Socio-ecological models combine environmental and social factors to explain the formation of animal groups. In anurans, tadpole aggregations have been reported in numerous species, but the factors driving this behaviour remain unclear. We conducted controlled choice experiments in the lab to determine whether green and golden bell frog (Litoria aurea) tadpoles are directly attracted to conspecifics (social factors) in the absence of environmental cues. Using repeated measures, we found that individual tadpoles strongly preferred associating with conspecifics compared to being alone. Furthermore, this preference was body size dependent, and associating tadpoles were significantly smaller than non-associating tadpoles. We suggest that small tadpoles are more vulnerable to predation and therefore more likely to form aggregations as an anti-predator behaviour. We demonstrate that tadpoles present an ideal model system for investigating how social and ecological factors influence group formation in vertebrates.
Citation: Leu ST, Whiting MJ, Mahony MJ (2013) Making Friends: Social Attraction in Larval Green and Golden Bell Frogs, Litoria aurea. PLoS ONE 8(2): e56460. doi:10.1371/journal.pone.0056460
Editor: Brock Fenton, University of Western Ontario, Canada
Received: October 19, 2012; Accepted: January 9, 2013; Published: February 12, 2013
Copyright: © 2013 Leu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: STL was funded by a Port Waratah Coal Services Limited (PWCS) grant while conducting the study. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have the following competing interest: STL was funded by a Port Waratah Coal Services Limited (PWCS) grant while conducting the study. There are no patents, products in development or marketed products to declare. This does not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials, as detailed online in the guide for authors.
Group formation in animals has evolved many times independently in response to a wide range of selective pressures . Understanding the ultimate factors and proximate mechanisms influencing group formation and ultimately sociality, is a major challenge in evolutionary biology. Both environmental and social factors have been identified to confer a fitness advantage and to influence the formation of animal aggregations and have been combined into socio-ecological models. Important environmental factors include the abundance, distribution, and quality of resources, such as discrete food patches , nest sites  and refuges .
The most commonly recognized social factors and benefits of being in a group include dilution of predation risk , increased vigilance and early predator detection , , thermoregulation  and enhanced feeding through the use of social information , . Nevertheless, there are also costs to group formation which are often proportional to group size and include increased within-group competition , home range size and travel distances , and susceptibility to parasites and pathogens , . These costs and benefits, in a species-specific combination, selected for the independent evolution of group living in numerous taxa .
Tadpole schooling has evolved multiple times in distantly related taxa and occurs in at least 10 families . Much of our understanding of the causes and consequences of tadpole schooling is based on observations in the field and a number of laboratory studies –. A key finding suggesting the importance of social cues for group formation is that tadpoles of Bufo americanus and several other species are capable of kin recognition and preferentially school with siblings –. Interestingly, the majority of aggregating species breed in ephemeral ponds , indicating that tadpoles may aggregate at scarce food resources. Conversely, it has been shown that daylight induces group formation , and water clarity increases group size and decreases inter-individual distances , which both suggest a spacing pattern that enhances anti-predator behaviour. Social foraging  and the use of social information to find food  have also been suggested as important social factors selecting for the formation of tadpole aggregations.
Taken collectively, the relative importance of ecological and/or social factors for the formation of tadpole aggregations is as yet unclear. The green and golden bell frog (Litoria aurea) is a large endangered Australian frog (adult snout-vent length 55–90 mm) . It is adapted to breeding in permanent ponds  and its tadpoles (snout-vent length 3–30 mm from hatching to metamorphosis)  can be under strong predation pressure, for example by the mosquitofish (Gambusia holbrooki) , . Litoria aurea tadpoles have occasionally been found in aggregations in the wild . We tested whether tadpoles form groups on the basis of social conspecific attraction in the absence of environmental resources.
Materials and Methods
All methods and procedures were formally approved by the University of Newcastle Animal Care and Ethics Committee (approval no A-2011-150) in compliance with the Australian Code of Practice for the use of animals for scientific purposes.
Adult green and golden bell frogs were collected from Kooragang Island in the Hunter Estuary in New South Wales (−32°51′48″/151°43′55″) during 2010 as part of an independent breeding project. For the experiment reported here, we randomly selected tadpoles from two clutches laid by two different females in captivity, in December 2011. Tadpoles were housed in two mixed sibship groups, with both group sizes varying repeatedly over time between 25–100 individuals (approximately 5–20 tadpoles/L water). Social experience through raising tadpoles in different group sizes does not influence their propensity to aggregate . Tadpoles were maintained in aged carbon-filtered water at room temperature on the natural light cycle (light-dark cycle of 14∶10 h) and they were fed three times/week with sprinkles of Goldfish Flake Plus (Marine Master). Water was changed as required, on average every four days (range 1–7, N = 9). Tadpoles were returned to the breeding project after the experiment.
Tests of social attraction
We tested a total of 48 tadpoles. Each tadpole was tested once, and all were at Gosner developmental stage 25 , . The study group was divided into two groups of 24 tadpoles, which were tested on consecutive days at similar times when tadpoles were active. Tadpoles were fed within 24 hours prior to the experiment to minimize the influence of the nutritional state on grouping behaviour. We used six test chambers at a time, which were visually isolated from each other. Each test chamber (figure 1) was partitioned into three compartments, the rectangular test arena (9.0×9.2×5.0 cm) in the middle and one adjoining compartment (2.8 cm×9.2×5.0 cm) on opposite sides. The test arena was divided in half again by a line on the base of the arena. We used three conspecifics as stimuli in each apparatus. Different stimuli tadpoles were used during the second experimental day. On each day focal and stimuli tadpoles were taken from one of the two mixed sibship groups. Focal and stimuli tadpoles were not size matched. Partitioning walls were transparent to allow visual contact between stimuli and focal tadpoles. Each day, we filled the six test chambers with water from a container that held all 42 tadpoles (24 focal tadpoles + 6×3 conspecific stimuli tadpoles) for 15 minutes, to minimize differences in tadpole chemical cues between trials. The partitioning walls were not watertight and tadpole-borne chemical cues could diffuse from the conspecific compartment into the arena. Visual and chemical cues were available to the focal tadpoles. We controlled for spatial preference in focal tadpoles by haphazardly assigning conspecifics to three left and three right compartments in the six test chambers used at a time. The water level in the test chambers was 2 cm. L. aurea tadpoles can occur in shallow water and often swim near the water surface . Furthermore, following similar experiments , , , we chose this relatively low water level to allow us to clearly determine the position and distance of the focal individual relative to conspecifics. After introducing a focal tadpole into each arena we allowed them two minutes to acclimate, inspect the arena with its two adjoining compartments, and make a decision. We then took four photographs at two minute intervals from behind a blind to establish the tadpole's position. At the end of each experimental day we measured the body size (snout-vent length) of each tadpole to the nearest 0.1 mm.
Arena with adjoining compartments on either side to hold conspecifics (test stimulus). A line on the base of the chamber (here dashed) divided the arena into halves.
We used the photographs to score whether tadpoles were present on the half of the test arena adjacent to the compartment with conspecifics or the empty compartment. We discarded five trials because they could not be accurately scored, leaving a total of 43 trials. Tadpoles were active and usually moved during the two minute interval between photographs. We used Cochran's Q test (an extension to the McNemar test) to ensure that repeated testing did not affect the decision making process in a systematic way , as the precision of decision making may improve with repetition , . We then allocated trials to one of two categories: (1) focal tadpole present on the conspecific side three or four times out of the four measures; and (2) focal tadpole present on the empty side two, three or four times. We used a chi-square test to determine whether option (1) occurred more frequently than expected by chance and hence whether tadpoles were attracted to conspecifics. Expected frequencies were calculated using the probability mass function for binominal distributions. Finally, we used a two-sample t-test to investigate whether social attraction strength was a function of body size. Data were reflect square root transformed due to negative skew to meet the assumption of normality. We used an alpha level of 0.05 to denote significant results. Statistical analyses were performed in SPSS 17.0.2.
Whether tadpoles chose to be alone or to associate with conspecifics was not significantly different among trials (Cochran's Q test: Q3 = 6.66, P = 0.08), and neither increased nor decreased through the course of multiple trials. This indicated a high repeatability among trials. Significantly more tadpoles than expected by chance chose the compartment containing conspecifics at least three out of four times (category 1;Chi-square test: χ21 = 29.69, P<0.0001; figure 2). Furthermore, social attraction was a function of body size. Tadpoles that were attracted to their conspecifics (category 1) were significantly smaller than tadpoles that showed no social attraction (category 2, unpaired t-test: t41 = −2.84, P = 0.007; figure 3).
Number of individuals that repeatedly chose the side with conspecifics over the empty side.
We conducted a relatively simple choice experiment in which isolated tadpoles had the option of spatially associating with an empty compartment or a compartment containing a group of three conspecific tadpoles. We found a very strong preference for the compartment containing conspecifics, which suggests that Litoria aurea tadpoles prefer to aggregate with conspecifics through social attraction. Whether visual or olfactory cues were used to locate the conspecifics remains to be determined. Alternatively, tadpoles may have avoided the empty compartment. However this is unlikely, because L. aurea tadpoles occupy structured habitats in the wild, yet have been observed to school .
Both environmental and social factors have been identified to select for the formation of animal aggregations. By conducting controlled experiments in the lab, we were able to exclude environmental factors such as food, temperature and light gradients that might influence tadpole grouping behaviour , . This indicates that aggregation behaviour in L. aurea may be driven by social factors. One such social factor is the benefit of schooling in the presence of a predator  and/or when tadpoles detect chemical cues from a predator . Predators may include larger conspecifics  and therefore represent another form of natural selection. This may result in relatively higher predation pressures on smaller individuals. Small tadpoles are most vulnerable to predation , and tadpoles generally become less vulnerable as they grow , because larger tadpoles are able to escape from predators more readily  or are too large for canibalistic conspecifics. However, during metamorphosis tadpoles are once again vulnerable because their emerging limbs may compromise locomotor ability . Interestingly, we found a relationship between social attraction and tadpole size in L. aurea tadpoles. Individuals that were attracted to conspecifics were significantly smaller than tadpoles that showed no social attraction. This may indicate that schooling in L. aurea could be an anti-predator behaviour. This hypothesis requires empirical testing, especially since the tadpoles in our study were born in captivity and not exposed to predatory cues.
Another social route to tadpole schooling is the benefit of social foraging and the transfer of social information to obtain food. For example, tadpoles of the spadefoot toad Spea multiplicata have been shown to form a vortex in response to introduced food particles in feeding experiments . It has been suggested that this collective behaviour allows access to food resources through agitating the pond substrate that otherwise would be inaccessible . Similarly, Bufo americanus tadpoles use social information through the presence of conspecifics, to locate food patches, rather than chemical cues from the food source itself .
A third social route to schooling appears to be relatedness. The tadpoles of several species preferentially school with kin and remain in schools even in the absence of an immediate predatory threat , . However, the mechanisms driving kin recognition and preference for siblings appear to be varied  and there is even debate that kin recognition might simply be a by-product of species and/or group member recognition and therefore not true kin recognition . Nevertheless, a popular argument in support of kin-based schooling is that if distasteful tadpoles are preyed upon, this cost is balanced against the improved survival of siblings . Hence, relatedness could be an alternative explanation for the observed association behaviour in L. aurea, which would indicate a genetic determination. However, we did not distinguish tadpoles on the basis of their clutch, so the effects of relatedness on schooling behaviour are difficult to gauge in our study.
There has been considerable debate and speculation about the factors and cues that drive group formation in tadpoles . While environmental drivers appear to be important for some species, social reasons are equally important for other species. Our study establishes the existence of tadpole schooling in L. aurea and indicates the importance of social factors, but more research is required to distinguish between the possible functions of this behaviour. For example, whether L. aurea tadpoles are attracted to conspecifics as an anti-predator behaviour or because conspecifics signal the presence of food remains to be determined. Nevertheless, the observed inverse relationship between tadpole size and aggregation preference supports the notion of an adaptive anti-predator behaviour.
Given the significance of the tadpole phase for successful reproduction in frogs, we need more studies examining social and environmental factors that potentially drive schooling behaviour, in order to construct a general theory for the evolution of social aggregations in tadpoles. Tadpoles represent an ideal model system for the understanding of socio-ecological factors that drive group formation during a vulnerable life stage.
We would like to thank J. Newson for his help during data collection. Three reviewers provided valuable feedback on the manuscript.
Conceived and designed the experiments: STL MJM. Performed the experiments: STL. Analyzed the data: STL MJW. Contributed reagents/materials/analysis tools: MJM. Wrote the paper: STL MJW.
- 1. Wilson EO (1975) Sociobiology: the new synthesis: Harvard University Press.
- 2. Schradin C (2005) When to live alone and when to live in groups: ecological determinants of sociality in the African striped mouse (Rhabdomys pumilio, Sparrman, 1784). Belg J Zool 135: 77–82.
- 3. Komdeur J (1992) Importance of habitat saturation and territory quality for evolution of cooperative breeding in the Seychelles warbler. Nature 358: 493–495. doi: 10.1038/358493a0
- 4. Visagie L, Mouton PlFN, Bauwens D (2005) Experimental analysis of grouping behaviour in cordylid lizards. The Herpetological Journal 15: 91–96.
- 5. Hamilton WD (1971) Geometry for the selfish herd. J Theor Biol 31: 295–311. doi: 10.1016/0022-5193(71)90189-5
- 6. Lanham EJ, Bull CM (2004) Enhanced vigilance in groups in Egernia stokesii, a lizard with stable social aggregations. J Zool 263: 95–99. doi: 10.1017/s0952836904004923
- 7. Roberts G (1996) Why individual vigilance declines as group size increases. Anim Behav 51: 1077–1086. doi: 10.1006/anbe.1996.0109
- 8. Hwang YT, Larivière S, Messier F (2007) Energetic consequences and ecological significance of heterothermy and social thermoregulation in striped skunks (Mephitis mephitis). Physiol Biochem Zool 80: 138–145. doi: 10.1086/509211
- 9. Duellmann WE, Trueb L (1986) Biology of amphibians. New York: McGraw-Hill.
- 10. Kurvers RHJM, Prins HHT, van Wieren SE, van Oers K, Nolet BA, et al. (2010) The effect of personality on social foraging: shy barnacle geese scrounge more. Proc R Soc B 277: 601–608. doi: 10.1098/rspb.2009.1474
- 11. Terborgh J, Janson CH (1986) The socioecology of primate groups. Annu Rev Ecol Syst 17: 111–136. doi: 10.1146/annurev.es.17.110186.000551
- 12. Chapman CA, Chapman LJ (2000) Determinants of group size in primates: the importance of travel costs. In: Boinski S, Garber P, editors. On the move: how and why animals travel in groups. Chicago: University of Chicago Press. pp. 24–42.
- 13. Cote IM, Poulin R (1995) Parasitism and group size in social animals: a meta-analysis. Behav Ecol 6: 159–165. doi: 10.1093/beheco/6.2.159
- 14. Freeland WJ (1976) Pathogens and the evolution of primate sociality. Biotropica 8: 12–24. doi: 10.2307/2387816
- 15. Alexander RD (1974) The evolution of social behavior. Annu Rev Ecol Syst 5: 325–383. doi: 10.1146/annurev.es.05.110174.001545
- 16. Wells KD (2007) The ecology and behavior of amphibians. Chicago: University of Chicago Press.
- 17. Wassersug R, Hessler CM (1971) Tadpole behaviour: aggregation in larval Xenopus laevis. Anim Behav 19: 386–389. doi: 10.1016/s0003-3472(71)80021-0
- 18. Foster MS, McDiarmid RW (1982) Study of aggregative behavior of Rhinophrynus dorsalis tadpoles: design and analysis. Herpetologica 38: 395–404.
- 19. O'Hara RK, Blaustein AR (1981) An investigation of sibling recognition in Rana cascadae tadpoles. Anim Behav 29: 1121–1126. doi: 10.1016/s0003-3472(81)80064-4
- 20. Waldman B, Adler K (1979) Toad tadpoles associate preferentially with siblings. Nature 282: 611–613. doi: 10.1038/282611a0
- 21. Blaustein AR, Waldman B (1992) Kin recognition in anuran amphibians. Anim Behav 44: 207–221. doi: 10.1016/0003-3472(92)90027-7
- 22. Waldman B (1982) Sibling association among schooling toad tadpoles: field evidence and implications. Anim Behav 30: 700–713. doi: 10.1016/s0003-3472(82)80141-3
- 23. Rödel M-O, Linsenmair KE (1997) Predator-induced swarms in the tadpoles of an African savanna frog, Phrynomantis microps. Ethology 103: 902–914. doi: 10.1111/j.1439-0310.1997.tb00132.x
- 24. Spieler M (2003) Risk of predation affects aggregation size: a study with tadpoles of Phrynomantis microps (Anura: Microhylidae). Anim Behav 65: 179–184. doi: 10.1006/anbe.2002.2030
- 25. Bazazi S, Pfennig K, Handegard N, Couzin I (2012) Vortex formation and foraging in polyphenic spadefoot toad tadpoles. Behav Ecol Sociobiol 66: 879–889. doi: 10.1007/s00265-012-1336-1
- 26. Sontag C, Wilson DS, Wilcox RS (2006) Social foraging in Bufo americanus tadpoles. Anim Behav 72: 1451–1456. doi: 10.1016/j.anbehav.2006.05.006
- 27. Pyke GH, White AW (2001) A review of the biology of the green and golden bell frog Litoria aurea. Aust Zool 31: 563–598. doi: 10.7882/az.2001.003
- 28. Hamer AJ, Lane SJ, Mahony MJ (2002) The role of introduced mosquitofish (Gambusia holbrooki) in excluding the native green and golden bell frog (Litoria aurea) from original habitats in south-eastern Australia. Oecologia 132: 445–452. doi: 10.1007/s00442-002-0968-7
- 29. Morgan LA, Buttemer WA (1996) Predation by the non-native fish Gambusia holbrooki on small Litoria aurea and L. dentata tadpoles. Aust Zool 30. doi: 10.7882/az.1996.005
- 30. Pyke GH, White AW (2000) Factors influencing predation on eggs and tadpoles of the endangered Green and Golden Bell Frog Litoria aurea by the introduce Plague Minnow Gambusia holbrooki. Aust Zool 31: 496–505. doi: 10.7882/az.2000.011
- 31. Nicieza AG (1999) Context-dependent aggregation in Common Frog Rana temporaria tadpoles: influence of developmental stage, predation risk and social environment. Funct Ecol 13: 852–858. doi: 10.1046/j.1365-2435.1999.00375.x
- 32. Gosner KL (1960) A simplified table for staging anuran embryos and larvae with notes on identification. Herpetologica 16: 183–190.
- 33. Anstis M (2002) Tadpoles of south-eastern Australia. A guide with keys. Sydney: Reed New Holland.
- 34. Baugh AT, Ryan MJ (2009) Female túngara frogs vary in commitment to mate choice. Behav Ecol 20: 1153–1159. doi: 10.1093/beheco/arp120
- 35. Hoeffler S, Ariely D (1999) Constructing stable preferences: a look into dimensions of experience and their impact on preference stability. Journal of Consumer Psychology 8: 113–139. doi: 10.1207/s15327663jcp0802_01
- 36. Brouwer R, Dekker T, Rolfe J, Windle J (2010) Choice certainty and consistency in repeated choice experiments. Environmental and Resource Economics 46: 93–109. doi: 10.1007/s10640-009-9337-x
- 37. Kennedy M, Gray RD (1993) Can ecological theory predict the distribution of foraging animals - a critical analysis of experiments on the ideal free distribution. Oikos 68: 158–166. doi: 10.2307/3545322
- 38. Beiswenger RE (1977) Diel patterns of aggregative behavior in tadpoles of Bufo americanus, in relation to light and temperature. Ecology 58: 98–108. doi: 10.2307/1935111
- 39. Black JH (1970) A possible stimulus for the formation of some aggregations in tadpoles of Scaphiopus bombifrons. Proc Oklahoma Acad Sci 49: 13–14.
- 40. Glos J, Erdmann G, Dausmann KH, Linsenmair KE (2007) A comparative study of predator-induced social aggregation of tadpoles in two anuran species from western Madagascar. Herpetological Journal 17: 261–268.
- 41. Sadler LM, Elgar MA (1994) Cannibalism among amphibian larvae: a case of good taste. Trends Ecol Evol 9: 5–6. doi: 10.1016/0169-5347(94)90222-4
- 42. Semlitsch RD (1990) Effects of body size, sibship, and tail injury on the susceptibility of tadpoles to dragonfly predation. Can J Zool 68: 1027–1030. doi: 10.1139/z90-149
- 43. Alford RA (1999) Ecology: Resource use, competition and predation. In: McDiarmid RW, Altig R, editors. Tadpoles The biology of anuran larvae. Chicago: University of Chicago Press. pp. 240–278.
- 44. Brown RM, Taylor DH (1995) Performance and maneuvering behavior through larval ontogeny of the wood frog, Rana sylvatica. Copeia 1: 1–7. doi: 10.2307/1446794
- 45. Grafen A (1990) Do animals really recognize kin? Anim Behav 39: 42–54. doi: 10.1016/s0003-3472(05)80724-9