Aquatic Mammals 2003, 29.3, 355359

Food sharing in wild bottlenose dolphins

Sharlene M. Fedorowicz, David A. Beard and Richard C. Connor

Department of Biology, University of Massachusetts, Dartmouth, North Dartmouth, MA 02747, USA

Abstract

In Golfo Dulce, Costa Rica, we documented the rst case of two bottlenose dolphins sharing a sh.The sh was shared between a male and femaledolphin (who was accompanied by a calf). Posses-sion of the sh changed 11 times during the 30-minobservation period. No obvious attempts weremade by any of the dolphins to steal or escape withthe sh. Interestingly, apparent consumption of the sh did not commence until possession of the shhad changed several times. The relationship be-tween the male and female is unknown, but sexualbehaviour between the two was observed the pre-vious day. Although food sharing has been reportedin a number of mammals in a variety of contexts,the repeated changes in ownership of the food itemreported here is, at best, extremely rare in othertaxa.

Key words: Food sharing, Bottlenose dolphins,Tursiops sp., Costa Rica, mothercalf interaction,behaviour, mating relationships

Introduction

The bottlenose dolphins, Tursiops spp., is the mostwell studied cetacean (Barros & Wells, 1998;Connor et al., 2000). They live in complex ssion-fusion societies where the size and composition ofgroups changes often as individuals join and leave(Wells et al., 1987; Connor et al., 1992a; Smolkeret al., 1992; Smolker et al., 1993; Mann & Smuts,1999). The bottlenose dolphin ssion-fusion society,like those primate ssion-fusion social systems towhich it is often compared (e.g., chimpanzees andspider monkeys, Goodall, 1986; Symington, 1990,see Connor et al., 2000), may have an ecologicalbasis in a food distribution that is spatially andtemporally patchy.

Types of feeding and foraging behavioursBottlenose dolphins feed on a wide variety of sh,as well as cephalopods (squid and octopus), shrimp(Gunter, 1951) and small rays and sharks (Mead &

Potter, 1990). They hunt schooling and solitaryprey using a variety of solitary and group feedingstrategies throughout the water column, on shore orabove the water surface (Connor et al., 2000). Forexample, bottlenose dolphins dive rostrum rst intothe sand up to their eyes after prey in the Bahamas(Rossbach & Herzig, 1997). In Shark Bay,Australia, bottlenose dolphins carry cone-shapedsponges on their rostra, possibly to protect theirrostra when they forage on bottom dwelling prey(Smolker et al., 1997). Bottlenose dolphins pursueprey onto mud banks in Georgia and SouthCarolina (Petricig, 1993) and some stun or kill shby whacking them with their ukes (Wells et al.,1987). Individual dolphins have been reported topursue their prey by swimming upside down nearthe surface (Leatherwood, 1975; Belkovich et al.,1991) or by swimming belly up to chase a sh alongthe surface then grabbing it as it jumps in the air(Connor et al., 2000). Solitary or groups of bottle-nose dolphins also have been observed circlingaround schools of sh and darting into the schoolsto feed (Morozov, 1970; Leatherwood, 1975;Belkovich et al., 1991). In spite of the dazzlingarray of feeding and foraging behaviours observedin numerous locations around the globe, food shar-ing, de ned as the joint use of a monopolizablefood item (Stevens & Gilby, in press), has not beenobserved in this genus. Here we detail the rst suchobservation.

Materials and Methods

The food sharing occurred in a group of threedolphins (a male, female and her calf) on 5 August1998 in Golfo Dulce, a bay on the south Paci ccoast of Costa Rica (8 23 N8 45 N). The gulf isabout 50 km long and 10 to 15 km wide containinga shallow outer slope with a deep inner basinreaching to a depth of 215 m (Acevedo-Gutirrez &Burkhart, 1998). The food sharing observation wasmade from a 4 m vessel with an outboard motorduring the course of a 4 month study of mothercalfinteractions, that included general surveys of groupcomposition and activity and focal follows on adult

? 2003 EAAM

females accompanied by calves (see Connor et al.,2000). Individual dolphins were identi ed photo-graphically (Wrsig & Wrsig, 1977). The behav-iour described here was recorded ad libitum using asmall hand-held cassette recorder and a video cam-era to supplement the observation that began whenwe encountered these dolphins at 0915 h. Whileusing the video camera, data were spoken into oneacoustic channel and a hydrophone recordingdolphin sounds was fed into the other acousticchannel.

Results

Previous observationsThree dolphins, Top Notch, Smash and a calf wereobserved during the described food sharing period.Top Notch was photographed (Wrsig & Wrsig,1977) during a previous eld season that year(March), when he was determined to be a male bythe absence of mammaries and a 2.5 cm gapbetween the genital and anal slits (Connor et al.,1992). Our rst observation of Smash was the daybefore the food sharing event when she was ob-served foraging and travelling with her calf (judgedto be about 3 years old based on size) in infantposition (Connor & Smolker, 1995; Mann &Smuts, 1999). In Shark Bay, infant position is areliable indicator that a calf is not weaned (Mannet al., 2000).Top Notch also was observed with Smash and

her calf on 4 August. This observation lasted forover 2 h, beginning at 1043 h. We rst encounteredTop Notch foraging alone for 26 min, about 2050 m south of Smash and her calf. Top Notchapproached, then joined Smash and her calf, andthe group then traveled for 17 min followed by a7-min period of foraging. A 77-min bout of social-izing and sexual activity followed. During thisperiod of social/sexual behaviour, Top Notch en-gaged in tail slaps and mounts (including invertedand side mounts) on Smash, who often rolledbelly up. Considerable whistling and clickingaccompanied the social behaviour. The grouptravelled for the last 8 min of the observation.

Food sharing observationObservations of Top Notch and Smash (her calfremained close) with the sh were made during a30-min period on 5 August 1998 beginning at0915 h. When we rst observed them, Smash, wasalready holding the sh in her mouth, visuallyestimated to be approximately 55 cm long and15 cm wide. The sh appeared to be a Carangid, butthe species could not be determined. Initially, the sh was whole, alive, and in good condition. Dur-ing the observation period, when visible at the

surface, all three dolphins spent a majority of theirtime within 2 m of each other.Initially, Smash held the live sh in her mouth

shaking it and swimming on her side causing con-siderable splashing as her calf and the other animal,Top Notch, trailed her within 2 m. She did notobviously attempt to break the sh apart or eat it.After holding the sh for approximately 90 s, shereleased it. At this point, the sh appeared dead,but was still whole. The sh oated for about 7 s atthe surface of the water until Top Notch took it,shook it back and forth in his mouth, and then dovewith the sh followed by Smash and her calf next tohim. Upon the next surfacing, Top Notch releasedthe sh, which was still whole, and Smash, who wasabout 1 m behind Top Notch, took the sh anddove with it in her mouth. The sh was releasedand changed possessions 11 times during theobservation period, and each possession lasted forapproximately 10 to 145 s (except for the lastsurfacings in which Top Notch had the sh (seeTable 1). The sh was left oating at the surface for218 s between possessions and was never passeddirectly between the dolphins (Table 1).On one occasion, Smash reclaimed the sh three

consecutive times within approximately 1 to 14 safter release (Table 1). During the second surfacingof this series, about 8 min into the observation, the rst damage was noticeable on the sh; it appearedslightly mangled and torn. Between the second andthird possession, splashing involved Smash, TopNotch, and the calf. In each case, Top Notchremained within 2 m of Smash but made no attemptto take the sh after she had released it. AfterSmash dove with the sh for the third consecutivetime, Top Notch resurfaced with it approximately165170 s later. At this stage, the sh was slightlymore damaged. This was the only observationwhere an underwater exchange occurred and theonly sequence that included any indication (i.e.,splashing) of social behaviour other than the shexchange itself.During the 23 min after we observed the rst

damage, the dolphins exchanged the sh seventimes slowly breaking it apart. The last dolphin seenwith the sh was Top Notch, who dove with thelast remaining piece, which was approximatelyone-fourth the size of the original sh.Whistles and clicks were heard on ve separate

occasions lasting from 5280 s each (Table 1).No obvious attempts were made by any of the

dolphins to steal or escape with the sh during the30-min period. It is interesting in this regard thatthe sh did not become ragged until part of the waythrough the observation period even though itchanged possession frequently. After the sh wasconsumed, the animals began to forage at a dis-tance greater than 10 m from each other, a typical

356 S. M. Fedorowicz et al.

distance for foraging behaviour. The calf never hadpossession of the sh, but remained close to theother two dolphins; the calf also did not forage orfeed on his/her own during the food-sharing event.

Discussion

Ours is the rst observation of food sharing in wildbottlenose dolphins. In Shark Bay, where dolphinfeeding behaviour has been observed for hundredsif not thousands of hours, no similar case has beenreported. Observations in Shark Bay suggest astrong sense of ownership where dolphins maytoss a sh up to 3 m, but other dolphins do notattempt to take it (Connor et al., 2000). In Sarasota,Florida, the longest running bottlenose dolphinstudy site, Randall Wells (pers. comm.) hasobserved dolphins feeding on Carangid sh withoutobvious sharing. In both sites, observations include

dolphins feeding in the company of other dolphinson large prey items that require considerable time tobreak up and consume.Food sharing may be a rare behaviour in bottle-

nose dolphins that just happened to be observed forthe rst time in Gulfo Dulce. A more interestingpossibility is that food sharing occurs more com-monly in Gulfo Dulce than in the more intensivelystudied Sarasota and Shark Bay sites, either amongparticular individuals or in the population gener-ally. This possibility is suggested by the increasinglyfrequent observation of site-speci c or individualforaging behaviours that have been reported inbottlenose dolphins and other marine mammals(Connor et al., 2000; Connor, 2001). That suchforaging specializations might include social com-ponents would not be surprising, either becausesocial behaviour and bonds also vary between sites(Connor et al., 2000).

Table 1. Order and duration of food sharing possessions between Top Notch and Smash in Golfo Dulce, Costa Rica.

Number ofswitch Dolphin with sh

Duration ofpossession (s) Splashing bouts Damage

Intitial Smash 96 01 Top Notch 46* 02 Smash 143* 0

Smash * Splashing (20 s) 0Smash * Splashing (15 s) 0Smash * Splashing (6 s) 0

3 Top Notch 49 04** Smash 56 0** Smash 8* 1** Smash * Splashing 1** Smash * 15 Top Notch * Resurfaces 165170 s later 26 Smash * 2

Smash * Splashing (13 s) 2Smash * Splashing (7 s) 2Smash * Splashing (5 s) 2Smash * Whistles & clicks (52 s) 2

7 Top Notch 10* 38 Smash 10 39 Top Notch 104 3

Top Notch * Splashing (*) 3Top Notch * Whistles & clicks (60 s) 3Top Notch * Splashing (*) 4

10 Smash 4* 511 Top Notch 5.5 min from here to 5

Top Notch last evidence of sh . . . Whistles & clicks (*) 5Top Notch * Splashing (*) 5Top Notch * Whistles & clicks (*) 5Top Notch Last evidence of sh Splashing & whistles & clicks (*)

Number of the switch and the duration of time each animal possessed the sh including splashing and sound events. The(*) indicates either close approximation times since exact times were not available, or unavailable times. The (**) indicatesSmashs multiple possession bouts where she released the sh at the surface and reclaimed it. Damage was on a scale from05; 0=no damage, 5=most damage.

357Food sharing in wild bottlenose dolphins

Obviously, with a single observation we can onlyspeculate about the function of the food sharingbetween Top Notch and Smash. We do not know ifthe two are related and we do not have any long-term association data that might suggest a bondbetween the two. The sexual behaviour observedbetween Top Notch and Smash on the previous daysuggests that the sharing may have been associatedsomehowwith a mating relationship.The size of thecalf with Smash suggests that Smash could havebeen in estrus; in Shark Bay three-year old calvesfrequently accompany their mothers during con-sortships (Connor et al., 2000). The observationsuggests a food for sex hypothesis, but we do notknow which dolphin caught the sh.Although food sharing has been reported in a

variety of taxa (reviewed by Stevens & Gilby, inpress), the behaviour described here may be unique.Food sharing may apply to a range of behav-iours where more than one individual consumesa monopolizable resource. Individuals may feedsimultaneously on a large food item, they may alertothers to the presence of food, or they may pas-sively or actively give food to another individual(for speci c examples see Stevens & Gilby in press).However, in no case that we are aware of, havetwo individuals been observed repeatedlyexchanging control of the same food item.

Acknowledgments

The rst author dedicates this paper to the memoryof her friend, Sally, who initially supported andencouraged her in this endeavour. She also thanksher parents, Chet and Felicia Fedorowicz, andsister, Loren, for all of their support and enthusi-asm. This research was conducted in collaborationwith the University of Costa Rica, San Jos, CostaRica. We are indebted to Dr Jos Vargas from theUniversity of Costa Rica (CIMAR) and Jefe ElicerLeal from INCOPESCA for the tremendousamount of help they provided us while in CostaRica. Special thanks are due to Dr AlejandroAcevedo and Priscilla Cubero-Pardo and her familyfor all of their support and guidance in establishingthis project. This research was funded in part bygrants from the Lerner Gray Fund for MarineResearch of the American Museum of NaturalHistory and the UMD Foundation/HealeyEndowment Fund.

Literature Cited

Acevedo-Gutirrez, A. & Burkhart, S. (1998) Seasonaldistribution of bottlenose (Tursiops truncatus) andpan-tropical spotted (Stenella attenuata) dolphins(Cetacea: Delphinidae) in Golfo Dulce, Costa Rica.Revista de Biologa Tropical 46, 91101.

Barros, N. B. & Wells, R. S. (1998) Prey and feedingpatterns of resident bottlenose dolphins (Tursiops trun-catus) in Sarasota Bay, Florida. Journal of Mammalogy79, 10451059.

Belkovich, V. M., Ivanova, E. E., Yefremenkova, O. V.,Kozarovitsky, I. B. & Kharitonov, S. P. (1991) Search-ing and hunting behavior in bottlenose dolphin(Tursiops truncatus) in the Black Sea. In: K. Pryor &K. S. Norris (eds.) Dolphin Societies: Discoveries andPuzzles. Berkeley, University of California Press.

Connor, R. C. (2001) Individual foraging specializationsin marine mammals: culture and ecology. Behavioraland Brain Sciences 24, 329330.

Connor, R. C. & Smolker, R. A. (1995) Seasonal changesin the stability of malemale bonds in Indian Oceanbottlenose dolphins (Tursiops sp.). Aquatic Mammals21, 213216.

Connor, R. C., Smolker, R. A. & Richards, A. F. (1992a)Dolphin alliances and coalitions. In: A. H. Harcourt &F. B. M. de Wall (eds.) Coalitions and Alliances inHumans and Other Animals. Oxford, Oxford UniversityPress.

Connor, R. C., Smolker, R. A. & Richards, A. F. (1992b)Two levels of alliance formation among bottlenosedolphins (Tursiops sp.). Proc. Natl. Acad. Sci. USA 89,987990.

Connor, R. C., Wells, R. S., Mann, J. & Read,A. J. (2000)The bottlenose dolphin: social relationships in a ssion-fusion society. In: J. Mann, R. C. Connor, P. L. Tyack& H. Whitehead (eds.) Cetacean Societies: Field Studiesof Dolphins and Whales. pp. 91126 The University ofChicago Press, Chicago.

Goodall, J. (1986) The Chimpanzees of Gombe: Patterns ofBehavior. Harvard University Press, Cambridge.

Gunter, G. (1951) Consumption of shrimp by the bottle-nose dolphins. J. Mammal 32, 465466.

Leatherwood, S. (1975) Some observations of feedingbehavior of bottle-nosed dolphins (Tursiops truncatus)in the northern Gulf of Mexico and (Tursiops cf. T.gilli)oV southern California, Baja California, and Nayarit,Mexico. Marine Fisheries Rev. 37, 1016.

Mann, J. & Smuts, B. B. (1999) Behavioral developmentin wild bottlenose dolphins newborns (Tursiops spp.).Behaviour 136, 529566.

Mann, J., Connor, R. C., Barre, L. M. & Heithaus, M. R.(2000) Female reproductive success in wild bottlenosedolphins (Tursiops sp.): Life history, habitat, provision-ing, and group size eVects. Behav. Ecol.

Mead, J. G. & Potter, C. W. (1990) Natural history ofbottlenose dolphins along the central Atlantic coast ofthe United States. In: S. Leatherwood & R. R. Reeves(eds.) The Bottlenose Dolphin, San Diego, AcademicPress.

Morozov, D. A. (1970) Dolphins hunting. RybnoeKhoziaistvo 46, 1617.

Petricig, R. O. (1993) Diel patterns of strand-feedingbehavior by bottlenose dolphins in South Carolina saltmarshes. Abstracts, Tenth Biennial Conference on theBiology of Marine Mammals, Galveston, TX.

Rossbach, K. A. & Herzing, D. L. (1997) Underwaterobservations of benthic feeding bottlenose dolphins(Tursiops truncatus) near Grand Bahama Island,Bahamas. Marine Mammal Science 13(3), 98503.

358 S. M. Fedorowicz et al.

http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0022-2372^28^2979L.1045[aid=3298708]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0140-525X^28^2924L.329[aid=5510306]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0027-8424^28^2989L.987[aid=523142]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0005-7959^28^29136L.529[aid=5510309]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0022-2372^28^2979L.1045[aid=3298708]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0140-525X^28^2924L.329[aid=5510306]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0027-8424^28^2989L.987[aid=523142]

Smolker, R. A., Mann, J. & Smuts, B. B. (1993) Useof signature whistles during separations and reunionsby wild bottlenose dolphin mothers and infants. Behav-ioral Ecology Sociobiology 33, 393402.

Smolker, R. A., Richards, A. F., Connor, R. C. & Pepper,J. W. (1992) Sex diVerences in patterns of associationamong Indian Ocean bottlenose dolphins. Behaviour123, 3869.

Smolker, R. A., Richards, A. F., Connor, R. C., Mann, J.& Berggren, P. (1997) Sponge-carrying by IndianOcean bottlenose dolphins: Possible tooluse by adelphinid. Ethology 103, 454460.

Stevens, J. R. & Gilby, I. C. (2003) A conceptual frame-work for non-kin food sharing: timing and currencybene ts. Animal Behavior in press.

Symington, M. (1990) Fission-fusion social organiz-ation in ateles and pan. Int. Journal of Primotology 11,4761.

Wells, R. S., Scott, M. D. & Irvine, A. B. (1987) The socialstructure of free-ranging bottlenose dolphins. In:H. Genoways (ed.) Current Mammalogy, vol. 1. NewYork, Plenum Press.

Wrsig, B. & M. Wrsig. (1977) The photographic deter-mination of group size, composition, and stability ofcoastal porpoises Tursiops truncatus. Science 198, 755756.

359Food sharing in wild bottlenose dolphins

http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0340-5443^28^2933L.393[aid=1508014]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0179-1613^28^29103L.454[aid=4982225]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0164-0291^28^2911L.47[aid=951473]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0340-5443^28^2933L.393[aid=1508014]http://www.ingentaselect.com/rpsv/cgi-bin/linker?ext=a&reqidx=/0164-0291^28^2911L.47[aid=951473]