--The White-tailed Kite (Elanus leucurus) of the Americas has been merged with the Black-shouldered (or Black-winged) Kite (E. caeruleus) of the Old World and the Australian Black-shouldered Kite (E. axillaris) by North American authorities (but not elsewhere), primarily because of similarity in plumage. However, American kites differ from Old World kites in greater size and weight, in proportions (relatively longer tail and smaller bill and feet), plumage pattern (particularly of juveniles), and in behavior. Here we argue that these characters are sufficiently distinctive to warrant recognition of E. leucurus at the species level. Received 19 May 1992, accepted 1 Sept. 1992.

The White-tailed Kite (Elanus leucurus) of the Americas was long con- sidered a species distinct from the Black-shouldered (or Black-winged) Kite (E. caeruleus) of the Old World, and the Australian Black-shouldered Kite (E. axillaris). We follow Schodde and Mason (1980) and McAllan and Bruce (1989) in the use of axillaris rather than notatus as the name for the Australian species. However, Parkes (1958) and Husain (1959) suggested that both leucurus and axillaris should be considered conspecific with caeruleus, and that the complex forms a superspecies with the Letter- winged Kite, E. scriptus, of Australia. Arguments for this merger were based primarily on the similarity of plumage. Parkes (1958) stated that, for any plumage character, forms of the combined species E. caeruleus could be arranged in a progressively intergrading series. Husain (1959) postulated a biogeographical scheme to account for the distribution of the forms, considering only the amount of black on the underwing. The American Ornithologists' Union (AOU 1983), Palmer (1988), and American field guides based on AOU (1983) follow Parkes (1958) and Husain (1959) and treat the American forms under the combined name 4554 Shetland Green Rd., Alexandria, Virginia 22312. 2 U.S. Fish and Wildlife Service, National Museum of Natural History, Washington, D.C. 20560. 571 572 THE WILSON BULLETIN ß VoL 104, No. 4, December 1992 caeruleus. On the other hand, all authorities outside of and many from North America have continued to recognize E. leucurus and E. axillaris as species distinct from, but closely related to, E. caeruleus (Vaurie 1965, Brown and Amadon 1968, Mayr and Short 1970, Glutz von Blotzheim 1971, Condon 1975, Stresemann and Amadon 1979, Cramp and Sim- mons 1980, Brown et al. 1982, Amadon and Bull 1988, Short et al. 1990, and Sibley and Monroe 1990). The three taxa are often treated as allospe- cies. Here we discuss differences in size, proportions, plumage, and behavior that lead us to the conclusion that Elanus leucurus of the Americas should be treated as a species distinct from caeruleus and its races. We believe that axillaris should similarly be considered distinct, although our study was directed less toward that form and we have no experience with it in the field. Mees (1982) suggested that several other recognized subspecies of caeruleus should be combined with E. c. hypoleucus and recognized at the same level as caeruleus, leucurus, and axillaris. METHODS The senior author has observed both E. leucurus and E. caeruleus in the field for many hours, leucurus in California, Texas, Mexico, Guatemala, and Panama, and caeruleus in Senegal, Kenya, Spain, and India. His notes on behavior and appearance are supplemented by information in the literature. Plumage characters were examined on and measurements of wing chord, tail length, culmen from cere, and hallux length were taken from specimens in the American Museum of Natural History, the British Museum (Natural History), and the U.S. National Museum of Natural History. Additional sets of measurement data were obtained from the literature, as was information on weight. RESULTS White-tailed Kites (E. leucurus) differ from Black-shouldered Kites (E. caeruleus) in size and especially in proportions, in characteristics of the plumage, especially in young birds, and in hunting behavior and flight. Australian Black-shouldered Kites (E. axillaris) are similar to caeruleus in proportions but resemble one or the other of the Old and New World forms in plumage characters and behavior. Size and proportions.--White-tailed Kites are considerably larger than Black-shouldered Kites (Table 1; Fig. 1), particularly in length of wing and tail. In all forms of the genus other than E. scriptus the sexes are similar in size and samples have been combined by many authors; females ofscriptus are larger than males (Table 1; Brown and Amadon 1968). We attribute differences in means in samples of leucurus to differences in method of measurement, as several of the samples may have included the same individual specimens (Palmer 1988:135). As noted by Bangs Clark and Banks ß TAXONOMIC STATUS OF WHITE-TAILED KITE 573 TABLE 1 MEASUREMENTS (MM) FROM POPULATIONS IN THE GENUS ELANUS Wing Tail Culmen N sex. Range œ N Range R N Range R Source ' E. l. majusculus, North America 10 300-324 309 10 170-181 177 9 18.0-19.1 18.7 1 25 300-325 316 25 170-188 179 25 16.5-21.0 19.2 2 14 302-328 314 14 174-186 181 14 18-19 18.7 3 20 m 316 19 19.1 4 15 f 318 15 19.5 4 9 m 309 9 184 9 18.5 5 7 f 307 7 183 7 19.2 5 E. l. leucurus, South America 10 288-304 295 10 157-169 162 10 16.6-17.9 17.4 1 14 290-310 301 14 149-177 162 14 16.5-19.5 17.7 2 E. c. caeruleus, Europe and Africa 13 m 249-292 271 18 108-136 118 17 15.8-17.9 17.1 6 13 f 262-297 273 13 108-127 118 15 15.9-18.2 17.2 6 E. c. caeruleus, Africa 7 248-272 260 7 116-127 120 7 15.2-17.0 16.2 1 E. c. vociferus, Southern Asia 9 248-278 264 9 113-130 123 9 15.2-17.9 16.7 1 E. c. sumatranus, Sumatra 11 m 290-303 294 12 132-146 137 12 17.8-19.0 7 13 f 285-305 294 13 130-146 138 13 18.5-20.3 7 E. c. hypoleucus, Java 16 m 292-310 298 18 132-150 140 18 18.8-19.8 7 16 f 294-310 302 18 131-151 142 18 19.0-20.8 7 E. c. wahgiensis, New Guinea -- 297 8 E. axillaris, Australia 6 291-298 294 6 136-142 139 1 -- m 280-302 294 -- 142-153 8 -- f 280-310 298 -- 142-154 8 E. scriptus, Australia -- m 292-296 -- 146-150 8 -- f 302-313 -- 156-162 8 "Sexes combined unless indicated (m for male, f for female). b Sources: 1, this study; 2, Bangs and Pennard ( 1920); 3, Friedmann ( 1950); 4, Snyder and Wiley ( 1976); 5, Hawbecker (1942); 6, Cramp and Simmons (1980); 7, Mees (1982); 8, Brown and Amadon (1968). 574 THE WILSON BULLETIN ß Vol. 104, No. 4, December 1992 FI. 1. Two specimens ofE. caeruleus (left) and two ofE. leucurus (righO showing dorsal (above) and ventral (below) surfaces. Note differences in size, relative lengths of tails, and back color. Clark and Banks ß TAXONOMIC STATUS OF WHITE-TAILED KITE 57 5 TABLE 2 WING/TAIL RATIOS OF SEVERAL POPULATIONS OF EZNWS (BASED ON D^T^ nq TI 1) Species Population Range of ratio Elanus leucurus North America 1.70-1.80 E. leucurus South America 1.75-1.86 E. caeruleus Africa 2.05-2.28 E. caeruleus Asia 2.05-2.28 E. "hypoleucus" Sumatra, Java 2.14 a E. aM#aris Australia 2.06-2.14 E. scriptus Australia 1.93-1.99 From data in Mees (1982). and Pennard (1920), North American White-tailed Kites, E. l. majusculus, are somewhat larger than South American birds, E. l. leucurus. The latter are nevertheless notably larger than Old World caeruleus and Australian axillaris. The populations constituting hypoleucus of Mees (1982) are close to the size of American birds in wing chord measurements but have much shorter tails (Table 1). The difference in tail length between leucurus and caeruleus is greater than the difference in wing length (Table 1). This difference in proportions may be best appreciated from calculation of the wing/tail ratio. In all American birds, this ratio is less than 2.0; in Old World birds and in those of Australia it is invariably more than 2.0 (Table 2). The ratio is intermediate in E. scriptus. This proportional difference is particularly noticeable on perched birds of both forms. On perched leucurus the wing tips do not extend to the tail tip, whereas on perched caeruleus they project far beyond the tail tip (Frontispiece). Lengths of the culmen (Table 1) and hallux (data not shown; sample means 15.7-16.5 ram) are similar throughout the complex and cannot be used to separate leucurus and caeruleus. However, as caeruleus is smaller, the bill and feet are proportionally larger; in particular, caeruleus appears larger headed than does leucurus (Frontispiece). Mendelsohn and Jaksic (1989) summarized weight data, and showed that American leucurus averages 50-65 g heavier than African caeruleus; Australian axillaris is intermediate, but is nearer to South American leu- CUrUS. Data summarized by Mendelsohn and Jaksic (1989) indicate that the length of the secondaries in Californian E. L majusculus and African E. caeruleus is similar (158 vs 146 mm, respectively), but as leucurus is larger 576 THE WILSON BULLETIN ß Vol. 104, No. 4, December 1992 and longer winged, the wing shapes of the two are different. This and the differences in relative tail length result in quite different shapes of flying White-tailed Kites and Black-shouldered Kites. According to Mendelsohn and Jaksic (1989), leucurus has both a higher aspect ratio and a much greater wing area than does caeruleus (1001 vs 843 cm2). Plumage. --E. leucurus and E. axillaris have a patch of black underwing coverts at the base of the primaries. This patch is lacking in E. caeruleus, although some individuals of that species have a few coverts with dusky tips (Husain 1959). The lower surface of the primaries is dark in leucurus, axillaris, and most caeruleus. The primaries are nearly white below, with a variable amount of gray on the wing tip, in E. c. hypoleucus and E. c. sumatranus (Salomonsen 1953), and apparently in E. c. wahgiensis (Mayr and Gilliard 1954). The outer five pairs of rectrices of young leucurus are white, with a variable amount of pale gray on the distal portion of the outer webs. On all but the outer pair the gray extends onto the inner web forming a continuous subterminal tail band. The central rectrices are entirely gray except for a narrow white tip. The gray on the rectrices is darker and seldom, if ever, forms a band across the tip of the tail in young caeruleus, usually being restricted to the outer webs. In E. c. hypoleucus and in E. axillaris, the gray on the tail is much reduced and except for the central rectrices the tail is essentially white. Behavior. -- White-tailed Kites hunt almost exclusively by hovering; Jaksic et al. (1987) reported seeing only four instances of hunting from perches during observations of 80 hover-hunts in Chile. Australian ax- illaris, like leucurus, hunts mostly by hovering (R. Schodde, pers. comm.). After reviewing the literature, Mendelsohn and Jaksic (1989) could report no observations of perch-hunting from North America. On the other hand, Old World Black-shouldered Kites hunt regularly from perches (Men- delsohn and Jaksic 1989), hunting from the hover only about 30% of the time (Tarboton 1978), although the proportional number of hunting events (as opposed to time spent) is not given. Clark and P. Bloom were able to capture perched Black-shouldered Kites in India easily using bal-chatri traps with mice. Bloom (pers. comm.) was able to capture only one White- tailed Kite in California using the same trap with mice out of more than 200 attempts. The one caught was undernourished and was captured outside of the breeding range of the kite. Differences in hunting methods may be related to differences in prey selection. The White-tail is a rodent specialist, whereas the Black-shoul- der, at least in some areas, takes a wider selection of prey including birds and lizards (Cramp and Simmons 1980, Palmer 1988:147). White-tails catch prey twice as quickly as do Black-shoulders from hovers (Mendel- sohn and Jaksic 1989). Clark and Banks ß TAXONOMIC STATUS OF WHITE-TAILED KITE 577 Black-shoulders often cock the tail up over their backs and then down (tail-bobbing of Palmer 1988:145); Tarboron (1978) reported tail-cocking during 14% of perch-hunting time, when intensely excited just before or after a strike. Although Steyn (1963) had previously linked tail-cocking to excitement, Mendelsohn and Jaksic (1989) suggested that it served to warn other birds away from the hunting area. Goriup (1981) reported tail-cocking during preening in Portugal. This behavior is seldom reported for White-tails (Brooks 1943). Despite many hours of observation, WSC has observed only one instance of tail-bobbing in American kites, by an adult female in California being harassed by her begging young. Black-shouldered Kites often perch with their folded wings drooped below the tail whereas White-tailed Kites usually perch with the wing tips folded over the tail (Frontispiece), but both can hold their wings either way. A perched Australian Black-shouldered Kite is illustrated with the wings folded over the tail (Macdonald 1973), the usual posture for that species (R. Schodde, pers. comm.). White-tailed Kites are somewhat communal and nest in fairly close proximity to one another and defend territories weakly. Watson (1940) reported some fighting by White-tails in the establishment of territories but noted that pairs often breed communally when prey are abundant. Peyton (1915) found two nests within 200 yards (184 m), and Pickwell (1930) reported three nests at points of a triangle measuring 320, 200, and 175 yards (294, 184, and 161 m). Hawbecker (1942:269) reported two pairs nesting within 400 yards (368 m) of one another, with only one aggressive encounter. Dixon et al. (1957) reported three nests within a radius of 500 feet (153 m), with two others in adjoining groves, and saw no territorial fighting. On the other hand, Black-shouldered Kites are strong in their territorial defense; Mendelsohn (1983) reported that South African birds defended their territories vigorously and persistently, where average territory size ranged from about 2.4 to about 4.5 km 2. However, Madden (1977) found two nests in South Africa that were approximately 200 m apart. DISCUSSION The inclusion of leucurus with caeruleus by the AOU was based mainly on Parkes (1958), and to a lesser extent on Husain (1959). Parkes (1958) made two points to support this merger. First, that one could arrange specimens of both forms progressively in an intergrading series and sec- ond, that these forms resemble each other more than do subspecies of some other raptors, specifically Accipiter gentilis, A. striatus, and Falco peregrinus. We have found that E. leucurus can always be separated from E. caerule- us by the presence of black carpal patches and by a wing/tail ratio less 578 THE WILSON BULLETIN ß Vol. 104, No. 4, December 1992 than 2.0; caeruleus lacks the black carpal patch and has a wing/tail ratio greater than 2.0. Clark has observed in the field on different continents two or more races of 12 species of diurnal raptors, including Accipiter gentills and Falco peregrinus. Although plumage can vary greatly between races of some of these species, none show the differences in behavior and overall propor- tions exhibited between leucurus and caeruleus as presented above. We believe that the characters of the American birds are sufficiently distinctive to warrant recognition of the White-tailed Kite E. leucurus at the species level. ACKNOWLEDGMENTS We thank the curatorial staffs of the American Museum of Natural History and the British Museum (Natural History) for allowing access to those collections. We thank M. R. Brow- ning, F. Gill, F. M. Jaksic, J. M. Mendelsohn, S. L. Olson, K. C. Parkes, and R. Schodde for comments on the developing manuscript. Clark's field work in India was sponsored by the Office of International Affairs of the U.S. Fish and Wildlife Service and the Bombay Natural History Society. LITERATURE CITED AMADON, D. AND J. BULL. 1988. Hawks and owls of the world: a distributional and taxonomic list. Proc. West. Found. Vert. Zool. 3:295-357. AMERICAN ORNITHOLOGISTS' UNION. 1983. Check-list of birds of North America. 6th ed. Amer. Ornith. Union, Washington, D.C. BANGS, O. AND T. E. PENNARD. 1920. Two new American hawks. Proc. New England Zool. Club 7:45-47. BROOKS, A. 1943. An unusual attitude in a hawk. Condor 45:119. BROWN, L. trD D. AM,DON. 1968. Eagles, hawks and falcons of the world. Vol. 1. McGraw Hill, New York, New York. BROWN, L. H., E. K. UAN, ta, rD K. NEWMAN. 1982. The birds of Africa. Vol. 1. Academic Press, London, England. CONDON, H.T. 1975. Checklist of the birds of Australia, Part 1. Non-passerines. Royal Australian Ornithologists Union, Melbourne, Australia. CRMP, S. teed K. E. L. SIMMONS (EDs.). 1980. Birds of the Western Palearctic. Vol. 2. Oxford Univ. Press, Oxford, England. DIXON, J. B., R. E. DIXON, AND J. E. DIXON. 1957. Natural history of the White-tailed Kite in San Diego County, California. Condor 59:156-165. FRIEDMANN, H. 1950. The birds of North and Middle America. Part 11. U.S. Natl. Mus. Bull. 50, Washington, D.C. GLUTZ VON BLOTZHEIM, U. N. (ED.). 1971. Handbuch der Vogel Mitteleuropas. Band 4. Academissche Verlag., Frankfurt, Germany. GoRIuP, P. D. 1981. Observations on a pair of Black-winged Kites (Elanus caeruleus) in eastern Portugal. Arq. Mus. Bocage, ser. B, 1:65-79. HAWBECKER, A.C. 1942. A life history study of the White-tailed Kite. Condor 44:267- 276. Husn, K.Z. 1959. Notes on the taxonomy and zoogeography of the genus Elanus. Condor 61:153-154. Clark and Banks ß TAXONOMIC STATUS OF WHITE-TAILED KITE 579 Jacsic, F. M., R. Rozzi, A. Lav, AND J. E. JIMENEZ. 1987. The hunting behavior of Black-shouldered Kites (Elanus caeruleus leucurus) in central Chile. Condor 89:907-911. MACDONALD, J.D. 1973. Birds of Australia. H. F. and G. Witherby, London, England. MADDEN, S.T. 1977. Notes on two nests of the Blackshouldered Kite. Ostrich 48:115- 116. MAYR, E. AND E. T. GILLIARD. 1954. Birds of central New Guinea. Bull. Amer. Mus. Nat. Hist. 103:315-374. -- L. L. SHORT. 1970. Species taxa of North American birds. Publ. Nuttall Orn. Club no. 9. McAn, I. A. W. M.D. BRUCE. 1989. The birds of New South Wales. A working list. Biocon Research Group, Sydney, Australia. MœEs, G.F. 1982. Birds from the lowlands of southern New Guinea (Merauke and Koembe). Zool. Verb. 191. MœrDELSOI-, J. M. 1983. Social behavior and dispersion of the Blackshouldered Kite. Ostrich 54:1-18. -- F. M. Jacslc. 1989. Hunting behaviour of Blackshouldered Kites in the Americas, Europe, Africa and Australia. Ostrich 60:1-12. PALMER, R. S. (ED.). 1988. Handbook of North American birds. Vol. 4. Yale Univ. Press, New Haven, Connecticut. Pvdcs, K.C. 1958. Specific relationships in the genus Elanus. Condor 60:139-140. PVTON, L. 1915. Nesting of the White-tailed Kite at Sespe, Ventura County, California. Condor 17:230-232. PiCKWt, G. 1930. The White-tailed Kite. Condor 32:221-239. SaOMONS, F. 1953. Miscellaneous notes on Philippine birds. Medd. fra Dansk Naturh. Foren. Bd. 115:205-281. SCHODDE, R. AND I. J. MASON. 1980. Nocturnal birds of Australia. Landsdowne Editions, Melbourne, Australia. SHORT, L. L., J. F. M. Hombre, C. MuvaGo-Gi. 1990. Annotated check-list of the birds of East Africa. Proc. West. Found. Vert. Zool. 4:61-245. SmLY, C. G. B. L. Mormon, JR. 1990. Distribution and taxonomy of birds of the world. Yale Univ. Press, New Haven, Connecticut. SNVœR, N. F. R. J. W. WiLdly. 1976. Sexual size dimorphism in hawks and owls of North America. Orn. Monogr. 20. SxY, P. 1963. The "wagtail" kite. Bokmakierie 15:9-10. STRESEMANN, E. AND D. AIADON. 1979. Order Falconiformes. Pp. 271-425 in Check-list of birds of the world, Vol. 1, 2nd ed. (E. Mayr and G. W. Cottrell, eds.) Mus. Comp. Zool., Cambridge, Massachusetts. TARBOTON, W. R. 1978. Hunting and the energy budget of the Black-shouldered Kite. Condor 80:88-91. V^URm, C. 1965. The birds of the Palearctic fauna. Non-Passeriformes. H. F. and G. Witherby, London, England. W^TSON, F. G. 1940. A behavior study of the White-tailed Kite. Condor 42:295-304.