The Auk 109(3):654-658, 1992 Centre d'Etudes Biologiques de Chizg, Centre National de la Recherche Scientifique, Villiers en Bols, 79360 Beauvo&-Sur-Niort, France Aptenodytes penguins are colonial seabirds that have no nest but incubate a single egg on their feet. There are two species in the genus: the Emperor Penguin (A. forsteri) breeds during the harsh antarctic winter on the sea ice, and the King Penguin (A. patagonicus) breeds in subantarctic regions on beaches. Both sexes in each species produce a mutual display call that facilitates the search for a partner at the pairing pe- riod. When these penguins return from the sea to resume responsibility for the egg or chick, they use the same call to achieve mutual recognition of the incubating partner and its mate (Stonehouse 1960, Prvost 1961, Jouventin 1982). King Penguins head towards the restricted area of the colony where their partners are incubating and call to be recognized. However, partner reunion would appear to be par- ticularly difficult for the Emperor Penguin because these birds exhibit a wandering incubation that en- ables them to regulate the microclimate of the colony by gathering in dense huddles, thus withstanding the INTENSITY A Fig. 1. Sonagrams (at a sampling rate of 6,512 Hz and a FFT of 256 points; frequency bandwidth 25.4 Hz) of the individual-specific calls in: (A) a male Emperor Penguin; and (B) a male King Penguin. On spectral slices performed at vertical line of sonagrams in long syllable of first series of syllables, arrows indicate the two voices. extreme weather conditions (PrEvost 1961). The birds also can move freely to more favorable places when necessary. This mobility demands that each time a bird returns from foraging, it may have to search among thousands of other birds, and the reunion is achieved only by vocalizations. Thus, in both species, but especially in the Emperor Penguin, the lack of a nest would be predicted to make the reunion of mates difficult. The Aptenodytes mutual display call is spe- cific to individuals (Jouventin 1982), and a bird must deal with the complexity of this specificity in order to distinguish its partner from conspecifics. Acoustic communication would be expected to exhibit marked adaptations to such extreme circumstances. The mutual-display call is composed of syllables separated by amplitude declines. These declines co- incide with falls in frequency for King Penguins, and are further pronounced to the extent that they appear as silences for Emperor Penguins (Fig. 1; Jouventin 1982, Brmond et al. 1990). Syllables are grouped in repetitive series and a series i defined as a group of syllables terminated by a long syllable. A call is com- posed of one to four series. Syllables of the Emperor Penguin call have two frequency bands with their respective harmonics (Brmond et al. 1990). The in- teraction of these frequencies generates a beat, which Brmond et al. (1990) suggested conveyed informa- tion of individual identity. Other studies (Robisson et al. 1989, Robisson 1990) have demonstrated the importance of the syllable structure for individual recognition in the genus Aptenodytes. Finally, bird species are well known to be able accurately to resolve minor differences in frequency (Dooling 1980, Kuhn et al. 1980, Stebbins 1983, Hulse et al. 1984), which has been suggested as a means to categorize notes (Weary 1990) and to discriminate among individuals (Brooks and Falls 1975, Falls 1982, Nelson 1989, Weis- man et al. 1990). This led me to study frequency at- tributes of the mutual-display call of Aptenodytes pen- guins. [ recorded Emperor Penguins on the Pointe Go- 1ogie Archipelago, Antarctica (66ø40'S, 140ø01'E) in the austral winter of 1987, and King Penguins on the Crozet Islands, Indian Ocean (46ø50'S, 37ø45'E) in the austral summer of 1989. I used an omnidirectional Beyer Dynamic M69 microphone connected to a Na- gra III or Uher 4000C tape-recorder ( 19.05 cm / s). Calls were displayed and analyzed on an Amiga microcom- puter associated with an analytic package that cal- culated a Fast Fourier Transform (FFT) with 256 con- secutive points and performed a spectrographic representation (Richard 1991). Figure 1 represented spectrograms with a frequency precision of + 12.7 Hz (sampling frequency of 6,512 Hz), whereas I mea- sured frequencies with a precision of +6.4 Hz (sam- pling frequency of 3,256 Hz). Frequency maxima and minima of each syllable of the first series were mea- sured. I calculated the average values for the lower TABLE 1. Means (+SD) of frequency features of mutual-display call in King and Emperor penguins. The following numbers of calls, each produced by a different bird, were analyzed: King Penguin, males 28, females 30; Emperor Penguin, males 23, females 24. All units in Hertz. Levels of significance two-tailed t-test. ***, P < 0.001; ns, P > 0.05. Penguin Species Sex Variable King Emperor differences Male Lower voice 456.4 _+ 29.9 370.7 + 24.2 10.92'** Upper voice 480.1 + 24.5 431.5 + 25.0 6.88*** Beat 23.6 + 17.1 60.4 + 13.7 8.45*** Female Lower voice 501.6 + 50.0 432.5 + 43.2 5.36*** Upper voice 528.0 + 46.4 528.3 + 53.7 0.02 ns Beat 27.1 + 22.0 95.9 + 25.3 10.51'** Sex differences Lower voice 6.30*** 8.54*** (t-values) Upper voice 8.18'** 10.04'** Beat 0.68 ns 0.99 ns and upper frequency bands, and the interval fre- quency between the two bands. Frequency features were compared between species and between sexes (t-test), and between individuals (single-factor anal- ysis of variance; F-test). Both Aptenodytes species produced a signal con- sisting of a continuous pattern of two simultaneous series of harmonically related bands of slightly dif- fering frequencies (Figs. 1A and B; see arrows). Such a signal corresponds closely with what other workers have termed a "two-voice" signal, which many birds are stated to produce through the elaboration of syr- inx anatomy. Located at the junction of the two pri- mary bronchi, each half of this two-part organ has an independent set of muscles and membranes thought to be involved in phonation and controlled separately by the tracheosyringealis branches of the right and left hypoglossus nerves. This double system enables birds to produce two voices simultaneously, as evi- denced by results of many studies on avian anatomy (Stein 1968, Gaunt et al. 1982, Gaunt 1983) and phys- iology (Nottebohm 1971, 1972, Nottebohm and Not- tebohm 1976, Nowicki and Capranica 1986, Suthers 1990), as well as by the spectrographic analyses of songs (Greenewalt 1968, Stein 1968, Latimer 1977, King and West 1983, Adret-Hausberger and Jenkins 1988) and calls (Stoddard and Beecher 1983, Beecher et al. 1985, Aubin 1986, Weisman et al. 1990). There- fore, the application of the two-voice theory (Greene- wait 1968) to the double-frequency structure of Ap- tenodytes calls is entirely reasonable. Between-species comparison showed that the pitch of the call was greater in King than in Emperor Pen- guins except for the upper voice for females, and beat frequency was significantly greater for Emperor than for King penguins (Table 1). This indicated that fre- quency characteristics were species-specific. Lower and upper frequencies were significantly higher for fe- males than for males in Emperor and King penguins, while no significant difference was found for the beat frequency. The sex differences in frequency features are clear in Table 1, though it is not these features that render the calls of the two sexes so distinct to the human ear. Rather, it is the differences in the temporal patterning of syllables (Jouventin 1982). I recorded seven male Emperor Penguins and seven male King Penguins from 5 to 10 times each. This gave a total of 47 Emperor Penguin calls and 53 King Penguin calls. An F-test showed significantly greater inter- than intraindividual variations for absolute fre- quencies and beat frequency in Emperor Penguins (lower voice, F = 80.1, df = 6 and 40, P < 0.001; upper voice, F = 99.5, df = 6 and 40, P < 0.001; beat, F = 98.6, df = 6 and 40, P < 0.001) and King Penguins (lower voice, F = 293.7, df = 6 and 46, P < 0.001; upper voice, F = 264.8, df = 6 and 46, P < 0.001; beat, F = 120.2, df = 6 and 46, P < 0.001). I suggest this represents a significant variability of frequency fea- tures, demonstrating their potential role as individual markers.' The sound structures of the functionally identical call in 10 other penguin species were analyzed (three Pygoscelis species, three Eudyptes species, two Sphen- iscus species, Megadyptes antipodes, and Eudyptula mi- nor), but only the Aptenodytes species employ two frequency bands. It is presumably more than coin- cidence that these 10 species, and not Aptenodytes spe- cies, breed on a fixed nest site that appears ideally suited as a rendezvous point. Therefore, I suggest that the exploitation of two acoustic sources represents a means whereby Aptenodytes penguins can increase the information content of the call regarding indi- vidual identity, which can facilitate the recognition process. This is in accordance with the model pro- posed by Schleidt (1976) where the number of fea- tures is a component of individual distinctivehess. A further possibility is that complexity of the call in- creased by two voices has evolved in parallel with loss of territoriality. The next step is to test experi- mentally whether birds actually use two acoustic sources to generate features relevant to the recogni- tion processes. I thank P. Jouventin (director of the Antarctic Mam- mals and Birds Research Group), T. Aubin, J.-C. Br- mond, V. Bretagnolle, and S. J. G. Hall for their help- ful comments on the manuscript. LITERATURE CITED ADRET-HAUSBERGER, M., AND P. F. JENKINS. 1988. Complex organization of the warbling song in the European Starling Sturnus vulgaris. Behaviour 107:138-156. AUBIN, T. 1986. Le phnomne deux voix et le co- dage de l'information dans les signaux acous- tiques d'oiseaux. Etude applique aux cris de d- tresse de l'tourneau (Sturnus vulgaris). C. R. Acad. Sci. Ser. III Sci. Vie 302:553-556. BELCHER, M. D., P. K. STODDARD, AND P. LOESCHE. 1985. Recognition of parents' voices by young Cliff Swallows. Auk 102:600-605. BRP2vlOND, J.-C., T. AUB1N, R. M. NYAMSI, AND P. ROBIS- SON. 1990. Le chant du manchot empereur (Ap- tenodytes forsteri): R;echerche des paramtres util- isables pour la reconnaissance individuelle. C. R. Acad. Sci. Ser. III $ci. Vie 311:31-35. BROOKS, R. J., AND J. B. FALLS. 1975. Individual rec- ognition by song in White-throated Sparrows. III. Song features used in individual,recognition. Can. J. Zool. 53:1749-1761. DOOLING, R.J. 1980. Behavior and psychophysics of hearing in birds. Pages 261-288 in Comparative studies of hearing in vertebrates (A. N. Popper and R. R. Fay, Eds.). Springer-Verlag, Berlin. FALLS, J. B. 1982. Individual recognition by sounds in birds. Pages 237-278 in Acoustic communica- tion in birds, vol. 2 (D. E. Kroodsma and E. M. Miller, Eds.). Academic Press, New York. GAUNT, A.S. 1983. On sonograms, harmonics, and assumptions. Condor 85:259-261. GAUNT, A. S., S. L. L. GAUNT, AND R. M. CASEY. 1982. Syringeal mechanics reassessed: Evidence from Streptopelia. Auk 99:474-494. GREENEWALT, C. H. 1968. Bird song: Acoustics and physiology. Smithsonian Inst. Press, Washing- ton. HULSE, S. H., J. CYNX, AND J. HUMPAL. 1984. Absolute and relative pitch discrimination in serial pitch perception by birds. J. Exp. Psychol. Gen. 113:38- 54. JOUVENT1N, P. 1982. Visual and vocal signals in pen- guins, their evolution and adaptive characters. Verlag Paul Parey, Berlin. KING, A. P., AND M. J. WEST. 1983. Dissecting cow- bird song potency: Assessing a song's geographic identity and relative appeal. Z. TierpsychoL 63: 37-50. KUHN, A., H. J. LEPPELSACK, AND J. SCHWARTZ- KOPFF. 1980. Measurement of frequency dis- crimination in the Starling (Sturnus vulgaris) by conditioning of heart rate. Naturwissenschaften 678:102. LATIMER, W. 1977. A comparative study of the songs and alarm calls of some Parus species. Z. Tierpsy- chol. 45:414-433. NELSON, D. A. 1989. Song frequency as a cue for recognition of species and individuals in the Field Sparrow (Spizella pusilia). J. Comp. Psychol. 103: 171-176. NOTTE8OHM, F. 1971. Neural lateralization of vocal control in a passerine bird. I. Song. J. Exp. Zool. 177:229-262. NOTTE8OHM, F. 1972. Neural lateralization of vocal control in a passerine bird. II. Subsong, calls, and a theory of the vocal learning. J. Exp. Zool. 179: 35-50. NOTTEIOHM, F., AND M. E. NoTTEOHM. 1976. Left hypoglossal dominance in the control of canary and White-crowned Sparrow song. J. Comp. Physiol. 108:171-192. NOWICK1, S., AND R. R. CAPRANICA. 1986. Bilateral interaction in vocal production of an oscine bird sound. Science 231:1297-1299. PRRVOST, J. 1961. Ecologie du manchot empereur. Hermann, Paris. RICHARD, J.P. 1991. Sound analysis and synthesis using an Amiga micro-computer. Bioacoustics 3: 45-60. ROBlSSON, P. 1990. The importance of the temporal pattern of syllables and the syllable structure of display calls for individual reqognition in the genus Aptenodytes. Behav. Processes 22:157-163. ROllSSON, P., T. AUIN, AND J.-C. BP, gMOND. 1989. La reconnaissance individuelle chez le manchot em- pereur (Aptenodytes forsteri): R61es respectifs du d6coupage temporel et de la structure syllabique du chant de cour. C. R. Acad. Sci. Ser. III Sci. Vie 309:383-388. SCHLEIDT, W. M. 1976. On individuality: The con- stituents of distinctiveness. Pages 299-310 in Per- spectives in ethology, vol. 2 (P. P. G. Bateson and P. H. Klopfer, Eds.). Plenum, New York. STEBB1NS, W.C. 1983. The acoustic sense of animals. Harvard Univ. Press, Cambridge, Massachusetts. STEIN, R. C. 1968. Modulation in bird sounds. Auk 85:229-243. STODDARD, P. K., AND M.D. BELCHER. 1983. Parental recognition of offspring in the Cliff Swallow. Auk 100:795-799. STONEHOUSE, B. 1960. The King Penguin (Aptenody- tes patagonicus) of South Georgia. I. Breeding be- haviour and development. Falk. Isl. Depend. Surv. Sci. Rep. 23. SUTHERS, R.A. 1990. Contributions to birdsong from the left and right sides of the intact syrinx. Nature (Lond.) 347:473-477. WEARY, D.M. 1990. Categorization of song notes in Great Tits: Which acoustic features are used and why? Anim. Behav. 39:450-457. WEISMAN, R., L. RATCLIFFE, I. JOHNSRUDE, AND T. A. HURLY. 1990. Absolute and relative pitch pro- duction in the song of the Black-capped Chick- adee. Condor 92:118-124. Received 22 April 1991, accepted 13 January 1992.