The Pale-faced Sheathbill (Chionis alba) is an opportunistic predator-scavenger. During spring in Antarctica it foraged in colonies of southern elephant seals, obtaining placentas, pup carcasses, milk from nursing cows, blood, and feces. Afterbirths and pup carcasses constituted the bulk of the food consumed. Daily consumption estimates averaged 67 g/bird for placenta and 11 g/bird for pup carcasses, which are 54% and 26% of daily energy requirements, respectively. Sheathbills spent 86% of the day foraging or displaying and 14% resting or preening. Actively feeding birds spent 38% of the time searching for food, 20% feeding, 23% resting, 14% on comfort activities, and 3% in agonistic behaviors.

Laboratorio de Vertebrados Departamento de Biologia Facultad de Clencias Exactas y Naturales Universidad Nacional de Mar del Plata Funes 3250 (7600) Mar del Plata, Argentina. LA ECOLOGA DE LA ALIMENTACION DE LA CHIONIS ALBA EN LOS HARENES DE ELEFANTES MARINOS EN LA ISLA KING GEORGE, ANTARCTICA Sinopsis.--La Paloma Antfirtica (Chionis alba) es una especie oportunista depredadora y carrofiera. Durante la primavera antfirtica se alimenta frecuentemente en los hateries de elefante marino obteniendo placentas, carcasas de crias muertas, leche de hembras, sangre y deyecclones. Las placentas y crlas muertas constituyeron lo mas importante en la dieta de la especie. Las estimaclones diarias de consumo pot ave promediaron 67 g de placenta y 11 g de tejido extra/do de carcasas, lo que represent6 un 54 y un 26% respectivamente de los requerimientos energtticos diarios de la especie. Las palomas antfirticas utilizaron el 86% del dla alimentfindose o interactuando con otros individuos y el 14% de d/a reposando o acicalfindose. Los individuos que se encontraron activamente alimentfindose emplearon un 38% del tiempo en la bfisqueda de alimento, 20% tomando alimento, 23% reposando, 14% en actividades de confort y un 3% en comportamientos de agresitn. Sheathbills (Chionis spp.) are opportunistic predators and scavengers breeding on the Antarctic Peninsula and subantarctic islands (Watson 1975). This species breeds in association with penguin colonies where high-quality food is abundant during summer (Burger 1981a, 1981b). However, their foraging behavior is sufficiently flexible to take advantage of other food sources (Verheyden andJouventin 1991). For example, pla- centas and carcasses of elephant seals Mirounga leonina (Burger 1981a, Ulrich-Peter 1988) and Weddell seals Leptonychotes weddelli (Jones 1963) are available between early September and mid-November, when the birds forage intensively amongst the seals. As reported by Jones (1963) and Ulrich-Peter (1988) for other localities in Antarctica, and by Favero (1993) for Stranger Point, during spring the number of Pale-faced Sheathbills increases as a result of the presence of elephant seals. Sheathbills concentrate their activity among seals until the beginning of the pygoscelid penguin breeding season that corresponds with the end of seal pupping (end of November for South Shetland Is- lands). Then sheathbills disperse to their breeding places, which are usu- ally located within penguin colonies. With the end of the penguin breed- ing season (early March), sheathbills disperse and residual populations that remain in Antarctica are found in association with permanent human habitation. Due to their small population size in relation to penguins, Imperial Cormorants offer a less important food source, although it is possible to find isolated sheathbills foraging in cormorant colonies (Fav- ero 1996). This paper reports a preliminary study of the food and for- aging behavior of Pale-faced Sheathbills (Chionis alba), associated with southern elephant seals in Antarctica. METHODS From 24 Sep.-22 Oct. 1992, information on Pale-faced Sheathbills as- sociated with reproductive groups of southern elephant seal was collected at Potter Peninsula (62ų14'S, 58ų38'W), King George Island, South Shet- land Islands, Antarctica. The study area comprised 8 km of shoreline between Potter Cove and Stranger Point. The number of sheathbills associated with elephant seal harems was estimated on the basis of transects carried out every 4-6 d throughout the study area. A total of 47 focal animal observations (Altmann 1974) was made on randomly selected sheathbills that were foraging actively within elephant seal harems. The activities recorded were: feeding, searching (for food), agonistic, resting (mainly standing), and comfort behavior. Agonistic behaviors were divided into three classes: postures, displacements, and stretch threats. Among others, "postures" include the forward and hunched display (Burger 1980), "displacements" include displacements and chases without contact between birds, "stretch threats" (Shaw 1984) include the aggressive upright posture and fighting (Burger 1980). Birds were observed for at least 10 min with a monocular telescope (12-36x). Observations were made throughout the day (0600-2000 h local time) and narrated into a tape recorder. Additional information was extracted from three film records carried out between 25 September and 15 October on birds scavenging afterbirths. Records started the moment of the expulsion of the placenta by the seal, and finished when it was totally consumed. To estimate the time spent resting and foraging from dawn to dusk, 84 instantaneous scan samples (Altmann 1974) were made at elephant seal harems covering different times of the day. The mass of the meals obtained from carcasses and placentas during focal animal observations was estimated based on trials performed around harems. Trials consisted of offering to Sheathbills known amounts of flesh and viscera (as carcasses) and different parts of an afterbirth, filming the birds feeding, and later counting the number of pecks used to consume the food. The consumption rates estimated were 13.4 g/100 pecks for afterbirths (SD = 4.5, 95% CI = 8.7-18.2, n = 6), and 5.1 g/100 pecks for carcasses (SD = 4.0, 95% CI = 0.9-9.3, n = 6). It was impossible to apply this methodology to milk because of difficulties in manipulation of this material. TABLE 1. Percentage occurrence in field observations, energy content, consumption, and energy contribution of different food items in the diet of Pale-faced Sheathbills. % Energy content Consumption occurrence (kcal/g fresh rate (g/bird % energy Item in field a mass) day) contribution e Placentas 19 1.13 b 67 54 Pup carcasses 26 3.30 c 11 26 Milk 3 4.30 d -- -- Seal excreta 52 <0.5 -- -- a Based on scan samples. b From Lavigne and Stewart (1979). c Estimations based on meat composition from Hamilton (1949), blubber energy content excluded. a From Carlini et al. (1994). e Using existence metabolic rate at 0ųC (Kendeigh et al. 1977), EM (kcal/day) = W ų-54 (where W is mass in grams). RESULTS Diet and consumption rates.--Elephant seal colonies are an attractive feeding area for Pale-faced Sheathbills during spring. Of a mean of 80 birds (n = 6 censuses) observed along an 8-km shoreline, at least 90% were observed around seal colonies. The number of sheathbills associated with seal harems was significantly correlated with the number of females (r = 0.87, P ( 0.001, n = 25) and the number of pups per harem (r s = 0.85, P ( 0.001, n = 25). In 52% of scan samples sheathbills were feeding on feces, 26% on pup carcasses, 19% on afterbirths, and 3% on milk obtained from nursing cows (Table 1). During 10-min focal sampling, consumption estimates averaged 12.8 g of placenta (SD = 3.1, 95% C1 = 9.7-20.3, n = 8) and 6.2 g of flesh from pup carcasses (SD -- 4.4, 95% C1 -- 1.1-11.42, n -- 6). Between September and November, a total of 594 elephant seals were born (G. Daneri, pers. comm.), representing 2,376 kg of afterbirth available. An average of 35 Pale-faced Sheathbills were associated with the biggest harem found on Potter Peninsula (about 70 females); in three film records done on birds foraging on placentas, a mean of 8 sheathbills (and a maximum of 24 at the same time, n = 46) were seen. Birds spent between 25-45 min con- suming an afterbirth, depending not only on the number of sheathbills foraging but on the presence of other species. Estimated total placenta intake over the whole season was 214 kg (95% C1 = 162-339 kg), which was 9% (95% C1 = 7-14%) of the total available. The mean total intake per bird for the whole seal breeding season (ca. 40 days) was 2.7 kg (2.0- 4.2 kg), which represents a daily intake of 67 g (50-105 g) (Table 1). Because sheathbills (and other bird species) required several days to consume pup carcasses, it was impossible to estimate the consumption rate on the basis of field trials. However, considering the minimal prof- itable mass of a small elephant seal pup to be 36 kg (excluding skin, TABLE 2. Activity budget (as percentage of the day) of Pale-faced Sheathbills foraging for different foods. FE = feces, PL = placentas, MK = milk, Unknown = undetermined item. Feces Pups Placenta Milk Unknown Significance a Resting 33.1 14.8 17.1 20.9 5.6 ns b Comfort 2.9 0.3 6.2 0.0 84.4 ns Searching 43.3 48.6 40.6 36.7 8.8 ns Foraging 17.6 31.5 31.0 31.6 0.0 FE < PL (P < 0.05) Agonistic 2.4 4.8 4.9 10.4 0.1 FE < PL (P < 0.05) FE < MK (P < 0.05) Other 0.7 0.0 0.2 0.4 1.1 ns Mann-Whitney U-test (undetermined items not tested). No significant differences. bones, and flippers from a total of 50 kg; Hamilton 1949), nine dead pups during the season represented 324 kg of food. Taking into account the interactions with other species and the fact that sheathbills in Ant- arctica play the role of "bone-picking" (see discussion), it was roughly estimated that sheathbills took 10% of the total profitable mass which means a daily intake of 11 g/bird (Table 1). Time-activity budget.--Instantaneous scan samples revealed that sheath- bills spent 86% of the day foraging or displaying, and 14% resting or preening. Sheathbills associated with harems averaged 38% of the day searching for food, 20% feeding, 23% resting, 14% on comfort activities (mainly preening and bathing), and 3% on intraspecific agonistic behav- iors (postures, displacements, chases, and stretch threats). No significant differences were found in time activity budget through the day (Mann- Whitney test, P > 0.05). From focal observations there were, however, differences in activity budgets in relation to the food source used, es- pecially for foraging and agonistic activities (Table 2). The mean time spent in foraging activities when birds were feeding on feces (18%) was less than when feeding on afterbirths (31%) (Mann-Whitney test, P < 0.05). The mean number of agonistic interactions between sheathbills was significantly greater when birds were feeding on placentas or milk than when foraging on dead pups or feces (Mann-Whitney ttest, P < 0.05) (Fig. 1). When sheathbills were resting, agonistic behaviors were much less frequent (Mann-Whitney ttest, P < 0.01) than when foraging ac- tively. Resting birds were included in the analysis because, as observed for Black-faced Sheathbill (Chionis minor) (Burger 1981b), some of the time spent in resting and comfort activities constituted sedentary food search- ing time. The large number of stretch threats observed in birds feeding on placentae ( = 2.6/10 min, n = 28) could have been due to the high concentration of sheathbills present in a small area (18 birds in 100 m ) generating high rates of encounters. In birds foraging on dead pups or 8 - Postures Displacements 0 i'ili:i?" Stretch-threats MK PL DP FE FIGURE 1. Number of postures, displacements, and stretch-threats observed per 10 min of observation in Greater Sheathbills foraging on different items (MK -- milk, PL = pla- centas, DP = dead pups, FE -- feces). on milk, no more than 10 or 3 birds, respectively, were observed actively feeding. Interactions with other scavenger species.--As mentioned previously, the time spent consuming an afterbirth depended on the number of birds and species foraging. In one film record done in mid-September, Pale- faced Sheathbill was the main species, peaking at 18 individuals foraging, with Kelp Gull (Larus dominicanus) as the only additional species present (maximum 14 birds). In a record done in mid-October, Kelp Gulls were the most abundant species reaching 33 foraging birds, followed by Pale- faced Sheathbills (maximum 14 birds), 2 Antarctic Skuas (Catharacta ant- arctica), and 2 Southern Giant Petrels (Macronectes giganteus) (Fig. 2). Table 3 shows the interactions between species while consuming an af- terbirth, illustrating the strong relationship between dominance and body mass of the species. Despite the large numbers of sheathbills present around the placentas, in general they were displaced by larger species. However, sheathbills rarely flew when chased, and usually immediately recovered their former position. DISCUSSION The number of sheathbills associated with elephant seals was highly correlated with the size of harems (expressed as the total number of females or as pups born). As in previous reports (Burger 1981a, Favero 1993, Jones 1963), these data, combined with observation of sheathbills foraging at other sites during the censuses, indicate the importance of 20 Diet of Pale-faced Sheathbills [297 15 o 10 t.. E i . . e 25 Sep. o// ų GSh I '"D.. KGI ' ..---. 0 0...''''''"'0 ,on o ų [] brqb..a...a...n.-'" a [] o o". [] ß 5 10 15 20 25 30 35 40 45 Time (min) 40 50 35 30 25 20 E 15 10 15 Oct, ß ,,,e,,. GSh '"... KGI '-o... ASk GPt / o 6'. 0 5 10 15 20 25 30 Time [min) FIcuP. 2. Species involved in consumption of placentas on different dates during elephant seal breeding season (GSh = Greather Sheathbill, KGI = Kelp Gull, ASk -- Antarctic Skua, GPt -- Southern Giant Petrel). Polynomial five-order method used for fit lines. elephant seal colonies as a source of food during spring. This diet is very different from that of Black-faced Sheathbills at the same time of the year, because this species feed mainly in the intertidal zone, and excreta, pla- centas, or carcasses from elephant seals are only rarely taken (Burger 1981a). T/mIE 3. Number of Intra~ and interspecific interactions (displacements, chases, rightings) observed while birds foraging on placenta during a birth on 15 October. (+) = "won," (-) = "lost." GSh = Greather Sheathbill, KG1 = Kelp Gull, ASk = Antarctic Skua, GPt = Southern Giant Petrel. GSh KGI ASk GPt Birds displaced per attack a % won (inter- ( - ) ( - ) ( - ) ( - ) GSh KGI ASk GPt specific) GSh (+) 31 1 0 0 1.0 1.0 0.0 0.0 4 KG1 (+) 20 36 3 0 1.4 1.3 1.0 0.0 64 ASk (+) 3 7 0 0 5.3 4.4 0.0 0.0 77 GPt (+) 4 5 0 0 7.5 8.4 0.0 0.0 100 Mean number of birds displaced by one bird of another species using one attack. Using the existence metabolic rate from Kendeigh et al. (1977) for non- passerines at 0ųC, M(kcal/d) = 4.142 W ų's4 (where W is mass in g), the daily energy requirement of a Pale-faced Sheathbill is 140.8 kcal/d. Con- sumption rates averaged 75.7 and 36.3 kcal/d for placentas and pup car- casses respectively, which means that these items can supply up to 80% of the daily energy requirements (Table 1). During spring it is also pos- sible to find in the study area isolated carcasses of adult elephant seal, Weddell seal, and Antarctic fur seal (Arctocephalus gazella), which usually are scavenged by the same species that forage among pupping areas. The activity budget observed for Pale-faced Sheathbill in this study is similar to that reported by Burger (1984) for Black-faced Sheathbills dur- ing winter (69% foraging [search included], 21% resting and 9% preen- ing). No differences were found in activity budgets through the day, and some nocturnal inspections revealed that sheathbills were still active dur- ing the night, as has been observed in other shorebirds (Robert and McNeil 1989, Robert et al. 1989). The presence of other bird species foraging on placentas together with sheathbills depends both on the site and time of births. Sheathbills were dominant when births occurred in the center of harems. When foraging areas were on the borders of harems, Kelp Gulls and Southern Giant Petrels were more abundant. Skuas arrived at the study area on 6 October, but they were not abundant until 22 October when a large increase in the numbers along the coast was observed (D. Montalti, pers. comm.). This is important because both skua species (Catharacta antarctica and C. maccormick2) are dominant over sheathbills and gulls, and displace these species from placentas and pup carcasses (Table 3). On 20 October, 366 births (62% of the total) were registered; these placentas were consumed mainly by sheathbills, gulls, and giant petrels. Although the presence of other species results in interspecific compe- tition, sheathbills could benefit when Giant Petrels, skuas, and gulls open the skin of the carcasses of pups and adult seals. Furthermore, much of the food eaten by sheathbills, such as pieces of flesh picked from skele- tons, was too small to be eaten profitably by larger birds. In this case Pale- faced Sheathbills could play the role of "bone-picking" as reported for Black-faced Sheathbills, and high levels of interspecific competition at placentas and pup carcasses was probably mediated by size differences between members of the predator-scavenger guild (Burger 1981a). ACKNOWLEDGMENTS I thank J.P. Croxall, K. Yasukawa, C. Ray Chandler, and two anonymous referees whose criticisms and suggestions improved the manuscript. Thanks to G. Soave, D. Montalti and G. Danneri for supplying unpublished data. I'm particularly grateful to P. Silva for her help and assistance in the field. LITERATURE CITED AkTMaN, J. 1974. Observational study of behavior: sampling methods. Behavior 49:227- 265ß BuP, CEP,, A. E. 1980ß An analysis of the displays of lesser sheathbills Chionis min Z. Tierp- sychol 52:381-396. ß 1981a. Food and foraging behavior of Lesser Sheathbills at Marion Island. Ardea 69:167-180ß --. 1981b. Time budgets, energy needs and kleptoparasitism in breeding Lesser Sheath- bills (Chionis minor). Condor 83:106-112. ß 1984ß Winter territoriality in Lesser Sheathbills on breeding grounds at Marion Island. Wilson Bullß 96:20-33ß CARLINI, A. R., M. E. I. MRQUEZ, G. SOAVE, D. F. VERGANL, AND P. A. RONAYNE DE FERRER. 1994ß Southern elephant seal, Mirounga leonina: composition of milk during lactation. Polar Biol. 14:37-42. FAVP,O, M. 1993. Asociaciones alimentarias de la Paloma Anfftrtica Chionis alba en las Islas 25 de Mayo (King George Is.) y Media Luna (Halfmoon Is.), Shetland del Sur. Actas II Jor. Corn. Inv. Cient. Ant. 2:55-58ß Buenos Aires. FARO, M. 1996. Kleptoparasitism of Imperial Cormorants Phalacrocorax atriceps by Pale- faced Sheathbills Chionis alba in Antarctica. Marine Ornithol. 23, in press. HAMILTON, J. E. 1949ß Weight, etc., of Elephant Seal. Nature 163:536ß JONES, N. V. 1963. The sheathbill Chionis alba (Gmelin) at Signy Island, South Orkney Islands. Br. Antarct. Surv. Bull. 2:53-71ß KENDEIGH, S.C., V. R. DOLNK, AND V. M. GAWULOV. 1977ß Avian energeticsß Pp. 27-204, in J. Pinowski and S.C. Kendeigh, eds. Granivorous birds in Ecosystems. Cambridge Univ. Press, London, United Kingdom. LAVGNE, D. M., aND R. E. A. STEWArt. 1979. Energy content of harp seal placentas. J. Mamm. 60:854-856. ROBERT, M., AND R. MCNEILß 1989ß Comparative day and night feeding strategies of shore- bird species in a tropical environment. Ibis 131:69-79. , --, AND A. LEDUC. 1989. Conditions and significance of night feeding in shore- birds and other water birds in a tropical lagoonß Auk 106:94-101. SHAW, P. 1984ß The relationship between dominance behavior, bill size and age groups in Greater Sheathbills Chionis alba. Ibis 128:48-56ß ULRICH-PETER, H., M. KMSER, A. GEBAUER, AND D. ZIPPEL. 1988. The dynamics of wintering groups of Chionis alba at King George Island (South Shetland Islands). Veitr. Vogelk. 34:205-220. VERHEIDEN, C., AND P. JOUVENTIN. 1991. Over-wintering strategies of the Lesser Sheathbills Chionis minm:. Oecologia 86:132-139. WATSON, C,. E. 1975. Birds of Antarctic and Sub-antarctic. Am. Geoph. Union, New York, New York. 350 pp. Received 25 Aug. 1995; accepted 19 Sep. 1995.