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 M39-00382
Dispersin de Mucuna urens por las guatusas (Dasyprocta punctata) en la zona de Monteverde en Costa Rica
Mucuna urens dispersal by agoutis (Dasyprocta punctata) in the Monteverde area of Costa Rica
Mucuna urens (Papilionaceae) is a large-seeded liana that is largely dispersed by scatterhoarding mammals. Agoutis (Dasyprocta punctata) are one of the most common scatterhoarding dispersers in Costa Rica and are a known disperser of M. urens. D. punctata in two locations in Monteverde, Costa Rica, were presented with 203 flagged M. urens seeds, in groups of twenty, over a period of 16 days. The seeds were recovered after the D. punctata had cached them or the flags were collected for those eaten. There was no significant difference between the number of seeds eaten and the number of seeds cached (p = 0.385 Scheffee). The weight of the cached seeds did not significantly change during the caching process, indicating that no consumption occurred before caching (Paired t-test, p = 0.0550). D. punctata did not preferentially choose bigger or smaller seeds for hoarding or consumption (ANOVA, p = 0.5765). The D. punctata cached most of the seeds within 20 m of the site (x = 5.49) in a contagious or clumped pattern. Although D. punctata did not disperse seeds a great distance from the site, they did cache viable seeds away from the parent tree and may play an important part in seed dispersal.
Mucuna urens (Papilionaceae) es una especie de liana con semillas grandes que es dispersada en gran parte por mamferos que esparcen y acaparan sus semillas. La guatusa (Dasyprocta punctata) es uno de los dispersores esparcidores y acaparadores ms comunes de Costa Rica y es un dispersor conocido de M. urens. D. punctata, en dos sitios en el rea de Monteverde (Costa Rica) fueron presentadas con 203 semillas de M. urens marcadas con cintas en grupos de veinte por un perodo de 16 das. Las semillas fueron recuperadas despus de que D. punctata las hubieran enterrado, o las cintas fueron recogidas para aquellas semillas que D. punctata devoro. No hubo diferencias significativas entre el nmero de semillas comidas y el nmero de semillas enterradas (P = 0.385 Scheffee). El peso de las semillas enterradas no cambi significativamente durante el proceso de enterramiento, lo que indic que ningn consumo ocurri antes del anterior (Prueba de t de estudiante pareada, p = 0.0550). D. punctata no escogieron preferencialmente las semillas ms grandes o ms pequeas para enterrar o para el consumo (ANOVA, p = 0.5765). D. punctata enterr la mayora de las semillas a 20 m de distancia del sitio en que las semillas fueron colocadas (x = 5.49) en una distribucin agrupada. Aunque D. punctata no dispers semillas a gran distancia, si enterr las semillas viables lejos del rbol materno.
Text in English.
Central American agouti
Costa Rica--Puntarenas--Monteverde Zone
Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology Fall 2004
Ecologa Tropical Otoo 2004
t Monteverde Institute : Tropical Ecology
1 Mucuna urens dispersal by agoutis ( Dasyprocta punctata ) in the Monteverde area of Costa Rica Melissa Senf Department of Animal Science, University of Nebraska Lincoln ABSTRACT Mucuna urens (Papilionaceae) is a large seeded liana that is largely dispers ed by scatterhoarding mammals. Agoutis ( Dasyprocta punctata ) are one of the most common scatterhoarding dispersers in Costa Rica and are a known disperser of M. urens . D. punctata in two locations in Monteverde, Costa Rica, were presented with 203 flagg ed M. urens seeds, in groups of twenty, over a period of16 days. The seeds were recovered after the D. punctata had cached them or the flags collected for those eaten. There was no significant difference between the number of seeds eaten and the number o f seeds cached (p = 0.385 Scheffee). The weight of the cached seeds did not significantly change during the caching process, indicating that no consumption occurred before caching (Paired t test, p = 0.0550). D. punctata did not preferentially choose bigg er or smaller seeds for hoarding or consumption (ANOVA, p = 0.5765). The D. punctata cached most of the seeds within 20 m of the site (x = 5.49) in a contagious or clumped pattern. Alhough D. punctata did not disperse seeds a great distance from the site, they did cache viable seeds away from the parent tree and may play an important part in seed dispersal. RESUMEN Mucuna urens (Papilionaceae) es una especie de liana con semillas grandes que es dispersada en gran parte por mamÃferos que esparcen y aca paran sus semillas. La guatusa ( Dasyprocta punctata ) es uno de los dispersores esparcidores y acaparadores mÃ¡s comunes de Costa Rica y es un dispersor conocido de M. urens . D. punctata, en dos sitios en el Ã¡rea de Monteverde (Costa Rica) fueron presentada s con 203 semillas de M. urens marcadas con cintas en grupos de veinte por un perÃodo de 16 dÃas. Las semillas fueron recuperadas despuÃ©s de que D. punctata las hubieran enterrado, o las cintas fueron recogidas para aquellas semillas que D. punctata devora rÃ³, No hubo diferencias significativas entre el nÃºmero de semillas comidas y el nÃºmero de semillas enterradas (P = 0.385 Scheffee). El peso de las semillas enterradas no cambiÃ³ significativamente durante el proceso de enterramiento, lo que indicÃ³ que ning Ãºn consumo ocurriÃ³ antes del entierror (Prueba de t de estudiante pareada, p = 0.0550). D. punctata no escogieron preferencialmente las semillas mÃ¡s grandes o mÃ¡s pequeÃ±as para enterrar o para el consumo (ANOVA, p = 0.5765). D. punctata enterrÃ³ la mayorÃa de las semillas a 20 m de distancia del sitio en que las semillas fueron colocadas (x = 5.49) en una distribuciÃ³n agrupada. Aunque D. punctata no dispersÃ³ semillas a gran distancia, si enterrÃ³ las semillas viables lejos del Ã¡rbol materno. INTRODUCTION Pl ants depend on various models of dispersal to spread their seeds. Dispersal may be an important mode of escape from density dependent mortality of seeds and seedlings by insect or rodent predation or by pathogen attack (Howe and Smallwood 1982). The fewe r seeds left in an area, the more likely those seeds are to escape the notice of predators and survive to maturity (Janzen 1970) making plants more reproductively successful if they move their offspring away from themselves and siblings. In dispersing seed s away from the base of the parent tree, the seedlings may experience less competition with the parent and sibling plants for factors such as light, nutrient, or water (Jansen et al 2002). In this way dispersal benefits both the parent plant and the offsp ring. The Colonization Theory of seeds dispersal emphasizes the importance of moving seeds widely to ensure that some of them will end up in a suitable habitat for germination such
2 as treefalls or disturbances (Howe and Smallwood 1982). A seedling is mor e likely to survive and thrive in this light filled environment. Scatterhoarding is a behavior common among rodents that plants exploit to disperse their seeds. It is a practice in which an animal carries seeds or fruits away from the parent plant and buries them for consumption when resources are not as readily available (Forget et al. 2002). This behavior links rodent seed predation to seed dispersal which has importance in plant species survival and vegetation structure within tropical forests (Wang serve as major seed predators for the same plants they disperse. This predation cannot overshadow the importance of these mammals as seed dispersers. Predation of these seeds can ins tead be seen as a cost of the dispersal service provided for the plant (Janzen 1970). Plants reliant on scatterhoarding as their mechanism for dispersal are adapted to encourage this behavior, and generally make few seeds that are large and nutritious in o rder to support the dispersers that feed on their seeds (Forget et al. 2002). In this way plants are adapted to their scatterhoarding dispersers and pay the price required by the disperser. Some scatterhoarders bury many seeds in one cache leading to gre ater density of seedlings sprouting from one cache; others cache each seeds separately, but this may also lead to an aggregated seedling configuration. Dasyprocta punctata is an example of a scatterhoarder that buries each seed individually carrying one seed at a time to a location 0.5 to 300 m from the original location. According to Silvius and Fragosus short distance dispersal by D. punctata , might be a cause of the observed aggregated dispersion of large seeded tropical trees (2003). This clumped pa ttern of tree dispersal may be a sign of ineffective dispersal and may lead to the previously discussed competition among related trees and density dependent predation that the tree tries to avoid through seed dispersal. Yet, D. punctata are the largest s catterhoarders, and as such, are considered to be especially important in dispersing large seeds and hard fruits in the tropics (Hallwachs 1986). The effectiveness of scatterhoarding as a mode of dispersal has been explored through studies of Myoprocta a couchy, the red acouchy, a common scatterhoarding rodent in French Guiana. During observations of the caching behaviors of the acouchy, it was found that the persistence of a cache was increased with greater isolation and distance (Jansen et al. 2002). T his would indicate that the seeds dispersed in the most favorable configuration for the plant are the seeds most likely to escape consumption by the acouchy and persist until germination. This would indicate that the acouchy is an effective disperser, and that scatterhoarding can be an effective means of dispersal, though the number of seeds dispersed into favorable conditions, may be small. Dasyprocta punctata is the most commonly encountered diurnal mammal in the low to middle elevation rain forest betw een southern Mexico and northern Argentina (Smythe 1983), and is considered one of the most important dispersers of large seeds in this area. Here I looked at the efficiency of D. punctata as a disperser of Mucuna urens seeds by looking at the amount of e ndocarp eaten in cached seeds and the number of seeds eaten compared to the number of seeds cached. I also examined whether the D. punctata preferentially took larger seeds, and to what distance and in what distribution pattern the seeds were taken. It w as believed that the D. punctata would cache more seeds than they ate to ensure their food supply in the future. It was also predicted that D.
3 punctata would preferentially eat smaller seeds, saving larger seeds for future consumption. Also, it was expec ted that D. punctata would cache the seeds far from the parent plant in isolation from original site, making them an efficient disperser for M. urens . METHODS This study was conducted from October 31 to November 15 in Monteverde, Costa Rica. Two sites w ere chosen within the territory of one or more D. punctata . D. punctata live in stable pairs that maintain a territory together, but may allow other D. punctata to enter the territory when food in not limited (Nowak and Paradiso 1983, Reid 1997). The fir st site of this study was on the property of the EstaciÃ³n BiolÃ³gica Monteverde at an elevation approximately 1535 meters. This site was on the forest edge bordering a field with fruiting Psidium guajava trees where D. punctata had been seen foraging. The second site was located at the Finca EcolÃ³gica in Cerro Plano, Costa Rica on the forest edge next to a clearing. D. punctata are often seen in the clearing, possibly looking for bananas or food scraps set out by the staff of the Finca EcolÃ³gica. Each sit e was approximately two meters by three meters in size. Pods were collected from one individual of Mucuna urens while still on the liana. Mucuna urens (Papilionaceae) is one of the most common leguminous liana found in ats and primary forests (Young 1983). The fruit of the plant is a dark brownish red pod that is covered with urticating hairs; each pod contains were split open using a kni fe, and the seeds were removed. Seeds were numbered and weighed. Seeds were prepared by attaching approximately one meter of nine pound test or less nylon fishing line and a ten centimeter piece of red flagging tape. The line was attached to the seed by threading it through a hole punched in the seed using a metal probe. The flagging tape of each seed was marked with the seed number and initial weight. All seeds were marked with the flagging tape so that they could be tracked. Marked seeds can be retrie ved even when cached away from the site because D. punctata bury only the seed, not the flagging tape or the fishing line. It was determined, in previous studies, that marking seeds in this way does not alter disperser behavior; however, changes in behavio r due to marking cannot be discounted completely (P.A. Jansen et al. 2002). M. urens seeds were placed randomly at each site. To prevent any differential effects caused by density dependent predation, the seeds were presented to D. punctata in a single g roup of 20 at each site. The seeds were left at the site for one or two days before they were checked. The first day the seeds were put out was left as discovery time for D. punctata at both sites. After the first day, all seeds were classified as eaten, cached, or remaining, when first checked. Eaten seeds were classified as such when the flagging and fishing line were found detached from the seed. It was generally apparent that the line had been broken and it was presumed that agoutis had eaten the att ached seed. Cached seeds were seeds that had been moved from their original position and were found buried shallowly in the soil. D. punctata digs a hole and buries the seed with .5 3 cm of soil. D. punctata are very careful to remove all appearance of ca ching by tamping down the soil and they have been seen placing leaves or twigs over the cached seed (Hallwachs 1986) this makes cached seeds very difficult to find when not marked by
4 flagging. Seeds that were classified as remaining were neither eaten nor cached. All such seeds were kept in place and were presented to the D. punctata as part of the following group of 20. Cached seeds, when found, were dug up and reweighed. In five trials, measurements were taken to determine the distance that each cac hed seed was taken from the site. The last trial took place on November fourteenth. Any seeds that were not found as of November 15 were considered unrecovered in the results of the experiment. RESULTS A total of 203 seeds were presented to D. punctata and 178 of them were recovered during the time of this study. Of these 178 seeds, 104 were found eaten, 73 were found cached, and one remained at the site (Table 1). In this study, all eaten seeds were consumed completely. In all cases when a seed was present. In most cases, it was apparent that the fishing line had been broken and pieces of the seed coat were found near the line with no remaining endosperm (Table 1) . This germination. Conversely, sixty eight cached seeds were weighed after recovery; there was no significant difference found in the before and after weights (Pair ed t test, p = 0.0550). All cached seeds were left virtually untouched, and there was no evidence that the teeth had punctured the seed coat. Cached seeds seem to have no endosperm loss and should therefore be viable when dispersed to other locations. The number of eaten, cached, remaining, and unrecovered seeds were analyzed using a one between the groups. This was based on eight trials in which the agoutis were presented with 20 seeds. In the trials included in these data, there was no significant difference between the number of seeds eaten and cached (p = 0.9529 Scheffee). When the number of seeds eaten was compared with the number of seeds remaining, the difference was sig nificant (p < 0.0001), as was the difference between the number of seeds cached and remaining (p < 0.0001) (Table 1 and Figure 1). The number of seeds eaten and the number of seeds cached versus those not recovered were both significantly different (p valu es of < 0.0001 and 0.0002 respectively). D. punctata did not preferentially chose larger seeds; there was no significant difference in the size of seeds eaten and those cached (ANOVA, p = 0.5765); (Figure 2). There was no significant size difference betwee n the unrecovered and cached seeds (p = 0.1617), or between the unrecovered and eaten seeds (p = 0.2816). This is evident when looking at the mean weights of each group. The mean weight of eaten seeds was 14.393 grams (standard deviation = 2.740). The average weight of the cached seeds was 14.178 grams (standard deviation = 2.224), and the unrecovered seeds averaged 15.0 grams (standard deviation = 2.399). The distance moved from the site was variable, however, only two of the recovered seeds were f ound more than 20 m away from the site (Figure 3, Figure 4). The range of seed dispersal was from 0 21.55 m and the mean was 5.49 m (standard deviation meaning that the seeds were cached in a clumped or aggregate pattern (s 2 =36.73 which is
5 greater than x = 5.49) (Southwood 1978). So D. punctata is not spreading the seeds out evenly, though they do not hoard many seeds in one cache. DISCUSSION The clumped dispersal of trees in tropical forests implies that D. punctata is not a very effective disperser. In this study, D. punctata was shown to eat as many seeds as it disperses. When dispersing seeds through caching, D. punctata was observed to move a large majority of the seeds less than five meters away from the parent plant and cache them in a clumped pattern. All of these practices are detrimental to plant dispersal throughout tropical forests, yet D. punctata has long been hailed as an important disperser of lar ge seeds. There are many possible reasons that D. punctata are often ineffective dispersers. One possibility is that D. punctata are finding many seeds and it is energetically efficient to carry the seeds a great distance when all they need is available under the parent tree. The number of cached seeds verses eaten seeds may vary greatly throughout the year depending on the availability of food. Although many studies reference the seasonality of scatterhoarding in the tropics, there is a dearth of real understanding of the effect of seasonal fruiting in the tropics on caching rates and on plant disperser interactions (Forget 2002). In D. punctata , the number of seeds eaten versus the number of seeds cached may vary throughout the year as different plant s have their fruiting seasons, but D. punctata response may not be linked to the number of seeds or fruit available in the area. D. punctata neither ate larger seeds nor cached larger seeds with greater frequency. It is possible that the D. punctata eat t he first seeds they come to without considering size. However Hallwachs (1986) showed that D. punctata preferentially ate small seeds and did not cache them. Nevertheless, the seeds classified as small within her study were much smaller than the smallest seed used in this study. The difference seen in this study may be caused by the relatively uniform size of M. urens seeds. According to other studies, D. punctata are not adequate dispersers of small seeds. This may again be a reflection of energy effi ciency. For D. punctata , it might not be energy efficient to bury small seeds or to return to look for small seeds. D. punctata seed dispersal is clumped and does not move the seeds far from the parent plant. The caching locations D. punctata chooses cou ld reflect the travel patterns or the individual territories of D. punctata . D. punctata have small territories and all limiting the placement of cached seeds are the paths that D. punctata tend to follow through the forest. Another possibility for why D. punctata are not ideal dispersers is that D. punctata eat many of the available seeds. In this experiment, it was observed that D. punctata never ate seeds partiall y. When a seed was eaten, the entire endosperm was consumed. Therefore, D. punctata are quickly sated and cached seeds were left entire. Even though the plant might benefit from a greater dispersal distance, D. punctata do not always take seeds a grea t distance. All but two of the seeds in this study were found cached within 20 meters of the original pile. However, secondary dispersal of the seeds, moving cached seeds to a second or a third cache may cause reduced seed
6 clumping over time (Wang and Smi th 2002) and may also result in seeds moving farther from the parent plant. Secondary dispersal is usually by the same agouti that first cached the seed, however, it is possible that other hoarding rodents may find the cached seeds and take them. Secondar y caching behaviors may be an important part of the dispersal of M. urens seeds since D. punctata failed to disperse seeds more than 20.55 meters away. Caching behaviors are important for the survival of the plants that rely on scatterhoarders for disper sal as well as the mammals that use the caches. In the future, studies could be conducted to look at the length of time a seed stays in its cache before being retrieved. Similarly, an investigation could be done as to how many seeds are overlooked by D. punctata long enough for germination to occur and what percent of seeds produced by M. urens are capable of germination. The mechanism by which D. punctata remembers the location of the cache could also be investigated. All of these studies address key a spects of animal behavior and plant dispersal and consequently address the diversity and functionality of a tropical ecosystem. ACKNOWLEDGEMENTS I would like to thank Alan Masters and Javier MÃ©ndez for their help in all aspects of my project, from the pro ject idea to the completion. I would also like to thank the EstaciÃ³n BiolÃ³gica Monteverde and the Finca EcolÃ³gica for allowing me to conduct my study on their property. Thank you to Hilary Wilson for her aid in the seed preparation process, and to Ollie Hyman and Matt Gasner for their help in getting all the materials needed for the setup. In addition, I would like to express my gratitude to Richard Laval and Federico Chinchilla for their help in previous project ideas, though they did not pan out. L ITERATURE CITED Forget, P.M., D.S. Hammond, T. Milleron, and R. Thomas. 2002. Seasonailty of fruiting and food hoarding by rodents in Neotropical forests: consequences for seed dispersal a nd seedling recruitment. In D.J. Levey, W.R. Silva, and M. Galetti, eds. Seed Dispersal and Frugiory: Ecology, Evolution and Conservation. CABI Publishing. Hallwachs, W. 1986. Agoutis ( Dasyprocta punctata ): The Inheritors of Guapinol ( Hymenaea courbaril: L eguminosae). In A. Estrada and T. H. Fleming, eds. Frugivores and Seed Dispersal. Dr. W. Junk Publishers, Dordrecht. Howe, H.F. and J. Smallwood. 1982. Ecology of Seed Dispersal. Annual Review of Ecological Systems. Annual Reviews Inc. 201 228. Jansen, P. A., M. Bartholomeus, F. Bongers, J.A. Elzinga, J. den Ouden, and S. E. Van Wieren. 2002. The Role of Seed Size in Dispersal by a Scatter hoarding Rodent. In D.J. Levey, W.R. Silva, and M. Galetti, eds. Seed Dispersal and Frugivory: Ecology, Evolution and Conservation. CABI Publishing. Janzen, D.H. 1971. Herbivores and the number of tree species in tropical forests. American Naturalist 104: 501 528. Mabberley, D.J. 1981. The plant book. Cambridge University Press, Cambridge. Nowak, R.M. and Paradiso, J.L University, Baltimore. 814 815.
7 Reid, F.A. 1997. A Field Guide to the Mammals of Central America & Southeast Mexico. Oxford Press, New York. 243 244. Southwood, T.R.E. 1976. Ecological Methods: with par ticular reference to the study of Insect Populations. Chapman and Hall, London. 26 28. Wang, B.C. and T.B. Smith. 2002. Closing the Seed Dispersal Loop. TREE 17: 379 385. .
8 Table 1. Number of total seeds in each categ ory (Eaten, Cached, Unrecovered, Remaining), the mean weight of seeds within each category, and the mean number of seeds put in each category per group of seeds presented to Dasyprocta punctata in Monteverde. Category Number of seeds Mean weight Mean nu mber per group Eaten 104 14.393 Â± 2.740 9.500 Â± 3.16 Cached 73 14.178 Â± 2.24 8.750 Â± 3.845 Unrecovered 25 15 Â± 2.399 1.750 Â± 1.488 Remaining 1 14.800 0 Figure 1. The average number of see ds eaten, cached, left, and unrecovered in each group of seeds that were presented to the Dasyprocta punctata in Monteverde, Costa Rica. This data incorporates the state of eight groups of 20 and one group of 32 seeds.
9 Figure 2. Mean before weights of the seeds eaten or cached by Dasyprocta punctata , and unrecovered seeds in Monteverde, Costa Rica. Left seeds are not included due to small numbers. There were no statistically significant differences between the three categori es. Figure 3. The number of seeds and distance taken from the original pile of seeds in Monteverde, Costa Rica by Dasyprocta punctata . Distance shown in increments of 5 meters.
10 Figure 4. The n umber of seeds D. punctata took from the original pile and the distance each of the seeds were taken in Monteverde, Costa Rica.