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 M39-00007
La dinmica de la captura de presas de la colonia de araas Metabus gravidus (Araneidae) en Monteverde, Costa Rica
The dynamics of prey capture of the colonial, orb-weaving spider Metabus gravidus (Araneidae) in Monteverde, Costa Rica
Prey capture dynamics of the colonial, orb-weaving spider Metabus gravidus (Araneidae)
are based on individuals. It is possible that prey caught in certain locations of the web
will stimulate more reactions, and that different types of prey will stimulate variable
reactions. Since the members of the colony live close to one another, it is also possible
that spiders exhibit competition over prey. Different insects were dropped into various
locations of individual orbs in order to observe reactions. The spiders did not seem to
prefer whether prey was dropped closer or further from the center of the web (Chisquared
goodness of fit, l2 = 0.684, P = 0.710, DF = 2); number of captures did not
depend on location either (Chi-squared goodness of fit, l2 = 0.651, P= 0.651, DF = 2).
Grasshoppers generated medium-speed reactions, but overall the spiders preferred moths
to grasshoppers or ants (chi-squared goodness of fit, l2 =14.88, P = 0.006, DF = 2). It
seems that variation in prey stimulates different behaviors in M. gravidus.
Las dinmicas de la captura de presas de la colonia de araas Metabus gravidus (Araneidae) se basa en los individuos. Es posible que las presas capturadas en determinados lugares de la telaraa provocaran ms reacciones. Ya que los miembros de la colonia viven cerca unos de otros, tambin es posible que las araas exhiban ms competencia sobre la presa.
Text in English.
Orb weavers--Costa Rica--Puntarenas--Monteverde Zone
Tejedores de orbe--Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology 2008
Prey capture dynamics
Ecologa Tropical 2008
Dinmicas de captura de presas
t Monteverde Institute : Tropical Ecology
The Dynamics of Prey Capture of the Colonial, OrbWeaving Spider Metabus gravidus (Araneidae) in Monteverde, Costa Rica. Signe Spencer Department of Biology, Portland State University ABSTRACT Prey capture dynamics of the colonial, orb-weaving spider Metabus gravidus (Araneidae) are based on individuals. It is possible that prey caught in certain locations of the web will stimulate more reactions, and that different t ypes of prey will stimulate variable reactions. Since the members of the colony live cl ose to one another, it is also possible that spiders exhibit competition over prey. Differe nt insects were dropped into various locations of individual orbs in order to observe re actions. The spiders did not seem to prefer whether prey was dropped closer or further f rom the center of the web (Chisquared goodness of fit, l2 = 0.684, P = 0.710, DF = 2); number of captures did not depend on location either (Chi-squared goodness of fit, l2 = 0.651, P= 0.651, DF = 2). Grasshoppers generated medium-speed reactions, but overall the spiders preferred moths to grasshoppers or ants (chi-squared goodness of fi t, l2 =14.88, P = 0.006, DF = 2). It seems that variation in prey stimulates different b ehaviors in M. gravidus. RESUMEN Las dinmicas de captura de presas en Metabus gravdius (Araneidae) depende de individuos, aunque vivan en colonias. Es possible q ue presas en diferentes lugares de la red provoquen reaciones differentes. Debido a que l os miembros viven en colonias, es posible que haya competencia por las presas. Se co locaron insectos en diferentes lugares en las telaraas y se observaron sus reacciones. La s araas no tienen preferencia por la distancia de la presa al centro de la red (Chi-squa red goodness of fit, l2 = 0.684, P = 0.710, DF = 2); Las capturas no dependen de lugares (Chi-squared goodness of fit, l2 = 0.651, P= 0.651, DF = 2). Los saltamontes ocacionar on reacciones de velocidad intermedia, pero las araas prefieren ms las marip osas que los saltamontes o las hormigas (chi-squared goodness of fit, l2 =14.88, P = 0.006, DF = 2). Parece que la variacin en el tipo de presa estimula diferentes e n comportamientos en M. Gravidus. INTRODUCTION
The colonial, orb-weaving spider Metabus gravidus (Araneidae) can be found in riparian environments in the Central American tropics. These spiders suspend their orbs over flowing river water forming a community network of webs. Despite their intricate web system, they are not truly social as the individual s are merely tolerant of one another (Downes 1995). Metabus gravidus communities are usually found near gentle slopes o r ledges and over a steady current. These diurnal spi ders feed daily on a variety of insects including moths, flies, dragonflies and bees (Buski rk 1975b). The communal web building of M. gravidus allows them to access a unique niche by bridging rivers. This niche is subject to a vas t amount of insects that travel near the water, and though the spidersÂ’ webs are connected, prey capture is dependent upon the individual orb of each spider (Uetz 1986). Metabus gravidus usually sit in the center of the web while awaiting prey and it is reasonable to predict that insects landing in different parts of the web would stimulate differen t reactions. Previous studies have observed prey capture dynamics in regards to the dy namics of the colony as a whole (Buskirk 1975b), yet little has been documented abo ut reactions to prey on an individual level. I hypothesized that spiders would respond different ly depending on where is prey is located within the web and what type of prey is caught in the web. I predicted that the spiders would respond quickest to prey dropped clos e to the center of the web and would prefer moths to grasshoppers or ants. Since M. gravidus live in colonies, I also expected to see some cases of competition over prey. MATERIALS AND METHODS The study site was along La Quebrada Maquina in the Monteverde Cloud forest of Costa Rica. The surrounding environment was dense forest with occasional light gaps and the river was narrow with a moderate flow. Ten differe nt M. gravidus colonies were found along this river and labeled accordingly. The numb er of individuals per colony varied between one and twelve. The spiders were always fo und around congregations of logs and/or rocks that they used for anchoring their web s. Data collection with observations consisted of a three-day period, beginning each day at 9:00 and ending at 13:00. Before experimentation could begin, three differen t types of prey were collected and held in their respective jars: ants, grasshoppe rs and moths. These insects would serve as prey variables. The ants had to be large enough to stick in the webs, while the grasshoppers and moths had to be small enough to av oid destroying the webs. Size of prey varied depending on the size of the spiders (s mall, medium or large), as the goal was to facilitate capture. Experimentation began immediately upon encounterin g a new spider colony. Each colony was first labeled, and then dusted with baby powder for visibility purposes so that individuals could be identified and counted Then, an individual was randomly chosen from within the colony to have an ant droppe d into the web. Location of drop was relative to the center of the web, where the spider waits for prey, and included: close to center, middle from center, or far from center. A s econd random individual was then chosen to have a grasshopper dropped close, medium and far from the center of the web. A third, and final, random individual was then chos en and the same process was used for
the moth. Each type of prey, ant, grasshopper, and moth were dropped at each location for a total of three drops/insect/colony. A new pr ey was dropped into the webs every 1525 minutes to allow a break between captures. Prey type, distance of prey form center of web, an d reaction of spider was recorded for each drop. Reactions included: no rea ction, slow-speed reaction, mediumspeed reaction, fast-speed reaction and retreating reaction (speeds based on relativity). Once these data were collected the materials were g athered and I continued to move down the river to the next colony; the same procedu re was carried out for each. Upon returning to the station, data were analyzed using chi-squared contingency tables and goodness of fit tests statistics. The statistics c ompared location of prey drop and spider reaction, location of prey drop and successful capt ures, and type of prey and spider reaction. Type of prey and successful captures wer e also compared. RESULTS There was no trend in number of reactions compared to prey location within the web (Chi-squared goodness of fit, l2 = 0.684, P = 0.710, DF = 2). There was also no tre nd in number of successful captures compared to prey loca tion within the web (Chi-squared goodness of fit, l2 = 0.651, P = 0.651, DF = 2). Only one significant trend was found for prey type compared to spider reaction (Table 1). The only behavior that differs between t ypes of prey was medium-speed reaction (Table 1). The spiders usually reacted at medium-speed when a grasshopper was dropped into the web (Figure 1). The final analysi s compared prey type versus successful captures. Spiders captured more moths than any othe r insect (Figure 2). DISCUSSION Metabus gravidus show no preference for prey dropped in different p arts of the web. The lack of preference for prey location could possibly be attributed to the quality of the webs and the spidersÂ’ sensitivity to vibrations througho ut the entire web (Foelix 1996). The data show that the spiders recognized the vibration s both close and far from the center of the web. Another observation, which also ran count er to my original prediction, is that prey dropped close to the center often generated a retreating reaction instead of a quick capture. Spiders tended to act at medium-speed when grassho ppers were dropped into the web. Perhaps because it is typical of spiders to sp end a few moments tugging on the web to locate prey before they run out to catch it, a m edium-speed reaction would then indicate that the spider had Â‘thoughtÂ’ about the pr ey before capture (Foelix, 1996). After detecting vibrations, spiders must have decided tha t grasshoppers were a Â‘goodÂ’ prey. Concerning prey preference, the original prediction was supported as M. gravidus preferred moths to grasshoppers or ants (Figure 2). Since M. gravidus build their webs over water, it was expected that prey with wings wo uld be common prey within the webs and easily recognized by the spiders. Though the sp iders often reacted quickly to ants they were rarely captured because the ants would re adily defend themselves upon encounter. Also, the grasshoppers did not stick v ery easily in the webs and were often able to break free. Though moths were captured more frequently than the other insects,
M. gravidus usually attempted to capture whatever landed withi n the web, and location of drop did not seem to matter. Although subsequent studies of M. gravidus have discovered cases of fighting and web take-over amongst members of a colony, there ha s been no documentation of one spider stealing anotherÂ’s prey (Buskirk 1975a). Th roughout my observations, I did not see evidence of fighting, web take-over, or stealin g of prey. Further studies may want to observe the reactions of M. gravidus individuals to different sizes of prey or vary the type of prey us ed. Also, data may have yielded better results if more individuals per colony were tested. However, since prey capture is the reason that spiders build their elaborate webs, stu dies of prey capture dynamics are essential to understanding how they fit into the gl obal ecosystem (Riechert 1984). ACKNOWLEDGEMENTS Thanks to Karen Masters and the CIEE program for al lowing me the opportunity to conduct this research and to the Biological Station of Monteverde for providing an area of private study within the cloud forest. Much app reciation goes to Tania Chavarria for assisting me with the development and statistical a nalysis of this project, and to Pablo Allen and Moncho Caldern for their helpful advice. Finally, thanks to all the students of this program for their support and enthusiasm! LITERATURE CITED Buskirk, R. E. 1975a. Aggressive display and orb de fense in a colonial spider, Metabus gravidus Animal Behavior 23: 560-567. Buskirk, R. E. 1975b. Coloniality, activity pattern s and feeding in a tropical orb-weaving spider. Eco logy 56: 1314-1328. Daz-Fleischer, F. 2005. Predatory behaviour and prey -capture decisionmaking by the web-weaving spider Micrathena sagittata Canadian Journal of Zoology 83: no2 268-73 F Downes, M.F. 1995. Australasian social spiders: wha t is meant by Â“socialÂ”?. Rec. West. Austral. Mus. Suppl.: 25 pages? Richert, S.E and T. Lockley. 1984. Spiders as biolo gical control agents. Annual Review of Entomology 29: 299-320 Uetz, G. W. 1986. Web building and prey capture in communal spiders. Shear W.A (Ed). Spiders: Webs, behavior, and evolution. Stanford University Press Stanford, CA.: 226-231.
Table 1: Chi-distribution and Probability Values fo r different reactions in Metabus gravidus spiders. These reactions were compared against thr ee different types of prey, but a medium-speed reaction was the only on w ith a value that reveals a significant trend between prey type and reaction. Reaction Type l2 P Value Overall 0.357 0.836 No reaction 0.230 0.883 Slow-speed reaction 2.20 0.247 Medium-speed reaction 5.20 0.074 Fast-speed reaction 4.12 0.127 Scared 1.60 0.449 nrr Figure 1. Chi-distribution statistics reveal a tren d between prey type and spider ( Metabus gravidus ) reaction. Spiders tend to move at a medium-speed toward grasshoppers more frequently than they will towards ants or moths (Ch i-squared goodness of fit, l2 = 5.20, P = 0.074, DF = 2).
nrr Figure 2. Chi-distribution statistics show trend be tween prey type and capture frequency for Metabus gravidus spiders (chi-squared goodness of fit, l2 = 14.88, P = 0.006, DF = 2). The spiders preferred moths (18 captures) to g rasshoppers (7 captures) or ants (2 captures). They preferred ants the least out of th e three prey types.