SEARCH

SEARCH BY CITATION

Keywords:

  • Atta spp. ;
  • disturbances;
  • gaps;
  • leaf-cutting ant nests;
  • tropical rain forests

Abstract

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

We discuss the role of leaf cutting ant nests (Atta sp.) on the formation of understory “bottom–up” gaps in tropical rain forests. We (i) describe the characteristics of Atta nests that qualify them as promoters of understory gaps; (ii) review the effects of Atta nests on the dynamics of systems other than tropical forests; and (iii) discuss the hypotheses that in neotropical rain forests Atta nests affect plant productivity, and also influence distribution and community composition of the flora. Finally, we discuss the importance of Atta nests in the dynamics of forests and compare their effects with the better-known effects of treefall gaps. Atta nests modify soil structure, soil nutrient content, light intensity at ground level, seed bank composition and understory structure. The presence of the nest and associated refuse also accelerates the soil nutrient cycles. The size, high frequency, turnover rate and duration of these nests could make them a good substrate for particular species of plants. In spite of their potential importance, little information exists on the effects of Atta nests on plant productivity and/or plant composition in neotropical rain forests.

“…Sometimes things are ahead and sometimes they are behind, sometimes one is up and sometimes down…”

Lao tsu, “Tao Te Ching”

In neotropical rain forests, treefalls are dominant examples of gaps, and the most studied. However, there are gaps made by small-scale disturbances other than treefalls that can be important in these systems. We believe that leaf-cutting ant nests are an under-appreciated source of such gaps in neotropical forests.

A gap is created when a disturbance clears space and promotes changes in resource availability. If different agents cause different types of gaps, it would follow that not all gaps are alike, and so their possible effects on neotropical forest dynamics would differ. Gap openings may originate at and reach different levels of the forest, affecting the type and intensity of resource modification. If we were to imagine a vertical section of the forest, our traditional concept of gap results in a disturbance that arises from the upper layer and reaches understory levels ( Fig. 1a). For example, treefalls create canopy gaps and thereby bring light down toward the forest floor ( Chazdon & Fetcher 1984). Many studies show how these “top–down” gaps affect forest structure, composition and dynamics ( Denslow 1980; Hartshorn 1980; Brokaw 1982, 1985; Sanford et al. 1986 ; Young & Hubbell 1991), and give gaps definitions such as “a hole in the forest extending through all levels down to an average height of two meters above ground” ( Brokaw 1982). But what about an opening that originates in the understory and has its most important effects below those two meters, or a “bottom–up” gap ( Fig. 1b)? Many neotropical animals cause “bottom–up” gaps through soil disturbance, including agoutis (Dasyprocta agouti), pacas (Agouti paca), wild pigs (Tayassu), armadillos (Dasypus novemcinctus), termites and leaf-cutting ants ( Clark 1990; Dangerfield et al. 1998 ). Of all of the above, leaf-cutting ants (mainly Atta spp.) create important soil disturbances, acting as ecosystems engineers (sensuJones et al. 1997 ). When Atta ants construct and maintain their nests, they create understory gaps by covering the forest floor with large quantities of excavated soil, clearing the area of understory vegetation and modifying the availability of resources ( Haines 1978; Alvarado et al. 1981 ; Cherrett 1989). When an Atta colony dies or abandons a nest, the area is no longer maintained free of plants, and thus becomes an important site for recolonization.

image

Figure 1. .  Schematic representation of “top–down” gaps (A) and “bottom–up” gaps (B) in a vertical cross-section of forest. Possible examples are (A) treefall and (B) abandoned or dead Atta ant nest. Light areas represent area of gap, dark areas represent forest.

Download figure to PowerPoint

A major question that remains unanswered is exactly how do Atta nests affect forest structure, composition and dynamics. Despite the fact that Atta nests, like treefall gaps, are common and conspicuous structural components of neotropical rain forests, data do not currently exist to answer this question. We begin to fill in this information gap from the bottom up by (i) summarizing the existing knowledge of Atta nests that demonstrates their role as understory gaps; (ii) reviewing the effects of Atta nests on the dynamics of tropical systems other than tropical rain forests; (iii) discussing the possible effects of Atta nest disturbances on neotropical rain forest dynamics, and (iv) comparing their effects on neotropical rain forest dynamics with the better-known effects of treefalls. We believe that, like treefalls, dead or abandoned Atta nests have an important influence on tropical rain forest dynamics and composition.

Leaf-cutting ant nests as disturbances that create gaps in forest habitats

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

Size, intensity and frequency are characteristics of disturbances that are relevant to plant responses ( Bazzaz 1983). If leaf-cutting ant nests are disturbances, then nest characteristics such as dimension, number per ha, demography and type of environmental modifications could likewise affect plant response.

Ant activities and nest characteristics

Atta nests are formidable structures that may house millions of ants and their fungus colonies for 10–20 years ( Weber 1982; Cherrett 1989). During the construction, expansion and lifetimes of Atta nests, there are at least three colony activities that are relevant to our discussion. First, leaf-cutting ants maintain an open area free of plants and debris. Attacephalotes, A. colombica, A. laevigata and A. sexdens (the principal tropical Atta species), can maintain an open area of 50–160 m2 ( Alvarado et al. 1981; Weber 1982; Cherrett 1989; Farji-Brener & Silva 1995a, b; personal observations). Secondly, leaf-cutting ants construct deep underground chambers and move large quantities of soil to the nest surface. An adult nest can contain 300–7800 chambers from near the surface to 7 m depth ( Stahel & Geijskes 1939; Jonkman 1979; Moreira et al. 1995 ; Forti & Moreira 1997). To construct this underground network, ants may move up to 23 m3 of soil ( Autori 1942). Finally, ants cut plant material from many plant species and process it in the nest for the cultivation of their fungus. The organic wastes produced by the ants and fungus are deposited either in underground chambers or in above-ground dumps. Autori (1942) found 296 refuse chambers with organic material at 1–6 m depth in a nest of Attasexdens. In an Attalaevigata nest, Coutinho (1984) found a large cavity containing organic matter 6 m below the surface, and in a nest of Attacapiguara, Amante (1964, 1967) discovered a chamber containing 500 kg of organic waste. Haines (1975, 1978) found that, unlike the above Atta spp., Attacolombica places its wastes in dumps on the surface.

Environmental modifications

Colony activities described in the preceding paragraph produce physical and chemical modifications in the nest area; such modifications may influence forest regeneration when the colony dies or the nest is abandoned ( Jonkman 1978; Bucher 1982; Weber 1982; Cherrett 1989). For example, by maintaining an open area, the ants can change light incidence at ground level and increase the availability of “unoccupied” space (see Appendix 1 and 2). Ant activities also result in changes in the physical properties of the soil (i.e. profile composition), and can promote an increment in soil porosity and density. In Costa Rican rain forests, Alvarado et al. (1981 ) found that Attacephalotes modified 85% of 27 soil profiles sampled in a 2.5 ha area.

Chemically, the ants modify the environment by concentrating nutrients and accelerating nutrient cycles in the nest area. In tropical parkland savannas, soils from Atta laevigata nests were 30%–60% higher in N, Mg, Ca and organic carbon than adjacent soils ( Farji-Brener & Silva 1995a, b). In tropical rain forests, samples from Atta colombica surface refuse were 40%–120% richer in N, P, K and S than leaf litter ( Haines 1978) and 1.3–80 times richer in N, P, K, Mg and Ca than forest soil ( Haines 1975). Additionally, for external refuse of Atta colombica, nutrient flow is 16–98 times faster than for leaf litter controls ( Haines 1978). With the return of organic matter back to the forest in the form of refuse, A. colombica has a net positive effect on forest energy balance ( Lugo et al. 1973 ).

Nest distribution and demography

The importance of the physical and chemical modifications created by a disturbance is directly related to the disturbance’s intensity and frequency, which, in the case of Atta nests, can be equated to nest density and demography. Nest density varies with habitat characteristics and with Atta species (0.04–16 nests ha−1). Generally, nest density is higher in second-growth forests and disturbed areas than in old-growth forests or undisturbed areas ( Haines 1978; Fowler et al. 1986 ; Cherrett 1989; Jaffe & Vilela 1989; Vasconcelos & Cherrett 1995). For example, in primary forests of Amazonia, densities of Attacephalotes nests are only 0.05 ha−1 ( Jaffe & Vilela 1989). In secondary forests of the Brazilian Amazon, Attalaevigata, A. cephalotes and A. sexdens can have nest densities 30 times higher than in primary forests ( Vasconçelos & Cherrett 1995). While Attacephalotes nest densities reach 3 ha−1 in Costa Rican old-growth rain forest, they reach 16 nests ha−1 in 10-year second-growth forest, and 7 nests ha−1 in disturbed forest areas along trails ( Perfecto & Vandermeer 1993; Bianchi 1998). Similarly, on Barro Colorado Island in Panama, Attacolombica nest density is around five times higher in disturbed areas near the laboratories than in less disturbed areas farther away (Farji-Brener, unpublished data).

Mortality and natality rates for adult nests are not well documented. The only study from tropical rain forest areas showed that 33 of 74 adult Attacephalotes nests died and 66 new nests appeared over a 2 year period in a 15 ha area ( Perfecto & Vandermeer 1993). Additionally, abandonment rates can be high. On Barro Colorado Island, 10% of Attacolombica colonies (four out of 40) abandoned their nests and built new ones during a 1-year period (H. Herz, personal communication).

Leaf-cutting ant nests affect vegetation dynamics of other systems

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

Many studies in nontropical rain forest ecosystems have shown that Atta nests can favour a set of plants different from that of the dominant vegetation. In Paraguayan savannas and Argentinean open woodlands, abandoned or dead Attavollenweideri nests are colonized largely by woody Prosopis spp. The ant nest is a favourable environment for these seedlings because it is free of competing grasses, rich in nutrients and higher in water content than the surrounding environment ( Jonkman 1979; Bucher 1982). In Venezuelan tropical savannas, Attalaevigata favours the germination and establishment of the tree Tapirira velutinifolia in their nest area by accumulating seeds in the nutrient rich soil of the nest, and by removing the pulp from the seeds, which prevents their infection by pathogens ( Farji-Brener & Silva 1996). In the Brazilian savannas, the exotic grass Melinis minutiflora mainly colonizes the open spaces provided by the Attalaevigata nests ( Coutinho 1982).

Ant nest–plant relationships can have repercussions at the landscape level. Abandoned or dead nests of Attavollenweideri become “woody nuclei”, accelerating the invasion of woody plants into the Paraguayan savannas and transforming them into open woodlands ( Jonkman 1978). Living Attalaevigata nests and some tree species have a synergistic relationship in Venezuelan parkland savannas, encouraging the presence of forest groves in a herbaceous matrix ( Farji-Brener & Silva 1995a).

Effects of leaf-cutting ant nests on neotropical rain forest dynamics

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

In neotropical forests, abandoned or dead Atta nests appear frequently and present a set of conditions favourable for plant establishment and growth that are uncommon in the forest understory: large free space, higher light incidence at ground level and greater nutrient concentrations. Given observations of the impact of Atta nests on other tropical or subtropical systems, we hypothesize that Atta nests can affect plant productivity, and species distribution and/or composition. We present preliminary evidence that supports these two nonexclusive hypotheses and we propose easily verified predictions. Finally, we discuss the possible ecological or evolutionary importance of Atta nests in tropical forest dynamics.

Hypothesis 1: Atta nests increase plant productivity

There is some preliminary evidence that abandoned or dead Atta nests affect plant productivity. As discussed earlier, Atta nest soils and/or external refuse dumps can be 1.3–80.0 times higher in nutrients than adjacent soils, an effect that remains after the colony dies or the nest is abandoned. Given that in forest soils of tropical America nutrients can limit plant growth ( Vitousek 1984; Sollins 1998), the increased concentration caused by the ant nests should have an important effect on plants.

Studies have demonstrated that plants can access the increased nutrients in the nest area. Density of fine roots is 4-fold higher in the external refuse dumps of Atta colombica than in control soils ( Haines 1978; Farji-Brener & Medina 2000). Tree roots were also found in subterranean refuse chambers of Attacephalotes and A. sexdens 3–6 m below ground ( Stahel & Geijskes 1939). Presence of roots in refuse dumps may increase soil nutrient transfer to the plant. Some studies of other non-Atta ant species show that plants growing in nutrient-rich nest areas increase their production of fruits and/or seeds ( Rissing 1986; Hanzawa et al. 1988 ).

Although ant nests enrich the soil, and plants can access these nutrients, does this result in an increase of the growth, survival and/or reproduction of plants that recolonize the nest area? Data do not currently exist to answer this question. If the answer is yes, we would expect plants on abandoned or dead nests to (i) grow faster, (ii) accumulate greater biomass, and/or (iii) produce more flowers/fruits and flowers/fruits for longer periods of time than plants in forest areas away from Atta nests. To test these predictions, field and greenhouse experiments measuring growth rates, total biomass and fitness for plants in nest soils and control soils would complement comparative field measurements.

Hypothesis 2: Atta nests affect plant species composition and/or distribution

The environmental conditions that Atta nests modify may be more important for some plant species or species assemblages than others. For example, the higher light intensity at ground level would especially favour species requiring more light for establishment and initial growth, such as small-seeded plants with few energy reserves. The litter layer may act as inhibitory light filter (Vazquez-Yanes et al. 1990), and thus the diminished leaf litter barrier in nest areas could favour certain species that need cleared soil for germination and/or establishment. Many seeds of tropical plant species were capable of germinating in response to disruption of the overlying litter layer and disturbance of the soil without simultaneous opening of the canopy ( Metcalfe & Turner 1998). The modified structure of the soil and the higher level of nutrients in the nest area could also affect species composition.

Possibly more important, Atta species can selectively collect fruits and seeds of tropical trees and carry them to their nests, thereby concentrating potential colonizers of specific plant species in the nutrient-rich nest area ( Roberts & Heithaus 1986; Kaspari 1993). The fate of these seeds, however, may vary with Atta species, because different species may dump seeds in different locations above or below ground. For example, in Panamanian forests, Attacolombica is one of the main dispersers of Miconia argentea because it actively harvests fruits and dumps their seeds in above-ground refuse piles ( Dalling & Wirth 1998). Miconia argentea also composed 86% of the total seed number found in refuse samples, making the nest area much less diverse than forest control soils in terms of seed-rain species. Greenhouse experiments showed that these seeds germinate in the refuse, suggesting that they are viable recolonizers of dead or abandoned Atta nests ( Farji-Brener & Medina 2000). In parkland savannas of Venezuela, Attalaevigata ants harvest fruits of Tapirira velutinifolia trees and carry them to the nest, where they strip the seeds of pulp. Stripped seeds, which germinate better than control seeds in laboratory experiments, are left on the nutrient-rich nest surface and pulp is carried below ground. The consequence of this ant’s activity is that forest groves with Attalaevigata nests have higher densities of Tapirira velutinifolia trees than forest groves without A. laevigata nests ( Farji-Brener & Silva 1996). However, some Atta species accumulate seeds together with other organic refuse in subterranean chambers, where it is less probable that the seeds germinate and colonize the abandoned nest area (but see Leal & Oliveira 1998). Atta nests can also affect the abundance and diversity of small understory plants < 10 cm in height. Garrettson et al. (1998 ) found that while active nests of Attacephalotes in Costa Rica have lower morphospecies richness and abundance of small understory plants relative to nearby forest, abandoned nests show 58% higher morphospecies richness and 73% greater abundance than adjacent forest.

Thus, Atta nests affect environmental conditions in forest understory, and ants can concentrate viable seeds of certain species in the nutrient-rich nest area. This can result in a change in species abundance and richness of small understory plants. Yet, does this significantly affect the plant species composition and/or distribution in neotropical rain forests? Again, data do not currently exist to answer this question. If the answer is yes, then we would expect to find different plant species or species assemblages on dead or abandoned nests than in adjacent forest areas. To test these predictions, field and greenhouse experiments measuring germination, establishment and growth of different plant species or mixed species assemblages in nest and non-nest conditions would complement comparative field measurements. In this sense, the descriptive results discussed above from Garrettson et al. (1998 ) are a good start.

Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

The importance of Atta nests can be better understood if compared with the well-known effects of treefall gaps. The purpose of utilizing treefall gaps as a reference is not to determine which type of disturbance is more important, rather it is to help to understand what classes of plants the nests could affect, and therefore how Atta nests affect neotropical rain forests. Compared to treefall gaps, a dead or abandoned Atta nest offers plants more space clearly of understory vegetation, higher nutrient concentrations and–in some circumstances–similar light levels at ground level ( Table 1, Appendix 1 and 2). For example, mean irradiance (and ranges) was 3–8 μmol m−2 s−1 for intact forest (0.5–22), 6–13 μmol m−2 s−1 for Atta nests (0.5–40), and 6–21 μmol m−2 s−1 (0.1–100) for treefall gaps, suggesting that irradiance in a recently abandoned or dead Atta nest can be as high as in some treefall gaps (Appendix 1, see also Chazdon & Fetcher 1984). Light intensity at ground level in a treefall gap may not necessarily be greater than in abandoned or dead Atta nests because the fallen tree and debris that it carries with it may cover the ground much more completely than a more diffuse undisturbed canopy. It is true that the light intensity above the gap may be much greater, but at ground level the treefall debris can block much of this light. However, in between the pieces of debris, there are also spots that receive almost full light. Therefore, treefall gaps promote a heterogeneous and (by mean) higher light level on the ground, while the top of an Atta nest offer a spatially more homogeneous high-intermediate light level. In addition, light levels in recently abandoned or dead Atta nest are among those levels at which pioneers tropical trees regenerate (Vazquez-Yanes et al. 1990; Veneklaas et al. 1998 ; Agyeman et al. 1999 ; Kyereh et al. 1999 ). Pioneer species are likely to remain dormant in the forest understory when the light is low, but can germinate in even small gaps with only 5 μmol m −2 s −1 ( Kyereh et al. 1999 ). This irradiance level is rarely found on the forest floor but is common at the surface of Atta nests (Appendix 1). The unique characteristics of Atta nests (space without a litter layer, higher nutrient concentrations and higher light levels at ground level) could be important for those plant species groups whose growth and development are strongly limited by the above-mentioned factors. For example, just as treefall gaps aid the long-term growth of seedlings of shade-intolerant species, it could be that Atta nests favour the germination and initial growth of intermediate shade-intolerant plant species, nutrient-limited species and/or litter-gap demanders (sensuMetcalfe & Turner 1998). Moreover, the turnover rates of Atta nests are relatively similar to those of treefall gaps ( Table 1), suggesting that Atta nests would likewise affect forest dynamics and/or composition. However, since Atta nests and the larger, older trees likely to cause gaps are most numerous in forests at different successional stages, they each may have their greatest influence at different times during the development of a forest. So the present hypothesis about abandoned nests of Atta ants as an important type of understory gaps would apply principally for secondary forests and/or at sites of earlier disturbances.

While it is generally agreed that the effects of treefall gaps are important at an ecological scale, it is debatable whether they are important at an evolutionary scale ( Denslow 1980; Whitmore 1989). We believe that Atta nests raise the same controversy: are Atta nests important agents for natural selection for some plant species, or are the previously discussed effects important only at an ecological scale? The first step toward understanding this would be to determine if the modifications generated by the nests are taken advantage of by a random selection of plant species from the surrounding matrix or by a consistently repeating set of species. If Atta nests are an important selective force, then certain plant species or phenotypes of a species should (i) depend disproportionately on the conditions generated by the nests for survival or improvement of fitness, or (ii) be selectively harvested by the ants for their seeds, which could increase the seeds’ viability and deposit them in sites that favour or improve their germination, establishment and survival rates, and (iii) possess dependent or beneficiary relationships with the ant nest sufficiently prolonged and consistent to allow for the action of natural selection.

Independent of the existence of effects at an evolutionary scale, Atta nests are involved in various ecological processes that could affect the productivity and composition of neotropical forests. Parts 2 and 3 above provide many examples of ecological importance. In addition, by increasing spatial heterogeneity, Atta nests can positively influence the species richness of plants and animals in an area ( Cherrett 1989; Garrettson et al. 1998 ). Although not described here, Atta nests can act as feeding, nesting and/or reproductive areas for certain animal species. For example, at least 111 animal species are found associated with Atta nests, and at least 30% have narrow associations with them ( Weber 1982; Cherrett 1989; Della Lucia et al. 1993 ). Furthermore, various species of reptiles use Atta nests as egg-laying sites ( Azevedo-Ramos & Mountinho 1994).

Stimulating remarks

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

Various characteristics of Atta nests could make them important influences on neotropical rain forest dynamics: (i) unlike other “bottom–up” gaps (but like treefall gaps), the size of the disturbance is large enough to stimulate plant response; (ii) they are a common component of neotropical rain forests; (iii) the nests die or are abandoned frequently (high turnover rate); and (iv) their effects on the soil are important for plants and remain after the colony dies or abandons them. The scale of these characteristics makes the abandoned or dead nests spatially, temporally and qualitatively a good “landing” for plants. We hope that the information presented in this work will stimulate future research on how beneficial this “landing” actually is.

Appendix 1

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

Light level (Mean ± 1 SD, in μmol m−2 s−1) at ground level on Attacephalotes nests, treefall gaps and adjacent forests (n = 3 in each case) measured in a cloudy day in an old-growth tropical rain forest at La Selva Biological Station, Costa Rica.

Light measurements were taken simultaneously in each time lapse on nests, treefalls and adjacent forest with light meters Extech â model 401025. Lux units were converted to μmol m−2 s−1 using the conversion factor proposed by Young et al. (1987 ). In each case, the mean value originate from 10 random points measured at 0 cm (ground level). Values between parentheses are the average coefficients of variation (%), and can be considered as an index of the spatial variation of the light level (higher values, higher spatial variation). Treefall gaps were less than 6-month-old (O. Vargas, personal communication). We used active leaf-cutting ant nests because they were clearly of understory vegetation, a similar situation that will characterize the top of a recently abandoned or dead Atta colony. In each column (separately for light levels and coefficient of variation), means which different superscripts are significantly different. *P < 0.10 for light level and +P < 0.10 for coefficient of variation (Kruskal–Wallis tests). The minimum–maximum daily values of light levels were 0.1–100, 0.5–40 and 0.5–22 μmol m−2 s−1 for tree-fall gaps, Atta nests and old-growth forest ground, respectively.

Appendix 2

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

Litter cover and litter biomass (mean ± 1 SD) at ground level on Attacephalotes nests, treefall gaps and adjacent forests (n = 3 in each case) in an old-growth tropical rain forest at La Selva Biological Station, Costa Rica.

Litter cover were measured in each nest, treefall gap and adjacent forest with three 1-m2 plots subdivided in 16 squares. In each plot we randomly choose four random subplots to measure litter biomass. In each row, values with different superscripts are significantly different. **P < 0.01, Kruskal–Wallis test.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References

To B. Loiselle, D.A. Clark, R. Montgomery, M. Aizen and J.P. Groover for important comments that improved the manuscript considerably. C. Chavez, F. Duran, E. Toral and N. Velazquez helped with the light and litter measurements in Costa Rica. We also thanks the Organization for Tropical Studies for the use of the facilities of La Selva Biological Station.

References

  1. Top of page
  2. Abstract
  3. Leaf-cutting ant nests as disturbances that create gaps in forest habitats
  4. Leaf-cutting ant nests affect vegetation dynamics of other systems
  5. Effects of leaf-cutting ant nests on neotropical rain forest dynamics
  6. Importance of leaf-cutting ant nests for neotropical rain forest dynamics: little evidence, big influence?
  7. Stimulating remarks
  8. Appendix 1
  9. Appendix 2
  10. BIOSKETCH
  11. Acknowledgements
  12. References
  • 1
    Agyeman, V., Swaine, M., Thompson, J. (1999). Responses of tropical tree seedlings to irradiance and the derivation of a light response index. J. Ecol., 87, 815 827.
  • 2
    Alvarado, A., Berish, C., Peralta, F. (1981). Leaf-cutter ant (Atta cephalotes) Influence on the morphology of andepts in Costa Rica. Soil Sci. Soc. Am. J., 45, 790 794.
  • 3
    Amante, E. (1964). Nota previa sobre a estrutura do ninho de una nova formiga sauva (Atta sp) (Hymenop., Formicidae). Biológico, 30, 96 97.
  • 4
    Amante, E. (1967). Prejuizios causados pela formiga sauva en plantacoes de Eucalyptos e Pinus no estado de Sao Pablo. Silvicultura Em Sao Pablo, 6, 355 363.
  • 5
    Autori, M. (1942). Contribucao para o conocimento da sauva (Atta spp) 3. Excavacao de un Sauverio. (Atta Sexdens). Archivos Do Instituo Biologico Do Sao Pablo, 13, 137 148.
  • 6
    Azevedo-Ramos, C. & Mountinho, C. (1994). Amphisbaenians (Reptilia: amphisbaenidae) in nest of Atta sexdens in eastern Amazonia. Brazil. Entomol. News, 18, 3 4.
  • 7
    Bazzaz, F.A. (1983). Characteristics of populations in relation to disturbance in natural and man-modified ecosystems. In: Disturbance and Ecosystems (eds Mooney, H. & Godron, M.). Springer-Verlag, New York, pp. 259 274.
  • 8
    Bianchi, G. (1998). Densidad poblacional de hormigas cortadoras (Atta cephalotes) en la reserva biológica La Selva, Costa Rica. In: Ecología tropical y conservación, 98–2. O.T.S. Course Book (eds Farji-Brener A.G. and Chinchilla F.). San José, Costa Rica, pp. 289 291.
  • 9
    Brokaw, N. (1982). The definition of treefall gap and its effect of forest dynamics. Biotropica, 14, 158 160.
  • 10
    Brokaw, N. (1985). Treefalls, regrowth, and community structure in tropical forests. In: The Ecology of Natural Disturbance and Patch Dynamics (eds Picket S. & White P.). Academic Press, San Diego, pp. 53 69.
  • 11
    Bucher, E. (1982). Chaco and Caatinga-South American arid savannas, woodlands and thickets. In: Ecology of Tropical Savannas (eds Huntley, B. & Walker, B.). Springer-Verlag, Berlin, pp. 48 79.
  • 12
    Chazdon, R. & Fetcher, N. (1984). Photosynthetic light environments in a lowland tropical rain forest in Costa Rica. J. Ecol., 72, 553 564.
  • 13
    Cherrett, J.M. (1989). Leaf-cutting ants, biogeographical and ecological studies. In: Ecosystem of the World 14b, Tropical Rain Forest Ecosystem (eds Lieth, H. & Werger, M.). Elsevier, New York, pp. 473 488.
  • 14
    Clark, D. (1990). The role of disturbance in the regeneration of neotropical moist forests. In: Reproductive Ecology of Tropical Forest Plants. Man and the Biosphere Series, 7 (eds Bawa, K. & Hadley, M.). Unesco, Paris, 291 315.
  • 15
    Coutinho, L. (1982). Aspectos ecológicos da saúva no cerrado. Os murundus de tierra, as características psamofíticas das especies de sua vegetacao e a sua invasao pelo Capim Gordura. Revista Brasilera de Biologia, 42, 147 153.
  • 16
    Coutinho, L. (1984). Aspectos ecológicos da saúva no cerrado. A saúva, as queimadas e sua possivel relacao na ciclagem de nutrientes minerais. Boletim Zoologico de la Universidad de Sao Paulo, 8, 1 9.
  • 17
    Dalling, J. & Wirth, R. (1998). Dispersal of Miconia argentea seeds by the leaf-cutting ant Atta colombica. J. Trop. Ecology, 14, 705 710.
  • 18
    Dangerfield, J., McCarthy, T., Ellery, W. (1998). The mound-building termite Macrotermes michaelseni as an ecosystem engineer. J. Trop. Ecology, 14, 507 520.
  • 19
    Della Lucia, T., Moreira, D., Oliveira, M. (1993). Inimigos naturais e organismos associados aos ninhos. In: As Formigas Cortadeiras, (eds Della Lucia, T.). Vicosa, Brazil, pp. 131 150.
  • 20
    Denslow, J.S. (1980). Gap partitioning among tropical rainforest trees. Biotropica Suppl., 12, 47 55.
  • 21
    Farji-Brener, A.G. & Sierra, C. (1993). Distribution of attacked plants along trails in leaf-cutting ants: consequences in territorial strategies. Revista Biología Tropical, 41, 891 896.
  • 22
    Farji-Brener, A.G. & Silva, J. (1995a). Leaf-cutting ants and forest groves in a tropical parkland savanna of Venezuela: facilitated succession? J. Trop. Ecology, 11, 651 669.
  • 23
    Farji-Brener, A.G. & Silva, J. (1995b). Leaf-cutting ant nests and soil fertility in a well-drained savanna in western Venezuela. Biotropica, 27, 250 253.
  • 24
    Farji-Brener, A.G. & Silva, J. (1996). Leaf-cutting ant Atta laevigata aid to the establishment success of Tapirira velutinifolia (Anacardiaceae) seedlings in a parkland savanna. J. Trop. Ecology, 12, 163 168.
  • 25
    Farji-Brener, A.G. & Medina, C. (2000). The importance of where to dump the refuse: seed banks and fine roots in nests of the leaf-cutting ants Atta cephalotes and Atta colombica. Biotropica, 32(1), 120 126.
  • 26
    Forti, L. & Moreira, A. (1997). Architecture of Atta nests. In: Mirmecologia Neotropical. Anais Do VI International Pest Ant Symposium (eds Delabie, J., Campiolo, S., Do Nascimiento, I. & Ferreira, C.). Ilheus, Brazil, pp. 91.
  • 27
    Fowler, H., Da-Silva, L., Forti, C., Saes, N. (1986). Population dynamics of leaf-cutting ants. In: Fire and Leaf-Cutting Ants: Biology and Management (eds Logfren, S. & Vander Meer, R.). Westview Press, Boulder, CO, pp. 123 145.
  • 28
    Garrettson, M., Stetzel, J., Halpern, B., Hearn, J., Lucey, B., Mckone, M. (1998). Diversity and abundance of understory plants on active and abandoned nest of leaf-cutting ants (Atta cephalotes) in Costa Rica rain forest. J. Trop. Ecology, 14, 17 26.
  • 29
    Haines, B. (1975). Impact of leaf-cutting ants on vegetation development at Barro Colorado Island. In: Tropical Ecological Systems, Trends in Terrestrial and Aquatic Research (eds Goley, F. & Medina, E.). Springer-Verlag, New York, pp. 99 111.
  • 30
    Haines, B. (1978). Element and energy flows through colonies of the leaf-cutting ant, Atta colombica, in Panama. Biotropica, 10, 270 277.
  • 31
    Hanzawa, F., Beattie, A., Culver, D. (1988). Directed dispersal: demographic analysis of an ant-dispersed mutualism. Am. Naturalist, 131, 1 13.
  • 32
    Hartshorn, G. (1976). Tree falls and tropical forest dynamics. In: Tropical Trees as Living Systems (eds Tomlison, P. & Zimmerman, M.). Cambridge University of Press, New York, pp. 617 630.
  • 33
    Hartshorn, G. (1980). Neotropical forest dynamics. Biotropica Suppl., 12, 23 30.
  • 34
    Jaffe, K. & Vilela, E. (1989). On nest densities of the leaf-cutting ant Atta cepahlotes in tropical primary forest. Biotropica, 21, 234 236.
  • 35
    Jones, C., Lawton, J., Shachak, M. (1997). Positive and negative effects of organisms as physical ecosystem engineers. Ecology, 78, 1946 1957.
  • 36
    Jonkman, J. (1978). Nests of the leaf-cutting ant Atta vollenweideri as accelerators of succession in pastures. Z Angewandte Entomologie, 86, 25 34.
  • 37
    Jonkman, J. (1979). Distribution and densities of nest of the leaf-cutting ant Atta vollenweideri Forel in Paraguay. Z Angewandte Entomologie, 88, 27 43.
  • 38
    Kaspari, M. (1993). Removal of seeds from Neotropical frugivore droppings. Ants response to seed number. Oecologia, 95, 81 88.
  • 39
    Kyereh, B., Swaine, M., Thompson, J. (1999). Effect of light on the germination of forest trees in Ghana. J. Ecol., 87, 772 783.
  • 40
    Leal, I. & Oliveira, P. (1998). Interactions between fungus-growing ants (Attini), fruits and seeds in cerrado vegetation in southeast Brazil. Biotropica, 30, 170 178.
  • 41
    Levey, D.J. (1988). Tropical wet forest treefall gaps and distributions of understory birds and plants. Ecology, 69, 1076 1089.
  • 42
    Lugo, A., Farnworth, E., Pool, D., Jerez, P., Kaufman, G. (1973). The impact of the leaf cutter ant Atta colombica on the energy flow of a tropical wet forest. Ecology, 54, 1292 1301.
  • 43
    Metcalfer, D. & Turner, I. (1998). Soil seed bank from lowland rain forest in Singapure: canopy-gap and litter-gap demanders. J. Trop Ecology, 14, 103 108.
  • 44
    Moreira, A., Vitorio, A., Forti, L. (1995). Arquitetura da colonia da sauva Atta laevigata (Hymen., Form.). In: Anais XII Encontro de Mirmecologia. Brazil, pp. 67.
  • 45
    Orians, G. (1982). The influence of tree-falls on tropical forests species richness. Trop. Ecol., 23, 255 279.
  • 46
    Vazquez-Yanes, C., Orozco-Segovia, A., Rincon, E., Sanchez-Coronado, M., Huarte, P., Toledo, J., Barradas, V. (1990). Light beneath the litter in a tropical forest: effect on seed germination. Ecology, 71, 1952 1958.
  • 47
    Perfecto, I. & Vandermeer, J. (1993). Distribution and turnover rate of a population of Atta cephalotes in a tropical rain forest in Costa Rica. Biotropica, 25, 316 332.
  • 48
    Putz, F. (1983). Treefall pits and mounds, buried seeds, and the importance of soil disturbance to pioneer trees on Barro Colorado Island, Panama. Ecology, 64, 1069 1074.
  • 49
    Putz, F. & Milton, K. (1983). Tree mortality rates on Barro Colorado Island. In: The Ecology of Tropical Forest: Seasonal Rhythms and Long-Term Changes (eds Leigh, E., Rand, S. & Windsor, D.). Smithsonian Institution Press, Washington, D.C., pp. 95 100.
  • 50
    Rissing, S. (1986). Indirect effect of granivory by harvester ants: plants species composition and reproductive increase near ant nest. Oecologia, 68, 231 234.
  • 51
    Roberts, J. & Heithaus, J. (1986). Ants rearrange the vertebrate-generated seed shadow of neotropical fig tree. Ecology, 67, 1046 1051.
  • 52
    Sanford, R.L., Braker, E., Hartshorn, G. (1986). Canopy openings in a primary neotropical lowland forest. J. Trop. Ecology, 2, 277 282.
  • 53
    Schupp, E., Howe, H., Augspurger, C., Levey, D. (1989). Arrival and survival in tropical treefall gaps. Ecology, 70, 562 564.
  • 54
    Sollins, P. (1998). Factors influencing species composition in tropical lowland rain forest: does soil matter? Ecology, 79, 23 30.
  • 55
    Stahel, G. & Geijskes, D. (1939). Ueber den bau der nester von Atta cephalotes L. & A. sexdens L. (Hymenop., Formicidae). Revista Entomología, 10, 27 78.
  • 56
    Uhl, C., Clark, H., Dezzeo, N., Maquirrino, P. (1988). Vegetation dynamics in Amazonian trefall gaps. Ecology, 69, 751 763.
  • 57
    Vasconçelos, H. & Cherrett, M. (1995). Changes in leaf-cutting ant populations (Formicidae; Attini) after the clearing of mature forest in Brazilian Amazonia. Studies Neotrop Fauna Environ., 30, 107 113.
  • 58
    Veneklaas, E. & Poorter, L. (1998). Growth and carbon partitioning of tropical seedlings in contrasting light environments. In: Physiological Mechanisms and Ecological Consequences (eds Lambers, H., Poorter, H. and Van Vuuren, M.). Bakhuys Publishers, the Netherlands, pp. 337 361.
  • 59
    Vitousek, P. (1984). Litterfall, nutrient cycling and nutrient limitation in tropical forests. Ecology, 65, 285 298.
  • 60
    Vitousek, P. & Denslow, J. (1986). Nitrogen and phosphorus availability in treefall gaps of a lowland tropical rainforest. J. Ecol., 74, 1167 1178.
  • 61
    Weber, A. (1982). Fungus ants. In: Social Insects, Vol. 4 (ed. Hermann, H.). Academic Press, London, pp. 255 363.
  • 62
    Whitmore, T. (1989). Canopy gaps and the two major groups of trees. Ecology, 70, 536 538.
  • 63
    Young, S., David, C.T., Gibson, G. (1987). Light measurement for entomology in the field and laboratory. Physiol. Entomol., 12, 373 379.
  • 64
    Young, T. & Hubbell, S. (1991). Crown asymmetry, treefalls, and the repeat disturbance of broad-leaved forest gaps. Ecology, 72, 1464 1471.