On-line version ISSN 0718-560X
Lat. Am. J. Aquat. Res. vol.40 no.1 Valparaíso Mar. 2012
Lat. Am. J. Aquat. Res., 40(1): 134-141, 2012
Territorial hypothesis predicts the trade-off between reproductive opportunities and parental care in three species of damselfishes (Pomacentridae: Actinopterygii)
La hipótesis territorial predice la compensación entre las oportunidades reproductivas y el cuidado parental en tres especies de peces damisela (Pomacentridae: Actinopterygii)
Eduardo Bessa1,2 & José Sabino3
1Laboratório de Ecologia Comportamental da Reprodução, Universidade do Estado de Mato Grosso Rod MT 358 km 7, Jd. Aeroporto, CEP 78300-000. Tangará da Serra-MT, Brazil.
2Depto. de Zoologia e Botânica, UNESP, Instituto de Biologia, Letras e Ciências Exatas Programa de Pós-Graduação em Biologia Animal.
3Universidade Anhanguera-Uniderp, Projeto Peixes de Bonito, 79037-280 Campo Grande, MS, Brazil.
ABSTRACT. Paternal care is rare in most animals, but common in fishes, including Pomacentridae. One way of explaining it is through the territorial hypothesis, which predicts that paternal care may evolve when caring for a territory with eggs will not cost more than territories without eggs. Here we describe the reproductive behavior of three pomacentrids from São Paulo, Brazil, and evaluate if the territorial hypothesis explains their behavioral variation through five predictions. We checked if females have multiple spawns (prediction 1), if they forage more (prediction 2) and if territory defence demands more when eggs are present (prediction 3). We also described how visible eggs are (prediction 4) and how available nesting sites are (prediction 5). Chromis multilineata differed from Abudefduf saxatilis and Stegastes fuscus by not guarding the eggs, spending less time with territoriality and fanning eggs; and more time feeding and foraging. Females of the three species are iteroparous. Females' A. saxatilis and S. fuscus foraged more than males, which spent the same amount of time in territorial defence before and after eggs were delivered. These two species also have very conspicuous egg clutches on clean rock surfaces defended by them. On the other hand, males and females of C. multilineata spent the same time foraging, while territoriality took more effort after eggs were present and it spawns cryptic eggs amid Sargassum, a common and disperse micro-environment in São Paulo rocky shores. Territorial hypothesis explains why paternal care by A. saxatilis and S. fuscus lasts longer than by C. multilineata.
Keywords: reproductive behavior, resource defence, rocky reef fish, behavioral ecology, sex role reversal, southwestern Atlantic.
RESUMEN. El cuidado paterno es raro en la mayoría de los animales, pero común en los peces, incluyendo los Pomacentridae. Una forma de explicar este fenómeno es la hipótesis territorial, que predice que el cuidado paterno puede evolucionar cuando la protección de un territorio con huevos no cuesta más que la de territorios sin huevos. Aquí se describe el comportamiento reproductivo de tres pomacentrídios de São Paulo, Brasil, y se evalúa si la hipótesis territorial explica la variación de su comportamiento a través de cinco predicciones. Se analizó si las hembras tienen múltiples desoves (predicción 1), si forrajean más (predicción 2) y si la defensa del territorio es mayor cuando los huevos están presentes (predicción 3). También se evaluó si los huevos eran visibles (predicción 4) y cuán disponibles eran los sitios de anidación (predicción 5). Chromis multilineata difería de Abudefduf saxatilis y Stegastes fuscus por no proteger a sus huevos, dedicar menos tiempo a la territorialidad y ventilación de sus huevos, y más tiempo para la alimentación y forrajeo. Las hembras de las tres especies son iteróparas. Las hembras de A. saxatilis y S. fuscus forrajean más que los machos, que gastan la misma cantidad de tiempo en la defensa territorial antes y después de liberar los huevos. Estas dos especies también defienden a desoves muy conspicuos adheridos en superficies de roca limpia. Por otra parte, las hembras y machos de C. multilineata pasaron el mismo tiempo forrajeando, mientras la territorialidad tomó más esfuerzo después, con los huevos presentes, y los desoves estaban escondidos en medio de Sargassum, un ambiente común y disperso en micro-ambientes de las costas rocosas de São Paulo. La hipótesis territorial explica el motivo por el cual el cuidado paterno de A. saxatilis y S. fuscus es más prolongado que el de C. multilineata.
Palabras clave: comportamiento reproductivo, defensa de recursos, peces de costa rocosa, comportamiento ecológico, reversión del papel sexual, Atlántico suroccidental.
Reproduction is the final target of any organism. It is through reproduction that the organism passes away its genetic information to keep present in the gene pool of its population. Despite of that, to reproduce is useless if the offspring itself does not reach the reproductive age, what may demand parental care. This behavior brings many costs and a trade-off between reproducing and taking care of the offspring is expected (Clutton-Brock, 1991). Parental care is usually accomplished by females, since they benefit more from the energy invested in the offspring than males (Queller, 1997), although, in many cases males exert this (Ridley, 1978).
There are various explanations for why males care for their offspring (Wright, 1998). Williams (1975) provided one of the most convincing explanations. He suggests that prolonged paternal care will evolve when males already defend territories including a rare spawning site to attract females as a resource defence polygyny (Emlem & Oring, 1977). The presence of eggs is also supposed to attract more females (Manica, 2010). Birds and fishes are especially prone to paternal care (Ridley, 1978); among them coral reef damselfishes are known to be devote fathers (Breder & Rosen, 1966).
The Pomacentridae family (damselfishes) belongs to suborder Labroidei, Perciformes, Actinopterygii (Nelson, 1994). They are divided in 28 genera and 384 species (Froese & Pauly, 2011) distributed through all the tropical oceans. Restricted to shallow regions with hard substrate, their main region of diversity is the South Pacific Ocean. In Brazil four genera and 14 species are known (Moura & Menezes, 2003). These species are explored for aquarium trade, from January 1995 to November 2000 a total of 6.800 individuals were traded in Ceará state (Monteiro-Neto et al., 2003). The present study focused three of the most frequent species in São Paulo State northern coast (Menezes & Figueiredo, 1985): Abudefduf saxatilis, Stegastes fuscus and Chromis multilineata. These species form territories in at least part of the reproductive season; males court females, fecundate the eggs and guard them. How long the paternal care lasts, what is guarded in the territory and what kind of substrate is used for spawning varies between species (Emery, 1973).
Although Williams (1975) explains most of the parental care in fishes, it is not known to explain the variation in paternal effort in pomacentrids. This paper aims to compare A. saxatilis, S. fuscus and C. multilineata, and then define what aspects of their biology may explain how much males care for their offspring. For that we test the following predictions drawn from territorial hypothesis: (1) female should be iteroparous, (2) females should forage and feed more than guarding males, (3) time budget dedicated to territoriality should not change from territorial to parental males, (4) egg-clutches should be easily spotted, (5) egg-clutches should be placed on rare defendable resources.
Iteroparity is expected since females will not benefit from paternal males if they cannot, ecologically or physiologically, lay multiple clutches per breading season (Tallamy, 2000). The second prediction is based on the idea that, by feeding more, females will increase their fecundities; this was reviewed by Armstrong & Witthames (2011). If taking care of an empty territory or one filled with eggs takes the same amount of time and offspring survival is greatly enhanced by parental care, then, we may expect that paternal care will occur, what takes us to prediction three. The fourth prediction that clutches will be easily seen has two consequences; it may be easily seen and evaluated by a female searching for a good father to deliver her eggs to (Sikkel, 1995), albeit it may also result in being easily spotted by a visual oriented egg predator, common on reefs (Lowe-McConnell, 1999). Finally, territoriality will only occur when a resource for which animals would compete has an aggregated pattern of distribution (Wilson, 2000), thus, a territorial explanation for paternal care requires first that the parents are territorials.
MATERIALS AND METHODS
The present study was carried out in São Sebastião and Ilhabela, northern shore of São Paulo, Brazil (23o49'S, 45o25'W). The region is characterized by the presence of São Sebastião Channel and by its predominantly rocky shore, most of which is covered with algae, soft corals and sponges. São Sebastião Island represents an obstacle for winds and more energetic events, with less turbulent water inside the channel.
Collects and fecundity analysis
Seventeen mature Abudefduf saxatilis females were collected with hook and bait and killed with benzocaine at 100 ppm for 20 min, animals were then injected with formalin at 10%. They were dissected and gonads were submitted to histology. We used haematoxylin-eosin (HE) method with the gonads included in paraffin and cut in 5 μιη slices. Only one female of Stegastes fuscus and no Chromis multilineata were collected. The same treatment given to A. saxatilis was done to S. fuscus but further data for C. multilineata and S. fuscus iteroparity were taken from literature (Myrberg et al., 1967; Souza et al., 2007). Observing how distended the visiting females' abdomens were after they left the nest, as commonly used in aquaculture (Woynarovich, 1989), offered an indirect and complementary approach to the presence of iteroparous females in the study area.
We did 114 h of SCUBA diving in a mean depth of 4 m during day time from November 2002 to February 2003, including finding individuals, nest marking, acclimatation period, evaluating algae coverage and egg location (for C. multilineata). Observed individuals were standardized for size (A. saxatilis ~150 mm, S. fuscus ~130, C. multilineata ~110 mm). Observations started after a 5 min acclimation interval.
We began by doing an ethogram listing and describing all the behaviors of interest with ad libitum method (Martin & Bateson, 2007). Meanwhile we also checked where eggs were placed and how easily seen they were. Nests were marked with a coloured buoy anchored to the bottom by a fishing weight and accompanied four times a day to evaluate parent desertion and egg hatching.
We than did 15 min sessions using focal animal with instant samples every 30 s (Martin & Bateson, 2007) measuring the frequency of each behavior.
Underwater notes were taken with soft pencils on polyethylene waterproof sheets with tables printed in laser printers. Focal animals were identified by species, life stage (with or without eggs) and sex.
Abudefduf saxatilis may be sexed by its color and behavior (males are bluish and guard a territory). Male and female Stegastes fuscus were identified by accompanying a marked nest previously filled with eggs (Since only males guard the eggs; Souza et al., 2007) or by their courting behavior. Chromis multilineata could be sexed by the presence of a swallowed genital papilla in females (Myrberg et al., 1967; Swerdloff, 1970).
An analysis of similarity using Euclidian distance (ANOSIM) (Costa et al., 2010) was used to compare the behaviors of the three species with 1000 permutations. A principal component analysis (PCA) was used to order the sampled individuals by the number of observations of each behavior (Manly, 2008) . A Mann-Whitney sum of ranks test (Zar, 1999) checked the difference in the time spent foraging between males and females within each species due to the heterocedastic nature of the data. Following, a t-test (Zar, 1999) compared the time dedicated to territoriality between males guarding territories with or without eggs, since these data were normally distributed. Type I error tolerance was α = 0.05.
Fourteen behaviors were observed (Table 1), although, they varied greatly among the three species (ANOSIM = 0.6248; N = 300; P < 0.001). Abudefduf saxatilis and Stegastes fuscus behave more similarly than Chromis multilineata. This may be observed both in the PCA (Fig. 1), where these two species overlay while C. multilineata is concentrated on the left of axis 2, and in the barplot (Fig. 2), where the frequency of the main behaviors are similar for A. saxatilis and S. fuscus, but differ from C. multilineata. The PCA pointed that the vectors 'guarding' and 'fanning' ordered A. saxatilis and S. fuscus, while 'feeding' and 'foraging' influenced C. multilineata (Fig. 1). The first and second axis explained 80.2% of the data dispersal.
|Table 1. Behavioral repertoire of the pomacentrids from São Paulo northern shore. |
Tabla 1. Repertorio comportamental de los pomacentrídeos de la costa norte de São Paulo.
Figure 1. Principal component analysis (PCA) of the behaviors of Abudefduf saxatilis, Stegastes fuscus and Chromis multilineata. Arrows represent the behaviors that most affect each species. Notice how A. saxatilis and S.fuscus overlay and are distinct of C. multilineata.
Figura 1. Análisis de componentes principales (PCA) de los comportamientos de Abudefduf saxatilis, Stegastes fuscus y Chromis multilineata. Las flechas representan los comportamientos que afectan más a cada especie. Notar que A. saxatilis y S. fuscus se sobreponen y son distintos de C. multilineata.
Figure 2. Frequency of the main behaviors executed by the damselfishes stressing the difference between the paternal species (Abudefdufsaxatilis and Stegastes fuscus) and the less territorial one (Chromis multilineata).
Figura 2. Frecuencia de los principales comportamientos hechos por los pomacentridios marcando las diferencias mientras las especies paternales (Abudefduf saxatilis y Stegastes fuscus) y la poco territorial (Chromis multilineata).
Female Abudefduf saxatilis presented multiple spawning with oocytes ready to be delivered and atretic follicles in their ovaries in the 17 collected females. The single mature female of Stegastes fuscus collected also presented atretic follicles and mature eggs, suggesting multiple spawnings. Furthermore, the remaining turgid abdomens seen on females leaving the nest site may be due to oocytes still available for future mates.
Females of A. saxatilis and S. fuscus, but not C. multilineata, spend more time foraging and feeding than males (Table 2). Males of A. saxatilis fed almost five times less than females, while S. fuscus males forage just about 39% less. This time budget is directed to other activities such as territoriality and parental care, while females certainly increase energy intake. Chromis multilineata males foraged just a little less than females (15%).
The same happened with time invested in territory defence before and after eggs were delivered to the territorial male. A. saxatilis and S. fuscus males spend the same time budget on territory defence with or without eggs (less than 40% increase for A. saxatilis and about 35% decrease for S. fuscus), but the investment in territory defence by C. multilineata raised significantly (more than 10-fold) for the short time these males held their territories (Table 3).
A. saxatilis and S. fuscus spawn on clear rock surfaces kept by their mates in the territory. A. saxatilis clutches are purple patches of eggs of around 250 cm2 adhered to rocky surfaces in the territory of a male. The eggs of S. fuscus form white elliptic clutches of around 200 cm2. Both these species have to build their nests clearing rock areas with their teeth. C. multilineata eggs are cryptic and disperse amid Sargassum tuffs (see Fig. 3 in Myrberg et al., 1967).
Pomacentrid males invest their time and energy in parental care. Parental care in Abudefduf saxatilis and Stegastes fuscus lasts until the larvae hatch, parents defend offspring from predators, fan and clean the eggs. These two species agree in every aspect with the general pattern suggested for pomacentrids by Breder & Rosen (1966) and Balon (1975). They present nuptial color (A. saxatilis) or no secondary sex character (S. fuscus), mate in pairs, spawn demersal adhesive eggs on rocks, which will be guarded and fanned by the male. Both are egg-guarders who belong to the nesting sub-guild and are litophyles. On the other hand, Chromis multilineata, despite showing no colour dimorphism, mating in pairs and spawning demersal adhesive eggs, does not guard the eggs for more than a few hours, hiding them amid Sargassum banks in São Paulo coast, disagreeing markedly from previous classifications of Pomacentridae.
Since parental care is probably a characteristic inherited from their sister-group, Cichlidae, and C. multilineata is a derived taxon within Pomacentridae (Cooper et al., 2009), its lack of paternal behavior is probably a secondary loss within Pomacentridae. If so, what are the possible causes for this loss? The predictions drawn from the classic book of Williams (1975) have pointed us some cues.
The first prediction suggested that, for paternal care to evolve, females should be iteroparous, so that they may benefit from guarding males (Tallamy, 2000). Results confirmed iteroparity in A. saxatilis and S. fuscus (presence of atretic follicles in fertile females), while abdominal turgidity suggested the same for all the three species. This was already expected from other studies (Myrberg et al., 1967 for C. multilineata; Bessa et al., 2007 for A. saxatilis; and Souza et al., 2007 for S. fuscus); what confirms the first prediction for all the three species.
Our second prediction, that females were expected to forage and feed more than males, was proven only for the two paternal species. Males of C. multilineata forage as much as females. This extra feeding time females have may be converted in more numerous or larger eggs (Tsadik & Bart, 2007), increasing both the fitness of females and males that reproduced with them through indirect fitness (Donelson et al., 2008).
Time invested in territory defence did not increase for A. saxatilis and S. fuscus (prediction 3), but rose significantly for C. multilineata. Both sexes of the latter species can forage equally because neither spends time on territory defence. This will only change when males start courting females, moment when their territorial behavior increases. Other species from the genus Chromis have been described as territorials and paternal (Fishelson, 1970; Tzioumis & Kingsford, 1999). Even C. multilineata, in reefs where Sargassum were less abundant, defended algae tuffs and guarded them after spawning (Myrberg et al., 1967). It is important to mention that S. fuscus territoriality is similar to A. saxatilis־' during the reproductive period only. This species will stay guarding a territory for its feeding value (algae garden) in the cold months of the year, while male A. saxatilis have left their territories and spend the time foraging in the water column.
The easily seen eggs of A. saxatilis and S. fuscus confirmed prediction four. This will allow females to use it to choose partners. Clutches were much more conspicuous in the two former species and egg condition is known to be a cue for male quality (Sikkel, 1995; Manica, 2010). In C. multilineata a series of female visits to a male nest were described to happen one after the other in a short interval (Myrberg et al., 1967), what we also observed in São Paulo. Therefore, eggs are probably not used by females to choose partners and paternal care is less interesting in terms of sexual selection. C. Chromis multilineata females probably use more cues from the male than from eggs he would be guarding.
The same conspicuous clutches that attract females can be spotted by visually oriented predators, increasing the need for nest protection. Some common visually oriented egg predators found in São Paulo rocky shores are Abudefduf saxatilis, Chaetodon striatus and Diplodus argenteus (Randall, 1967), amid many others.
Rare nest sites distributed in patches are a defendable resource to be guarded in territories (Maher & Lott, 1995). Our last prediction suggests that paternal males are supposed to be territorials since before reproductive period. Territorial males will have to spend time guarding a territory as much as they will spend on defending its offspring, reducing the costs of paternal care and, thus, explaining the evolution of this kind of sex role reversal. As long as A. saxatilis has to guard the clear rock nest he built and S. fuscus has the nest and an algae garden to guard in its territory, the extra cost driven to parental care is paid by the benefits of enhancing offspring survival. Since in rocky shores with large Sargassum banks covering up to 80% of the substrate, a common situation in São Paulo coast (Paula & Oliveira-Filho, 1980), algae tuffs are not a rare resource distributed in patches. It is less likely that C. multilineata will develop territoriality for guarding this nest site, augmenting the costs of parental care by the male.
We conclude that the variation in male parental care in the pomacentrids of São Paulo northern coast is explained by the territorial hypothesis, mainly in accordance to the availability of nest sites and the need to guard it as a scarce resource.
The authors would like to thank the financial support of Conselho Nacional de Desenvolvimento Científico e Tecnológico, CNPq (process N°130812/2002-1) and Coordenadoria de Pessoal de Nível Superior through the program of research support, PROAP-CAPES; also the Centro de Biologia Marinha of Universidade de São Paulo, CEBIMar-USP, for logistic support. Further thanks are due to two anonymous referees for their valuable suggestions, Flavia Krauss and Judy Finlayson for Spanish and English review, respectively. EB thanks José Renato Leite for his continuous support and Ana Maria de Souza for supervision. JS is fellow of CNPq (process 306169/2008-8).
Armstrong, M.J. & P.R. Witthames. 2011. Developments in understanding of fecundity of fish stocks in relation to egg production methods for estimating spawning stock biomass. Fish. Res. (in press). [ Links ]
Balon, E.K. 1975. Reproductive guilds of fishes: a proposal and definition. J. Fish. Res. Board Can., 32(6): 821-863. [ Links ]
Bessa, E., J.F. Dias & A.M. Souza. 2007. Rare data on a rocky shore fish reproductive biology: sex ratio, length of first maturation and spawning period of Abudefduf saxatilis (Linnaeus, 1758) with notes on Stegastes variabilis spawning period (Perciformes: Pomacentridae) in São Paulo, Brazil. Braz. J. Oceanogr., 55: 199-206. [ Links ]
Breder Jr., C.M. & D.E. Rosen. 1966. Modes of reproduction in fishes: how fish breed. TFH Publications, Jersey City, 941 pp. [ Links ]
Clutton-Brock, T.H. 1991. The evolution of parental care. Princeton University Press, Princeton, 368 pp. [ Links ]
Cooper, W.J., L.L. Smith & M.W. Westneat. 2009. Exploring the radiation of a diverse reef fish family: Phylogenetics of the damselfishes (Pomacentridae), with new classifications based on molecular analyses of all genera. Mol. Phylogenet. Evol., 52: 1-16. [ Links ]
Costa, D.A., L.C.G. Vieira, L.C.F. Silva, R.A. Carvalho, M.V.C. Vital, G.F. Lima-Filho, A.V.T. Silveira & B.B. Santos. 2010. Métodos para comparação de comunidades: um estudo da fauna de invertebrados de duas áreas da região metropolitana de Goiânia. Estudos, 37(2): 369-379. [ Links ]
Donelson, J.M., M.I. Mc Cormick & P.L. Munday. 2008. Parental condition affects early life-history of a coral reef fish. J. Exper. Mar. Biol. Ecol., 360: 109-116. [ Links ]
Emery, A.R. 1973. Comparative ecology and functional osteology of fourteen species of the damselfish (Pisces: Pomacentridae) at Alligator Reef, Florida Keys. Bul. Mar. Sci., 23: 649-770. [ Links ]
Emlem, S.T., & L.W. Oring. 1977. Ecology, sexual selection and the evolution of mating systems. Science, 197: 215-223. [ Links ]
Fishelson, L. 1970. Behavioral observations on eniwetok damselfishes (Pomacentridae: Chromis) with special reference to the spawning of Chromis caeruleus. Copeia, 1970(2): 371-374. [ Links ]
Lowe-McConnell, R.H. 1999. Estudos ecológicos de comunidades de peixes tropicais. EdUSP, São Paulo, 535 pp. [ Links ]
Maher, C.R. & D.F. Lott. 1995. Definitions of territoriality used in the study of variation in vertebrate spacing systems. Anim. Behav., 49: 1581-1597. [ Links ]
Manica, A. 2010. Female scissortail sergeants (Pisces: Pomacentridae) use test eggs to choose good fathers. Anim. Behav., 79: 237-242. [ Links ]
Manly, B.J.F. 2008. Métodos estatísticos multivariados: uma introdução. Artmed, Porto Alegre, 230 pp. [ Links ]
Martin, P. & P. Bateson. 2007. Measuring behaviour, an introductory guide. Cambridge University Press, Cambridge, 176 pp. [ Links ]
Menezes, N.A. & J.L. Figueiredo. 1985. Manual de peixes marinhos do sudeste do Brasil. V Teleostei (4). São Paulo, 105 pp. [ Links ]
Monteiro-Neto, C., F.E.A. Cunha, M.C. Nottingham, M.E. Araujo, I.L. Rosa & G.M.L. Barros. 2003. Analysis of the marine ornamental fish trade at Ceara State, northeast Brazil. Biodivers. Conserv., 12: 1287-1295. [ Links ]
Moura, R.L. & N.A. Menezes. 2003. Pomacentridae. In: N.A. Menezes, P.A. Buckup, J.L. Figueiredo & R.L. Moura (eds.). Catálogo das espécies de peixes marinhos do Brasil. Museu de Zoologia USP, São Paulo, pp. 91-92. [ Links ]
Myrberg Jr., A.A., B.D. Brahy & A.R. Emery. 1967. Field observations on reproduction of the damselfish, Chromis multilineata (Pomacentridae), with additional notes on general behavior. Copeia, 1967(4): 818-827. [ Links ]
Nelson, J.S. 1994. Fishes of the world. John Wiley & Sons, New York, 600 pp. [ Links ]
Paula, E.J. & E.C. Oliveira-Filho. 1980. Phenology of two populations of Sargassum cymosum (Phaeophyta - Fucales) of São Paulo State coast, Brazil. Bol. Inst. Bot., 8: 21-39. [ Links ]
Queller, D.C. 1997. Why do females care more than males?. Proc. Roy. Soc. Lond. Ser. B., 264: 1555-1557. [ Links ]
Randall, J. 1967. Food habits of reef fishes of the West Indies. Stud. Trop. Oceanogr. Miami, 5: 665-847. [ Links ]
Ridley, M. 1978. Paternal care. Anim. Behav., 26: 904-932. [ Links ]
Sikkel, P.C. 1995. Diel periodicity of spawning activity in a permanently territorial damselfish: a test of adult feeding hypothesis. Environ. Biol. Fish., 42: 241-251. [ Links ]
Souza, L.L.G., S. Chellappa & H.C.B. Gurgel. 2007. Biologia reprodutiva do peixe-donzela, Stegastes fuscus Cuvier, em arrecifes rochosos no nordeste do Brasil. Rev. Bras. Zool., 24(2): 419-425. [ Links ]
Swerdloff, S.N. 1970. Behavioral observations on Eniwetok damselfishes (Pomacentridae: Chromis) with special reference to the spawning of Chromis caeruleus. Copeia, 1970(2): 371-374. [ Links ]
Tallamy, D.W. 2000. Sexual selection and the evolution of exclusive paternal care in arthropods. Anim. Behav., 60: 559-567. [ Links ]
Tsadik, G.G. & A.N. Bart. 2007. Effects of feeding, stocking density and water-flow rate on fecundity, spawning frequency and egg quality of Nile tilapia, Oreochromis niloticus (L.). Aquaculture, 272: 380-388. [ Links ]
Tzioumis, V. & M.J. Kingsford. 1999. Reproductive biology and growth of the temperate damselfish Parma microlepis. Copeia, 1999(2): 348-361. [ Links ]
Williams, G.C. 1975. Sex and evolution. Princeton University Press, New Jersey, 210 pp. [ Links ]
Wilson, E.O. 2000. Sociobiology, the new synthesis. Harvard University Press, Cambridge, 697 pp. [ Links ]
Woynarovich, E. 1989. Manual de piscicultura. CODEVASF, Brasília, 71 pp. [ Links ]
Wright, J. 1998. Paternity and paternal care. In: T.R. Birkhead & A.P. Moller (eds.). Sperm competition and sexual selection. Academic Press, London, pp. 117-145. [ Links ]
Zar, J.H. 1999. Biostatistical analysis. Prentice Hall, New Jersey, 663 pp. [ Links ]
Received: 9 March 2011; Accepted: 9 January 2012.
Corresponding author: Eduardo Bessa (email@example.com)