Morphological caste differences in polybia aequatorialis (hymenoptera: vespidae: polistinae: epiponini), a social wasp of the highlands of Costa Rica.
Article Type: Report
Subject: Paper wasps (Physiological aspects)
Insect societies (Physiological aspects)
Authors: Pizarro, Laura Chavarria
Alvarado, Mauricio Roverssi
Noll, Fernando Barbosa
Pub Date: 04/01/2009
Publication: Name: Entomologica Americana Publisher: New York Entomological Society Audience: Academic Format: Magazine/Journal Subject: Biological sciences; Science and technology Copyright: COPYRIGHT 2009 New York Entomological Society ISSN: 1947-5136
Issue: Date: April, 2009 Source Volume: 115 Source Issue: 2
Geographic: Geographic Scope: Costa Rica Geographic Code: 2COST Costa Rica
Accession Number: 257814197
Full Text: Abstract.--Many aspects of the biology and organization of Neotropical social wasps in the highlands are unknown. Polybia aequatorialis is a highland wasp of Costa Rica distributed between 1,150 and 3,200 m in altitude, and little information on this species is recorded. We investigated the size of a colony of P. aequatorialis in the Cerro de la Muerte region of Costa Rica, and studied the morphological differences between queens and workers. Measures were taken from 248 reproductive and non-reproductive females, and caste differentiation was analyzed by Discrimination Function Analysis. We did not find a highly pronounced caste distinction in P. aequatorialis, even though ANOVA showed that queens and workers differed in all morphometric measures. The morphological differences between the reproductive and non-reproductive females probably results from a developmental switch, which is a characteristic caste syndrome of Polybia.

Key words: Polistinae, Epiponini, Polybia aequatorialis, colony size, caste differences.

Resumen.--Muchos aspectos sobre la biologia y la organizacion social de las avispas sociales de las regiones de altura del Neotropico son desconocidos. Polybia aequatorialis es una avispa de las regiones altas de Costa Rica que se distribuye desde los 1,150 hasta los 3,200 metros de altura. Se conoce muy poco sobre esta especie, razon pot la cual el objetivo de este estudio rue investigar el tamano de una colonia de P. aequatorialis en el Cerro de la Muerte en Costa Rica, asi como estudiar las diferencias morfologicas entre reinas y obreras para determinar que tipo de casta presentaba la colonia. Tomamos medidas de 248 hembras reproductivas y no reproductivas y la diferenciacion de castas rue analizada por medio de un Analisis Discriminante. El tamano de la colonia se determino contando uno por uno los individuos de la colonia. No encontramos una alta distincion de castas entre los miembros de la colonia de P. aequatorialis, sin embargo el ANOVA mostro que reinas y obreras diferian en todas las medidas morfometricas. Las diferencias morfologicas encontradas entre las hembras reproductivas y no reproductivas son probablemente el resultado deundesarrollo en "switch" un sindrome de casta caracteristico del genero Polybia.


Social organization in insects is complex and varies across the main social groups; the study of social organisms across taxa has been of great importance in understanding the evolutionary trends in animal societies. Within Hymenoptera, the degree of social organization ranges from small societies formed by a few individuals without morphological differentiation of castes (Polistes and Mischocyttarus) and weak reproductive division of labor, to societies composed of millions of individuals with morphological castes and a strong associated reproductive division of work (Formicidae and Apidae: Apini) (Bourke, 1999).

Colony size seems to be related to several of features that characterize these eusocial hymenopteran societies. Species that have colonies made-up of relatively few individuals tend to have little or no morphological differentiation between queens and workers; the worker caste tends not to be polymorphic; the nests and communication system are relatively simple (Bourke, 1999); and per capita productivity is high (Karsai and Wenzel, 1998). In contrast, in species that produce large colonies queens and have workers are morphologically distinct, there is high polymorphism among the worker caste; the nests and systems of communication are complex (Bourke, 1999), and per capita productivity is low (Wenzel and Karsai, 1998). According to Wenzel and Karsai (1998), species with larger size colonies offers advantages despite the fact that species with smallest colonies produce descendents in a more efficient way. Larger colonies allow for a greater protection against enemies, smaller variation in the productivity of the colony, greater labor capacity, greater ability to manipulate the environment, greater and a more predictable reproductive output that guarantees perennial survival, and enhanced resistance in seasonal climates (Bourke, 1999).

Behavioral and morphological differentiation of individuals in social groups can clearly be beneficial as individuals maximize their performance in the society. In large groups, the evolution of morphologically distinct individuals favors a consolidated defense by some members of the group without the need to carry out an emergency change of caste in the adult state when one of the castes is scarce (West-Eberhard, 1981). However, this should be true only when 1) caste differentiation is not associated with age and reproductive development (Metapolybia aztecoides Richards [West-Eberhard et al., 1995]), 2) females size, color and physiological characteristics are constant through the seasons (some temperate zone Polistes [West- Eberhard, 1969]); 3) there is no social selection for queens with an intermediate phenotype (Noll and Zucchi, 2000).

In some species, caste differentiation is plastic and is determined by colony cycle. In these cases, there is a fluctuation in queen number caused by their removal by workers who are constantly testing the queens and determining whether or not they should be removed from the colony (West-Eberhard, 1978).

Colony sizes in species of social wasps in the tribe Epiponini generally vary from hundreds to thousands of individuals. The largest colony size comes from the Epiponini Agelaia vicina (de Saussure), which can have up to one million individuals (Zucchi et al., 1995). Colonies in species from the genus Polybia generally do not exceed a few thousand individuals (Table 1), with the exceptions of P. liliacea (F.), P. scutellaris (White) (Noll et al., 2004) and P. dimidiata (Olivier) (Shima et al., 1996) which have colonies with more than 20,000 individuals. Interestingly, castes in these three species are clearly more distinct than in other Polybia (Noll et al., 2004). This suggests a possible relationship between colony size and caste distinction.

Polybia aequatorialis (Zavattari) is a species that is found in the high regions of Costa Rica between the 1,150 and 3,200 m (Richards, 1978; O'Donnell, 2000). Little information is known about this species, but colonies found in the Monteverde region at 1,500 m are generally not larger than 10,000 individuals (Hanson pers. comm.). In this work, we studied caste differences in a colony of Polybia aequatorialis with a colony size larger than previously recorded for this species.


Even though the study of several colonies is beneficial in terms of obtaining additional information on colony cycles (Noll & Zucchi, 2000), the information gained from examining a single colony provides important and useful information regarding caste syndromes and is an important starting point for future studies (Noll et al., 2004). For this study, we collected a single P. aequatorialis colony approximately 3,000 m above sea level at the patramo of the Cerro de la Muerte biological research station (9[degrees]56'17 10"N and

83[degrees]74'06 29"W) in August of 2004. The Cerro de la Muerte paramo is characterized has a mean temperature of 12[degrees]C and ranges from -2[degrees]C to 22[degrees]C; the average annual precipitation is 2,500 mm and the zone is often cloudy and exposed to cold rain and strong winds (Boza and Cebo, 1998).

Adults of the colony were censed and collected into 70% ethyl alcohol at approximately 19:00h, when we expected that most of the foragers had returned to the nest. We then collected the nest to allow the juveniles to mature. Each comb of the nest was separated and placed into sealed boxes where the adults eventually emerged. For each comb, we counted the number of cells and recorded their contents as empty, with eggs, larva or pupa.

We measured 248 wasps and classified them as either 'queen' or 'worker' according to reproductive development. We measured MSW (width of mesoscutum), HW (head width), TBW (maximum width of the tergite II) and TMW (minimum width of the tergite II), which have been shown to differ significantly between castes (Noll et al., 2004). MSW and HW were proxies of individual body size, and TBW and TMW show variation among same-classed individuals.

Classification into queens or workers were based on the degree of ovarian development and the color of the anterior edge of the second tergite of the mesostoma, which according to West-Eberhard (pers. comm.) turns darker with age. Individuals classified as queens had completely developed ovaries and an almost black anterior edge. Individuals classified as workers had little or no ovarian development and with a light brown or medium dark anterior edge. Measurements and classifications were taken by one individual to avoid inter-observer variation. Specimens were preserved in 70% ethyl alcohol for voucher.

Discriminant Function Analysis was used on MSW, HW, TBW and TMW to determine whether there was morphometric differentiation between queen and worker castes. To determine which morphometric measurements differed significantly between castes we carried out an ANOVA with a Bonferroni posterior test.


Nest architecture and colony composition

The P. aequatorialis nest hung 1.5 m above the ground from the branches of a bush belonging to the family Ericaceae, and was approximately 55 cm long (Fig. 1). The nest contained 8 combs, although the bottom of the nest was disintegrated and there may have been several more. The combs were concave and stacked one above the other. The nest had approximately 12 layers of fiber covering the combs, with air bags between them. The external layer of the nest was an almost uniform light brown color, and the surface had very few irregularities and giving it an almost smooth appearance. Each comb had an average of five entrances. The pupae were located in the central part of the combs, whereas the combs along the edges were used for storing honey. Nonetheless, fewer were used for storage than for oviposition (Table 2). Another colony of P. aequatorialis found by Weng and Ramirez on Cerro de la Muerte also contained dead insects in some of the cells (unpublished data), what is a common trait found in other Polybia species (Noll et al., 1997).

The colony had a total of 22,093 adults. In a subsample of 2,000 individuals, 2% (N = 45) were queens and the nest was likely to have housed a total of 500 queens. The Weng and Ramirez nest of P. aequatorialis) was slightly smaller and contained 15,300 adults, 9,659 pupas, 1,820 larvae and 1,150 eggs (unpublished data.)

Morphological differences between workers and queens

Wilks' lambda values for the morphometric measurements ranged from 0.44 to 0.54 (Table 3). These values indicate that the distinction between queen and worker castes in P. aequatorialis is not highly pronounced, but is similar to other studied species of Polybia (Noll et al., 2004). These values indicate an overlap in the size distribution of the castes, although there is some distinction between the two castes. HW and TBW explained the differences between queens and workers the best (Table 3; Fig. 2).

The ANOVA showed that queens and workers differed in all morphometric measures (F = 77.38 df = 243 P < 0.0001). Queens were consistently larger than the workers (Table 4). Ninety-three percent (174 correct, 5 incorrect), of the worker sample and 91% (57 correct, 12 incorrect) of the queens were classified correctly in the DA.



Queen and worker morphology

Castes in P. aequatorialis are similar in distinction to other Polybia previously studied, in which all measurements differ between queens and workers (Noll & Zucchi, 2000; Noll et al., 2004). Because the type of caste difference (Fig. 2) is similar to what was previously found in other epiponine species (Polybia, Protonectarina, Agelaia and Apoica; Noll et al., 2004), caste determination in P. aequatorialis may be caused by a developmental switch. In this 'caste syndrome', the allometric differences correspond to two size classes, in which the large individuals are differentiated from the small ones in shape or size and this generates discrete and disjunctive variation (Noll et al., 2004) (Fig. 2). In species with this syndrome, the workers and queens are clearly distinct. In P. aequatorialis the same pattern occurs because castes are highly correctly classified by their morphology (93% of workers and 91% of queens). It is possible that the absence of queens with an intermediate phenotype in P. aequatorialis indicates that there may be social selection on queen phenotype, which has been observed in other Polybia species (Noll and Zucchi, 2000). Even though P. aequatorialis presents pre-imaginal caste determination, according to the Wilks' lambda values from 0.4 to 0.5 (Table 3) the differences we found are not as strikingly distinct as in other Polybia with more pronounced dimorphism and higher colony size (Wilks' lambda values from 0.1 to 0.2; Noll et al., 2004).


Nest architecture and colony size

As in perennial nests of epiponines (Richards, 1978), the architecture of the P. aequatorialis nest also seems to be designed to minimize the external effects of the environment, in particular the large quantity of fiber layers that cover the combs. The nest studied here is larger (about 1.5 m in length) than other nests from the same species but with smaller colony size. If colonies of P. aequatorialis in Monteverde do not surpass 10,000 individuals (Hanson pers. com.), it is possible that the conditions in Cerro de la Muerte may favor larger colony sizes.

A colony consisting of a large number of individuals can generate certain benefits for P. aequatorialis. For example, a greater number of foragers would allow for more food reserves to be stored in the nest thereby allowing the colony a better chance of surviving periods of inclement weather. According to Jeanne (1991) large colonies can store large quantities of honey and build strong nests in which it is easier to maintain homeostasis. Greater survival allows for more a permanent colony, which continues growing through several years (perennial cycles instead of annual cycles), which in turn permits the construction of a more resistant and structurally complex nest (Bourke, 1999).

In summary, our study shows that P. aequatorialis may have a caste syndrome that is relatively similar to other Polybia, although caste differentiation not as highly pronounced as species of Polybia with larger colony sizes. Species of social insects that form average maximum large colonies tend to have a strong morphological dimorphism and workers with low reproductive potential (Bourke, 1999). As an exploratory study, only one colony was used for this study. However, with the collection and study of multiple colonies of P. aequatorialis, we may be able to determine whether morphological differentiation between castes varies in relation to the colony size and also whether and how elevation plays a role in colony social dynamics.


Special thanks to Paul Hanson for his helpful suggestions during the accomplishment of this work and comments of the final manuscript. We also thank Julin Weng, William Eberhard, Gilbert Barrantes, Mary Jane West-Eberhard (University of Costa Rica) and Christine Johnson (The American Museum of Natural History) for their comments, suggestions and helpful ideas; Guido Saborio, Cesar Sanchez and Federico Bolanos for their help in collecting of the Nest. The authors acknowledge the financial support by FAPESP (grant 07/08633-1) and CNPq (grant 304247/2007-3).

Received 11 July 2008; accepted 12 June 2009


Bourke, A. F. G. 1999. Colony size, social complexity and reproductive conflict in social insects. Journal of Evolutionary Biology 12: 245-257.

Boza, M. A. and J. H. Cevo. 1998. Parques Nacionales y otras areas protegidas en Costa Rica. Incafo Costa Rica. S.A. San Jose, Costa Rica, 222 pp.

Jeanne, R. L. 1991. The swarm-founding Polistinae. pp. 191-231. In: K. G. Ross and R. W. Matthews (eds.), The Social Biology of Wasps. Cornell University Press, New York, 678 pp.

Karsai, I. and J. Wenzel. 1998. Productivity, individual-level and colony-level flexibility, and organization of work as consequences of colony size. Proc. Natl. Acad. Sci. U.S.A. 95:8665-8669.

Noll, F. B., R. Zucchi and S. Mateus. 1997. Morphological caste differences in the Neotropical swarm-founding and polygynous Polistinae wasp, Polybia scutellaris. Studies on Neotropical Fauna and Environment 32: 76-80.

Noll, F. B. and R. Zucchi. 2000. Increasing caste differences related to life cycle progression in some neotropical swarm-founding polygynic wasps. (Hymenoptera: Vespidae: Epiponini). Ethology, Ecology and Evolution 12(1): 43-65.

Noll, F. B., J. Wenzel and R. Zucchi. 2004. Evolution of caste in Neotropical swarm-founding wasps (Hymenoptera: Vespidae; Epiponini). American Museum Novitates 3467: 1-24.

O'Donnell, S. 2000. Eusocial wasps (Vespidae: Polistinae). pp. 129-133. In: N. M. Nadkarni and N. T. Wheelwright (eds.), Monteverde, Ecology and Conservation of a Tropical Cloud Forest. Oxford University Press, New York, 573 pp.

Richards, O. W. 1978. The Social Wasps of the Americas, Excluding the Vespinae. British Museum (Natural History), London.

Shima, S. N., S. Yamane and R. Zucchi. 1996. Morphological caste differences in some Neotropical swarm-founding polistine wasps II. Polybia dimidiata (Hymenoptera, Vespidae). Japanese Journal of Entomology 64(1): 131-144.

West-Eberhard, M. J. 1969. The social biology of Polistinae wasps. Miscellaneous Publications of the Museum of Zoology, University of Michigan 140: 1-101.

West-Eberhard, M. J. 1978. Temporary queens in Metapolybia wasps: non-reproductive helpers without altruism? Science 200: 441443.

West-Eberhard, M. J. 1981. Intragroup selection and the evolution of insect societies, pp. 3-17 in R. D. Alexander and D. W. Tinkle (eds.), Natural selection and Social Behavior. Chiron Press, New York.

West-Eberhard, M. J., J. M. Carpenter and P. E. Hanson. 1995. The vespid wasps (Vespidae), pp. 561-587. In: P. E. Hanson and I. D. Gauld (eds.), The Hymenoptera of Costa Rica. The Natural History Museum, Oxford University Press, Oxford, 893 pp.

Zucchi, R., S. F. Sakagami, F. B. Noll, M. R. Mechi, M. V. Baio, S. Mateus and S. N. Shima. 1995. Agelaia vicina, a swarm-founding Polistinae with the largest colony size among wasps and bees (Hymenoptera, Vespidae). Journal of the New York Entomological Society 103(2): 129-137.


(1) Departamento de Biologia, Faculdade de Filosofia Ciencias e Letras de Ribeirao Preto, Universidade de Sao Paulo, Laboratorio de Vespas Sociais Departamento de Zoologia e Botanica, IBILCE--UNESP, Rua Cristovao Colombo, 2265 CEP 15055-240, Sao Jose do Rio Preto--SP, Brasil.

(2) Directoria del Programa Jovenes Triunfadores, Asociacion para la Defensa de la Vida, Zapote, San Jose, Costa Rica, 200 metros este de la rotonda de la Y griega, Zapote, San Jose, Costa Rica.

(2) E-mail address for correspondence: laurachp@

(3) Laboratorio de Vespas Sociais Departamento de Zoologia e Botanica, IBILCE--UNESP, Rua Cristovao Colombo, 2265 CEP 15055-240, Sao Jose do Rio Preto--SP, Brasil.
Table 1. Colony size (only adults) for colonies of Polybia in
Latin America according to Richards (1978).

    Species       Colony size            Place            Altitude

P. striata           7,000      Brasil, Mato Grosso       200-600 m
P. simillima          350       Barra Colorado, Panama      200 m
P. guadricincta      2,000      Brazil, Mato Grosso       200-600 m
P. bistriata        50-100      Brazil, Mato Grosso       200-600 m
P. juruana        2,000-3,000   Valle Chanchamayo, Peru     751 m
P. scutellaris    3,500-4,300   Brazil, Piracicaba          554 m
P. paulista       3,000-5,000   Brazil, Sao Paulo         0-1,135 m
P. parvulina        100-900     Guyana                      600 m

Table 2. Use of cells in the different combs of the
studied nest of Polybia aequatorialis.

Comb    Cells with cocoons   Cells with honey   Total

  1            3,200              1,144          4,344
  2            3,000                800          3,800
  3            3,384                416          3,800
  4            2,760                424          3,184
  5            2,960                384          3,344
  6            1,584                192          1,776
  7            1,256                208          1,464
  8              528                216            744
Total         1,8672               3784         2,2456

Table 3. Wilks' lambda values for the Stepwise
Discriminant Function Analysis of the sample of females
from Polybia aequatorialis, N = 248.

Measure   Wilks' lambda     F      Probability

AC            0.489       27.362     >0.0001
AT            0.445        3.065      0.081
AMAXP         0.544       57.384     >0.0001
AMINP         0.444        2.540      0.112

Table 4. Mean and standard deviation (mm) for the
morphometric measures of queens and workers of
Polybia aequatorialis.

                   Queens              Workers
                  Standard            Standard
Measure   Mean    deviation   Mean    deviation

AC        2.504     0.051     2.391     0.075
AT        2.571     0.045     2.459     0.076
AMAXP     1.019     0.036     0.901     0.058
AMINP     0.365     0.022     0.347     0.022
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.