calEcology,p.136140. DiscoveryBay Sides, E. M. 1975. A study of niche separation in t h r e e s p e c i e s o f O p h i o c o m a (Echinodermata: Ophiuroidea) in Jamaica. M.Sc.Thesis,Univ.WestIndies. Sides, E. M. 1981. Aspects of space utilization in shallowwater brittlestars (Echinodermata: Ophiuroidea) of Discovery Bay, Jamaica. Ph.DThesis,Univ.WestIndies. Sides,E.M.,andJ.D.Woodley.1985.Nichesepara tion in three species of Ophicoma (Echinodermata: Ophiuroidea) in Jamaica, WestIndies.BulletinofMarineScience,36: 701715. Tran, J. K., and B. Whited. 2004. Patterns of distri bution of three brittlestar species (Echinodermata:Ophiuroidea)oncoralreefs. Dartmouth Studies in Tropical Ecology, p. 177180.
BENEFITSOFMIXEDSPECIESSCHOOLSFORTHREEFISHSPECIES
TIMOTHYR.MATSUURA,MELISSAA.BARGER,GABRIELH.CALVIAND DANIELJ.MADIGAN
Abstract:Themechanismsthatdrivemixedspeciesschoolingarenotwellunderstood,andmayvary with different fishspecies. We examined schoolingbehaviorand the benefitsof schoolingin three fish species (spotlight parrotfish, doctorfish, and bluehead wrasse) at Discovery Bay, Jamaica. We followed individuals of each species, recording time spent schooling, average school size, feeding efficiencyand damselfishattacks in and out of schools. Bluehead wrasse spent less timeschooling than the other two species, although all species benefited from schools through increased foraging efficiency, decreased damselfish attacks, or both. Because food resources are neither limiting nor completelydefendedinDiscoveryBay,wesuggestthatsomemechanismotherthanincreasedforag ingefficiency,suchaspredatoravoidance,drivestheamountoftimeeachspeciesspentschooling.
KeyWords:blueheadwrasse,doctorfish,foragingefficiency,predatoravoidance,stoplight parrotfish INTRODUCTION Forming heterospecific foraging as sociations is a common behavior of many tropical reeffish species. These mixed speciesschoolscomprisecomplexandrela tively understudied interactions between species, and many mechanisms have been suggestedtoexplainthem.Themainbene fits of schooling include predator avoid ance and increased foraging rates (Lukoschek and McCormick 2000). Other studies show that groups of fish can help fish circumvent the territoriality of com petitors(Robertson1976).Aschooloffish canswampa damselfishterritory and fish cangraze onalgaewhilethedamselfishis occupied chasing other members of the school(Foster1985),thusallowinganindi vidual fish within a school to take more bitesandreceivefewerdamselfishattacks. We chose three species of fish com
165
DartmouthStudiesinTropicalEcology2005
monin the backreefofDiscovery Bay, Ja maica to compare the relative benefits gained by schooling across different fish species. Our three focal species were the stoplight parrotfish (Sparisomaviride), doc torfish(Acanthuruschirurgus),andtheblue head wrasse (Thalassoma bifasciatum). These fish are in different families and havedistinctlifehistories,andallofthese speciesareknowntoaggregateinloosely knit feeding groups. Both the doctorfish and parrotfish are generalist feeders that consumealgae,seagrassandmicroinverte brates in the sediment. Terminal male blueheadwrasseforageuponinvertebrates andzooplankton(Deloach1999),andhave been shown to prefer damselfish eggs whenavailable(Foster1987). Wewantedtoassessanydifferences in schooling benefits between these spe cies.Wemeasuredthedifferencesinfeed ingratesanddamselfishattackratesinand outofschoolsforeachofthesefishspecies. We also measured the proportion of time each fish spent in and out of schools. We predicted that the fish species with the greatest increase in feeding rate and the greatest decrease in damselfish attack rate would spend the greatest proportion of timeschooling. METHODS We observed fish from 410 March 2005 on the west back reef at Discovery Bay,Jamaica.Welocatedfocalindividuals of bluehead wrasse, stoplight parrotfish, and doctorfish by snorkeling over the reef untilanindividualwassightedinamixed
speciesschool.Sincewewereinterestedin schooling behavior, we chose fish that were initially in schools; any bias towards schooling should be seen across all three fish species and not affect the comparison betweenspecies.Tocontrolforsize,allob served individuals fell into the following size ranges: bluehead wrasse, 911 cm (terminal male); stoplight parrotfish, 912 cm (juvenile); doctorfish, 912 cm (juvenile).Sincesizerange and coloration were kept approximately equal for each fish species, the observed fish within each species were at approximately equal life stages. We followed each fish for 10 min, recording the time spent in schools and time spent alone. We counted feeding bitesandattacksbydamselfish(damselfish attackswereanyactsofdamselfishaggres sion that elicited a response in the focal fish).Wealsocountedthenumberoftimes an individual left and rejoined a school. Eachtimeanindividualjoinedaschool,we recorded the number of individuals in the school; these values were averaged to ob tainthemeanschoolsizeforeachindivid ualduringtheobservationperiod. Proportion of time spent schooling, inschoolandsolitaryfeedingratesandin schoolandsolitaryratesofattackbydam selfishwerecomparedfordifferentspecies usingonewayANOVAs(JMP5.0.1). RESULTS Bluehead wrasse spent significantly less time in schools than doctorfish and parrotfish (F = 50.22, df = 2, 66, P < 0.001,
166
DiscoveryBay
Tukeys = 0.05; Fig 1). Doctorfish and parrotfish did not differ in the proportion oftimetheyspentinschools. There was an effect of species and schoolingonfeedingrate(2wayANOVA, F=7.76,df=5,127,P<0.0001;SpeciesF= 8.76,df=2,P=0.0003;SchoolF=15.90,df =1,P=0.0001;InteractionF=2.27,df=2,P = 0.11; Fig. 2). The lack of an interaction shows there was no difference between speciesinthemagnitudeofthechangebe tween feeding rates in and out of schools. Foreachspecies,feedingratesweresignifi cantly higher when schooling than when solitary(DoctorfishF=22.36,df=1,41,P< 0.0001; Parrotfish F= 2.02, df = 1, 42, P = 0.16;BlueheadWrasseF=3.71,df=1,44,P =0.06;Fig.2). Therewasaneffectofschooling,but not species, on rate of damselfish attacks (2way ANOVA, F = 2.16, df = 5, 127, P = 0.06; Species F = 0.16, df = 2, P = 0.85; SchoolingF=5.34,df=1,P=0.02;Fig.3). The magnitude of the change between damselfishattacksmadeonfishinandout parrotfish ofschoolswasthesamefordoctorfishand but it differed significantly be tween these species and the bluehead wrasse (2way ANOVA, Interaction F = 2.67,df=2,P=0.07;Fig.3).Doctorfishand parrotfishbothexperiencedareducedrate ofdamselfishattackswhenschoolingthan when solitary; there was no difference for bluehead wrasse (Doctorfish F = 3.11, df = 1,44,P=0.09;ParrotfishF=7.28,df=1,42, P=0.001;BlueheadWrasseF=0.27,df=1, 44,P=0.61;Fig.3).
DISCUSSION All three fish species showed bene fitsfrommixedspeciesschools.Doctorfish and bluehead wrasse exhibited higher feeding rates in schools (Fig 2.), and the doctorfish and parrotfish experienced lower attack rates in schools (Fig. 3). The proportionoftimespentschoolingwasnot thesameamongthethreespecies,withthe bluehead wrasses schooling onethird as muchasthedoctorfishandparrotfish(Fig. 1). The main benefits suggested for mixedspecies schools are reduced risk of predation and higher foraging efficiency (Lukoschek and McCormick 2000). The proposed mechanisms by which foraging efficiency has been shown to increase in
Proportion of time spent in mixed species school
100
80
60
40
20
0 Parrotfish Doctorfish Wrasse
Figure 1. Proportion + 1 SE of time spent in mixed species schools. Doctorfish and Stoplight parrotfish spent more time in schools than out of schools. Bluehead wrasse spent more time out of schools than in schools. 167
DartmouthStudiesinTropicalEcology2005
mixedspeciesschoolsare1)resourceloca tion,inwhichschoolmembersexpeditethe findingofarareorlimitedresource,and2) swarming of territorially defended re sources,suchasfisheggsordamselfishal gal mat territories (Foster 1985, 1987). In Discovery Bay, we saw doctorfish and parrotfish foraging on algal and turtle grass substrates, food resources that are neither limited nor entirely defended. Therefore, enhanced feeding is unlikely to be the only mechanism directly driving schoolingbehaviorinthesespecies. Predator avoidance may be an im portant benefit of mixedspecies schooling inthesefish.Previousstudieshaveshown that when in groups, fish spend less time showing vigilant (predatordetecting) be
18 16
1.6
Damselfish attacks / min
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
in-school solitary
Parrotfish Doctorfish Wrasse
Figure 3. In-school and solitary damselfish attack rates + 1 SE for stoplight parrotfish, doctorfish, and bluehead wrasse in Discovery Bay, Jamaica (n = 69). Attack rates for parrotfish and doctorfish were higher when solitary. There was no difference in damselfish attack rates on bluehead wrasse while schooling or solitary.
Feeding rate (bites/min)
14 12 10 8 6 4 2 0
in-school solitary
Parrotfish Doctorfish Wrasse
Figure 2. In-school and solitary feeding rates + 1 SE of stoplight parrotfish, doctorfish, and bluehead wrasse in Discovery Bay, Jamaica (n = 69). All fish showed greater feeding rates in schools than when solitary. There was an affect of species with parrot fish and doctorfish having higher overall feeding rates than bluehead wrasse (Tukey = 0.05). There was no difference in the magnitude of change between in-school and solitary feeding rates across species.
havior, and thus more time foraging (Lukoschek and McCormick 2000). This was also supported by our behavioral ob servationsofthedoctorfishandparrotfish. In groups these fish fed in the open, but when solitary, they hid under rocks and fedlittle.Thissuggeststhatthedoctorfish andparrotfishschooltodecreasetheirrisk ofpredation,andthatwhileschooling,de creased time is spent on predator detec tion, leading to increased foraging effi ciency. This could be a response to natu rallyoccurringpredationthreats;itismore likely to be an ingrained, evolutionarily favoredbehavior,aspredatorsarethought to be relatively uncommon ...