Vol. 110 The Auk A Quarterly Journal of Ornithology No. 1 January 1993 The Auk 110:1-8 + frontispiece, 1993 RELIC OF A LOST WORLD: A NEW SPECIES OF SUNANGEL (TROCHILIDAE: HELIANGELUS) FROM "BOGOTA" GARY R. GRAVES Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, USA AISTRACT.--A new species of hummingbird, the BogotJ Sunangel (Heliangelus zusii), is described from a unique specimen purchased in 1909 in BogotJ, Colombia. Hellangelus zusii is intermediate in plumage between ungorgeted H. regalis and the typical gorgeted species in the genus. Now possibly extinct because of habitat destruction, H. zusii may have inhabited cloud forest and forest edge in the Eastern Cordillera of the Colombian Andes. Received 10 April 1992, accepted 6 July 1992. FUELED BY the demands of a fashion-conscious public, millions of hummingbirds destined for the millinery trade in Europe and the United States were exported from South America be- fore bird protection laws were passed in the early 20th century. The staggering size of these shipments can hardly be overemphasized. One London auction house alone sold 152,000 hum- mingbirds between 1904 and 1911 (Doughty 1975), a figure that undoubtedly exceeds the number of hummingbird specimens now housed in the world's museums. As expected, commerce in avian plumage had a profound effect on science. The golden age of hummingbird taxonomy and systematics was attained much earlier than in other Neotropical bird families (Gould 1861). More than a quarter of all hummingbird species were described dur- ing the 1840s and, by 1890, more than 91% had been discovered (Table 1). By comparison, only 73% of all antbird species were described by 1890. As the rate of discovery declined (1860- 1895), systematics scoured imports for rarities and described dozens of new trochilid taxa from specimens labeled "Bogota," Colombia, some of which were actually taken elsewhere. Many of these have now been determined to be hybrids, genetic plumage variants, artifacts, or subspe- cies of previously described species (e.g. Taylor 1909, Simon 1921, Berlioz and Jouanin 1944, Greenway 1978, Graves 1990). A tantalizing few (e.g. Isaacson's Puffieg, Eriocnemis i aacsonii [Par- zudaki] 1845) may represent valid species, pos- sibly surviving in some ornithologically unex- plored locality or perhaps extinct. In 1947, Brother Nicforo Maria sent a spec- tacular fork-tailed hummingbird purchased in Bogota in 1909 to Rudolphe Meyer de Schauen- see at the Academy of Natural Sciences of Phil- adelphia (ANSP) for identification. After re- ceiving the specimen, Meyer de Schauensee asked several of his colleagues to examine the specimen (literature among the archives of ANSP). Although there was little agreement among the correspondents, their ideas consti- tute some important hypotheses concerning the identity and origin of the specimen. James L. Peters (in litt., 10 April 1947) wrote, he uk A Quarterly Journal of Ornithology ol. No. 1 January 1993 The Auk 110:1-8 + frontispiece, 1993 LIC ST ORLD: W IES ANGEL ( CHILIDAE: IANGELUS) M " GOTA' ES Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, USA ABSTRACT.?A new species of hu ingbird, the Bogota Sunangel {Heliangelus zusii), is described fro a unique specimen purchased in 1909 in Bogota, Colo bia. Heliangelus zusii is intermediate in plumage between ungorgeted H. regalis and the typical gorgeted species in the genus. Now possibly extinct because of habitat destruction, H. zusii ay have inhabited cloud forest and forest edge in the Eastern Cordillera of the Colo bian Andes. Received 10 April 1992, accepted 6 July 1992. ands s ion-conscious public, illions of hu ingbirds destined for the illinery trade in Europe and the United States ere exported fro South A erica be- fore bird protection laws were passed in the early 20th century. The staggering size of these ship ents can hardly be overemphasized. One ndon tion l e l , - ingbirds between 1904 and 1911 (Doughty 1975), a figure that undoubtedly exceeds the nu ber of hu ingbird speci ens now used e orld's useums. As expected, commerce in avian plumage had a profound effect on science. The golden age of hu ingbird taxonomy and systematics was attained much earlier than in other Neotropical bird fa ilies (Gould 1861). ore than a quarter of all hu ingbird species were described dur- ing the 1840s and, by 1890, more than 91% had been discovered (Table 1). By comparison, only 73% of all antbird species were described by 1890. As the rate of discovery declined (1860- 1895), systematics scoured imports for rarities d ribed zens ch lid a speci ens labeled "Bogota," Colo bia, so e of which were actually taken elsewhere. any of these have now been deter ined to be hybrids, genetic plumage variants, artifacts, or subspe- cies of previously described species (e.g. Taylor 1909, i on 1921, erlioz a J a i 1944, reen ay 1978, Graves 1990). A tantalizing few (e.g. Isaacson's Puffleg, Eriocnemis isaacsonii [Par- zudaki] 1845) may represent valid species, pos- sibly surviving in some ornithologically unex- plored locality or perhaps extinct. In 1947, Brother Niceforo aria sent a spec- tacular fork-tailed hu ingbird purchased in Bogota in 1909 to Rudolphe eyer de Schauen- see at the Acade y of atural Sciences of Phil- adelphia (ANSP) for identification. fter re- ceiving the specimen, eyer de Schauensee asked several of his colleagues to exa ine the speci en (literature a ong the archives of ANSP). lthough there was little agree ent among the correspondents, their ideas consti- tute some i portant hypotheses concerning the identity and origin of the specimen. Ja es L. Peters (in lift., 10 pril 1947) rote, Frontispiece. Aerial pursuit of male Bogota Sunangels (Heliangelus zusii sp. nov.) above flowering Brachyo- turn mcrodon, a common melastome in the Eastern Cordillera of the Colombian Andes. Painting by Jon Fjelds/i. "*/ '. * t Frontispiece. Aerial pursuit of male Bogota Sunangels (Heliangelus zusii sp. nov.) above flowering Brachyo- tum microdon, a common melastome in the Eastern Cordillera of the Colombian Andes. Painting by Jon Fjeldsa. 2 GARY R. GRAVES [Auk, Vol. 110 TABLE 1. Number of hummingbird and antbird (Thamnophilidae and Formicariidae) species de- scribed as new to science by decade (taxonomy of Sibley and Monroe 1990). Species Decade Hummingbirds Antbirds 1750 13 0 1760 5 3 1770 0 1 1780 15 13 1790 1 0 1800 3 1 1810 15 9 1820 26 24 1830 47 14 1840 87 15 1850 33 44 1860 21 33 1870 14 11 1880 11 10 1890 9 8 1900 3 19 1910 4 7 1920 2 12 1930 1 5 1940 2 2 1950 2 3 1960 2 1 1970 2 1 1980 1 7 "Assuming your bird to be a hybrid, one of the parents is probably one of the Colombian pur- ple-tailed forms of Aglaiocercus; the shape of the bill, color and general shape of the tail and luminous plaques on forehead and throat could quite conceivably result from a union between Aglaiocercus and some form of Heliangelus, but in such a case I should expect the body plumage of the product to be green .... Now what  crossed with a C of predominately green col- oration (or vice versa) would produce a steel- blue offspring? Damned if I know. If Helianthea [Coeligena] prunellei is involved I should expect a longer billed result at least; Eriocnemis nigri- vestis can be dismissed as a geographic impos- sibility. Suppose Neolesbia to be distinct, should not the resultant product have at least a slightly decurved bill?... It's all very puzzling and I am afraid I haven't been any help." Meyer de Schauensee (in litt., 19 May 1947 to Alexander Wetmore) reported the observa- tions of John T. Zimmer of the American Mu- seum of Natural History, "Zimmer saw it and thought it could possibly be a hybrid between Aglaiocercus kingi and Heliangelus quamigularis." Finally, Alexander Wetmore (in litt., 5 May 1947) wrote, "I have never seen a bird like it .... I am inclined to doubt any hybrid origin for this bird, on the contrary I would suppose that it is a specimen of an unknown species .... If I were planning a description of this specimen I would compare it closely with Agelaiocercus [ ic] to determine whether it should be placed in that genus or named as a distinct genus." Meyer de Schauensee (1947:113) concluded, "Were it not for the fact that I have examined Metallura purpureicauda and Zodalia thaumasta nd found them to be virtually identical in color- ation and pattern, but differing in the shape of the bill, I would not hesitate to describe Brother Nicfiforo's bird as a new species and perhaps even a new genus. However, there are so many points of similarity between our specimen and Neolesbia that without seeing the type it seems the wisest course to identify it as a third ex- ample of Neolesbia. It should be noted that both the type and the American Museum specimen have longer bills (19.25, 17 mm) than ours (15.5)." Recently, Hinkelmann et al. (199 !) reaffirmed the identification of the ANSP specimen as an example of the problematic Nehrkorn's Sylph (Neolesbia nehrkorni), which they judged to be a hybrid between the Long-tailed Sylph (Aglaio- cercus kingi) and the Fork-tailed Woodnymph (Thaluraniafurcata). Here I present evidence that the hybrid origin of Brother Nicforo's speci- men can be rejected and that it does indeed represent an undescribed species. MATERIALS AND METHODS I compared the specimen (ANSP 159261) directly with all hummingbird taxa in the ANSP and National Museum of Natural History, Smithsonian Institution. Detailed notes and color transparencies of the spec- imen (Fig. 1) were compared with the trochilid col- lections of the Museum of Natural Sciences, Louisiana State University and the American Museum of Nat- ural History (AMNH), including a specimen with a molting tail, provisionally identified as Neolesbia nehr- korni (AMNH 484177), the type of which cannot now be found (Hinklemann et al. 1991). The specimen was compared with the description and color plate of Neo- lesbia nehrkorni (Berlepsch 1887), and color transpar- encies of a specimen in the Museum Heineanum Hal- berstadt, Germany (courtesy of B. Nicolai), that was recently identified as N. nehrkorni (Hinkelmann et al. 1991). All color comparisons were made under Ex- amolites (Macbeth Corporation). Descriptions of structural colors are unusually subjective and actual GARY R. GRAVES [ uk, ol. 110 TABLE 1. Number of hu ingbird and antbird (Tha nophilidae and For icariidae) species de- scribed as ne to science by decade (taxonomy of Sibley and onroe 1990). Species Decade Hummingbirds Antbirds 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 "Assu ing your bird to be a hybrid, one of the parents is probably one of the Colombian pur- ple-tailed forms of Aglaiocercus; the shape of the bill, color and general shape of the tail and lu inous plaques on forehead and throat could quite conceivably result from a union between Aglaiocercus and so e for of Heliangelus, but in such a case I should expect the body plumage of the product to be green. . . . Now what 6 crossed with a 9 of predo inately green col- oration (or vice versa) ould produce a steel- blue offspring? Damned if I know. If Helianthea [Coeligena] prunellei is involved I should expect a longer billed result at least; Eriocnemis nigri- vestis can be dis issed as a geographic i pos- sibility. Suppose Neolesbia to be distinct, should not the resultant product have at least a slightly decurved bill? . . . It's all very puzzling and I am afraid I haven't been any help." eyer de Schauensee (in litt., 19 ay 1947 to Alexander etmore) reported the observa- tions of John T. i er of the erican u- seu of atural istory, "Zi er saw it and thought it could possibly be a hybrid between Aglaiocercus kingi and Heliangelus squamigularis." Finally, Alexander etmore (in litt., 5 ay 1947) r te, "I have e er see a ir like it.... I am inclined to doubt any hybrid origin for this bird, on the contrary I would suppose that it is a speci en of an unkno n species. ... If I ere planning a description of this speci en I would compare it closely with Agelaiocercus [sic] to deter ine hether it should be placed in that genus or na ed as a distinct genus." eyer de Schauensee (1947:113) concluded, ere t r t t e ined etallura purpureicauda and Zodalia thaumasta and found the to be virtually identical in color- ation and pattern, but differing in the shape of t ill, I uld t it te t ri r ther iceforo's bird as a ne species and perhaps even a new genus. However, there are so many points of similarity between our specimen and Neolesbia that without seeing the type it seems the wisest course to identify it as a third ex- a ple of Neolesbia. It should be noted that both the type and the American useum specimen have longer bills (19.25, 17 mm) than ours (15.5)." Recently, Hinkel ann etal. (1991) reaffirmed the identification of the ANSP specimen as an example of the problematic Nehrkorn's Sylph (Neolesbia nehrkorni), which they judged to be a hybrid between the Long-tailed Sylph (Aglaio- cercus kingi) and the Fork-tailed oodnymph (Thalurania furcata). Here I present evidence that the hybrid origin of Brother iceforo's speci- en can be rejected and that it does indeed represent an undescribed species. ERIALS HODS I co pared the speci en (ANSP 159261) directly with all hu ingbird taxa in the ANSP and National useu of atural History, S ithsonian Institution. etailed notes and color transparencies of the spec- imen (Fig. 1) were compared with the trochilid col- l ti s f t use f atural i s, isia State University and the American useum of Nat- ural History (A NH), including a specimen with a molting tail, provisionally identified as Neolesbia nehr- korni (A NH 484177), the type of hich cannot no be found ( inkle ann et al. 1991). The speci en was co pared ith the description and color plate of Neo- lesbia nehrkorni (Berlepsch 1887), and color transpar- encies of a speci en in the useu eineanu Hal- berstadt, Ger any (courtesy of B. Nicolai), that was recently identified as N. nehrkorni ( inkel ann et al. 1991). All color co parisons ere ade under Ex- a olites ( acbeth Corporation). Descriptions of structural colors are unusually subjective and actual anuary 1993] New Species of Sunangel Fig. 1. Dorsal. ventral. and lateral view of holotype of Heliangelus zusii. January 1993] New Species of Sunangel >S Fig. 1. Dorsal, ventral, and lateral view of holotype of Heliangelus zusii. 4 GARY R. GRAVES [Auk, Vol. 110 color varies with angle of inspection. For this reason I use general color descriptions throughout. Measurements of wing chord, bill length from an- terior edge of nostril, and rectrix length from point of insertion of central rectrices to the tip of the longest rectrix of each pair (from innermost to outermost), were made with digital calipers to the nearest 0.1 mm. Measurements of the Heineanum Halberstadt speci- men of Neolesbia nehrkorni were provided by B. Ni- colai. Diagnostic assumptions and methods of hybrid diagnosis based on plumage characters and mor- phology follow Graves (1990). Heliangelus zusii sp. nov. Bogot Sunangel Holotype.--Academy of Natural Sciences of Philadelphia, No. 159261; adult ( ?); purchased in Bogota, Colombia in 1909 by Brother Nic- foro Maria. Diagnosis.--Heliangelus zusii is a dark bluish- black hummingbird with a deeply forked, dark purple tail, brilliant green frontlet and gorget, and straight bill (see frontispiece). It differs from H. regalis in having a frontlet and gorget, a pur- plish rather bluish-black tail, and pale rather than bluish-black undertail coverts. Other spe- cies of Heliangelus have green or bronzy-green body plumage and violet, purple, or orange gor- gets. Heliangelus zusii differs from Neolesbia nehr- korni in having a gorget and frontlet, purple rather than bluish-black tail, and a straight rath- er than decurved bill. Generic relationships.--The new species can be assigned to the genus Heliangelus by a combi- nation of characters (Figs. 1 and 2): (1) short (15.1 mm), straight, unmodified, uniformly dark bill; (2) brilliant frontlet and gorget; (3) un- modified remiges; (4) unmodified rectrices, flat in cross section; (5) lack of puffy tarsal plumes (present in Eriocnemis pp.); and (6) nonbrilliant undertail coverts (brilliant in most Eriocnemis spp.). Heliangelus zusii appears to bridge the gap between males of the ungorgeted, bluish-black H. regalis and the more typical members of the genus with contrasting gorgets and predomi- nately green body plumage. Description of holotype.--Taxidermy mount with glass eyes. Nasal feathers, lores, auriculars, and margins of crown are glittering steel-blue. A brilliant pale golden-green frontlet extends from the forecrown posteriorly to an imaginary line drawn between the centers of the orbits. Feathers of the frontlet are adpressed as in H. exortis. Nasal feathers extend 0.85 mm distal to the anterior edge of the nasal flanges. Traces of a white spot are found posterior to the eyes. The hindneck, back, and wing coverts are deep bluish-black, slightly iridescent, turning deep greenish-blue on the lower back and rump and returning to bluish-black on the upper tail co- verts. Primaries and secondaries are dark brown with a purplish tint. Secondaries have acumi- nate rather than broadly rounded tips. The deeply forked tail (fork 54% of length) is glit- tering dark purple above, duller and less iri- descent on the lower surface. Basal portions of the rectrices, obscured in the folded tail, are the same color as the exposed distal tips (unlike species of Aglaiocercus and Lesbia). Shafts of rec- trices are dark brown distally, fading to light brown basally. Rectrices are flat in cross section. The ventral plumage is duller than the dorsum. The chin is dull bluish-black, bordered poste- riorly by a brilliant golden-green gorget (same color as frontlet). Gorget feathers have golden- green tips separated from the gray base by a narrow violet band. Gorget and frontlet feath- ers are broadly rounded rather than narrow or pointed. The gorget is bordered below and on the sides by lustrous bluish-black. When viewed head-on in direct light, the head and breast ap- pear black and contrast greatly with the golden- green gorget and frontlet. Feathers of the upper breast are bluish-black with narrow buffy-gray margins, becoming progressively paler on the lower breast and belly. The midline of the belly is medium gray but faint traces of bluish-black can be observed on barbules under magnifica- tion. Sides and flanks are dark bluish-black (same as back). There are no traces of a pectoral band or subterminal white spots on breast feathers (present in some Heliangelus hybrids; Graves 1990, Graves and Zusi 1990). Undertail coverts are pale creamy-white with a single lanceolate, subterminal spot at the midline (brownish-gray with blue reflections). Tibial feathers are dull grayish-brown. The tarsi are now bare (former- ly feathered?). Soft-part colors in dried speci- men: bill black; scutes on feet brown; foot pads yellowish-brown. Measurements (mm).--See Table 2 for standard measurements. Greatest width of outermost rec- trices, 9.2. Gorget length x width, ca. 14 x 10. Sex, age, and molt.--Although the sex of the holotype is not known with certainty, all plum- age characters resemble components of male plumage in Heliangelus species. The fully de- veloped gorget and frontlet, unstriated maxil- lary ramphothecum, and lack of molt in the holotype indicate that it is an adult. RY . RAVES [ uk, ol. 110 color varies with angle of inspection. For this reason I use general color descriptions throughout. easurements of wing chord, bill length from an- terior edge of nostril, and rectrix length from point of insertion of central rectrices to the tip of the longest rectrix of each pair (from innermost to outermost), were made with digital calipers to the nearest 0.1 mm. easurements of the Heineanum Halberstadt speci- en of Neolesbia nehrkorni ere provided by B. Ni- colai. Diagnostic assumptions and methods of hybrid diagnosis based on plu age characters and or- phology follow Graves (1990). eliangelus zusii sp. nov. Bogota Sunangel Holotype.?Acade y of atural Sciences of Philadelphia, No. 159261; adult (6 ?); purchased in Bogota, Colo bia in 1909 by Brother Nice- ro aria. Diagnosis.?Heliangelus zusii is a dark bluish- black hu ingbird ith a deeply forked, dark purple tail, brilliant green frontlet and gorget, and straight bill (see frontispiece). It differs fro H. regalis in having a frontlet and gorget, a pur- plish rather bluish-black tail, and pale rather than bluish-black undertail coverts. ther spe- cies of Heliangelus have green or bronzy-green body plu age and violet, purple, or orange gor- gets. Heliangelus zusii differs fro Neolesbia nehr- korni in having a gorget and frontlet, purple rather than bluish-black tail, and a straight rath- an curved l. Generic relationships.?The ne species can be assigned to the genus Heliangelus by a co bi- nation of characters (Figs. 1 and 2): (1) short (15.1 mm), straight, unmodified, uniformly dark bill; (2) brilliant frontlet and gorget; (3) un- modified remiges; (4) un odified rectrices, flat in cross section; (5) lack of puffy tarsal plu es (present in Eriocnemis spp.); and (6) nonbrilliant undertail coverts (brilliant in ost Eriocnemis spp.). Heliangelus zusii appears to bridge the gap between males of the ungorgeted, bluish-black H. regalis and the ore typical e bers of the genus with contrasting gorgets and predomi- nately green body plumage. Description of holotype.?Taxidermy mount with glass eyes. Nasal feathers, lores, auriculars, and argins of cro n are glittering steel-blue. A brilliant pale golden-green frontlet extends from the forecrown posteriorly to an imaginary ne a n t en ters its. Feathers of the frontlet are adpressed as in H. tis. asal athers tend . 5 stal the anterior edge of the nasal flanges. Traces of a hite spot are found posterior to the eyes. The hindneck, back, and ing coverts are deep bluish-black, slightly iridescent, turning deep greenish-blue on the lower back and ru p and returning to bluish-black on the upper tail co- rts. i aries ndaries rk rown ith a purplish tint. Secondaries have acu i- nate rather than broadly rounded tips. The deeply forked tail (fork 54% of length) is glit- tering dark purple above, duller and less iri- descent on the lo er surface. Basal portions of t r tri s, s r i t f l ed tail, r t sa e color as the exposed distal tips (unlike species of Aglaiocercus and Lesbia). Shafts of rec- trices are dark bro n distally, fading to light brown basally. Rectrices are flat in cross section. The ventral plu age is duller than the dorsu . The chin is dull bluish-black, bordered poste- riorly by a brilliant golden-green gorget (sa e color as frontlet). Gorget feathers have golden- green tips separated fro the gray base by a narro violet band. Gorget and frontlet feath- ers are broadly rounded rather than narro or pointed. The gorget is bordered below and on the sides by lustrous bluish-black. hen viewed head-on in direct light, the head and breast ap- pear black and contrast greatly with the golden- green gorget and frontlet. Feathers of the upper breast are bluish-black with narrow buffy-gray margins, becoming progressively paler on the lo er breast and belly. The idline of the belly is ediu gray but faint traces of bluish-black can be observed on barbules under agnifica- tion. Sides and flanks are dark bluish-black (sa e as back). There are no traces of a pectoral band or subter inal hite spots on breast feathers (present in so e Heliangelus hybrids; Graves 1990, raves a usi 1990). ertail coverts are pale crea y- hite ith a single lanceolate, subter inal spot at the idline (brownish-gray it l e reflections). ibial feat ers are ll grayish-bro n. The tarsi are no bare (for er- ly feathered?). Soft-part colors in dried speci- en: bill black; scutes on feet bro n; foot pads yellowish-brown. e s re ents . ?See le r t ard easurements. reatest idth termost - trices, 9.2. Gorget length x width, ca. 14 x 10. Sex, age, and molt.? lthough the sex of the holotype is not kno n ith certainty, all plu - age characters rese ble co ponents of ale plu age in Heliangelus species. The fully de- veloped gorget and frontlet, unstriated axil- lary ramphothecum, and lack of molt in the holotype indicate that it is an adult. January 1993] New Species of Sunangel 5 Fig. 2. Dorsal and lateral view of head and bill of the holotype of Heliangelus zusii. Distnlution.--Origin and range unknown. See discussion. Etymology.--I take great pleasure in naming this hummingbird for my friend and colleague, Richard L. Zusi, in recognition of his contri- butions to the systematics of hummingbirds. The English common name commemorates the nominal origin of Heliangelus zusii and many other Andean species. DISCUSSION Because H. zusii is represented by a unique specimen, extraordinary care must be taken to January 1993] New Species of Sunangel Fig. 2. Dorsal and lateral view of head and bill of the holotype of Heliangelus zusii. Distribution.? rigin and range unkno n. See no inal origin of Heliangelus zusii and scu sion. other Andean sDecies. many i Etymology.?I take great pleasure in na ing this hu ingbird for my friend and colleague, Richard L. Zusi, in recognition of his contri- butions to the systematics of hu ingbirds. The English common name commemorates the pe ISCU SION Because H. zusii is represented by a unique specimen, extraordinary care must be taken to 6 GARY R. GRAVES [Auk, Vol. 110 TABLE 2. Measurements (mm) of holotype of Helian- gelus zusii and Heineanum Halberstadt specimen of Neolesbia nehrkorni (courtesy of B. Nicolai). Heliangelus Neolesbia Characters zusii nehrkorni Wing 66.8 63.5 Bill 15.1 18.0 Rectrix 1 (innermost) 30.1 25.5 Rectrix 2 36.6 30.5 Rectrix 3 45.7 37.5 Rectrix 4 56.2 50.5 Rectrix 5 (outermost) 64.8 66.5 rule out or reject alternate taxonomic hypoth- eses. Below I address the possibility that the specimen represents: (1) Neolesbia nehrkorni; (2) a genetic variant or undescribed subspecies of some other species; (3) a hybrid; or (4) an arti- fact. Neolesbia nehrkorni?--As noted in the diag- nosis, H. zusii differs from H. nehrkorni in having a brilliant gorget and frontlet. Could H. zusii represent the mature male plumage of N. nehr- korni? Several factors argue against this being the case. Neolesbia nehrkorni (wing, 63.5, tail 67.5; Berlepsch 1887) is approximately the same size as H. zusii (wing, 66.8, tail 64.8). Because tail length is correlated with age in males of long- tailed hummingbirds (e.g. Aglaiocercus spp.), one would expect the type specimens of N. nehrkorni and H. zusii to be roughly the same age. Yet, N. nehrkorni lacks a brilliant gorget and frontlet, characters that are acquired at maturity in He- liangelus and are present in H. zusii. Signifi- cantly, the bill of N. nehrkorni s slightly de- curved as opposed to straight in H. zusii (Berlepsch 1887). Although intraspecific varia- tion in bill curvature in hummingbirds has not been studied, I know of no example where bill curvature varies with age. Heliangelus zusii and the Heineanum Halber- stadt specimen of N. nehrkorni also differ sig- nificantly in body proportions (Table 2). De- spite having a longer wing, H. zusii has a shorter bill and less deeply forked tail; the rectrices of H. zusii are more evenly graduated in length. In sum, H. nehrkorni and H. zusii do not rep- resent different age classes of the same taxon. Discussion of the systematic status of N. nehr- korni will appear elsewhere (Graves in prep.). Genetic plumage variant or undescribed subspe- cies?--Plumage color polymorphism and mel- anism, uncorrelated with size and shape vari- ation, has been documented in a number of trochiline species (Hartert 1922, Greenway 1978, Bleiweiss 1985). Because of its unique size and shape (e.g. forked tail, straight bill, etc.), H. zusii could not represent a simple genetic plumage variant of another species. I also considered the possibility that H. zusii represents a well-differentiated allopatric form of the ungorgeted Royal Sunangel (H. regalis). Heliangelus zusii is larger in all dimensions (ca. 25% in wing and tail length) than H. regalis. This difference exceeds the magnitude of geo- graphic size variation found within sexes of other species of Heliangelus (Bleiweiss 1985; pets. observ.). Significantly, no Heliangelus pecies (males) has "gorgeted" and "ungorgeted" pop- ulations or subspecies. Because the differences between H. zusii and H. regalis are both quali- tative (see diagnosis) and quantitative, I believe the two taxa are more genetically differentiated than are many of the component taxa of Andean superspecies (Graves 1980, 1985, 1991). Hybrid?--Determining whether a unique specimen represents a hybrid or a valid species can be very difficult depending on the circum- stances (Graves 1990). Complicating this partic- ular case is the fact that the precise origin of H. zusii is unknown. For the purpose of hybrid diagnosis (Graves 1990), the species pool must include all trochiline hummingbird that occur in Colombia, about 120 species (Hilty and Brown 1986). However, the task of determining wheth- er H. zusii is a hybrid is simplified by its dis- tinctive characters: (1) deeply forked tail; (2) straight bill; (3) brilliant gorget and frontlet; and (4) bluish-black body plumage. In correspondence cited above, Peters and Zimmer suggested that H. zusii might represent a hybrid of Aglaiocercus sp. x Heliangelus p. Several combinations of species from these gen- era could have produced a hybrid with a deeply forked tail, straight bill, and brilliant gorget and frontlet. However, no species in either genus, other than H. regalis, has bluish-black body plumage. Furthermore, H. zusii exhibits no trac- es of the peculiar awn-shaped bill, tapered rec- trix tips, or bicolored rectices of Aglaiocercus spp. In conclusion, there appears to be no combi- nation of species, considered two at a time, that collectively exhibit the characters of H. zusii and bracket its morphological dimensions (Graves 1990). Artifact?--Many artifacts, specimens created from parts of two or more species, are known ARY . RAVES [ uk, ol. 110 TABLE 2. easurements (mm) of holotype of Helian- gelus zusii and Heineanum Halberstadt specimen of Neolesbia nehrkorni (courtesy of B. Nicolai). Heliangelus Neolesbia aracters i ing 66.8 63.5 ill . Rectrix 1 (inner ost) 30.1 25.5 trix . . trix . . trix . . ectrix 5 (outer ost) 64.8 66.5 rule out or reject alternate taxonomic hypoth- eses. Below I address the possibility that the speci en represents: (1) Neolesbia nehrkorni; (2) a genetic variant or undescribed subspecies of some other species; (3) a hybrid; or (4) an arti- t. Neolesbia nehrkorni??As noted in the diag- nosis, H. zusii differs fro H. nehrkorni in having a brilliant gorget and frontlet. Could H. zusii represent the ature ale plu age of N. nehr- korni? Several factors argue against this being the case. Neolesbia nehrkorni ( ing, 63.5, tail 67.5; Berlepsch 1887) is approximately the same size as H. zusii (wing, 66.8, tail 64.8). Because tail length is correlated ith age in ales of long- tailed hu ingbirds (e.g. Aglaiocercus spp.), one would expect the type specimens of N. nehrkorni and H. zusii to be roughly the same age. Yet, N. nehrkorni lacks a brilliant gorget and frontlet, characters that are acquired at aturity in He- liangelus and are present in H. zusii. Signifi- cantly, the bill of N. nehrkorni is slightly de- curved as opposed to straight in H. zusii (Berlepsch 1887). Although intraspecific varia- tion in bill curvature in hu ingbirds has not been studied, I kno of no exa ple here bill curvature varies ith age. Heliangelus zusii and the Heineanu Halber- stadt speci en of N. nehrkorni also differ sig- nificantly in body proportions (Table 2). De- spite having a longer wing, H. zusii has a shorter bill and less deeply forked tail; the rectrices of H. zusii are more evenly graduated in length. In su , H. nehrkorni and H. zusii do not rep- resent different age classes of the sa e taxon. Discussion of the systematic status of N. nehr- korni ill appear else here (Graves in prep.). Genetic plumage variant or undescribed subspe- cies??Plumage color poly orphism and mel- anis , uncorrelated ith size and shape vari- ti , ented i ber f trochiline species (Hartert 1922, Greenway 1978, Bleiweiss 1985). Because of its unique size and shape (e.g. forked tail, straight bill, etc.), H. zusii could not represent a si ple genetic plu age variant of another species. I also considered the possibility that H. zusii represents a ell-differentiated allopatric for of the ungorgeted Royal Sunangel (H. regalis). Heliangelus zusii is larger in all di ensions (ca. 25% in wing and tail length) than H. regalis. This difference exceeds the agnitude of geo- graphic size variation found ithin sexes of other species of Heliangelus (Bleiweiss 1985; pers. observ.). Significantly, no Heliangelus species (males) has "gorgeted" and "ungorgeted" pop- ulations or subspecies. Because the differences between H. zusii and H. regalis are both quali- tative (see diagnosis) and quantitative, I believe the two taxa are more genetically differentiated than are any of the co ponent taxa of ndean superspecies (Graves 1980, 1985, 1991). Hybrid?? eter ining hether a unique specimen represents a hybrid or a valid species can be very difficult depending on the circum- stances (Graves 1990). Complicating this partic- ular case is the fact that the precise origin of H. zusii is unknown. For the purpose of hybrid diagnosis (Graves 1990), the species pool must include all trochiline hu ingbird that occur in Colombia, about 120 species (Hilty and Brown 1986). However, the task of deter ining wheth- er H. zusii is a hybrid is simplified by its dis- tinctive characters: (1) deeply forked tail; (2) straight bill; (3) brilliant gorget and frontlet; and (4) bluish-black body plumage. In correspondence cited above, Peters and Zimmer suggested that H. zusii ight represent a hybrid of Aglaiocercus sp. x Heliangelus sp. Several combinations of species from these gen- era could have produced a hybrid with a deeply forked tail, straight bill, and brilliant gorget and frontlet. o ever, no species in either genus, other than H. regalis, has bluish-black body plu age. Further ore, H. zusii exhibits no trac- es of the peculiar a n-shaped bill, tapered rec- trix tips, or bicolored rectices of Aglaiocercus spp. In conclusion, there appears to be no co bi- nation of species, considered t o at a ti e, that collectively exhibit the characters of H. zusii and bracket its orphological di ensions (Graves 1990). Artifact?? any artifacts, specimens created fro parts of two or ore species, are kno n January 1993] New Species of Sunangel 7 among 19th century collections (pers. observ.). These were created for the amusement of the preparators or to dupe eager collectors of nat- ural-history specimens. The possibility that H. zusii is an artifact was rejected after a careful examination of the feathers and skin under magnification (7-30 x) and x-radiographs of the specimen. These procedures revealed no ev- idence of composite construction of the speci- men. Biogeography.--I speculate that the holotype of H. zusii originated from the Eastern Cordil- lera of the Colombian Andes within a few hun- dred kilometers of Bogota, or possibly in the Central Cordillera. The fact that only a single specimen is known suggests that the species had a relictual or restricted geographic distri- bution when first collected. Sunangels (Helian- gelus) inhabit Andean cloud forest and shrub- lands from 1,200 to 3,400 m elevation, a zone that has been largely deforested in the Eastern Cordillera for cultivation of coffee, maize, po- tatoes, and other crops (Chapman 1917, Insti- tuto Geographico "Agustin Codazzi" 1982, Hil- ty 1985). Currently, only a single species of sunangel, the Amethyst-throated Sunangel (H. amethysticollis), occurs over most of the Eastern Cordillera; two others, the Orange-throated (H. mavors) and Tourmaline sunangels (H. exortis), have restricted distributions (Hilty and Brown 1986). Although the likelihood that H. zusii still survives seems remote, efforts should be made to find it in remnant patches of forest, including second growth, on both slopes of the Eastern Cordillera between 1,400 and 2,200 m. If extinct, H. zusii would be the first such instance for a hummingbird species in South America. The only other species of hummingbird that is sus- pected of being extinct is Chlorostilbon bracei, the endemic emerald of New Providence Is- land, Bahamas (Graves and Olson 1987). The effects of commercial collecting on 19th- century hummingbird populations in Colombia are unknown, but nearly all species with his- torically large geographic ranges (> 30,000 km 2) are common today in pristine habitats (Hilty and Brown 1986). This suggests that habitat de- struction, rather than intensive collecting, is the principal threat to most hummingbird species (see Hilty 1985). For example, the Black Inca (Coeligena prunellei) was one of commonest spe- cies in "Bogota" collections (Berlioz 1937). Re- grettably, the species is known presently from only a few remnant patches of cloud forest in the Eastern Cordillera, where it is common (Snow and Snow 1980, Hilty and Brown 1986, J. Fjeldst in litt.). It, the Bogota Sunangel, and many other presently endangered species are relics of a lost world that existed in the Andes before the arrival of human civilization. ACKNOWLEDGMENTS I thank R. C. Banks, L. C. Binford, R. E. Bleiweiss, C. Hinkelmann, B. L. Monroe, J. V. Reinsen, and R. L. Zusi for comments on the manuscript, as well as the curators and staff of the American Museum of Natural History, Louisiana State University Museum of Natural Science, and the National Museum of Nat- ural History, Smithsonian Institution for assistance in their collections. I am grateful to Jon Fjeldst for his beautiful and innovative frontispiece. Special thanks are due to M. B. Robbins and F. B. Gill for loaning the unique specimen and providing access to the archives of the Academy of Natural Sciences of Philadelphia and to B. Nicolai for color transparencies of Neolesbia. Research was partially supported by grants from the Frank M. Chapman Memorial Fund of the American Museum and the Research Opportunities Fund of the Smithsonian Institution. LITERATURE CITED BERLEPSCH, H. VON. 1887. Beschreibung eines neuen Colibri und Bemerkungen tiber eine Collection yon Vogelbilgen aus der Umgegend yon Sta. F de Bogota in Colombia. Zeitschrift ftir die ge- saturate Ornithologie 4:177-187. BERLIOZ, J. 1937. Three new cases of presumed nat- ural hybrids among Trochilidae. Ibis 1 (14th set.): 105-109. BERLIOZ, J,, AND C, JOUANIN, 1944, Liste de Trochil- ids trouvs dans les collections commerciales de Bogota. Oiseau 14:126-155. BLEBVEISS, R. 1985. Iridescent polychromatism in a female hummingbird: Is it related to feeding strategies? Auk 102:701-713. CHAPMAN, F.M. 1917. The distribution of bird-life in Colombia: A contribution to a biological sur- vey of South America. Bull. Am. Mus. Nat. Hist. 36:1-729. DOUGHTY, R. W. 1975. Feather fashions and bird preservation: A study in nature protection. Univ. California Press, Berkeley. GOULD, J. 1861. An introduction to the Trochilidae or family of humming-birds. Published by the author, London. GRAVES, G.R. 1980. A new species of metaltail hum- mingbird from northern Peru. Wilson Bull. 92: 1-7. GRAVES, G. R. 1985. Elevational correlates of speci- ation and intraspecific geographic variation in January 1993] New Species of Sunangel a ong 19th century collections (pers. observ.). ese ere ted r usement preparators or to dupe eager collectors of nat- ural-history specimens. The possibility that H. zusii is an artifact was rejected after a careful amination athers in der agnification (7-30 x) and ^-radiographs of the speci en. These procedures revealed no ev- idence of co posite construction of the speci- en. Biogeography.?I speculate that the holotype of H. zusii originated fro the Eastern Cordil- ra olombian ndes ithin n- dred kilo eters of Bogota, or possibly in the Central Cordillera. The fact that only a single specimen is known suggests that the species had a relictual or restricted geographic distri- bution hen first collected. Sunangels (Helian- gelus) inhabit ndean cloud forest and shrub- l s fr , t , l ation, a that has been largely deforested in the Eastern Cordillera for cultivation of coffee, aize, po- tatoes, and other crops (Chap an 1917, Insti- tute Geographico "Agustin Codazzi" 1982, Hil- ty 1985). Currently, only a single species of sunangel, the A ethyst-throated Sunangel (H. amethysticollis), occurs over most of the Eastern Cordillera; two others, the range-throated (H. mavors) and Tour aline sunangels (H. exortis), have restricted distributions ( ilty and Brown 1986). lthough the likelihood that H. zusii still s r i es see s r t , eff rts s l e to find it in re nant patches of forest, including second gro th, on both slopes of the Eastern r illera t een , , . If ti ct, . ould irst stance r hu ingbird species in South A erica. The only other species of hu ingbird that is sus- pected of being extinct is Chlorostilbon bracei, emic erald ew vidence land, Baha as ( raves and lson 1987). The effects of commercial collecting on 19th- century hu ingbird populations in Colo bia are unkno n, but nearly all species ith his- torically large geographic ranges (> 30,000 k 2) are co on today in pristine habitats ( ilty and Brown 1986). This suggests that habitat de- struction, rather than intensive collecting, is the principal threat to most hu ingbird species (see ilty 1985). For exa ple, the Black Inca (Coeligena prunellei) was one of co onest spe- cies in "Bogota" collections (Berlioz 1937). Re- grettably, the species is known presently from only a few re nant patches of cloud forest in t t r rdillera, ere it is on (Snow and Snow 1980, ilty and Brown 1986, J. Fjeldsa in litt.). It, the Bogota Sunangel, and many other presently endangered species are ics rld t isted ndes fore rival an ilization. OWLEDGMENTS I t . . s, . . i ford, . . l i eiss, . inkel ann, B. L. onroe, J. . emsen, and R. L. Zusi for co ents on the anuscript, as ell as tors erican useum Natural History, Louisiana State University useum f atural i , t ational useu f at- ural History, S ithsonian Institution for assistance in their collections. I a grateful to Jon Fjeldsa for his beautiful and innovative frontispiece. Special anks . bbins i l r loaning the unique specimen and providing access to the archives of the Acade y of atural Sciences of Philadelphia and to B. Nicolai for color transparencies of Neolesbia. Research was partially supported by grants fro the Frank . Chap an e orial Fund of the A erican useu and the Research pportunities nd ithsonian stitution. RE BERLEPSCH, H. VON. 1887. Beschreibung eines neuen Colibri und Bemerkungen iiber eine Collection von Vogelbalgen aus der Umgegend von Sta. Fe de Bogota in Colombia. Zeitschrift fur die ge- sammte rnithologie 4:177-187. BERLIOZ, J. 1937. Three ne cases of presu ed nat- ural hybrids among Trochilidae. Ibis 1 (14th ser.): 5-109. BERLIOZ, J., AND C. JOUANIN. 1944. Liste de Trochil- es ves s lections mercials Bogota. Oiseau 14:126-155. BLEIWEISS, R. 1985. Iridescent polychro atism in a female hummingbird: Is it related to feeding strategies? Auk 102:701-713. MAN, . M. . i tri tion f ir -life in Colo bia: A contribution to a biological sur- vey of South America. Bull. Am. us. Nat. Hist. :1-729. HTY, . . 1975. eat er fas i s a ir preservation: A study in nature protection. Univ. alifornia Press, Berkeley. GOULD, J. 1861. n introduction to the Trochilidae or family of humming-birds. Published by the t r, ndon. GRAVES, G. R. 1980. A new species of metaltail hum- mingbird from northern Peru. ilson Bull. 92: 7. GRAVES, G. R. 1985. Elevational correlates of speci- ation and intraspecific geographic variation in 8 GARY R. GRAVES [Auk, Vol. 110 plumage in Andean forest birds. Auk 102:556- 579. GRAVES, G.R. 1990. Systematics of the "green-throat- ed sunangels" (Aves: Trichilidae): Valid taxa or hybrids? Proc. Biol. Soc. Wash. 103:6-25. GRAVES, G. R. 1991. Bergmann's rule near the equa- tor: Latitudinal clines in body size of an Andean passerine bird. Proc. Natl. Acad. Sci. USA 88:2322- 2325. GRAVES, G. R., AND S. L. OLSON. 1987. Chlorostilbon bracei Lawrence, an extinct species of humming- bird from New Providence Island, Bahamas. Auk 104:296-302. GRAVES, G. R., AND R. L. ZIJSI. 1990. An intergeneric hybrid hummingbird (Heliodoxa leadbeateri x He- liangelus amethysticollis) from northern Colombia. Condor 92:754-760. GREENwAY, J. C., JR. 1978. Type specimens of birds in the American Museum of Natural History. Part 2. Bull. Am. Mus. Nat. Hist. 161:1-305. HARTERT, E. 1922. Types of birds in the Tring Mu- seum. B. Types in the general collection. Trochili. Novitat. Zool. 29:403-412. HILTY, S.L. 1985. Distributional changes in the Co- lombian avifauna: A preliminary blue list. Pages 1000-1012 in Neotropical ornithology (P. A. Buckley, M. S. Foster, E. S. Morton, R. S. Ridgely, and F. G. Buckley, Eds.). Ornithol. Monogr. No. 36. HILTY, S. L., AND W. L. BROWN. 1986. A guide to the birds of Colombia. Princeton Univ. Press, Prince- ton, New Jersey. HINKELMANN, C., B. NICOLAI, AND R. W. DICKERMAN. 1991. Notes on a hitherto unknown specimen of Neolesbia nehrkorni (Berlepsch, 1887; Trochilidae) with a discussion of the hybrid origin of this 'species.' Bull. Br. Ornithol. Club 111:190-199. INSTITUTO GEOGRAPHICO "AGUSTIN CODAZZI." 1982. Atlas regional Andino. MEYER DE SCHAUENSEE, R. 1947. New or little-known Colombian birds. Proc. Acad. Nat. Sci. Phila. 99: 107-126. SIBLEY, C. G., AND B. L. MONROE, JR. 1990. Distri- bution and taxonomy of birds of the world. Yale Univ. Press, New Haven, Connecticut. SIMON, E. 1921. Histoire naturelie des Trochilidae (synopsis et catalogue). Encyclopedia Roret, L. Mulo, Paris. SNOW, D. W., AND B. K. SNOW. 1980. Relationships between hummingbirds and flowers in the An- des of Colombia. Bull. Br. Mus. Nat. Hist. (Zool.) 38:105-139. TAYLOR, W.P. 1909. An instance of hybridization in hummingbirds, with remarks on the weight of generic characters in the Trochilidae. Auk 26: 291-293. COLOR PLATE Publication of the color plate was supported by the Donald L. Bleitz Fund of The American Ornitholo- gists' Union. GARY R. GRAVES [ uk, ol. 110 plu age in ndean forest birds. uk 102:556- 9. GRAVES, G. R. 1990. Systematics of the "green-throat- ed sunangels" (Aves: Trichilidae): Valid taxa or hybrids? Proc. Biol. Soc. ash. 103:6-25. GRAVES, G. R. 1991. Bergmann's rule near the equa- tor: Latitudinal dines in body size of an Andean passerine bird. Proc. atl. cad. Sci. S 88:2322- 5. S, . ., . . . . l til bracei Lawrence, an extinct species of humming- ir fr r idence Isl , s. :296-302. GRAVES, G. R., AND R. L. ZUSI. 1990. An intergeneric hybrid hu ingbird (Heliodoxa leadbeateri x He- liangelus amethysticollis) fro northern Colo bia. ndor : 54-760. GREENWAY, J. , JR. 1978. Type speci ens of birds in the American useum of Natural History. Part . ll. . us. at. ist. :1-305. HARTERT, E. 1922. Types of birds in the Tring u- seum. B. Types in the general collection. Trochili. ovitat. ol. :403-412. HILTY, S. L. 1985. Distributional changes in the Co- lombian avifauna: A preli inary blue list. Pages 1000-1012 in Neotropical ornithology (P. A. Buckley, M. S. Foster, E. S. orton, R. S. Ridgely, and F. G. Buckley, Eds.). Ornithol. Monogr. No. . HILTY, S. L., AND . L. BROWN. 1986. A guide to the ir s f l ia. ri ceton iv. ress, ri ce- ton, New Jersey. I , , . I I, . . I . 1991. Notes on a hitherto unkno n speci en of Neolesbia nehrkorni (Berlepsch, 1887; Trochilidae) with a discussion of the hybrid origin of this 'species.' Bull. Br. rnithol. Club 111:190-199. TO S IN ZZI." . Atlas regional Andino. NSEE, . . e r little-known lombian rds. c. cad. at. . ila. 7-126. SIBLEY, . ., AND B. L. R E, JR. 1990. istri- bution and taxonomy of birds of the orld. Yale iv. ress, e a en, ecticut. I , . . istoire t relle s rochilidae (synopsis et catalogue). Encyclopedia Roret, L. ulo, ris. SNOW, D. ., AND B. K. SNOW. 1980. Relationships bet een hu ingbirds and flowers in the An- des of olo bia. ull. Br. us. at. ist. ( ool.) :105-139. TAYLOR, . P. 1909. An instance of hybridization in hummingbirds, with remarks on the weight of generic characters in the Trochilidae. Auk 26: 1-293. Publication of the color plate was supported by the onald . leitz nd he erican rnith lo- gists' Union.