Kin recognition in the social spiders Anelosimus sp. in Monteverde, Costa Rica Mary Kathryn Busby Department of Integrative Biology, University of Te xas at Austin ABSTRACT Anelosimus spiders are social and use pheromonal and vibratio nal cues to distinguish conspecifics from predators and prey. Juveniles live in their natal w eb with adults until maturity, at which time they d isperse. Given that juveniles are more likely to disperse to locations near to their natal web and that Anelosimus has developed mechanisms for kin recognition, these spi ders would be expected to exhibit aggressive behavi or less frequently in the presence of intruding spider s from nearby webs than towards intruders from dist ant webs. This hypothesis was tested by introducing spi ders to new webs from distant locations as well as from nearby locations. There was no significant relation ship found between web distance and level of aggres sion ( C2 = 5.47, df = 2, p = 0.06) nor between colony size and level of aggression ( C2 = 2.43, df = 2, p = 0.30). A linear regression comparing colony size to level of aggression yielded no significant results for ig noring behavior (F = 0.94, df =1 and 18, p = 0.34), approa ching behavior (F = 3.36, df = 1 and 18, p = 0.08), nor aggressive behavior (F = 0.29, df = 1, p = 0.59). A ccording to this data there is little indication th at spiders change their level of aggression due to colony size or distance between webs. This study provides data that can be used to explain the purpose and function of kin-recognition behaviors. RESUMEN Las araas Anelosimus son individuos sociales que utilizan pistas como vi braciones y feromonas para distinguir entre conespecficos entre depredadores y presas. Los juveniles viven en sus telas natales hasta que maduran, tiempo al cual se dispersan. Con esto los juveniles tienden ms a dispersarse a telas ubicadas cerca de la tela natal y estas especies ti enen un mecanismo de reconociemiento de parentela, se espera en estas arenas un comportamiento agresivo m enos frecuente con la presencia de araas cercanas a la tela que con araas intrusas de telas lejanas. Esta hiptesis se prov colocando araas en nuevas telas desde distinas distancias as como araas de telas cercanas. No existe una relacin entre la distanci a y el nivel de agresin ( C2 = 5.47, df = 2, p = 0.06) tampoco entre el tamao de la colonia y el nivel de agresin ( C2 = 2.43, df = 2, p = 0.30). Una regression linear para comparar el tamao de la colonia y el nivel de agresin no mostr ningn resultado como comportami ento evasivo (F=0.94, df=1 and 18, p=0.34), comportamiento de aproximacin (F=3.36, df=1 and 18 p=0.08),y comportamiento agresivo (F=0.29, df=1, p=0.59). De acuerdo a estos datos no hay un i ndico de que las araas cambien su nivel de agresi n debido al tamao de la colonia o a la distancia ent re telas. Este estudio prove datos que expliquen los comportamientos de reconocimiento de parentela. INTRODUCTION Sociality has evolved independently several times i n spiders and to differing degrees. Periodic-social species are those that have associa tions with relatives that disintegrate prior to the mating season. In permanent-social spe cies the juveniles often remain at their natal web year-round. It is thought that permanent social traits evolve by suppressing this phase of dispersal of juveniles at a certain age (S eibt et al. 1988).
Anelosimus spiders are among the few spiders that practice tru e sociality (Foelix 1996). While they are less social than the eusocial insects in that all members are capable of reproducing and there is no real hierarchy or sy stem of rank, they form communal webs in which all members share duties (Foelix 1996 ). Anelosimus contains four permanently-social species (Avils et al. 1998). Anelosimus web colonies are built in disturbed areas such as roadsides or forest edges and clearings. They build these webs as a collectiv e effort and the webs increase in size as the number of members increases (Tietjen 1986). There may be up to thousands of members per colony or there may be colonies consist ing of only one female (Vollrath 1988). A colony may exist in one location for years (Seibt et al. 1998). Anelosimus spiders share web building, community maintenance, and prey capture tasks among members (Foelix 1996). Juveniles remain at their na tal web until adulthood, at which point most of them disperse. Many, however, stay in their natal webs as adults and sometimes forgo their own reproduction to aid in th e rearing of relativesÂ’ offspring (Jones 2001). This kin-selection behavior increases indivi dual fitness when spiders are highly related. Parental care (which exists in Anelosimus ) and tolerance of juveniles result in fitness trade-offs. Competition for food resources during p eriods of rapid growth is a drawback that in many species has caused juveniles to disper se during their high-growth period (Powers et al. 2003). Another drawback is that disp ersing juveniles frequently disperse to locations near to their natal web. While some juven iles disperse beyond local areas, many stay close enough to result in inbreeding depressio n (Avils et al. 1998). The benefits of communal living must outweigh the c osts for this trait to have evolved. Ultimately, kin selection increases the ch ances of an individualÂ’s own genes being perpetuated by close relatives. Proximately, parents and juveniles fill different feeding niches: juveniles may capture a wider varie ty of prey sizes and help keep the web clean of smaller prey (Tollins 2008). Other fitness benefits to having a communal lifestyle in spiders include having less demand on individual silk production and better defense against predation (Jones 2001). Because Anelosimus spiders live communally, there must be some system in place by which spiders can recognize their kin. Evolutionari ly this trait should have evolved to prevent spiders from killing conspecifics carrying their own genes, as Â“an individualÂ’s decision to attack or tolerate a stranger would be governed by a cost/benefit ratioÂ” (Seibt et al. 1988). These recognition mechanisms have bee n shown to be both vibrational and pheromonal (Foelix 1996). In the species Anelosimus jucundus spiders frequently disperse to locations within five meters of their natal web (Powers et al. 2003) Due to this factor, one would expect nearby webs to be more likely than distant webs to contain related spiders. One would also expect spiders to be more likely to show aggre ssive behavior towards spiders from distant webs than towards spiders from nearby webs. This study attempts to determine if there is a relationship between web distance and ag gressive behavior. If a spider from a distant web is introduced to a new web, it should b e more likely to cause an aggressive reaction than the introduction of a spider from a n earby web. One would also expect webs with a higher colony pop ulation to show higher levels of aggression towards intruders because there are more individuals capable of defending the web.
METHODS Twenty-two webs of Anelosimus spp. (Theridiidae) were located down the length of the road leading to the Cerro Amigos tower in Monte verde, Costa Rica from July 15 until August 2, 2008. Twenty of these were manipulated an d used for data collection. The other two were used to supply additional spiders. D ata was recorded for both of the two extra webs, but no experiments were performed on th ese webs. Webs were marked with numbered flagging tape, designating each web with a number from one to 22. The size of each web was recorded as length by width in centime ters, with length being the length down the supporting branch on which the web was bui lt and width being the horizontal dimension perpendicular to the length. Distance bet ween each web was measured in paces, which were later converted to meters. The nu mber of individual spiders inhabiting each web was also counted and recorded. Two foreign spiders were introduced to each of the 20 experimental webs. Of these two, one originated from a web near the web to whic h it was introduced and the other originated from a web distant to the web to which i t was introduced. A nearby web almost always meant a next-door neighbor, ten meter s away or fewer. For instance, a spider introduced to web number nine from web numbe r eight would be considered a next-door neighbor and may be used as a Â“nearÂ” web. Distant webs were chosen in such a way as to maximize the distance between the two web s and to minimize damage to the web population size by removing spiders from small web colonies. Observations were recorded in terms of the level of aggression displa yed by the defending spider (Â“defensive spiderÂ” is defined as the spider who is in its own web when the Â“intruderÂ” is introduced). Aggression levels were grouped into three categorie s: Â“ignoring,Â” Â“approaching,Â” and Â“aggressive.Â” Â“IgnoringÂ” means the defensive spider made no movement or indication of awareness that an intruder had been introduced. Â“Ap proachingÂ” means the defensive spider moved from its place and may have climbed to a location very close to the intruder but made no attempt to chase, push, or bite the int ruder. Chasing, pushing, or biting behaviors were dubbed Â“aggressive.Â” Spiders were transferred from web to web using the simple method of forceps-andvial. The spiders were taken from their home web us ing forceps, transferred to a glass vial marked with their web number of origin, and th eir length was measured in centimeters. Since these spiders stay in an outstre tched position, their lengths were measured from the tip of the front leg to the tip o f the back leg. They were then carried to one of the other 20 webs, either nearby or far from their original web, and placed in the new web using forceps. Efforts were made to prevent anything other than the intruding spider from touching the new web. That is, forceps were held near enough for the spider to climb onto the new web but not near enough for t he forceps to cause warning vibrations in the web. The relationship between aggression behavior and di stance between webs was analyzed using a Chi-squared test. The relationship between colony size and aggressive behavior was analyzed using regression analysis wit h the statistical software STATISTICA. By dividing colony sizes into Â“bigÂ” and Â“small,Â” signifying colonies with populations greater than nine and less than nine re spectively, relationship between colony size and level of aggression was tested using anoth er Chi-squared test.
RESULTS There was no significant difference found between w eb distance and aggression level ( C2 = 5.47, df = 2, p = 0.06). Based on another Chi-squ ared test there was also no significant difference between colony size and aggression level ( C2 = 2.43, df = 2, p = 0.30). Linear regressions for colony size and ignoring behavior ( F = 0.94, df = 1 and 18, p = 0.34), approaching behavior (F = 3.36, df = 1 and 18, p = 0.08), and aggressive behavior (F = 0.29, df = 1, p = 0.59) were not significant. DISCUSSION This study indicates that there is little relations hip between the effects of colony size or distance of webs and level of aggression against intruders. This requires more studies to ascertain why the relationship between these var iables is minimal. Perhaps with a larger sample size of webs the data may prove signi ficant. Similarly, the relationship between web distance and aggression behavior is clo se to being significant and may prove to be so if more data is collected (Fig. 1). Another hypothesis to explain the lack of significa nce is that there is either a lack of or overabundance of relatedness between nearby webs If juveniles regularly disperse to distances further than the ones used in this study, then the likelihood that spiders from the distant webs are related is unknown. There may be a need for repeating this experiment with a wider distance between webs to find out if t his relationship is significant. Personal observation suggested that there might be a relationship between aggression of spiders and the presence of an egg sac. Adult sp iders clutching egg sacs appeared more active and more likely to display aggressive behavi ors. This factor would most likely depend on the time of year in which the study is co nducted. Other experimenters have noticed dramatic changes in web composition and beh avior of members across different seasons. For instance during the dry season spiders tend to be less active, since activity causes them to desiccate more quickly. The number o f adults, juveniles, and eggs is highly variable throughout the year. Therefore if t his experiment is repeated in a different time of year it may yield different results (Vollra th 1988). Along those lines, different weather conditions on data collection days may have played a role in spider behavior. Several days were misty or raining, while others we re sunny. This study could have been improved with more infor mation regarding the meaning of different spider behaviors. Primarily, it is unc ertain whether Â“approachingÂ” behaviors should be considered aggressive or benign. Also, ba sed on personal observation, there is a wide variation in the behavior of individual spid ers. This is especially evident when capturing spiders. There were many different respon ses on the part of Anelosimus spiders to being pursued by a pair of forceps, including hi ding, dropping to the ground, and playing dead. Further investigation may clarify the connection between morphology and behavioral response to threat.
ACKNOWLEDGMENTS I would like to thank Tania Chavarra for the inval uable expertise, advice, and time she contributed to this project. I would also like to t hank Karen Masters, CIEE, and the Estacin Biolgica de Monteverde for providing insi ght and resources.
LITERATURE CITED Avils, L. and G. Gelsey. 1998. Natal dispersal and demography of a subsocial Anelosimus species and its implications for the evolution of sociality in spiders. Zoology 76: 2137-2147. Foelix, R.F. 1996. Biology of Spiders pp 258-264. Oxford University Press. New York, New York. Jones, T. 2001. Delayed juvenile dispersal benefits both mother and offspring in the cooperative spider Anelosimus studiosis (Araneae: Theridiidae). Behavioral Ecology 13:142-148. Powers, K. and L. Avils. 2003. Natal dispersal pat terns of a subsocial spider Anelosimus cf. Jucundus (Theridiidae). Ethology 109: 725-737. Seibt, U. and W. Wickler. 1988. Interspecific toler ance in social Stegodyphus spiders (Eresidae, Araneae). Arachnol. 16: 35-39. Tietjen, W. 1986. Social Spider Webs, with Special Reference to the Web of Mallos gregalis Shear W.R(ed). Spiders: Webs, Behavior, and Evol ution pp 172-206. Stanford University Press. Stanford, California. Tollins, M. 2008. A Study of Social Interactions an d Web Dynamics in the Spider Anelosimus sp ., (Theridiidae) in Monteverde, Costa Rica. CIEE. S pring 2008. Vollrath, F. 1986. Environment, reproduction, and t he sex ratio of the social spider Anelosimus eximius (Araneae, Theridiiae). Arachnol.14: 267-281.
0 2 4 6 8 10 12 14 16 18 ignoredapproachedaggresive Behavior near far Figure 1. Black bars show number of instances of each behavio r per nearby spider that was introduced. White bars show number of instances of each behavior per far away spider that was introduced ( C2 = 5.47, df = 2, p = 0.06). 0 2 4 6 8 10 12 14 ignoredapproachedaggresive Behavior small big Figure 2. Black bars show number of instances of each behavio r observed per small colony. White bars show number of instances of each behavior observed per big colony ( C2 = 2.43, df = 2, p = 0.30).
-0.5 0 0.5 1 1.5 2 2.5 0510152025303540 # Approached Figure 3. Regression line shows relationship between number of times approaching behavior was observed and population of web colony (F=3.36, df=1 and 18, p=0.08). Table 1. Linear regression results show relationship betwee n number of times each behavior was observed and the population of web col ony. Regression F Degrees Freedom p Colony size vs. ignored 0.94 1 and 18 0.34 Colony size vs. approached 3.36 1 and 18 0.08 Colony size vs. aggressive 0.29 1 and 18 0.59
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 M39-00010
Busby, Mary Kathryn
Reconocimiento familiar en las araas sociales Anelosimus sp. en Monteverde, Costa Rica
Kin recognition in the social spiders Anelosimus sp. in Monteverde, Costa Rica
Anelosimus spiders are social and use pheromonal and vibrational cues to distinguish conspecifics from
predators and prey. Juveniles live in their natal web with adults until maturity, at which time they disperse.
Given that juveniles are more likely to disperse to locations near to their natal web and that Anelosimus has
developed mechanisms for kin recognition, these spiders would be expected to exhibit aggressive behavior
less frequently in the presence of intruding spiders from nearby webs than towards intruders from distant
webs. This hypothesis was tested by introducing spiders to new webs from distant locations as well as from
nearby locations. There was no significant relationship found between web distance and level of aggression
(C2 = 5.47, df = 2, p = 0.06) nor between colony size and level of aggression (C2 = 2.43, df = 2, p = 0.30).
A linear regression comparing colony size to level of aggression yielded no significant results for ignoring
behavior (F = 0.94, df =1 and 18, p = 0.34), approaching behavior (F = 3.36, df = 1 and 18, p = 0.08), nor
aggressive behavior (F = 0.29, df = 1, p = 0.59). According to this data there is little indication that spiders
change their level of aggression due to colony size or distance between webs. This study provides data that
can be used to explain the purpose and function of kin-recognition behaviors
Las araas Anelosimus son individuos sociales que utilizan pistas como vibraciones y feromonas para distinguir conspecficos entre los depredadores y las presas. Los juveniles viven en sus telas natales con los adultos hasta que maduran, tiempo al cual se dispersan. Con esto, los juveniles tienden mas a dispersarse a telas ubicadas cerca de la tela natal y estas especies tienen un mecanismo de reconocimiento de parentela, se espera en estas araas un comportamiento agresivo menos frecuente con la presencia de araas cercanas a la tela que con las araas intrusas de telas lejanas.
Text in English.
Kin recognition in animals
Reconocimiento de parientes en los animales
Tropical Ecology 2008
Ecologa Tropical 2008
t Monteverde Institute : Tropical Ecology