TY - JOUR
AU - Gilardi, James, D.
AB - Abstract Parrots are vocal learners and many species of parrots are capable of learning new calls, even as adults. This capability gives parrots the potential to develop communication systems that can vary dramatically over space. St. Lucia Parrot (Amazona versicolor) flight vocalizations were examined for geographic variation between four different sites on the island of St. Lucia. Spectrographic cross-correlation analysis of a commonly used flight vocalization, the p-chow call, demonstrated quantitative differences between sites. Additionally, the similarity of p-chows decreased as the distance between sites increased. Flight call repertoires also differed among sites; parrots at the Des Bottes and Quilesse sites each used one flight call unique to those sites, while parrots at the Barre de L'Isle site used a flight call that Quilesse parrots gave only while perched. It is unclear whether the vocal variation changed clinally with distance, or whether there were discrete dialect boundaries as in a congener, the Yellow-naped Parrot (Amazona auropalliata, Wright 1996). The geographical scale over which the St. Lucia Parrot's vocal variation occurred was dramatically smaller than that of the Yellow-naped Parrot. Similar patterns of fine-scale vocal variation may be more widespread among other parrot species in the Caribbean than previously documented. Variación Geográfica de las Vocalizaciones de Vuelo de Amazona versicolor Resumen. Los loros aprenden vocalizaciones y los individuos de muchas especies son capaces de aprender nuevas llamadas incluso cuando son adultos. Esta habilidad otorga a los loros el potencial de desarrollar sistemas de comunicación que pueden variar notablemente en el espacio. Se estudiaron las vocalizaciones de vuelo de Amazona versicolor para determinar la ocurrencia de variaciones geográficas entre cuatro lugares diferentes de la isla de Santa Lucía. Un análisis espectrográfico de correlación cruzada de la llamada común, conocida como “p-chow”, demostró diferencias cuantitativas entre los distintos lugares. Además, la similitud entre estas llamadas disminuyó con incrementos en la distancia entre los lugares. Los repertorios de llamadas de vuelo también difirieron entre lugares: los loros de las localidades de Des Bottes y Quilesse emplearon una llamada de vuelo exclusiva para cada uno de esos sitios, mientras que los de Barre De L'Isle usaron como llamada de vuelo una vocalización que los de Quilesse sólo emitieron cuando estaban posados. No queda claro si las vocalizaciones cambian clinalmente con la distancia, o si existen distintos tipos discretos de vocalizaciones, como en su congénere A. auropalliata (Wright 1996). La escala geográfica a la que se da la variación vocal en A. versicolor es bastante menor que aquella a la cual se presenta en A. auropalliata. Patrones similares de variación vocal a pequeña escala pueden ser más comunes en otras especies de loros del Caribe de lo que se ha documentado hasta ahora. Introduction Geographic variation of vocalizations has been widely documented in oscine songbirds (Order Passeriformes, Mundinger 1982), but less so in two groups of non-oscine avian vocal learners: parrots (Order Psittaciformes) and hummingbirds (Order Apodiformes, Baptista and Schuchmann 1990). Among parrots, there is evidence for geographic variation of vocalizations in White-tailed Black Cockatoos (Calyptorhynchus funereus, Saunders 1983), Orange-winged Parrots (Amazona amazonica, Nottebohm 1970), and Ringneck Parrots (Barnardius zonarius, Baker 2000). Extensive studies in Costa Rica found that the contact calls of Yellow-naped Parrots (Amazona auropalliata) could be classified into three geographically distinct dialects (Wright 1996, 1997), while the contact calls of the Orange-fronted Parakeets (Aratinga canicularis) from the same region varied clinally, but lacked dialects with discrete boundaries (Bradbury et al. 2001). Although the concept of avian vocal dialects has existed for over 50 years, no single definition is accepted by all researchers (Mundinger 1982). An early definition was based on shared vocal features or patterns that represented a local population (Marler 1952, Marler and Tamura 1962). Recent definitions emphasize distinct boundaries that separate two or more vocal traditions within a contiguous population (Marler and Tamura 1964, Baker 1975, Baptista 1975). The parrots of the Caribbean are thought to have originated from Central and South America, with the four extant species of Amazona that inhabit the Lesser Antilles most likely coming from South America (Forshaw 1989). The Amazona species of the Caribbean, especially those in the Lesser Antilles, have small, geographically restricted populations compared to many of their mainland counterparts. Parrot extinctions have occurred throughout the Caribbean; currently the region supports only 12 of a minimum of 28 species that existed in 1490 (Wiley 1991). In the mid-1970s it was estimated that there were only 100 St. Lucia Parrots (Amazona versicolor) left in the wild (Jeggo 1975). The population has grown in the intervening years largely because forest reserves have been created to preserve habitat, hunting of all wildlife has been outlawed, and an effective environmental education program promoted the plight of the parrot. As part of a larger study on the communication system of the St. Lucia Parrot, we investigated geographical variation of their flight vocalizations. Breeding St. Lucia Parrots from the Quilesse Forest of St. Lucia have a vocal repertoire of 26 call types, five of which are used during flight (Kleeman 1999). In this paper we test whether St. Lucia Parrot flight vocalizations vary between the Quilesse Forest and three other nearby sites on St. Lucia by quantitatively analyzing a commonly used flight vocalization, the p-chow call, and by comparing the repertoire of flight calls used at each site. Methods St. Lucia is the second largest Windward Island in the Lesser Antilles. It lies 29 km south of Martinique and 50 km north of St. Vincent (Fig. 1). The parrot's range is limited to the remaining central primary and secondary rainforest on the 622 km2 island (Forshaw 1989). The interior of the island consists of a disordered topography of steep-sided canyons and ridges capped with residual volcanic plugs. Figure 1. Open in new tabDownload slide The West Indies, showing St. Lucia (inset), and four sites on St. Lucia where parrot vocalizations were recorded in 1996 and 1997. The extent of rain forest was based on Beard's (1949) approximation. Both Des Bottes and Saltibus were in secondary rainforest Figure 1. Open in new tabDownload slide The West Indies, showing St. Lucia (inset), and four sites on St. Lucia where parrot vocalizations were recorded in 1996 and 1997. The extent of rain forest was based on Beard's (1949) approximation. Both Des Bottes and Saltibus were in secondary rainforest We recorded flight vocalizations of parrots at four locations on the island during March and April 1997: the Quilesse Forest (5.5 hr observation, 64 min of recordings), Saltibus (8 hr obs., 6 min rec.), Des Bottes (14 hr obs., 60 min rec.), and Barre de L'Isle (8.5 hr obs., 24 min rec., Fig. 1). We collected vocalizations opportunistically as we encountered parrots. Although we were not able to identify individual parrots, we tried to ensure that individual parrots were not recorded more than once per day by noting the origin and destination of their flights. Extensive recordings of breeding pairs of parrots in the Quilesse area during 1996 and 1997 (Kleeman 1999) demonstrated that brief recordings of flight vocalizations made for this study were representative of flight vocalizations used by parrots in that area. We recorded all vocalizations on DAT tapes (Ampex 467 R-124) with a DAT recorder (Sony, TCD-D7), coupled with a Sennheiser (Old Lyme, Connecticut) ME67 shotgun microphone and LC-6 powering module. We prefiltered recordings with a graphic equalizer to eliminate frequencies above 8.0 kHz. This ensured that no spurious frequencies were introduced when we digitized the calls into an Apple iMac as SoundEdit 16 files (V. 2.0.7, Macromedia, San Francisco, CA) at 16 bits with a sampling rate of 22.2 kHz. We converted these files into Audio Interchange Format Files and imported them into Canary software (V. 1.2.1; Charif et al. 1995) to create spectrograms of the calls. Spectrograms were analyzed with the following parameters: filter bandwidth of 352.9 Hz, grid resolution of 1.4 msec with 87.5% overlap, 43.5 Hz, FFT size of 512 points, Hamming window, logarithmic amplitude, and clipping level of −75.0 dB (Charif et al. 1995). Spectrograms were further filtered in Canary to remove noise below 0.5 kHz and above 6.0 kHz. For spectrogram descriptions, we defined a note as a continuous sound separated by a pause of at least 15 msec from other sounds, whereas an element referred to a portion of a note. Statistical Analyses To test for geographic variation of p-chow calls among the four sites, we chose 4–12 (9.5 ± 4.0, n = 38) of the highest quality flight p-chows from each region. We compared spectrograms using the batch correlation process (with normalized cross correlations) in Canary which calculated a correlation coefficient between all pairs of spectrograms by sliding them past each other in time and measuring the amount of temporal and frequency overlap between spectrograms (Charif et al. 1995). Results were displayed in a symmetrical n × n matrix in which each cell contained a correlation value between zero and one (zero represented the least similarity, one represented the greatest). Only the triangular lower left-hand corner of this matrix was used in the following analysis since the upper right hand corner merely duplicated this information. The main diagonal of the matrix represented autocorrelations with values of 1, and was also disregarded. Parrots represented by more than one call had their correlation values averaged against all other birds to yield an individual by individual matrix with an array size of 13. We tested whether p-chow calls given at any particular site (intrasite comparisons) were more similar than p-chows from different sites (intersite comparisons) by creating a second matrix of the same size and axes that was coded for group membership (Wright 1996, Gaunt et al. 1994). Dyads in this matrix that represented intrasite comparisons were coded with 1 n−1i where n represented the number of parrots from each site (Luo and Fox 1996). Dyads that represented intersite comparisons were coded with 0's. The degree of correlation between these two matrices was tested with a Mantel test (Mantel 1967) in R statistical program (V. 4.0, Casgrain and Legendre 2001). The Mantel statistic calculated from these two matrices was then compared to a sampling distribution constructed by randomly reallocating the order of elements in the coded matrix 4999 times and recalculating a Mantel statistic for each iteration. The significance of the original test statistic was determined by comparing it to the sampling distribution of test statistics. For these and other statistical tests we used a significance level of P < 0.05 and values reported as means ± SD. Similarity distances between the p-chows from the 13 parrots were derived in S-plus (V. 4.0, 1998, Mathsoft Inc.) from the averaged correlation values between each possible pair of individuals and were graphically represented in two dimensions with multidimensional scaling. We tested whether call similarity was negatively correlated with geographical distance by comparing the call correlation matrix with a third matrix that was coded for distance between sites (Gaunt et al. 1994). We used the same methods as above to determine the correlation between matrices. Results St. Lucia Parrot flight p-chows varied among the four sites sampled on the island. Additionally, certain other call types differed structurally between sites, or were used in different contexts. The Mantel test showed a significant difference between intrasite and intersite correlation values of p-chows (r = 0.70, P < 0.001), and call similarity among parrots decreased as geographic distance between individuals increased (r = −0.60, P < 0.001). Multidimensional scaling of averaged correlation values between each possible pair of individuals showed that p-chows differed between sites with the possible exception of Saltibus and Quilesse (Fig. 2). The point that represents the Saltibus parrot lays quite close to the group of points that represent Quilesse parrots (Fig. 2), and a spectrogram of a p-chow call from the Saltibus parrot was similar to that from a Quilesse parrot (Fig. 3A, B). Of all sites, Saltibus and Quilesse were separated by the least geographical distance (3.9 km). Spectrograms of p-chows from Barre de L'Isle did not have the concave frequency sweeps in the main body of the call as seen in p-chows from the other sites (Fig. 3C). Instead, they were more angular and had a unique element that initiated the main body. P-chows from the Des Bottes parrots had complex first notes with rapid frequency-modulations (Fig. 3D), unlike p-chows from other sites. Figure 2. Open in new tabDownload slide Multidimensional scaling of averaged spectrographic cross-correlation values of St. Lucia Parrot p-chow calls recorded at four sites on St. Lucia in 1997: Quilesse (crosses), Saltibus (circle), Des Bottes (triangles), and Barre de L'Isle (squares). Each point represents one parrot, and points from each site are enclosed by polygons, except for Saltibus because it represents only one parrot. The degree of call similarity among parrots is shown by the proximity of points to each other Figure 2. Open in new tabDownload slide Multidimensional scaling of averaged spectrographic cross-correlation values of St. Lucia Parrot p-chow calls recorded at four sites on St. Lucia in 1997: Quilesse (crosses), Saltibus (circle), Des Bottes (triangles), and Barre de L'Isle (squares). Each point represents one parrot, and points from each site are enclosed by polygons, except for Saltibus because it represents only one parrot. The degree of call similarity among parrots is shown by the proximity of points to each other Figure 3. Open in new tabDownload slide Spectrograms of St. Lucia Parrot flight p-chows recorded from: (A) Saltibus, (B) Quilesse, (C) Barre de L'Isle, and (D) Des Bottes. All recordings collected in 1997 from St. Lucia Figure 3. Open in new tabDownload slide Spectrograms of St. Lucia Parrot flight p-chows recorded from: (A) Saltibus, (B) Quilesse, (C) Barre de L'Isle, and (D) Des Bottes. All recordings collected in 1997 from St. Lucia Vocalizations from parrots at Barre de L'Isle and Des Bottes had structural and contextual differences when compared to call types from the Quilesse repertoire. Parrots from Quilesse used five call types during flight: p-chow, rek, ayo, whistle, and bubble-up. At Barre de L'Isle, parrots used p-chow, ayo, and whistle in flight, as well as calls that resembled the Quilesse rek and squeak-rek. However, they did not use bubble-up. Both the rek and squeak-rek from Barre de L'Isle were strongly pulsed compared to their counterparts from Quilesse (Fig. 4). Quilesse parrots did not use squeak-rek in flight, and the Quilesse squeak-rek spectrogram (Fig. 4) was recorded from a perched bird. Figure 4. Open in new tabDownload slide Representative spectrograms and total number of flight calls recorded from St. Lucia Parrots at four locations on St. Lucia in 1997. The frequency of use of each call type is given below each spectrogram. Dashes indicate that we did not hear any calls, thus, we do not have spectrograms for these vocalizations. Parrots at Saltibus may have had more flight calls than shown since few parrots were recorded at this location Figure 4. Open in new tabDownload slide Representative spectrograms and total number of flight calls recorded from St. Lucia Parrots at four locations on St. Lucia in 1997. The frequency of use of each call type is given below each spectrogram. Dashes indicate that we did not hear any calls, thus, we do not have spectrograms for these vocalizations. Parrots at Saltibus may have had more flight calls than shown since few parrots were recorded at this location Parrots at Des Bottes used both p-chow and ayo, but did not use rek, whistle or bubble-up. However, they did use one call type, gleep, during flight that was not heard elsewhere. Additionally, two vocalizations recorded from perched parrots at Des Bottes differed structurally from those heard at Quilesse (Kleeman 1999). Parrots recorded at Saltibus used p-chow, rek, and ayo flight calls that were quite similar to those call types heard at Quilesse, however, we encountered less than a dozen parrots at Saltibus and made few recordings there. Flight calls from the four sites are summarized in Figure 4. Discussion Quantitative analysis of flight p-chows from four sites in the parrot's range showed geographical variation in this call type. This conclusion was based on recordings of a limited number of parrots at each site, but these calls were representative of other parrots that we heard at each of the sites. For example, in 9 months of study between 1995 and 1997 in Quilesse we heard parrots use the same five call types in daily flight as our 64 min of recordings used in this study. On only three occasions in nine months of field work did we hear parrots use flight calls that differed from those described above (Kleeman 1999). The parrots we heard at Saltibus sounded quite similar to those from Quilesse and it is possible that a greater sampling of p-chows from Saltibus parrots would have revealed considerable overlap between the two sites. On the other hand, infrequently used call types may have been missed because of limited encounters with Saltibus parrots. There was a greater amount of vocal activity at Quilesse, Des Bottes, and Barre de L'Isle than at Saltibus thus, we are more confident in the completeness of our description of flight call repertoires for these sites. Consequently, the following discussion focuses on flight calls of Quilesse, Des Bottes, and Barre de L'Isle parrots. The pattern of variation in flight calls at Quilesse, Des Bottes, and Barre de L'Isle occurred on two levels. The first level was evidenced by structural variation in a shared call type (p-chow) among sites. Ayos were also shared among all sites, and qualitative inspection of spectrograms suggested that they also varied among sites, although we did not test this quantitatively. The second level of variation between sites concerned which call types were used during flight. Parrots from each of the three sites differed lexically by using a call type that parrots at the other sites did not use in flight: bubble-up at Quilesse, gleep at Des Bottes, and squeak-rek at Barre de L'Isle. Not only did the structure of squeak-rek change between Barre de L'Isle and Quilesse, but the context in which the call was used changed as well; Quilesse parrots only gave the call when perched, while Barre de L'Isle parrots used squeak-rek in flight. It is unclear whether the geographical variation found in St. Lucia Parrots has discrete dialect boundaries or changes clinally with distance, as in House Finches (Carpodacus mexicanus, Bitterbaum and Baptista 1979) and Orange-fronted Parakeets (Bradbury et al. 2001). The contact calls of Orange-fronted Parakeets were homogenous over distances of 7–10 km, but call duration and structure varied at greater distances. Radio-telemetered parakeets had range lengths (the maximum distance between any two points in the parakeets' home range) of 2–9 km, closely matching the distance of vocal homogeneity, and suggesting that diffusion and drift might play an important role in their patterns of vocal variation (Bradbury et al. 2001). Although they are sympatric with the parakeets, Yellow-naped Parrots had dialects with distinct boundaries at which four of the five call classes examined went through dramatic changes in acoustic structure (Wright 1996, 1997). Dialects of the Yellow-naped Parrot span large distances, 110 km for one dialect, and 40 km for another (extrapolated from Wright 1996). This contrasts greatly to the St. Lucia Parrot's situation where the vocal variation we observed occurred within distances of 6–9 km, and the species' entire range is roughly 100 km2 (Butler 1980). The variation in flight p-chows and lexical differences between Quilesse and nearby Des Bottes (5.5 km NW of Quilesse) suggest that a dialect boundary might have existed somewhere between the two sites. Lexical differences in flight vocalizations during courtship were used to distinguish dialects in Brown-headed Cowbirds (Molothrus ater, Warren 2002) and Bronzed Cowbirds (Molothrus aeneus, Rothstein and Fleischer 1987). Although there did not appear to be any physical barriers on the island that would isolate parrots from any part of their range, Quilesse and Des Bottes were separated by the island's highest peak, Mt. Gimie (951 m). Mt. Gimie might have been enough of an obstacle to limit movements between the drainages on either side of it. Our most recent observations indicated that the parrots were not rapidly expanding into the northern rainforests of the island where they were historically found, even though apparently suitable habitat still exists (JDG, pers. obs.). This lack of long distance movements is consistent with the fine-scale vocal variation we observed, and together they indicate a pattern of a sedentary parrot. Many parrot species are able to learn vocalizations as adults in captivity (Forshaw 1989), but little is known about the ontogeny of vocal learning in the St. Lucia Parrot or any parrot. Equally little is known about this parrot's dispersal patterns. Year-old birds do appear at nest sites with their parents in some years (N. F. R. Snyder, pers. comm.), indicating that substantial early learning is done very close to home. A fuller understanding of the St. Lucia Parrot's vocal variation would benefit from additional sampling throughout their range, a telemetry study examining the movements of adults and dispersing young, and a study of the genetic structure of the population. Observations from other researchers suggest that geographical variation of vocalizations from parrots inhabiting Caribbean islands might be more common than previously thought. Currently in the literature, only the Orange-winged Parrot of Trinidad (Nottebohm 1970) is reported to have geographic variation in vocalizations, but St. Vincent Parrots (Amazona guildingii) may also use a different contact call on the leeward versus the windward side of the island of St. Vincent (JDG, pers. obs.), and flight vocalizations of the Hispaniolan Parrot (Amazona ventralis) apparently vary geographically in the Dominican Republic (J. W. Wiley, pers. comm.). These observations combined with our findings support the idea that geographical variation of vocalizations is widespread among Amazona parrots and is likely to be an important aspect of their social biology. Acknowledgments Wildlife Preservation Trust International provided recording equipment and financial assistance. We thank N. Snyder and J. Dawson for their guidance while working with the St. Lucia Parrot Project. The St. Lucia Forest and Lands Department provided housing and logistical support. L. Baptista provided use of sound analysis equipment at the California Academy of Sciences (San Francisco, CA). G. Cruz kindly translated the abstract into Spanish, and G. Fellers assisted with figures. PMK thanks his graduate committee: D. Kitchen, L. George, J. Waters, and L. Baptista for their support, and T. Wright, J. Wiley, L. Kleeman, and two anonymous reviewers for their valuable comments on earlier versions of this work. Literature Cited Baker , M. C. 1975 . Song dialects and genetic differences in White-crowned Sparrows (Zonotrichia leucophrys). Evolution 29 : 226 – 241 . Google Scholar Crossref Search ADS PubMed WorldCat Baker , M. C. 2000 . Cultural diversification in the flight call of the Ringneck Parrot in Western Australia. Condor 102 : 905 – 910 . Google Scholar Crossref Search ADS WorldCat Baptista , L. F. 1975 . Song dialects and demes in sedentary populations of the White-crowned Sparrow (Zonotrichia leucophrys nuttalli). University of California Publications in Zoology 105 : 1 – 52 . WorldCat Baptista , L. F. , and K. L. Schuchmann . 1990 . Song learning in the Anna Hummingbird (Calypte anna). Ethology 84 : 15 – 26 . Google Scholar Crossref Search ADS WorldCat Bitterbaum , E. , and L. F. Baptista . 1979 . Geographical variation in songs of California House Finches (Carpodacus mexicanus). Auk 96 : 462 – 474 . WorldCat Bradbury , J. W. , K. A. Cortopassi , and J. R. Clemmons . 2001 . Geographical variation in the contact calls of Orange-fronted Parakeets. Auk 118 : 958 – 972 . Google Scholar Crossref Search ADS WorldCat Butler , P. J. 1980 . The St. Lucia Amazon (Amazona versicolor): its changing status and conservation, p. 171–181. In R. F. Pasquier [ed.], Conservation of New World parrots. Smithsonian Institution Press, Washington, DC . WorldCat Casgrain , P. , and P. Legendre . 2001 . The R package for multivariate and spatial analysis, version 4.0. Departement de sciences biologiques, Universite de Montreal, Montreal, Quebec, Canada. <http:// www.fas.umontreal.ca/biol/casgrain/en/labo/R/v4/ index.html> (16 September 2004) . WorldCat Charif , R. A. , S. Mitchell , and C. W. Clark . 1995 . Canary 1.2 user's manual. Cornell Laboratory of Ornithology, Ithaca, NY . WorldCat Forshaw , J. M. 1989 . Parrots of the world. 3rd ed. Landsdowne Editions, Willoughby, Australia . WorldCat Gaunt , S. L L. , L. F. Baptista , J. E. Sanchez , and D. Hernandez . 1994 . Song learning as evidenced from song sharing in two hummingbird species (Colibri coruscans and C. thalassinus). Auk 111 : 87 – 103 . Google Scholar Crossref Search ADS WorldCat Jeggo , D. F. 1975 . A report on the field study of the St. Lucia Parrot Amazona versicolor during 1975. Jersey Wildlife Preservation Trust Annual Report 12 : 34 – 41 . WorldCat Kleeman , P. M. 1999 . Vocalizations of the St. Lucia Parrot. M.Sc. thesis, Humboldt State University, Arcata, CA . WorldCat Luo , J. , and B. J. Fox . 1996 . A review of the Mantel test in dietary studies: effect of sample size and inequality of sample sizes. Wildlife Research 23 : 267 – 288 . Google Scholar Crossref Search ADS WorldCat Mantel , N. 1967 . The detection of disease clustering and a generalized regression approach. Cancer Research 27 : 209 – 220 . Google Scholar PubMed WorldCat Marler , P. 1952 . Variations in the song of the Chaffinch (Fringilla coelebs). Ibis 94 : 458 – 472 . Google Scholar Crossref Search ADS WorldCat Marler , P. , and M. Tamura . 1962 . Song ‘dialects’ in three populations of White-crowned Sparrows. Condor 64 : 368 – 377 . Google Scholar Crossref Search ADS WorldCat Marler , P. , and M. Tamura . 1964 . Culturally transmitted patterns of vocal behavior in sparrows. Science 146 : 1483 – 1486 . Google Scholar Crossref Search ADS PubMed WorldCat Mathsoft, Inc. 1988 . S Plus user's guide. Version 4.0, Mathsoft, Inc., Seattle, WA . WorldCat Mundinger , P. C. 1982 . Microgeographic and macrogeographic variation in the acquired vocalizations of birds, p. 147–208. In D. E. Kroodsma, E. H. Miller, and H. Oeullet [eds.], Acoustic communication in birds. Vol. 2. Academic Press, New York . WorldCat Nottebohm , F. 1970 . The ontogeny of bird song. Science 167 : 950 – 956 . Google Scholar Crossref Search ADS PubMed WorldCat Rothstein , S. I. , and R. C. Fleischer . 1987 . Vocal dialects and their possible relation to honest status signaling in the Brown-headed Cowbird. Condor 89 : 1 – 23 . Google Scholar Crossref Search ADS WorldCat Saunders , D. A. 1983 . Vocal repertoire and individual vocal recognition in the Short-billed White-tailed Black Cockatoo, Calyptorhynchus funereus latirostris. Australian Wildlife Research 10 : 527 – 536 . Google Scholar Crossref Search ADS WorldCat Warren , P. S. 2002 . Geographic variation and dialects in songs of the Bronzed Cowbird (Molothrus aeneus). Auk 119 : 349 – 361 . Google Scholar Crossref Search ADS WorldCat Wiley , J. W. 1991 . Status and conservation of parrots and parakeets in the Greater Antilles, Bahama Islands, and Cayman Islands. Bird Conservation International 1 : 187 – 214 . Google Scholar Crossref Search ADS WorldCat Wright , T. F. 1996 . Regional dialects in the contact call of a parrot. Proceedings of the Royal Society of London Series B 263 : 867 – 872 . Google Scholar Crossref Search ADS WorldCat Wright , T. F. 1997 . Vocal communication in the Yellow-naped Amazon (Amazona auropalliata). Ph.D. dissertation, Universtiy of California, San Diego, CA . WorldCat Author notes Address for correspondence: Western Ecological Research Center, USGS, Point Reyes National Seashore, Point Reyes, CA 94956. pkleeman@usgs.gov Present address: World Parrot Trust, 725 Peach Pl., Davis, CA 95616. gilardi@worldparrottrust.org © The Cooper Ornithological Society 2005
TI - Geographical Variation of St. Lucia Parrot Flight Vocalizations
JF - Condor: Ornithological Applications
DO - 10.1093/condor/107.1.62
DA - 2005-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/geographical-variation-of-st-lucia-parrot-flight-vocalizations-VMyUHvw20x
SP - 62
VL - 107
IS - 1
DP - DeepDyve
ER -