shrub Cordia multispicata Cham. as a succession facilitator in an abandoned pasture, Paragominas, Amazonia. Vegetatio 115: 91-99.

Vitousek, P. M. 1984. Litterfall, nutrient cycling and nutrient limitation in tropical forests. Ecology 65: 285-298.

-- & R. C. Sanford. 1986. Nutrient cycling in moist tropical forest. Annual Review of Ecology and Systematics 17:131-167.

Waller, G. R., D. Kumari, J. Friedman, O. Friedman & C. H. Chou. 1986. Caffeine autotoxicity in Coffea arabica L. Pp. 243-269 in A. R. Putnam & C.-S. Tang (eds.), The science of allelopathy. Wiley, New York.

Webb, L. J., J. G. Tracey & W. T. Williams. 1972. Regeneration and pattern in the subtropical rain forest. Journal of Ecology 60: 675-595.

Wellman, F. L. 1961. Coffee: Botany, cultivation and utilization. Interscience Publishers, New York.

Whiteman, P. C. 1980. Tropical pasture science. Oxford Univ. Press, Oxford.

Zahawi, R. A. & C. K. Augspurger. 1999. Early plant succession in abandoned pastures in Ecuador. Biotropica 31:123-129.

Zimmerman, J. K., J. B. Pascarella & T. W. Aide. 2000. Barriers to forest regeneration in an abandoned pasture in Puerto Rico. Restoration Ecology 8: 350-360.

Zou, X. & G. Gonzalez. 1997. Changes in earthworm density and community structure during secondary succession in abandoned tropical pastures. Soil Biology and Biochemistry 29: 627-629.

RANDALL W. MYSTER

Department of Biology
University of Central Oklahoma
Edmond, OK 73034, U.S.A.