THE WILSON BULLETIN A QUARTERLY MAGAZINE OF ORNITHOLOGY Published by the Wilson Ornithological Society VOL. 92, No. 1 MARCH 1980 PAGES 1-148 Wilson. Bull., 92(l), 1980, pp. l-7 A NEW SPECIES OF METALTAIL HUMMINGBIRD FROM NORTHERN PERU . GARY R. GRAVES Recent fieldwork in the isolated cordilleras of Peru by personnel of the Louisiana State University Museum of Zoology and other institutions has resulted in the discovery of several new high Andean birds (Blake and Hocking 1974, Weske and Terborgh 1974, Lowery and Tallman 1976, O?Neill and Parker 1976). In October 1977 during a preliminary LSUMZ expedition to the Divisoria de Huancabamba, Department of Piura, Peru, I collected 2 specimens of a previously undescribed metaltail humming- bird. The specimens were stolen in Peru before they could be critically compared with museum material but field sketches of the birds showed sufficient differences from other species of Metallura to suggest they rep- resented an undescribed form. A LSUMZ field party returned to the region in 1978 and fortunately obtained 5 skins and 1 complete skeleton of this new form. In the genus Metallura, the related forms recisa, williami, primolinus, atrigularis, baroni, theresiae, eupogon, aeneocauda and malagae form a superspecies group (Fig. 1) of uncertain relationships (M. tyrianthina and M. phoebe (Black Metaltail) are locally sympatric with members of this group). Peters (1945) merged primolinus and atrigularis into M. williami and reduced baroni to an isolated subspecies of eupogon. Consequently Zimmer (1952) combined malugue and aeneocauda, but considered baroni to be specifically distinct and states (Zimmer 1952:lO): ?The remaining three forms (baroni, theresiae, eupogon) are more divergent from the gen- eral pattern of the series, and it is probably best to give each of them individual specific status until specific relationships can be more positively demonstrated.? Wetmore (1970) considered the newly described recisa as a subspecies of primolinus, which he raised to specific level on the basis 1 THE WILSON BULLETIN - Vol. 92, No. I, March 1980 FIG. 1. Distribution of the aeneocauda superspecies: (1) Metallura williami [including (a) recisa, (b) williami, (c) primlinus, (d) atrigularis]; (2) M. baroni; (3) M. odomae; (4) M. theresiae; (5) M. eqmgon; (6) M. aeneocauda [including (a) Aeneocauda and (b) malagael. of tail coloration. Since aeneocauda (Gould) 1846 has priority, the group can be called the aerzeocauda superspecies (see Zimmer 1952). The pos- sible relationship of M. iracunda (Perija Metaltail) and M. phoebe to the ueneocuudu group should be investigated. The new species clearly belongs to the aeneocuuda superspecies, and fills what had been an apparent gap in the range of that superspecies (see frontispiece). Graves * NEW METALTAIL HUMMINGBIRD 3 TABLE 1 RANGES AND MEANS (MM) OF SELECTED METALLCIRA OF THE AENEOCAUDA SUPERSPECIES N Wing (chord) Tail Culmen from feathers baroni 3 66 56.4-58.0 (57.1) 37.9-38.8 (38.4) 12.7-13.4 (13.2) l? 54.8 37.2 13.0 odomne 3 86 59.2-60.7 (59.7) 40.2-%0.9 (40.5) 13.8-14.4 (14.1) 490 54.6-57.7 (56.3) 38.441.0 (40.2) 13.5-14.3 (13.8) there&e 4 6d 60.5-65.1 (62.6) 40.743.4 (42.3) 12.5-12.8 (12.6) 3 9P 58.1-60.4 (59.2) 40.341.1 (40.8) 12.4-13.3 (12.8) 7 66 62.545.4 (64.4) 41.043.8 (42.9) 11.6-13.6 (12.8) uqogon 3 99 59.3-59.4 (59.3) 39.640.7 (40.2) 11.9-12.6 (12.3) Metallura odomae sp. nov. NEBLINA METALTAIL HOLOTYPE.-Louisiana State University Museum of Zoology no. 87547; adult male from Cerro Chinguela, ca. 5 km northeast of Sapalache, elevation approximately 2900 m (9500 ft), 05?07?S, 79?23?W; Department of Piura, Peru; 10 June 1978; netted and prepared by Richard D. Semba; original number 535. DIAGNOSIS.-A medium-sized, sexually dimorphic Metallurn of the aeneocauda super- species. Gorget in males reddish purple, not green, orange, violet, or black as in other members of the superspecies. In females, chin, throat and upper breast speckled with Scarlet discs (capitalized color names are from Ridgway 1912) not restricted to central throat as in theresine and eupogon. Both sexes lighter below than other members of the arnrocauda superspecies. Bill of odomae longer than those of nearest allopatric of baroni, thwesiae and eupogon; not apparently overlapping with respective sexes (Table 1). DESCRIPTION OF HOLOTYPE.-Crown, hind neck, back, wing coverts and rump iri- descent dark green; chin and gorget Amaranth Purple; superorbital and postorbital areas hronzy olive grading to rich coppery bronze on auriculars, lateral throat and sides of neck surrounding gorget; underparts olive-bronze with buffy white subterminal bars giving the lower breast and belly a spotted aspect. Plumes at vent white and fluffy; feathers on lower belly edged terminally with tan; undertail coverts bronzy green, broadly edged with buffy cinnamon. Remiges dark purplish brown; small coverts on carpal region of wing cinnamon- buff. Rectrices seen from above and against the light, dark olive; with the light, shining dark steel-blue. From below rectrices are shining golden green; outermost pair tipped on inner and outer web with pale gray. Soft part colors in life: bill black; iris dark brownish black; feet black. SPECIMENS EXAMINED.--M. W. williami (FMNH) 2 6 6 Purace, Department of Cauca, 19 Malvasa, Department of Cauca and (LSUMZ) 18 El Crucero, Department of Cauca, Colombia; M. w. atrigularis (LSUMZ) 2 $ d Loma Magdalena, Prov. Morona-Santiago, Ec- uador; M. baroni (AMNH) 3d 6, 10 Cuenca, Ecuador; M. odonae (LSUMZ) Id, 49 P Cerro Chinguela (FMNH) 26 d ?above Huancabamba,? Peru; M. theresiae (LSUMZ) 8 6 6, 6P P Cordillera Colon, 18, I? above Balsas, Department of Amazonas, 3d 6, 2P 0 Zapa- 4 THE WILSON BULLETIN . Vol. 92: No. 1, March 1980 tagocha, Department of Hu6nuc0, Peru; M. eupogon (FMNH) 36 8, 39 0 Huhnuco Mts. (AMNH) 2 6 6 Maraynioc (LSUMZ) 3 6 6 NE Comas, Department of Junin; M. a. aeneo- caudn (LSUMZ) 2 $8, 10 near Ollantaytambo, Department of Cuzco, Peru, M. n. malagm (LSUMZ) Id, 49 0 Chapare, and 16 Incachaca, Bolivia. DISTRIBUTION. -So far, known only from the edge of uppermost forest into paramo from ca. 2600 m-3000 m of the Divisoria de Huancabamba, Department of Piura-Cajamarca, Peru. ETYMOLOGY.-1 take pleasure in naming this new form for Babette M. Odom of Orange, Texas, who has generously supported the LSUMZ field program in Peru as part of her long- standing interest in avian natural history. The proposed English name, Neblina Metaltail, refers to the foggy conditions on the wet, windswept shrublands above timberline on Cerro Chinguela where this hummingbird lives. REMARKS M. odomae differs from other members of the aeneocauda group in various details consistent at the superspecific level. M. odomae females are lighter below than females of M. w. williami (Viridian Metaltail), M. baroni (Violet-throated Metaltail), M. theresiae (Coppery Metaltail), M. eupogon (Fire-throated Metaltail), and M. a. aeneocauda and M. a. mal- ague (Scaled Metaltail). Females of all forms have buffy throats covered with scattered iridescent discs. The breast and belly feathers, with buff or buffy white subterminal bars, have terminal green discs which are smallest (thus making the underparts lighter) on M. odomae, M. baroni and M. a. aeneocauda progressing on a graded series to wider, darker green discs on M. theresiae and M. eupogon. Dorsally female M. odomae are nearly indistinguishable from examples of M. a. aeneocauda, M. a. malagae and M. w. williami, but lack the olive-bronze highlights of M. baroni, the coppery red colors of M. theresiae and the bronzy reflections of M. eupogon. Males of M. odomae are lighter below than the other allospecies. The subterminal buffy white bars of the underparts give a scaled appearance to M. odomae similar to that of M. a. aeneocauda. M. w. williami, M. w. atrigularis, M. baroni, M. theresiae and M. eupogon have uniformly dark green underparts with very little suggestion of barring or scaling. Dorsally, M. odomae males are intermediate in color between the bronzy-backed M. eupogon and the green-backed M. w. williami, M. w. atrigularis, M. a. aeneocauda and M. a. malugae. Gorget colors in males of the aeneo- cauda superspecies vary from green (recisa, williami, primolinus, there- siae, aeneocauda and malagae) to orange (Grenadine Red) in M. eupogon (gorget narrower than depicted in frontispiece), reddish purple in M. odomae, violet in M. baroni and black in M. w. atrigularis. Gorget discs in females are not noticeably different in color from gorgets of males ex- cept in M. odomae (see DIAGNOSIS). Pale tips are present in the outer rectrices of all female Metallura, but in males of the aeneocauda group, Gravrs * NEW METALTAIL HUMMINGBIRD 5 pale tips are conspicuous only in M. odomae (faintly present in M. a. aeneocauda of Cuzco). The longer bill of M. odomae suggests a niche shift relative to its nearest allopatric relatives (baroni, theresiae and eupogon). HABITAT AND FIELD OBSERVATIONS Two male specimens in the Field Museum of Natural History identified as ?Metallura eupogon baroni? proved to be referable to the new form. The collecting locality recorded on the tags by C. Kalinowski in 1954, ?Huancabamba, Department of Piura, Peru, 2600 m,? probably refers to the crest of the cordillera northeast of Huancabamba along the Huanca- bamba-Namballe mule trail (= LSUMZ ?Cerro Chinguela?). On Cerro Chinguela, M. odomae was found in pajonal similar to treeline habitats found in many other locations in Peru along the wet eastern slope of the Andes. The LSUMZ specimens of the type series were captured in a series of nets placed along lanes cut through tongues of stunted forest extending upward along ravines into the paramo and isolated clumps of elfin forest well separated from the forest below (Fig. 2). The odomae collected in 1977 were discovered during a wet snowstorm (ca. 5?C with moderately heavy winds). One of the individuals was foraging on the white flowers of storm flattened, dwarf ericaceous shrubs partly concealed by grass. The bird flew low from shrub to shrub, often alighting on prostrate sheaths of grass to probe protruding flowers. During the snowstorm, which lasted for over an hour (most snow melted on contact with the rain drenched slope), I observed Pterophanes cyanopterus (Great Sapphire- wing), Coeligena lutetiae (Buff-winged Starfrontlet), Eriocnemis vestitus (Glowing Puffleg), M. tyrianthina (Tyrian Metaltail) and Chalcostigma herrani (Rainbow-bearded Thornbill) foraging in more sheltered ravines and the leeward sides of elfin forest clumps. No call notes, songs or flight displays of M. odomae were noted by any expedition member. No M. odomae examined had enlarged gonads (6-24 June, 22 October). TAXONOMIC RELATIONSHIPS The relationships of Andean hummingbirds are poorly understood, and the existence of questionably isolated taxa hinders the application of the biological species definition (Mayr 1963). Morphological and behavioral characters and geographical isolation, however, permit inferences con- cerning species limits between allopatric populations whose potential for interbreeding cannot be determined with certainty. Members of the aeneocauda superspecies are restricted to wet treeline- paramo habitats between 2600? and 4000 m elevation from the Andes of Colombia south to Bolivia (Fig. 2). M. odomae is separated geographically by unsuitable habitat from the nearest populations of baroni (200 km), 6 THE WILSON BULLETIN - Vol. 92, No. 1, March 1980 FIG. 2. Isolated grove of elfin cloud forest in piramo on Cerro Chinguela near the type locality of Metallura odomae (photograph by Theodore A. Parker, III). theresiae (120 km) and eupogon (640 km). Recent work on the Pleistocene speciation of Andean birds (e.g., Vuilleumier 1969) has shown similar patchy distributions of other high Andean hummingbirds (Oreotrochilus, Chulcostigma). Present-day distribution patterns of the aeneocauda su- perspecies may represent relictual interglacial refugia, with much reduced or no gene flow between them. Although the paucity of information on the biology of Metallura does not yet permit a thorough systematic revision of the aeneocauda superspecies, evidence from central Peru suggests iso- lating mechanisms have developed between several forms. Specimens of theresiue and eupogon, from the Cordillera Carpish and the Huinuco Mountains, respectively (separated by the narrow arid intermontane valley of the Rio Huallaga, Department of Huanuco, Peru), show no evidence of intergradation. Unfortunately, no specimens have been taken in the area between the known ranges of eupogon (Cordillera Vilcabamba) and aeneo- cauda (Urubamba Valley) in the Department of Cuzco, Peru. Until contradictory information is available, I consider the genetically isolated, morphologically differentiated taxa as allospecies (M. baroni, M. odomue, M. theresiae, M. eupogon) and follow Zimmer (1952) in merging recisu, williami, primolinus and utrigularis (=M. williami) and aeneo- caudu and mulugue (=A!. aeneocuudu). I recommend a north to south Graves . NEW METALTAIL HUMMINGBIRD 7 linear arrangement of taxa, with M. odomae placed between M. baroni and M. theresiae. ACKNOWLEDGMENTS J. W. Eley, T. S. Schulenberg, T. A. Parker, 111, R. D. Semba and D. R. Hunter provided moral support and companionship during the 1977 and 1978 expeditions. 1 appreciate the opinions and discussion freely offered by J. V. Remsen, H. D. Pratt, J. P. O?Neill and G. L. Graham. O?Neill, Pratt, Graham, S. Y. Graves and M. D. Williams commented on several early drafts of this paper. The referees K. C. Parkes and J. W. Fitzpatrick made valuable comments on the manuscript. 1 thank J. W. Fitzpatrick (Field Museum of Natural History) and J. Farrand, Jr. (American Museum of Natural History) for providing comparative ma- terial. 1 commend John O?Neill for his artistic and scientifically useful frontispiece. Arturo and Helen Koenig, Manuel and Isabel Plenge and Gustav0 de1 Solar continue their gracious ayl~da to all LSUMZ personnel during necessary stays in Lima and Chiclayo. 1 am especially indebted to Manuel Sanchez S. whose help on the 1976, 1977 and 1978 expeditions has been indispensable. 1 am grateful to John S. Mcllhenny, H. Irving and Laura R. Schweppe, and Babette M. Odom for their support of the LSUMZ fieldwork. Antonio Brack E., Marc Dourojeanni R., Susana Moller H. and Carlos Ponce P. of the Direction General Forestal y de Fauna of the Ministerio de Agricultura, Lima, Peru, continue their support of the LSUMZ field studies and issued the necessary permits. 1 also acknowledge the continued collaboration of Aero Peru. LITERATURE CITED BLAKE, E. R. AND P. HOCKING. 1974. Two new species of tanager from Peru. Wilson Bull. 86:321-516. LOWERY, G. H., JR. AND D. A. TALLMAN. 1976. A new genus and species of nine-primaried oscine of uncertain affinities from Peru. Auk 93:415428. MAYR, E. 1963. Animal species and evolution. Harvard Univ. Belknap Press, Cambridge, Massachusetts. O?NEILL, J. P. AND T. A. PARKER, Ill. 1976. New subspecies of Schizoeclcafuli~inosn and Uromyias agra&ia from Peru. Bull. Br. Omithol. Club 96:136-141. PETERS, J. L. 1945. Checklist of birds of the world, Vol. 5. Harvard Univ. Press, Cambridge, Massachusetts. RIDGWAY, R. 1912. Color standards and color nomenclature. Washington, D.C. (Published by the author.) VUILLEUMIER, F. 1969. Pleistocene speciation in birds living in the high Andes. Nature 223:1179-1180. WESKE, J. S. AND J. W. TF,RBORGH. 1974. Hrmispingus pnrotlii, a new species of tanager from Peru. Wilson Bull. 86:97-103. WETMORE, A. 1970. Descriptions of additional forms of birds from Panama and Colombia. Proc. Biol. Sot. Wash. 82:767-776. ZIMMER, J. T. 1952. Studies of Peruvian birds, No. 62. The hummingbird genera Patagonn, Sappho, Polyonymus, Rumphomicron, Metallum, Chulcostigma, Tuphrolrsbia, and Aglaiocrrcus. Amer. Mus. Novit., No. 1595:1-29. MUSEUM OF ZOOLOGY, LOUISIANA STATE UNIV., BATON ROUGE, LOUISIANA 70893. ACCEPTED 16 OCT. 1979.