Marine Biology (2000) 137: 1105±1110
Ó Springer-Verlag 2000
Y. Ropert-Coudert á J. Baudat á M. Kurita á C.-A. Bost A. Kato á Y. Le Maho á Y. Naito
Validation of oesophagus temperature recording for detection of prey ingestion on captive AdeÂlie penguins (Pygoscelis adeliae)
Received: 24 April 2000 / Accepted: 20 September 2000
J. Baudat á C.-A. Bost á Y. Le Maho Centre d'Ecologie et de Physiologie EnergeÂtiques, Centre National de la Recherche Scienti®que, 23, rue Becquerel, 67087 Strasbourg, France
ous relatively small loggers monitoring dierent parameters as a function of time have been developed to be deployed on animals foraging at sea (e.g. Le Maho 1994). Among the types of data collected, the time of ingestion and the quantity of prey ingested represent key parameters in investigations on the feeding behaviour of predators. In the Antarctic food web, top predators are endothermic and ingest ectothermic prey; thus, irrespective of local adjustments of the internal temperature (Handrich et al. 1997), ingestion events can be detected by units measuring drops in the internal body temperature of the predators (Wilson and Culik 1991; Wilson et al. 1992). Based on this phenomenon a number of studies have been conducted using stomach temperature loggers in marine mammals and birds (e.g. Wilson et al. 1992; Kato et al. 1996). However, the utility of these loggers is limited because the detection rate of ingestion events decreases as the stomach ®lls and prey items cover the sensor (Wilson et al. 1995). Furthermore, the accuracy of detection depends on the position of the logger inside of the stomach (GreÂmillet and PloÈs 1994; Wilson et al. 1995), because local changes in the temperature of the abdomen and the stomach sometimes occur entirely due to diving activity in cold water (Handrich et al. 1997) even when no prey are ingested. To monitor ingestion events more accurately, a recording system was proposed that measures oesophageal temperature (Ancel et al. 1997). Here, prey items are thought not to cover the sensor, and the time lag between ingestion and the detection of the drop is also considered minimal (Wilson et al. 1995). The aim of this experimental study was to test the eciency and the reliability of this new methodology to detect ingestion events. For this, captive AdeÂlie penguins (Pygoscelis adeliae) were equipped with the recording unit; they were fed prey of various sizes and supplied at various frequencies.
M. Kurita Nagoya Public Aquarium, 1-3 Nagoyashi Minato-ku 455-0033, Japan
Materials and methods
Abstract The eciency of oesophagus and stomach temperature loggers to detect ingestion of prey items was studied in captive AdeÂlie penguins (Pygoscelis adeliae) fed on land in Antarctica and in an aquarium in Japan. On land, the detection rate was studied for dierent masses of prey delivered at various frequencies, while in the pool the delay between capture and swallowing was investigated. The rate at which food items were detected and the magnitude of the temperature drops induced were higher in the oesophagus than in the stomach. Where small food items were delivered at a high frequency, birds collected prey items in the beak before swallowing them. Thus, oesophagus sensors may underestimate the number of prey swallowed if the system is used in the wild. In the oesophagus temperature recordings, the magnitude of drops was weakly, but positively, correlated to the mass of single, ingested prey (R2 0.40).
Introduction The feeding behaviour of animals foraging underwater at sea is dicult to observe directly. To elucidate this, variCommunicated by T. Ikeda, Hakodate Y. Ropert-Coudert (&) Department of Polar Science, The Graduate University for Advanced Studies, National Institute of Polar Research, 1-9-10 Kaga, Itabashi-ku, Tokyo 173-8515, Japan Fax: +81-3-39625743 e-mail:
[email protected]
A. Kato á Y. Naito National Institute of Polar Research, 1-9-10 Kaga, Itabashi-ku, Tokyo 173-8515, Japan
One ®rst set of experiments was conducted during the austral summer 1998/99 on captive AdeÂlie penguins (Pygoscelis adeliae)
1106 from a sub-colony of the Ile des PeÂtrels, Dumont d'Urville, AdeÂlie Land (66.7°S; 140.0°E). The response of the oesophageal temperature sensor to dierent masses of food items and to various frequencies of ingestion was studied on seven captive birds. We used 12 bit resolution, 2 Mbytes memory, two channels UME-TT loggers (Little Leonardo, Tokyo, Japan) to record oesophageal temperature. These devices consisted of a titanium cylinder (68 mm long ´ 15 mm diameter, M 30 g in the air) containing the electronic components and battery from which two soft plastic cables, 27.5 and 2 cm long (1.2 mm diameter), respectively, emerged from one end. A temperature sensor (accuracy: 0.1 °C) was incorporated at the end of each cable and set to record at a frequency of 1 Hz. The longest cable was attached along its length to a thin ®lament, 50 cm long (0.5 mm diameter), itself also emerging from the logger with the cables. Birds were caught in the morning, on the shore, while departing for foraging trip with their stomachs presumed empty, and were induced to swallow the moistened logger. The throat was gently rubbed until the logger reached the stomach, and the ®lament emerging from the beak was glued (Loctite) on the feathers at three dierent points: 1 cm from the mouth on the cheek and two points on the back of the head. The ®lament was multistranded to resist traction and thus avoid injury at the beak rictus. Using this system, the sensor at the end of the 27.5 cm cable was held in the oesophagus lumen at the back of the throat, while the other sensor recorded stomach temperature (Fig. 1). The correct cable length had been determined by measuring the length of the oesophagus on AdeÂlie penguin carcasses. The position of the logger in the stomach was also veri®ed by X-ray photography in one trial. Each feeding session consisted of delivery of ®ve pieces of dead shrimp kept in icy water (0±2 °C). The masses of each piece re¯ected as closely as possible the masses and volumes of juvenile to adult euphausids, the main prey of AdeÂlie penguins (Ridoux and Oredo 1989; Wilson et al. 1989, 1991; Trivelpiece et al. 1990; Watanuki et al. 1993, 1994, 1997; Kerry et al. 1995). The mass of prey delivered ranged from 0.2 to 5.4 g (n 325). During the sessions, penguins were held between the knees of an experimenter who opened the beak at speci®c times to allow the ingestion of pieces of shrimp. The time at which food was swallowed was noted. The feeding sessions alternated between high and low frequencies; high-frequency sessions involved introduction of food items at intervals 5 min to let oesophageal temperature reach normal values. At the end of the experiment, the logger was retrieved by pulling gently on the ®lament until the bird regurgitated the logger.
Fig. 1 Pygoscelis adeliae. Attachment technique of oesophageal temperature logger to captive AdeÂlie penguins
In a second set of experiments in July 1999, three captive AdeÂlie penguins from the Nagoya Public Aquarium, Japan, were equipped with UME-TT loggers and released in a 21 ´ 4 m pool, 2.1±2.3 m deep, where they fed freely underwater on dead Euphausia superba. One bird was equipped at a time to facilitate observation. During feeding sessions, bird behaviour was monitored by video-camera and the time of prey capture recorded. During feeding sessions, food items were delivered by throwing handfuls of krill into the pool while several penguins were swimming. These feeding sessions ended when the basket of krill, containing 10±15 handfuls, was empty. The mass of the individual pieces of krill delivered was not measured but globally pieces ranged from 0.5 to 1.2 g. The temperature of the water was 6.2±6.4 °C. After the retrieval of the loggers in both series of experiments, the data were downloaded to a computer and analysed with IGOR software (Wavemetrics, USA). The data obtained in AdeÂlie Land were used to examine the response of the oesophagus temperature logger to various masses and frequencies of feeding. The experiments in Japan were used to examine the delay between capture and swallowing by comparing the time of capture recorded by video-camera with the time a drop in the oesophagus temperature occurred.
Results Feeding experiments on land A total of 36 low- and 27 high-frequency feeding sessions were performed on the seven birds which corresponded to an average amount of krill delivered per bird of 51.8 41.5 g (total number of pieces of shrimps seen swallowed by the seven birds 309). The feeding experiments lasted a mean of 1.3 0.4 h. During the experiments, the body temperature of the birds increased by about 1 °C in response to the stress of being restrained (Boyd and Sladen 1971; Regel and PuÈtz 1997). For three birds, the stomach temperature remained essentially constant at 40.8 0.4 °C, 39.5 0.4 °C and 40.9 0.6 °C, respectively, so that prey ingestion could not be detected. The logger of another bird did not record reliable data for stomach temperature. Finally, in the three last birds 63 temperature drops corresponded to 129 prey swallowed (49%). The average magnitude of these stomach temperature drops was 2.4 0.5 °C (n 63). Additionally, there was a substantial delay between the time the bird was seen swallowing and the initiation of the following temperature drop in the stomach. This delay was calculated for only one bird (5.6 4.5 s, n 32), because it was impossible to relate the time of swallowing and the initiation of following temperature drop reliably. The recording of oesophagus and stomach temperatures revealed several drops of various magnitudes (Fig. 2). Drops recorded by the sensors appeared to have various origins: ingestion of a prey at a lower temperature, variability in the internal temperature or variability in the measurements due to the logger itself. For each drop in the internal temperature, three parameters were collected: duration, magnitude and slope of drops. In the oesophagus, when these parameters were compared between drops directly following prey ingestion to drops due to other sources of variability,
1107 Fig. 2 Pygoscelis adeliae. Oesophagus temperature recorded in a captive AdeÂlie penguin showing various drops. Precipitous drops corresponding to the ingestion of cold pieces of shrimps are indicated by arrows
magnitude proved the most discriminating, with 96% of temperature drops with a magnitude >0.25 °C directly following prey ingestion being detected (Fig. 3). However, 12.7% of temperature drops >0.25 °C did not correspond to prey ingestion. In order to separate drops due to ingestion of prey from others, a mathematical method was applied to the data bird-by-bird (Ropert-Coudert et al. 2000). This method detects events that depart substantially from the norm. All decreasing events in the oesophageal temperature were ®rst identi®ed and, for all birds, were counted for dierent thresholds. The number of minor temperature drops was very high, re¯ecting the noise due to variability, but the frequency of temperature drops decreased substantially with increasing temperature drops. The form that this took was a sharp decrease in numbers before remaining roughly constant. This stabilisation corresponded to the detection of pronounced temperature drops that were due to food ingestion and distinct from normal variability. When the limit between noise and clear peaks was dicult to estimate for all birds, the dierentiation procedure was repeated on a bird-by-bird basis, giving a threshold for each bird. Finally, the time of all peaks above a speci®c threshold indicated the position of the pronounced temperature drop events in each bird, allowing accurate determination of drops. Using this method, from the 309 food items seen swallowed, 176 drops in the oesophagus temperature were detected whatever the mass of prey or the frequency of ingestion. Of these, 3% (n 5) was not associated with prey ingestion. Therefore, 171 temperature drops corresponding to food ingestion were taken into account by the mathematical method, corresponding to 55.3% of the 309 food items seen swallowed. The average magnitude of temperature drops was 8.2 3.3 °C, and the delay between the time of swallowing and the start time of the temperature drop was 4.0 5.5 s (n 171). Firstly, the success of detection was compared between high (3 s. The bird captured on average 3.5 1.9 prey bout)1. Comparison of the number of catches recorded by video-camera and the number of temperature drops recorded by the logger showed that 44.7% of the ingestion events were detected. Eleven isolated captures were followed by temperature drops after a delay of 3.8 2.5 s. Seventeen bouts of captures were followed by a temperature drop after a delay of 4.8 3.6 s. Five bouts had a temperature drop that occurred before the end of the bout. Finally, two bouts did not relate to any drop.
Fig. 4 Pygoscelis adeliae. Percentage of detection by oesophageal temperature sensor of dierent masses of cold food items. The frequency of delivery was low (³20 s)
On average, 2.2 1.0 apparent ingestion events (range 1±4) corresponded to one temperature drop, which shows that prey were gathered in the beak before swallowing. The mean magnitude of temperature drops following a single capture event (0.8 0.5 °C, n 9) was statistically less than those following several catches (1.8 1.1 °C, n 23; one-way ANOVA, F1 6.6, P < 0.016). Finally, the magnitude of the drop was independent of the pre-ingestion oesophageal temperature when the ingestion event occurred (Spearman's rank correlation coecient, q 0.20, P 0.17).
Discussion The impact of oesophageal temperature logger on the behaviour of AdeÂlie penguins is dicult to assess. In our study, equipped birds were nervous when released in the cage prior to feeding and, in some instances, tried to regurgitate the device. However, after some hours of isolation, the birds remained relatively quiet and did not exhibit any behaviour that could indicate a discomfort due to the logger. Moreover, after retrieval of the device, the birds did not have any injuries at the mouth corner or in the oesophagus. Stomach temperature sensors have helped substantially to elucidate feeding behaviour of seabirds and marine mammals (Weimerskirch and Wilson 1992; Wilson et al. 1992, 1993, 1994, 1995; GreÂmillet and PloÈs 1994; PuÈtz 1994; PuÈtz and Bost 1994; Wilson and Wilson 1995; Wilson 1995). However, temperature sensors placed in the stomach have highly variable responses to prey ingestion for a variety of reasons, such as the degree of fullness of the stomach, the amount of stomach churning, prey size, etc. (Wilson et al. 1995). In addition, the position of the sensor within the stomach can aect the likelihood that ingested prey will be detected (GreÂmillet and PloÈs 1994; Wilson et al. 1995). In our experimental study, the stomach temperature recordings of three birds did not show any substantial drops, indicating that the temperature sensor was likely directly in contact with a portion of the stomach wall and thus unlikely to come into contact with recently ingested prey. In the case of birds where drops were recorded in the stomach temperature, the detection rate might have been greater than that obtained on free-ranging animals, since the stomach of our captive birds was considered empty. This would lead to a greater probability that prey would touch the temperature sensor. In addition, the birds were on land, and, thus, stomach contents would not have been mixed due to swimming activity. Finally, since the birds were not in the water, there was no cooling of the internal tissues (Handrich et al. 1997) that could complicate the detection of prey ingestion. We consider it likely that prey were warmed up during the descent from the mouth to the stomach, so that the magnitude of temperature drops measured in the stomach was smaller than that recorded in the oesophagus.
1109 Fig. 5 Pygoscelis adeliae. Mass of cold food items swallowed and magnitude of the corresponding temperature drop recorded in the oesophagus of captive AdeÂlie penguins
The detection rate by the oesophagus sensor was considerably higher than that of the stomach sensor but still showed some limitations in the case of high feeding frequency. As krill are patchily distributed in the wild (Nicol and de la Mare 1993), free-ranging penguins may be picking up prey at a high rate. Falla (1937, cited in Zusi 1975) quoted Sir Douglas Mawson who watched AdeÂlie penguin feeding by saying: ``their heads darting from time to time to the left and to the right (...) and their beaks were going out just about as fast as barn-yard fowl feed on grain thrown on the ¯oor''. Similar behaviour was observed during the aquarium experiments, where the prey delivery method led to a patchy distribution of krill. The observation that birds collected several prey items over a few seconds, keeping them in the beak before swallowing, has super®cial similarities to pun foraging behaviour (Harris 1984) and leads to several prey being registered as a single item. Furthermore, because prey gathered in the bill would be likely warmed up, the subsequent temperature drop measured by the sensor would lead to an underestimation of the mass ingested. In some instances, prey items were seen swallowed oneby-one during high-frequency feeding sessions, but the warming of the oesophageal temperature between two pieces was probably too slow to be registered, leading once again to a single temperature drop recorded for several prey items swallowed. This phenomenon depends mainly on the speed at which oesophageal temperature changes during the warming process and the rate at which the cold prey might pass through it. Although the detection rate for isolated ingestion events was high for prey weighing >0.4 g, it was low in the case of small mass items, especially, for prey ranging from 0.2 to 0.4 g [converted into length (Miller 1986) this corresponds to krill 30±40 mm long]. Thus, oesophageal temperature loggers deployed on freeranging AdeÂlie penguins would tend to underestimate the number of small, isolated prey captured. This is likely to be the case for Euphausia crystallorophias, ranging from 14 to 41 mm in length (Paulin 1975; Ridoux and Oredo 1989; Davis and Miller 1990; Coria et al. 1995), immature euphausids (Miller and Hampton
1989) and some cephalopods (Ridoux and Oredo 1989; Coria et al. 1995). However, in our experiment, the temperature of prey was probably higher than in the wild and the magnitude of the temperature drops correspondingly reduced. Nevertheless, our results indicate that around 70% of isolated prey larger than 0.4 g are likely to be detected. In most localities, Euphausia superba is the dominant prey of AdeÂlie penguins, with a body size ranging from 32 to 59 mm (Ridoux and Oredo 1989; Wilson et al. 1989, 1991; Trivelpiece et al. 1990; Watanuki et al. 1993, 1994, 1997; Coria et al. 1995; Kerry et al. 1995), and a mass of between 0.8 and 1.5 g for gravid females (Hosie 1994). In some instances, AdeÂlie penguins may switch from krill to primarily ®sh (Hopkins 1987; Ridoux and Oredo 1989; Watanuki et al. 1993; Kerry et al. 1995), although, since the main ®sh prey is Pleuragramma antarcticum with lengths up to 75 mm (Paulin 1975; Oredo et al. 1985; Watanuki et al. 1994), it is unlikely that the detection of this would be problematic. Thus, oesophageal temperature sensors may provide substantial information about the ingestion of large prey in other species, such as king penguin (Aptenodytes patagonicus), feeding mainly on myctophid ®sh (Cherel and Ridoux 1992), or Spheniscus penguins feeding primarily on pelagic school ®sh (for review of penguin feeding habits see Croxall and Lishman 1987; Williams 1995). This study demonstrates the potential of an oesophageal temperature sensor for the detection of feeding in AdeÂlie penguins. Compared to stomach temperature recordings, the percentage of ingestion events detected, as well as the clarity of the temperature drops, has improved. Despite this, it is still impossible to separate temperature drops due to prey from those due to ice or water ingestion and to determine accurately the mass of the prey ingested (Hedd et al. 1996). The use of oesophageal temperature sensors in tandem with other loggers, such as time-depth recorders, might help eliminate some of these problems. Acknowledgments The authors wish to express their gratitude to the Institut FrancËais pour la Recherche et la Technologie Polaires
1110 (I.F.R.T.P) and the Terres Australes and Antarctiques FrancËaises (T.A.A.F.) for their ®nancial and logistics support during the experiments. This work was also ®nancially supported by the Grantin-Aid for International Scienti®c Research from the ministry of Education, Science, Sports and Culture of Japan. Many thanks also to all the members of the 49th mission in Dumont d'Urville and especially the logistics sta who provided us with help and support. In addition, we would like to thank C. Duchamp and F. Denjean for their invaluable friendship. Extra thanks to M. Kuroki for sharing her experience of oesophagus logger deployment on penguins.
References Ancel A, Horning M, Kooyman GL (1997) Prey ingestion revealed by úsophagus and stomach temperature recordings in cormorants. J Exp Biol 200: 149±154 Boyd JC, Sladen WJL (1971) Telemetry studies of the internal body temperatures of AdeÂlie and emperor penguins at Cape Crozier, Ross Island, Antarctica. Auk 88: 366±380 Cherel Y, Ridoux V (1992) Prey species and nutritive value of food fed during summer to king penguin, Aptenodytes patagonica, chicks at Possession Island, Crozet archipelago. Ibis 134: 118± 127 Coria NR, Spairani H, Vivequin S, Fontana R (1995) Diet of AdeÂlie penguins, Pygoscelis adeliae, during the post-hatching period at Esperanza Bay, Antarctica, 1987/88. Polar Biol 15: 415±418 Croxall JP, Lishman G (1987) The food and feeding ecology of penguins. In: Croxall JP (ed) Seabirds, feeding ecology and role in marine ecosystems. Cambridge University Press, Cambridge, pp 101±133 Davis LS, Miller GD (1990) Foraging patterns of AdeÂlie penguins during the incubation period. In: Kerry KR, Hempel G (eds) Antarctic ecosystems, ecological change and conservation. Springer, Berlin Heidelberg New York, pp 203±207 Falla RA (1937) Birds. B.A.N.Z. Antarct Res Exped 1929±31 Rep, Ser B, No 2, 1±288 GreÂmillet D, PloÈs AL (1994) The use of stomach temperature records for the calculation of daily food intake in cormorants. J Exp Biol 189: 105±115 GreÂmillet D, Wilson RP, Storch S, Gary Y (1999) Three dimensional space utilization by a marine predator. Mar Ecol Prog Ser 183: 263±273 Handrich Y, Bevan RM, Charrassin J-B, PuÈtz K, Butler PJ, Lage J, Le Maho Y, Woakes AJ (1997) Hypothermia while hunting in a bird diver: the king penguin (Aptenodytes patagonicus). Nature 388: 64±67 Harris MP (1984) Food and feeding of puns. In: Harris MP (ed) The pun. Poyser, Staordshire, pp 114±129 Hedd A, Gales R, Renouf D (1996) Can stomach temperature telemetry be used to quantify prey consumption by seals? Polar Biol 16: 261±270 Hopkins TL (1987) Midwater foodweb in MacMurdo Sound, Ross Sea, Antarctica. Mar Biol 96: 93±106 Hosie GW (1994) Multivariate analyses of the macrozooplankton community and euphausiid larval ecology in the Prydz Bay region, Antarctica. ANARE report 137, Australian Antarctic Division, Hobart, Tasmania Kato A, Naito Y, Watanuki Y, Shaughnessy PD (1996) Diving pattern and stomach temperature of foraging king cormorants at subantarctic Macquarie Island. Condor 98: 844±848 Kerry KR, Clarke JR, Else GD (1995) The foraging range of AdeÂlie penguins at Bechervaise Island, Mac Robertson Land, Antarctica as determined by satellite telemetry. In: Dann P, Norman I, Reilly P (eds) The penguins. Surrey Beatty, Kingston, Tasmania, pp 216±243 Le Maho Y (1994) New perspectives for research on Antarctic birds and mammals. Polar Biol 14: 315±318 Miller DGM (1986) Results from biological investigations of krill (Euphausia superba) in the southern Indian Ocean during SIBEX I. Mem Natl Inst Polar Res, 40 (Spec Issue): 117±139
Miller DGM, Hampton I (1989) Biology and ecology of the Antarctic krill (Euphausia superba dana): A Review. Biomass Handbook Series 9: 1±166 Nicol S, de la Mare W (1993) Ecosystem management and the Antarctic krill. Am Sci 81: 36±47 Oredo C, Ridoux V, Clarke RM (1985) Cephalopods in the diets of Emperor and AdeÂlie penguins in AdeÂlie land, Antarctica. Mar Biol 86: 199±202 Paulin CD (1975) Feeding of the AdeÂlie penguin Pygoscelis adeliae. Mauri Ora 3: 27±30 PuÈtz K (1994) Aspects of the feeding ecology of Emperor penguins (Aptenodytes forsteri) and King penguins (Aptenodytes patagonicus). Rep Polar Res 136: 1±139 PuÈtz K, Bost C-A (1994) Feeding behaviour of free ranging king penguins (Aptenodytes patagonicus). Ecology 75(2): 489±497 Regel J, PuÈtz K (1997) Eect of human disturbance on body temperature and energy expenditure in penguins. Polar Biol 18: 246±253 Ridoux V, Oredo C (1989) The diets of ®ve summer breeding seabirds in AdeÂlie Land, Antarctica. Polar Biol 9: 137±145 Ropert-Coudert Y, Sato K, Charrassin J-B, Bost C-A, Le Maho Y, Naito Y (2000) Preliminary investigations of prey pursuit and capture by king penguins at sea. Polar Bioscience 13: 102±113 Trivelpiece WZ, Trivelpiece SG, Reupel GR, Kjelmyr J, Volkman NJ (1990) AdeÂlie and Chinstrap penguins: their potential as monitors of the Southern Ocean marine ecosystem. In: Kerry KR, Hempel G (eds) Antarctic ecosystems, ecological change and conservation. Springer, Berlin Heidelberg New York, pp 191±202 Watanuki Y, Kato A, Mori Y, Naito Y (1993) Diving performance of AdeÂlie penguins in relation to food availability in fast sea-ice areas: comparison between years. J Anim Ecol 62: 634±646 Watanuki Y, Mori Y, Naito Y (1994) Euphausia superba dominates in the diet of AdeÂlie penguins feeding under fast sea-ice in the shelf areas of Enderby Land in summer. Polar Biol 14: 429±432 Watanuki Y, Kato A, Naito Y, Robertson G, Robinson S (1997) Diving and foraging behaviour of AdeÂlie penguins in areas with and without fast sea-ice. Polar Biol 17: 296±304 Weimerskirch H, Wilson RP (1992) When do wandering albatrosses (Diomedea exulans) forage? Mar Ecol Prog Ser 86: 297±300 Williams TD (1995) The penguins. Oxford University Press, Oxford Wilson RP (1995) Foraging ecology. In: Williams TD (ed) The penguins. Oxford University Press, Oxford, pp 81±106 Wilson RP, Culik BM (1991) The cost of a hot meal: facultative speci®c dynamic action may ensure temperature homeostasis in post-ingestive endotherms. Comp Biochem Physiol 100A: 151± 154 Wilson RP, Wilson M-P (1995) The foraging behaviour of the African penguin Spheniscus demersus. In: Dann P, Norman I, Reilly P (eds) Penguin biology: advances in research and management. Surrey Beatty, Sydney Wilson RP, Nagy KA, Obst BS (1989) Foraging ranges of penguins. Polar Rec 25: 303±307 Wilson RP, Culik BM, Adelung D, Spairani HJ, Coria NR (1991) Depth utilization by breeding AdeÂlie penguins, Pygoscelis adeliae, at Esperanza Bay, Antarctica. Mar Biol 109: 181±189 Wilson RP, Cooper J, PloÈtz J (1992) Can we determine when marine endotherms feed? A case study with seabirds. J Exp Biol 167: 267±275 Wilson RP, PuÈtz K, Bost C-A, Culik BM, Bannasch R, Reins T, Adelung D (1993) Diel dive depth in penguins in relation to diel vertical migration of prey: whose dinner by candlelight? Mar Ecol Prog Ser 94: 101±104 Wilson RP, Culik BM, Bannasch R, Lage J (1994) Monitoring Antarctic environmental variables using penguins. Mar Ecol Prog Ser 106: 199±202 Wilson RP, PuÈtz K, GreÂmillet D, Culik BM, Kierspel M, Regel J, Bost C-A, Lage J, Cooper J (1995) Reliability of stomach temperature changes in determining feeding characteristics of seabirds. J Exp Biol 198: 1115±1135 Zusi L (1975) An interpretation of skull structure in penguins. In: Stonehouse B (ed) The biology of penguins. Macmillan, London, pp 59±84
