xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00288
Abundancia de las esporas micorrizas vesiculo-arbusculares y las caractersticas del suelo a lo largo de un gradiente sucesional del bosque neotropical premontano
Vesicular-Arbuscular Mycorrhizae (VAM) spore abundance and soil characteristics along a neotropical premontane forest successional gradient
Tropical soils are generally nutrient poor, though they support high biodiversity and productivity. Most tropical plants are able to thrive in these soils because they form a symbiosis with Vesicular-Arbuscular Mycorrhizae (VAM). VAM increase the nutrient-absorption capabilities, and therefore the fitness, of the host plant. Previous studies suggest that as a pasture is allowed to regenerate, the soil should become less compact, more nutrient-rich,
and the pH should become more neutral. These soil characteristics, in conjunction with spore number, may drive the rate and direction of some successional communities following disturbance. This study examines how VAM spore abundance, macronutrient levels (nitrogen, phosphorus, and potassium), bulk soil density, and pH change between plots at various stages of regeneration from pastures (ages 0-45, and primary forest, N=15) in the Premontane Wet
Forest near San Luis, Costa Rica. The data show substantial variation within and between sites in spore number and other soil characteristics. For instance, the three pasture sites sampled varied from 4 to 83 spores ( 28), with nitrogen levels from 16.81 to 46.7 kg/ha (16.46). Within each individual site there was also variation: the 25 year old forest ranged from 46 to 132 spores ( 45). Regeneration age did not significantly affect VAM spore number,
macronutrient levels, or bulk density, though pH was positively correlated with increasing site age. Spore abundance was not significantly affected by variations in the other soil characteristics. These findings suggest that VAM spores may not be evenly distributed throughout the soil, instead having patchy distributions determined by
soil heterogeneity and location of mycotrophic hosts. VAM do not appear to be limiting in the young pastures of San Luis, suggesting that VAM are unlikely to direct forest regeneration.
Los suelos tropicales son generalmente pobres en nutrientes, aunque apoyan una alta biodiversidad y productividad. La mayora de las plantas tropicales son capaces de prosperar en estos suelos debido a que forman relaciones simbiticas con las micorrizas. Las micorrizas aumentan la capacidad de absorber los nutrientes, y por lo tanto el xito reproductivo de la planta hospedera. Estudios anteriores sugieren que conforme se deja regenerar a los pastizales, el suelo debera ser menos compacto, ms rico en nutrientes y el pH debera ser ms neutral. Estas caractersticas del suelo, en conjunto con el nmero de esporas, pueden llevar la tasa y la direccin de algunas comunidades en sucesin despus de los disturbios. Este estudio examina como la abundancia de las esporas de las micorrizas, los niveles de los macro-nutrientes (nitrgeno, fosforo y potasio), la densidad aparente del suelo y el pH cambian entre las diferentes parcelas en los pastizales en las diversas etapas de la regeneracin de los pastos (edades de 0 a 45 aos, y el bosque primario, N = 15) en el bosque premontano hmedo cercano a San Luis, Costa Rica. Mis datos sugieren una variacin sustancial entre y dentro de las parcelas en el nmero de esporas y otras caractersticas del suelo. Los tres sitios varan entre 4 y 83 esporas ( 28), con niveles de nitrgeno de 16.81 a 46.7 kg/ha (16.46). Dentro de cada sitio individual existe tambin una variacin; por ejemplo, la abundancia de esporas en el fragmento de 25 aos vara de 46 a 132 ( 45). La edad de la regeneracin no afecta el nmero de esporas de las micorrizas, los niveles de macro-nutrientes o la densidad del suelo, aunque el pH esta correlacionado positivamente al aumentar la edad de la regeneracin. La abundancia de las esporas no se ve significativamente afectada por las variaciones en las caractersticas del suelo. Estos descubrimientos sugieren que las micorrizas pueden no estar distribuidas equitativamente a lo largo del suelo, en cambio tiene una distribucin por parches determinado por la heterogeneidad y la ubicacin de los hospederos. Las micorrizas parecen no verse limitadas en los pastizales ms jvenes en San Luis, sugiriendo que ests no estn directamente relacionadas con la regeneracin del bosque.
Text in English.
Costa Rica--Puntarenas--Monteverde Zone--San Luis
Costa Rica--Puntarenas--Zona de Monteverde--San Luis
Tropical Ecology Fall 2010
Ecologa Tropical Otoo 2010
t Monteverde Institute : Tropical Ecology
Vesicular Arbuscular Mycorrhizae (VAM) Spore Abundance and Soil Characteristics along a Neotropical Premontane Forest Successional Gradient Kristin Schroder Department of Ecology and Evolutionary Biology, Department of Environmental Studies, University of Colorado at Boulder ABSTRACT Tropical soils are generally nutrient poor, though they support high biodiversity and productivity. Most tropical plants are able to thrive in these soils because they form a symbiosis with Vesicular Arbuscular Mycorrhizae (VAM). VAM increase the nutrient absorption capabilities, and therefore the fitness, of the host plant. Previous studies suggest that as a pasture is allowed to regenerate, the soil should become less compact, more nutrient rich, and the pH should become more neutral. These soil characteristics in conjunction with spore number may drive the rate and direction of some successional communities following distur bance. This study examines how VAM spore abundanc e, macronutrient levels (nitrogen, phosphorus, and potassium ), bulk soil density, and pH change between plots at various stages of regeneration from pasture (ages 0 45, and primary forest, N=15) in the Premo ntane Wet For est near San Luis, Costa Rica. The data show substantial variation within and between sites in spore number and other soil characteristics. For instance, t he three pastu re sites sampled varied from 4 to 83 spores ( 28) with nitrogen levels from 16.81 to 46.7 kg/ha ( 16.46 ). Within each individual site there was also variation : the 25 year old forest ranged from 46 to 132 spores ( 45 ). R egeneration age did not significantly affect VAM spore number, macronutrient levels, or bulk density, though p H was positively correlated with increasing site age. Spore abundance was not significantly affected by variations in the other soil characteristics These findings suggest that VAM spores may not be evenly distributed throughout the soil, instead having patchy distributions determined by soil he terogeneity and location of mycotrophic hosts. VAM do not appear to be limiting in the young pastures of San Luis, suggesting that VAM are unlikely to direct forest regeneration. RESUMEN Los suelos tropicales son generalmente pobres en nutrient e s, aunque soportan una alta diversidad y productividad. La mayora de las plantas tropicales son capaces de prosperar en estos suelos debido a relaciones simbiticas con micorrizas. Las micorrizas aumentan la capacidad de absorber nutrientes, y por lo tanto el xito reproductivo de la planta hospedera. Estudios previos sugieren que conforme se deja regenerar los pastizales, el suelo debera ser menos compacto, ms rico en nutrientes y el pH debera se ms neutro. Estas caractersticas del suelo, en conjunto con el nmero de esporas, puede llevar la tasa y direccin de algunas comunidades en sucesin despus de disturbios. Este estudio examina como la abundancia de esporas de micorrizas, niveles de macro nutrientes (N, P, K), la densidad aparente del suelo y el pH cambian entre diferentes parcelas en pastizales con diferente tiempo de regeneracin (edades de 0 a 45 aos, y bosque primario, N = 15) en el bosque hmedo premontano cercano a San Luis, Costa Rica. Mis datos sugieren una variacin sustancial entre y dentro de las parcelas en el nmero de esporas y ot ras caractersticas del suelo. L os tres sitios varan entre 4 y 83 esporas ( 28), c on niveles de nitrogeno de 16.81 a 46.7 kg/ha (16.46). Dentro de cada sitio individual existe tambin variacin; por ejemplo, la abundancia de esporas en el fragmento de 25 aos vara de 46 a 132 ( 45). La edad de regeneracin no afecta el nmero de esp oras de micorrizas, los niveles de macro nutrientes o la densidad del suelo, aunque el pH esta correlacionado positivamente al aumentar la edad de regeneracin. La abundancia de esporas no se ve significativamente afectada por las variaciones en las carac tersticas del suelo. Estos descubrimientos sugieren que las micorrizas pueden no estar distribuidas equitativamente a lo largo del suelo, en cambio tiene una distribucin por parches determinado por la heterogenidad y ubicacin de los hospederos. Las mi corrizas parecen no verse limitadas en los pastizales ms jvenes en San Luis, sugiriendo que ests no est directamente relacionadas con la regeneracin del bosque.
INTRODUCTION Deforestation rates are dramatically high in some parts of the world, such a s in South America, which suffered a net forest loss of about 4.3 million hectares per year from 2000 to 2005 (FAO ). However, other tropical areas are actually experiencing net forest gains. Such is the case in Costa Rica, which from 2000 to 2005 increas ed its total forest cover by abou t 15,000 hectares (FAO ). As Costa Rica shifts from an agriculture and cattle ranching economy to one of industry and tourism, farmland is being abandoned and allowed to regenerate into secondary forest. Though conservatio n efforts tend to focus on preserving old growth forests, secondary forests are also an asset to conservation. Many important reserves in Costa Rica, such as Santa Rosa National Park, would not exist today had it not been for the foresight of conservation ists (such as ecologist and conservationist Dr. Daniel Janzen) to purchase cheap pasture and let it regenerate naturally (UNEP ) Now, only a few decades later, these reserves act as important refuges of bi odiversity (UNEP ). Thus, from a conservation stan dpoint, it is important to understand how land naturally regenerates from anthropogenic disturbance. Soil characteristics are one especially influential ecological force affecting the rate and direction of forest succession ( Janos 1983 Neale 1997, Smith and Read 1997 ). Tropical rainforests are known for their biodiversity and high productivity. However, a shallow, nutrient poor, acidic, and phosphor us deficient (Janos 1983). luxurious growth is in part due to a very common mutualism with Vesicular Arbuscular Mycorrhizae (VAM) (Janos 1983). Though there are many species of mycorrhizae, most tropical plants associate specifica lly with species of VAM (Family Glomales, Division Zygomycota). These fungi are dependent upon their hosts for the products of photosynthesis (energy rich carbon compounds like glucose). In return, mycorrhizae provide host plants with increased nutrient and water absorption potential, thus allowing for a higher level of photosynthetic productivity (Janos 1983, Smith and Read 1997). This mutualism increases the competitive abilities of the host plant, and is therefore influential in determining dynamics of both intact, old growth forests and the trajectory of successional communities (Janos 1980). Mycotrophic hosts (plants that depend on this mutualism), may not be able to grow to maturity (or even germinate successfully, in the case of many tropical can opy trees) without the presence of fungal inocula (Janos 1983, Johnson et al. 1991). This is especially true in nutrient poor soils, like those in the Tropics, where these plants cannot absorb sufficient nutrients or water from the soil without VAM (Janos 1980). In these conditions, such plants may be easily outcompeted (at least temporarily) by non mycotrophic species, which are frequently pioneer species adapted to rapid growth in low nutrient conditions (Janos 1983). Thus, mycor rhizae spore number an d other soil characteristics l ike nutrient composition may work together to determine which plants can or cannot inhabit a regenerating area. As a forest is allowed to regenerate, the soil should become less compact and acquire more leaf litter (Guariguat a and Ostertag 2001) This leaf litter acts as a pH buffer (encouraging a more neutral pH), incr eases humus in the soil, and enriches the soil as the detritus decomposes (Guariguata and Ostertag 2001) Abandoned pastures, for example, are usually nutrient poor (as they are not norm ally fertilized), and have compacted soil due to frequent trampling by cattle (Neale 1997 ) One study in the Monteverde area found that the soils of abandoned pastures dramatically increased in their leaf li tter and humu s content and became less c ompacted over time (Neale
1997) However, some intensive land use activities, in some cases rangeland, may degrade the soil to the point where the original habitat type will never be able to regenerate (Guariguata and Ostertag 2 001 ) Not only do spore level s potentially influence the variety and quantity of plants that can live in a given area, plants may conversely affect spore abundance Because VAM produce few spores that are fragile and do not last long in the soil, if a dominantly non mycotrophic plant community inhabits an area for long enough, it may deplete the rhizosphere of spores. This, in turn, could affect the composition of future plant communities (Janos 1980). However, it is not allowed to regenerate from pasture to secondary forest, or how this impacts the quantity of VAM spore s in the soil. Janos (1980) found that poor soils usually favor either obligatory mycotrophic species dominant or non mycotrophic species dominant communities. He theorizes that if VAM are initially present in poor soils, obligatory mycotrophic speci es should be the most competitive. This would support a continuous abundance of VAM spores and ensure the long term dominance of mycotrophic species. However, if the soil is both nutrient poor and lacking VAM, it should favor non mycotrophic species, oft en pioneer species, which can out compete mycotrophic species when VAM are not present. Without a host, the VAM (as well as the short lived spores) would soon die, ensuring the continued dominance of non mycotrophic species in the area (Janos 1980). In c ontrast, Rogers (1998) hypothesized that forest succession following disturbance is not determined as much by original soil characteristics as it is by the success of pioneer species. Pioneer species are adapted to poor soils and are usually not dependent on mycorrhizae, but they may also outcompete mycotrophic plants in soil rich in nutrients and VAM spores. Maybe the trajectory of succession depends on the creation of better growing conditions through these pioneer species, which alter the microclimate they block the wind, decrease the temperature, storage potential (Guariguata and Ostertag 2001) This provides a better habitat for understory and later su ccession plants, as well for the primary dispersal agents of spores, such as burrowing vertebrates like moles and rabbits (Killham 1994, Kwan 1995, Wolf 1998). Thus, it is possible that mycorrhizal spore abundance in the soil does not affect pioneer commu nities as much as it does later successional stages. These initial conditions in nutrient levels and VAM spore abundance may result from the type of land use or the previous plant community prior to forest regeneration. For example, the farming of an o bligately mycotrophic crop like Lactuca sativa (common lettuce) could maintain a high VAM abundance in the soil (Janos 1983, Miller and Jackson 1998). Conversely, if the soil contains an excess of nutrients (such as from intensive fertilizer input), this could render abundant nutrients themselves, rather than trade away their photosynthetic products for this function. This means few VAM spores in the soil, because w ithout their host, mycorrhizae inocula soon become inviable. Pastureland is also usually found to have low numbers of VAM spores, because most grasses, though technically mycotrophic, do not frequently form VAM mycorrhizal associations, even in situations of low fertility (Johnson et al. 1991). In light of this literature, and with the premise that pastur es should be nutrient poor and have few VAM spores, I expect to find that spore number increases with regeneration as the plot is colonized first by no n mycotrophic pioneer species, which over time will attract mycorrhizal
dispersal agents and be colonized by later successional mycotrophic species. These changes should also be reflected by other changes in the soil as the area regenerates from pasturela nd, such as decreasing bulk density, increasing macro nutrient concentrations, and a more neutral pH. Alternatively, the spore abundance could stay low with regeneration time if the community continues to be outcompeted by n on mycotrophic plants, in which case the soil characteristics would remain relatively degraded and the climax community would differ from forest. METHODS I collected soil samples in the P rem ontane Neotropical Wet Forest near San Luis, Costa Rica on November 1 st 15th, 2010, at an elevation of 1055 to 1200 m. I first met with landowners in the San Luis area, as well as with staff members of the University of Georgia Research Station to I selected 13 sites of varying regeneration time (0 to 45 years) from the pasture s and secondary forest s of San Luis and on the trails around the UGA campus. I also selected three additional plots of primary forest to compare with the pastureland and seco ndary growth. I defined primary forest in this study as old growth forest for which the age is unknown and which is thought to be historically undisturbed by humans (excluding trails). y have presumably never been disturbed, it seems reasonable to assume that if the soil composition follows some trend with increasing regeneration time, then the plots would eventually converge on the p rimary forests characteristics I removed three soil samples at each site with a soil borer (volume of 365 cm 3 ) t o a depth of 15 cm after brushing away surface debris and plant material T his depth is commonly used to capture the be st representation of spor e numbers, ( Johnson et al. 1991 ) All sampl es were taken from level area s with at least 15 cm of topsoil and that were 1 0 m or more from the nearest road, stream, trail, or other habitat type. Each sample was kept separate for individual analysis. E ach soil sample was dried for 12 hours in a dry ing oven set to 63 C (145 F) and was then weighed to obtain the dry weight This was (a measure of soil compaction). I then used a wet sieving and centrifugation technique (slightly adapted from Wolf 1998) to separate the VAM spores from 7.5 grams of each soil sample. I stirred each soil sample into 500 mL of water to dislodge spores and clumps of soil, and rinsed the solution through a series of nested sieves (250 m and 125 m ) Therefore, I counted spores with a diameter of roughly 125 250 m which captures the midrange of most VAM spores (Wolf 1998). I removed and distributed the sievate with water soil sample at 3600 rpm for 45 seconds. I removed the supernatant from each centrifuge tube with a pipette, and examined it drop by drop on a Petri dish under a dissecting microscope (Figure 1) to count the number of spores. The dried soil samples were t hen tested us ing a LaMotte soil testing kit for pH and for the major soil macronutrients nitrate nitrog en, potassium, and phosphorus
RESULTS VAM Spore Abundance and Chemical Analysis The number of spores in each 7.5 g subsample varied substantially within site s (mean spore number 54 22 N = 15 ), thus for each site I averaged the VAM spore numbers from each of the three separate soil samples for my calculations (see Table 1). The sites ranged from an average spore count of about 4 to 97 spores, with the lowest from a pastureland site, and the highest found in the soil of 25 year old pastureland re growth. These averaged spore numbers also varied substantially between the three separate pastureland and primary forest sites, respectively (the average spore count in pastureland ranged from 4 to 83 spores, and in primary forest from 33 to 71 spores). Spore number was not significantly affected by regeneration age (regression, p FIGURE 1. Three Vesicular Arbuscular Mycorrhizae (VAM) spores, circled in red, as seen under a dissecting microscope (magnification 40X). Most neotropical plants depend on their association with VAM to overcome the nutrient deficiency of tropical soils. These spo res were extracted from 7.5 g dry weight of soil from a regenerating pasture near San Luis, Costa Rica using a wet sieving and centrifugation method.
= 0.6382, R 2 = 0.0174, N=13). A correlation table (Table 2) was constructed to determine how the physical and chemical soil characteristics were inter related. However, none of the factors, except regeneration age with pH and K with bulk density, showed any correlation (p valu es < 0.05). Spore number was not significantly affected by bulk soil density (p=0.2513, R 2 =0.1291), N (p=0.8423, R 2 =0.0042), P (p=0.7814, R 2 =0.0081), K (p=0.9785, R 2 =0.0001), or pH (p = .2556, R 2 = 0.127; multiple regression, N=13). TABLE 1. Average s oil c haracteristics of the 15 study site s of pasture, regenerating forest, or primary forest near San Luis, Costa Rica. N=3 for each average. Total mean spore abundance = 54 22 per 7.5 g dry weight of soil; total mean bulk density = 0.46 0.09 g/cm 3 ; total mean N = 35.62 16.46 g/ha; total mean P = 108.73 45.83 g/ha; total mean K = 0.56 0.11 mL 1 total mean pH = 6.23 0.3. Site Age/ Type Avg. Spore # Avg. Bulk Soil Density (g/cm 3 ) Avg. N (kg/ha ) Avg. P (kg/ha ) Avg. K ( mL 1 )* Avg. pH Pasture 1 3.67 0.50 16.81 106.48 0.44 5.43 Pasture 2 83.33 0.38 14.95 78.46 0.50 6.17 Pasture 3 62.00 0.59 46.70 117.69 0.65 5.83 5 years 26.33 0.62 26.15 74.73 0.59 6.27 10 years 81.33 0.48 44.84 37.36 0.54 6.30 15 years 24.33 0.41 18.68 102.75 0.47 6.67 18 years 56.33 0.44 59.78 186.82 0.66 6.30 20 years 42.67 0.40 63.52 46.70 0.47 6.23 25 years 97.00 0.32 13.08 140.11 0.43 6.53 30 years 71.33 0.53 44.84 121.43 0.54 6.30 37 years 40.67 0.45 37.36 95.28 0.59 6.40 45 years 50.67 0.58 39.23 84.07 0.83 6.33 Primary 1 33.00 0.33 18.68 168.13 0.44 6.50 Primary 2 71.67 0.44 44.84 84.07 0.62 6.10 Primary 3 66.33 0.38 44.84 186.82 0.64 6.10 The measurement for K ( mL 1 ) is an expression of relative abundance based on the soil testing procedure in the LaMotte Soil Testing Kit This does not directly correlate to lbs/acre of K at each site.
TABLE 2. Correlations between VAM spore number, regeneration age, and the other soil properties of 13 pasture and secondary forest sites. Only site age with pH and bulk density with K were significantly correlated (p < 0.05). Spore abundance was not significantly correlated with any other soil characteristic. Spore # Age Bulk Density N P K pH Spore # -------Age + 0.13 ------Bulk Density 0.35 0.05 -----N + 0.06 + 0.26 + 0.24 ----P + 0.09 + 0.10 0.16 0.01 ---K + 0.02 + 0.43 + 0.62* + 0.42 + 0.11 --pH + 0.36 + 0.55 0.37 + 0.02 + 0.01 + 0.05 -* indicate s p < 0 .05 FIGURE 2. Number of average VAM spores per 7.5 gram subsample with increasing natural regeneration time since pastureland. Spore number does not significantly increase with time (p = 0.6832, R 2 = 0.0174, N = 13, line of best fit: y = 0.2406 x + 49.195). 0 20 40 60 80 100 120 0 10 20 30 40 50 Average Number of Spores per 7.5 g dry weight Regeneration Time (yrs)
Qualitative Observations I observed a typical pattern of Neotropical community succession (previous personal observations), with grass dominated pastureland first being colonized by ground sprawling vines, small shrubs and herbaceous plants, and Cecropia and early successional tre e seedlings. This seemed to progress to dominance by Heliconia and vines as the seedlings increased in height. The understory continued to grow denser with Melastomataceae and aroid species, and some understory palms. Finally, the older sites seemed to be dominated by large trees (and no Cecropia ), with lianas, Piperaceae, Melastomataceae, ferns and understory palms present I also observed that pasture soil samples were extremely muddy and compacted from being trampled by cows, only covered in a thin layer of grass and little humus. Humus and leaf litter seemed to increase with regeneration time, and compaction seemed t o decrease, though level ing out at a mid regeneration time of approximately 25 years. This qualitative observation, therefore, contr adicts my measure of compaction (bulk density) for which there was no trend with increasing regeneration age. I did not notice any large differences in macro or micro invertebrate soil composition, except that the pasture soils seemed to contain more ann elids. DISCUSSION Trends in age were difficult to discern because of high within site and between site variability in spore number. This was obvious by looking at the three pasture sites (average spore numbers: 4 83, average N: 14.91 46.7 kg/ha etc). Moreover, the sites sometimes varied dramatically (such as the 25 year old secondary forest, which had from 46 132 spores) between the three samples at each site, demonstrating that even one site with the same regenerational history was heterogeneous in the number of spores it contained. Because of this soil heterogeneity, the data collected from each site could have reflected different regenerational trajectories, rather than a point on a single successional path. A longer term experiment would have be en able to judge the interaction of mycorrhizae and nutrient composition along a successional gradient while holding all baseline conditions constant (rather than assuming these starting conditions to be equivalent). In terms of my predicted results, the se observations support neither the hypothesis that VAM spores were consistently uncommon due to the dominance of non mycotrophic pioneer species, nor the hypothesis that spore numbers increased as mycotrophic species colonize d the area. There was no trend between number of spores and regeneration age, and it seems that soil compaction, N, P, K, and pH are not acting as confounding variables. It is possible that these trends do exist in the natural regeneration of pastureland, but that the starting spore or nutrient levels were different in the in itial pastureland of each site. However, the macronutrient content and the bulk density of the soil also did not follow a trend with regeneration age. Though I qualitatively observed an increase in leaf litter and humus with regeneration, and with it a decrease in soil compaction, this was not represented in the data. The bulk soil density may be a flawed measure of soil compaction, as a more sandy or heavy soil would have a greater mass per volume of the soil borer, possibly obfuscating the results. PH did increase with regeneration time (becoming more neutral), which may reflec t this increase in leaf litter, which sometimes acts as a pH buffer Ultimately, the soil in most sites s eemed to be heterogeneous both in spore abundance and macronutrient levels. One possible explanation for the observed lack of trend between spore number and regeneration time is that spores may not always linearly affect the infection rate of their h ost
plants, and therefore spore levels may not increase or decrease predictably with regeneration time. For example, the successful infection of some mycotrophic species (like canopy trees) may require a high density of VAM inocula in the soil, while othe r species may require a relatively low inocula density (Johnson et al. 1991). Therefore, looking directly at mycorrhizal root infection percentage might be more strongly correlated with forest regeneration time (Johnson et al. 1991). Additionally, spores are not the only representation of VAM inocula, thus using spores as a proxy for fungal spores, hyphae, and othe r kinds of inocula may confound results. Furthermore, the absence of trend in spore number with regeneration time could suggest that spores are mostly distributed randomly or in clumps throughout tropical soils. One study found a patchy distribution of fungal spores in sites due to both proximity to mycotrophic hosts and amount of organic matter (Carvalho et al. 2003). Another possible cause for this heterogeneity could be random nutrient deposition into the soil from cattle feces. These patches of greater nutrient availability could offer nearby plants enough nutrients to allow them to forgo their usual mycorrhizal association, creating occ asional pockets of lower VAM abundance (Wolf 1998). If VAM spore abundance is determined by patchy spatial distributions of soil micro climates and mycotrophic hosts, then this heterogeneity could imply the creation of several distinct successional commu nity patches resulting from a single disturbance. However, b ecause of this widespread but patchy distribution, it is unlikely that VAM are a limiting factor in successional commu nities and it is not clear how they m ight drive succession. It would be beneficial for further research to explore proxi mity to mycotrophic host plants and proximity to undisturbed habitat The effects of VAM spore abundance may be more pronounced for sites that are distant from living mycorrhizae infections, beca use this distance may limit certain mycotrophic plants from colonizing a regenerating area. Also mycorrhizae could potentially be important drivers of succession in areas with a more homogeneous soil composition, or they could be critically important to the recovery of sites that have un dergone severe disturbance, such as losing topsoil due to years of industrial agriculture practice. Such degraded sites are abundant, but could regenerate to a field of grasses and weedy non mycotrophs, never t o regenerat e the canopy these sites carried before transformation. ACKNOWLEDGMENTS Thank you staff of CIEE, especially to Ala n, Moncho, and Raquel for helping me with this research project! Thanks to the University of Georgia for letting me spend so much time in the lab and on the trails at the rese arch station in San Luis, and thanks to Lucas for all your biological knowledge. Thanks to Sarah for keeping me sane. I am very grateful to my Costa Rican host family, Mario, Anabelli, and Cristel for feeding me, teaching me that most wisdom is acquired outside of the classroom, and for insisting that I get enough sleep! Thank you fellow students for m aking it a great semester, and thanks to the wild, messy, beautiful forest for inspiring me, challenging me, and keeping me passionate
LITERATURE CITED C ARVALHO L., C ORREIA P., R YEL R., AND M. M ARTINS L OUCAO 2003. Spatial variability of arbuscular mycorrhizal fungal spores in two natural plant communities. Plant and Soil 251 : 227 236. FAO. 2006. Global Forest Resources Assessment 2005: Progress towards sustainable forest management Food and Agric ulture Organization of the United Nations, Rome. G UARIGUATA M. AND R. O STERTAG 2001. Neotropical secondary forest succession: changes in structural and functional characteristics. Forest Ecology and Management 148: 185 206. J ANOS D. 1980. Mycorrhizae influence tropical succession. Biotropica 12(supp): 54 64. J ANOS D. 1983. Vesicular Arbuscular Mycorrhizal Fungi. In D. H. Janzen (Ed.). Costa Rican Natural Hist ory, pp. 340 5. University of Chicago Press, Chicago. J OHNSON N. Z AK D., T ILMAN D., AND F. P FLEGER 1991. Dynamics of vesicular arbuscular mycorrhizae during old field succession. Oecologia 86( 3 ): 349 358. K ILLHAM K. 1994. Soil Ecology pp. 77 203 Cambridge University Press, Cambridge. K WAN D. 1995. Dispersal of mycorrhizal fungus spores by rodents. In EAP: Tropical Biology Program (Spring 1995). M ILLER R AND L. J ACKSON 1998. Survey of vesicular arbuscular mycorrhizae in lettuce production in relation to management and soil factors. Journal of Agricultural Science 130: 173 182. N EALE E 1997. Forest succession of regenerating pastures in Monteverde, Costa Rica. In CIEE: Tropical Ecology an d Conservation (Fall 1997), pp. 276 288. R OGERS S 1998. Rates of regeneration of pastures in San Luis, Costa Rica. In CIEE: Tropical Ecology and Conser vation (Spring 1998), pp. 271 81. S MITH S. AND D. R EAD 19 9 7. Mycorrhizal Symbiosis, pp. 9 126. Academic Press, San Diego. UNEP. Area de Conserva ci n Guanacaste, Costa Rica Forest Restoration Information Service: Case Studies. Retrieved from . W OLF R. 1998. The abundance of vesicular arbuscular mycorrhizal spores in secondary forest and pasture. In EAP: Tropical Bio logy Program (Spring 1998), pp. 85 91.