ISSeptembci 1998 PROCEEDINGS OF THK BIOLOGICAL SOCIETY OF WASHINGTON 1IK3):511-520. 1998. Diagnoses of hybrid (Aves: Trochilidae). 6. An intergeneric hybrid, Aglaiocercus kingi x Metallura tyrianthina, from Venezuela

Gary R, Graves Department of Vertebrate Zoology. National Museum of Natural History. Smithsonian Institution. Washington, DC. 20560. U.S.A.

Abstract.—An intergeneric hybrid . Aglaiocercus kingi X Me- tallura tyrianthina. is described. External measurements of the hybrid are in- termediate of those of the parental . Back plumage iridescence is bluer {511 nm) in the hybrid than in either of the parental species (553-571 nm). This color shift is thought to he caused by a developmental aberrancy or mu- tation which affects melanin granules thai produce iridescence in feather ker- atins.

Under certain circumstances, interspecif- Here I describe an intergeneric hybrid com- ic hybridization may be an important source bination, Aglaiocercus kingi x Metallura of genetic exchange among avian lineages tvriaiuhina. that may create favorable conditions for rapid and significant evolutionary change Materials and Methods (Grant & Grant 1992). From an analysis of data in Panov's (1989) catalog of avian hy- The unsexed specimen (American Mu- brids. Grant & Grant (1992) reported that seum of Natural History [AMNH] 146645) 19,1% (6t of 319) of hummingbird species was collected by S. Gabaldon in Esiado has hybridized in nature. A surprising Men da. Venezuela. The exact locality, ele- 69.2% (36 of 52) of the hybridizing pairs vation, and date of collection are unknown. is intergeneric ( of Sibley & Mon- The specimen appears to he a male in sub- roe 1990). a finding consistent with Prager definitive plumage as evidenced by the I aim & Wilson's (1975) thesis thai interspecific strialions on the maxillary ramphotheca hybridization potential is slowly lost during (see Ortiz-Crespo 1972) and by its elongat- avian evolution. The true extent of hybrid- ed tail (Fig. 1 & 2). Five different identifi- ization among hummingbirds, however, is cations have been written in ink and pencil imperfectly known. Panov's (1989) compi- (in quotations below) on the two attached lation includes numerous poorly document- AMNH labels since the specimen was cat- ed or erroneous records, as did its antece- aloged in 1927 (in probable order of occur- dent (Gray 1958). Moreover, many new hy- rence): (a) "Cyanolesbia" [—Aglaiocer- brid combinations have been reported re- cus]: (b) "Aglaiocercus Icaudata" cently (e.g.. Graves 1990, 1996a, 1998a: [=Aglaiocercus kingi caudatux]: (c) "Me- Graves & Zusi 1990: Hinkelmann 1996: rallura purpureicauda" [ = Chalcostigma Welter & Schuehmann 1997). A definitive purpureicauda]: (d) "?Hybrid?, Aglaiocer- analysis of hybridization and phyletic retic- cus caudata X Ramphomicron" [=Agtaio- ulation must await a robust phylogeny and cercus kingi caudarus X Ramphomicron a systematic survey of purported hybrids, microrhynchum\\ and (e) "Aglaiocercus type specimens, and museum collections. emmae caudata, (melanistic aberration), 512 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

Fig. 1. Vciural views of mule Agtaiocenux kingi ctiudittux I lop 1. Metullurtt tyritmthinu oreopoiu (bottom), and a probable hybrid, A. kingi caudatus x M. ryrumihina oreopolti (AMNH I46h45).

Fig. 2, Probable hybrid. AgUuocerais kingi caudatus X Metallura lyrimuhina onapofa (AMNH I4h(vt5j. VOLUME 11). NUMBER 1 513 fide J. Berlioz, Apr. 1975" \=Ag!aiocercus ues are dark, high values are light. In other kingi caudattts]. words, the more light reflected from the I compared the oft-identified specimen plumage the higher the L value will he. It with series of all species in the subfamily should be noted that visual systems in hum- Trochilinae, the typical hummingbirds (Zusi mingbirds (e.g.. Goldsmith & Goldsmith & Bentz 1982. Sibley & Monroe 1990. 1979) differ significantly from those of hu- Bleiweiss el al. 1997), in the collections of mans. The relevance of opponent color co- the American Museum of Natural History ordinates to colors perceived by humming- and the National Museum of Natural His- is unknown. Dominant wavelengths tory (USNM), Smithsonian Institution. For re lice ted from plumage surfaces are listed the purpose of hybrid diagnosis. I consid- for comparison. Data in Table I were com- ered all hummingbirds (Trochilinae: taxon- piled from the averages of five independent omy of Sibley & Monroe 1990) that occur measurements (specimen moved from ap- in Estado Merida as potential parental spe- erture between trials) for each plumage area cies {Phelps & Phelps 1958, Meyer de per specimen. Schauensee & Phelps 1978) (Appendix I). Measurements of wing chord, bill length In addition, I compared the specimens di- (from anterior extension of feathers), and rectly with the hoiotypes of Chalcostigma rectrix length (from point of insertion of the purpureica uda (A M N H 483931), Lesbia central rectriccs to the tip of each rectrix) ortoni (AMNH 156651), Zodalia thattmas- were taken with digital calipers and round- ta (USNM 173911), and Aeronympha pro- ed to the nearest 0.1 mm (Table 2). Mea- xantis (Field Museum of Natural History, surements and least squares regression lines FMNH 11852), and with notes, photo- were projected on bivariate plots to illus- graphs, and videotape of the holoiype of trate size differences (Wilkinson 1989). Hefiangelus zuxii (Academy of Natural Sci- There are three alternatives to consider- ences of Philadelphia, ANSP 159261). the specimen represents an aberrant color Color descriptions given in Appendix 2 morph of A. kingi or some other species, a were made under natural light. 1 evaluated hybrid, or an tmdescribed species. The the color of dorsal plumage (center of back) specimen differs significantly in size and and the ventral surfaces of the rectrices with shape from all species in Appendix I. In a reflectance spec trap ho to meter equipped particular, the rec trices of the specimen are with a 11.0 mm aperture (Color Mate Col- considerably wider, flatter in cross section, orimeter. Milton Roy). The colorimetric and more iridescent on the ventral surfaces characters were described in terms of op- than in A. kingi, indicating that it is not sim- ponent-color coordinates (L, a, b) (Hunter ply a melanistic example of that species as & Harold 1987). This system is based on suggested by Berlioz on the specimen label. the hypothesis that signals from the cone Because hybrids have no standing in zoo- receptors in the human eye are coded by the logical nomenclature, the burden of proof brain as light-dark (L), red-green (a), and rests on the investigator to refute this pos- yellow blue (/;). The rationale is that a color sibility before bestowing species status on cannot be red and green or yellow and blue a unique specimen. Because the evidence at the same time. Therefore "redness" and points to hybridization, I refer to the spec- "greenness" can be expressed as a single imen as a hybrid in the remainder of the value a, which is positive if the color is red paper. and negative if the color is green. Likewise, The diagnosis was approached in a hi- "yellowness" or "blueness" is expressed erarchical manner. The presumed parental by b for yellows and —b for blues. The species of the hybrid first were hypothe- third coordinate L, ranging from 0 to 100, sized through the comparative analysis of describes the "lightness" of color; low val- plumage pattern, as well as from feather 514 !>RC>| I I DINliS , »l THI UK II OCJICAl. SUClliTY OI- WASHINGTON

Table I.—Ranges and means (rstandard deviation) of opponent color coordinates (L a. b) and dominant wavelength reflected from dorsal plumage (enter of back) and the venmil surface of reel rices in male Aglaio- cercus kingi ctmcJtilits. Mrlallum tyriamhinu orwpttitt. and their probable hybrid (AMN'H 146645).

M Ivntinltiiuii Vtritbta (n - IZ hi - IZ) IMmJ

Back plumage L (LightNcssi (i) 23.6-30.1 21.5-26.6 24,1 27.1 ± 1.8 24 2 r 1.4 a (Red | + |/Grcen [-]) (a) - l4.8-(-6.6) -7.l-t-0.6l -7.9 - 10.9 + 2.9 -3.7 ± 1.8 t (Yellow [+yBkie (-]> (A) 15.4-20.2 12.6-20.9 2.9 17.8 t 14 17.3 ± 2.3 Dominant Wavelength Inml 553.1-563.8 562.8-570.5 511.3 557.8 ± 3.5 566.7 ± 2.6 Ventral surface of reel rices t (Lightness) CM 18.8-21.1 20.3-23.9 20.6 20.1 ± 0.7 21,7 ± 1.0 a (Red [ +(/Green [-)) fa) 3.8-5.9 11.5-17.0 10.6 4.9 ± 0.6 14.6 ± 1.6 ft (Yellow | +I/Blue (-]) (*) -4.3-1.5 5.5-14.9 -7.1 -0.9 ± 1.9 9.5 ± 2.9 Dominant Wavelength Inm) 487.2-652.1 588.9-622.7 541.0 520.7 ± 47.7 602.5 ±11.2 and bill shape. The restrictive hypothesis port for the hypothesis (Graves 1990. then was tested with a quantitative analysis Graves & Zusi 1990). of size and external proportions. Concor- dance of results is regarded as strong sup- Results and Discussion Plumage characters,—Salient characters Table 2.—Ranges and means (±standard deviation) of the hybrid that permit its parental species of measurements (mm) of males of Agkiiinvnus kingi to be identified include: (a) moderately caudatus isuhdefinitive plumage, see Appendix 2). elongated outer rectrices (fork depth = 23.7 Metatlura lyritintfiino oreapola. and their presumed hvhrid(AMNH 146645). mm), nearly flat in cross section; (b) un- marked rectrices exhibiting metallic irides- M (ynnnttiiint cence on the dorsal and ventral surfaces; (c) Chaiaurr (n »3t) Ml HyhnJ short tibial plumes (not extending to hal- Wing chord 59,7-63.6 54.0-61.5 61.4 lux); and (d) short straight bill (11.1 mm). 61.8 ± 1.0 57.7 * 2.2 Two species in the pool of potential paren- Bill 11.0-13.6 9.1-10.6 11. 1 tal species (Appendix 1) possess elongated 12.3 T 0.7 9.9 ± 0.4 tails (length of rectrix 5 > 55 mm): Ocrea- Rectrix 1 22.8-25.3 33.7-38.9 32.1 24.1 r 0.8 36.1 4 16 tus undenvoodii and Aglaiocerctis kingi. Rectnx 2 29.0-33.6 34.3-41.3 36.9 Ocreatus can be deleted from the list of 31.3 ± 1.2 37.9 + 1.8 possibilities because the hybrid lacks evi- Reclrix 3 38.9-46.7 36.0-43,6 43.1 dence of spatulate rectrices or lengthened 42.8 ± 1,9 39.9 ± 1.8 tibial plumes. Aglaiocerctis kingi is thus Rectrix 4 53.7-61.5 38.4-45..; 49.0 identified as one of the parental species. 58.0 ± 2.2 41.5 ± 2.1 Determination of the other parental spe- Rectrix 5 675-102.7 37.9-46.3 55.8 cies is equally straight forward. The inten- 81.5 ± 8.7 42.: : 2 7 sity of the metallic iridescence reflected VOLUME I I I, NUMBER 3 515

600

20 25 30 -10 -5 0 5 10 15 lightness (L) of Back Pkjmane Yellowness (b> or Bkjeness t-b) ol Ventral Surlace or Rectriees Fig. 3. Bivariaie plots o( spec trap hotometric data from male hummingbirds: Agtaiocercus kingi amdatuf (circles): Metailura tyrianthina oreopola (diamonds); and a probable hybrid. A. kmgi caudatus x M. tyrianthina orcopnia (triangle: AMNH 146645). from the ventral surfaces of the hybrid's 0.15u. thick. Briefly summarized, they rectriees is matched or exceeded onJy in found that granules contain a fairly uniform Metailura tyrianthina. Details of plumage layer of gas-filled vacuoles that resemble a pattern and feather shape are sufficient to monolayered foam. The melanin matrix and suggest that the parentage of the hybrid is gas-filled vacuoles have refractive indices Aglaiocercus kingi X Metailura tyrianthina of ~2.0 and 1.0, respectively. The color of (see Appendix 2). None of the other species iridescence varies according to the thick- in Appendix ], considered two at a time, ness of the granule and the amount of gas can account for the characters observed in in the vacuoles. Iridiscent colors change the hybrid. In particular, the ventral rectri- from blue to green to orange and finally to cial surfaces of the hybrid are metallic red- red, as the effective refractive index of dish-purple as opposed to dull black or pur- granules advances from 1.45 to 1.90 (figure plish-black in both Ramphomicron micro- 4 of Greenewalt et al. 1960a). Melanin rhynchum and A. kingi, effectively elimi- granules in nontridescent parts of feathers nating this pair of species from contention. lack vacuoles. The question of plumage color.—Irides- The pattern of bluish-green iridescence in cence in hummingbirds is caused by the in- the hybrid corresponds precisely to that of terference of light reflected from the upper green iridescence in the parental species, and lower surfaces of gas-filled vacuoles in suggesting a single mutational or develop- melanin granules in the keratin of feather mental aberrancy that affects plumage col- barbules, which are compactly stacked in or. The dominant wavelength reflected from 7-15 layers in the barbule keratins (Dorst dorsal plumage is shorter in the hybrid (511 1951; Greenewalt et al. 1960a. 1960b; Lu- nm) than in the parental species: Aglaio- cas & Stcttenheim 1972). Carotenoid pig- cercus kingi (553-564 nm) and Metailura ments have not been extracted from irides- ryrianthina (563-571 nm) (Table I, Fig. 3). cent feathers. Employing transmission The premise that "hybridization produces electron microscopy and micro-speetropho- no traits characteristic of genera or species tometry, Greenewalt et al. (1960a. 1960b) other than those involved in the particular found melanin granules to be elliptical in cross" (Banks & Johnson 1961:3) was ex- shape, about 2.5u, long, 1.5u. wide, and tended to spectropho to metric measures of 516 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

30 10 50 BO Length Reclrjx 4

Fig. 4. Bivariale plots of mensural characters of male hummingbirds: Aglaiacernis kingi caudtwm (triangles); Merallurti ivrUmthina orenpiihi (circles); and a probable hybrid, A. kingi cmtdatUS x M. lyrkinthiiiLi ottXrpota (diamond; AMNH 146645). Least squares regression lines are projected for comparison. VOLUME 111. NUMBER 3 517 plumage color {Graves 1996b). Both as- ues predicted by least squares regression on sumptions are violated in the present ex- bivariate plots (Table 2, Fig. 4). In sum- ample. mary, plumage pattern, distribution and in- Elucidation of the micro-structure and tensity of iridescence, rectrix shape, and spectrophotometric properties of melanin mensural characters provide strong support granules in Aglaiocercus kingi, Metallura for the hypothesis of hybridity (Aglaiocer- tyrianthina. and the hybrid is beyond the cus kingi X Metallura tyrianthina). scope of this paper. Several observations on Previous records.—A hybrid of Aglaio- hummingbird plumages, however, are cercus kingi and Metallura tyrianthina was worth noting. A variety of plumage aber- reported once before by Meyer de Schauen- rations, unassociated with hybridization, see (1947:108), who described a specimen has been observed in hummingbirds, in- (No. 134) obtained in Bogota, circa 1909, cluding leucism, albinism, schizochroism, from the Brother Niceforo Maria collection: erythrism, and melanism (Salvin 1892. "... fore-crown glittering brassy green, hind crown Banks & Medina 1963, Greenway 1978, and back dark bluish green, rump and upper tail Graves 1998b), Subtle within-population covens bluer; chin dusky, throat patch shaped as in variation in iridescent color is commonly Metallura tyriemthlna but blue instead of green; observed whenever large series of species breast dark bluish green, bases and edges of the are assembled. Although post-mortem ef- feathers huffy; belly dark green, the bases of the feathers white, showing through and giving a some- fects may be partially responsible in some what barred appearance; tail purple, deeply forked, cases (Graves 1986, 1991), most of the ob- the outermost tail feathers 50 mm., the central ones served variation in iridescence among in- 30 mm., wing 61 mm., culmen 12,5 mm." dividuals, factoring out the effects of sex The brief description of Niceforo's spec- and age (see Bleiweiss 1992), is due to ge- imen differs in minor details from the Ven- netic and developmental factors. Pro- ezuelan specimen (AMNH 146645). The nounced color shifts of the magnitude ob- two specimens are similar in size. Nicefo- served in this hybrid are rare but not un- ro's specimen possesses a bluish gorget as known (e.g., Salvin 1892, Greenway 1978). might be expected in an adult male hybrid The example described here seems to be the of Metallura t. tyrianthina and Aglaiocer- first in which a hybrid hummingbird exhib- cus k. kingi from the Cordillera Oriental of its a major plumage aberrancy. the Colombian Andes. Whereas I charac- External measurements.—One of the terized the back color of the Venezuelan guiding principles of hybrid diagnosis is specimen as "greenish-blue," Meyer de that hybrids are not larger or smaller than Sehauensee used the term "bluish-green" their parental species (Graves 1990). Mor- for Niceforo's specimen. This and other dis- phological luxuriance or dwarfism in hybrid crepancies might reflect semantics or real hummingbirds has not been recorded. Male differences in color. Unfortunately, the Aglaiocercus kingi and Metallura tyrianthi- whereabouts of Niceforo's specimen is un- na are similar in bill length (cumulative known, although another mentioned in range, 9,1—13.6 mm) and wing chord (cu- Meyer de Schaucnsee's paper was deposit- mulative range, 54.0—63.6 mm), but differ ed in the Academy of Natural Sciences of markedly in tail size and shape (Table 2, Philadelphia (Niceforo no. 148, now ANSP Fig. 4). Bivariate plots of rectrix length of 159261; Graves 1993). the parental species exhibit positive (1 vs. 2) or negative (1 vs. 3, 1 vs. 4, I vs. 5) Acknowledgments allometry. Except for rectrix 3, measure- ments of the hybrid fall between the char- I thank Richard Banks, Robert Bleiweiss, acter means for A. kingi and M. lyrianthina, Kenneth C. Parkes, and Richard Zusi for and, in several cases, approximate the val- critiques of the manuscript. I thank the cu- 518 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON rators and staffs of the American Museum nemix soderstromi Butler,—Proceedings of the of Natural History, New York, the Field Biological Society of Washington 109:764-769. Museum of Natural History, Chicago, and . 1996b. Hybrid wood warblers, Dendrr/ica stiiata X Dendroicu castaneu (Aves: Fringilli- the Academy of Natural Sciences of Phil- dae: Tribe Parulini) and (he diagnostic predict- adelphia, for permitting me to examine ability of avian hybrid phenotypes.—Proceed- specimens in their care and for specimen ings of the Biological Society of Washington loans. Photographic prints were provided 109:373-390. by Smithsonian photographic services. Mu- -, 1998a. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 5. Probable hybrid origin seum work was supported by the Alexander of Amtizititi distant Wetmore & Phelps.—Pro- Wetmore Fund and the Department of Ver- ceedings of the Biological Society of Washing- tebrate Zoology, National Museum of Nat- ton 111:28-34. ural History, Smithsonian Institution. , 1998b. Taxonomic notes on hummingbirds (Aves: Trochilidae). I. Erincneinis dyxetius El- liot. 1872 is a melanislic specimen ai Uriocnem- Literature Cited is cupreeventris (Eraser. 1840).—Proceedings of the Biological Society of Washington 111: Banks, R, C, & N. K. Johnson. 1961. A review of 420-424. North American hybrid hummingbirds.—Con- , & R. L. Zusi. 1990. An intergeneric hybrid dor 63:3-28. hummingbird iHeiiodoxis leitdbeateri X HelLm- , & D. R Medina. 1963. An albinistic Anna gehts ameihysiicollix) from northern Colom- Hummingbird.—Condor 65:69-70. bia.—Condor 92:754-760. Bleiweiss. R. 1992. Reversed plumage ontogeny in a Gray. A. P. 1958. hybrids. Commonwealth Agri- female hummingbird: implications for the evo- cultural Bureaux. Bucks. England, 390 pp. lution of iridescent colours and sexual dichro- Greenewalt, C. H.. W. Brandt. & D. D. Friel. 1960a. m at ism.—Biological Journal of the Linnean So- Iridescent colors of hummingbird feathers.— ciety 47:183-195. Journal of the American Optical Society 50: . J. A. W. Kirsch, &. J. C. Matheus. 1997. DNA 1005-1016. hybridization evidence for the principal lineages . 1960b. The iridescent colors of hummingbird of hummingbirds (Aves: Trochilidae).—Molec- feathers.—Proceedings of the American Philo- ular Biology and Evolution 14:325-343. sophical Society 104:249-253. Dorel, J. 1951. Recherches sur la structure des plumes Green way. J. C, Jr. 1978. Type specimens of birds in des Trochilides.—Memo:res du Museum Na- the American Museum of Natural History. Pan tional D'Histoire Nalurelle (Serie A. Zoo logic) 2.—Bulletin of the American Museum of Nat- 1:125-260. Goldsmith. T. H.. & K. M. Goldsmith. 1979. Discrim- ural History 161:1-305. ination of colors by the black-chinned hum- Hinkelmann. C. 1996. Evidence for natural hybridi- mingbird. Archilochus alexaridri.—Journal of sation in hermit hummingbirds iPhaeiftornia Comparative Physiology A I 30:209-220. spp,).—Bulletin of (he British Ornithologists' Grant, P. R . & B. R. Grant. 1992. Hybridization of Club 1 16:5-14. bird species.—Science 256:193-197. Hunter. R. S., & R. W. Harold. 1987. The measurement Graves. G. R. 1986. Systematics of the Gorgeted of appearance. 2nd edition. Wiley, New York, Wood stars (Trochilidae: Aceiirara).—Proceed- 41 I pp. ings of the Biological Society of Washington Lucas, A. M . & P. R. Steltenheim. 1972. Avian anat- 99:218-224, omy. Integument, Part 2.—United Slates De- . 1990. Systematic: of the "green-throated sun- partment of Agriculture. Washington. DC, Ag- angels" (Aves: Trochilidae): valid laxa or hy- ricultural Handbook 362:341-750. brids'.'—Proceedings of the Biological Society Meyer de Schauensee, R. 1947. New or little-known of Washington 103:6-25. Colombian birds.—Proceedings of the Acade- . 1991. Taxonomic status of the Sword-billed my of Natural Sciences of Philadelphia 99:107- Hummingbird (Ensifera ensifera caerules- 126. cenx),—Bulletin of the British Ornithologists' . & W. H. Phelps. Jr. 1978. A guide to the birds Club I 11:139-140. of Venezuela. Princeton University Press. 424 . 1993. Relic of a lost world: a new species of pp. sunangel (Trochilidae: Heliangelux) from Bo- Ortiz-Crespo, F I. 1972, A new method to separate gota—Auk 110:1-8. immature and adult hummingbirds.—Auk 89: . 1996a. Diagnoses of hybrid hummingbirds 851-857. (Aves: Trochilidae). 2, Hybrid origin of Erioc- Panov. E. N. 1989. Natural hybridisation and etholog- VOLUME 111. NUMBER 3 519

ical isolation in birds tin Russian). Nauka, Mos- retain a few strialums on the maxillary rampholhecum. cow, 510 pp. a character usually interpreted as a sign of immaturity Phelps. W. H.. & W. H Phelps. Jr. 1958, l.ista de las iOrti/-Crcspo 1972), The descriptions of Agtaiocerciu aves de Venezuela con su distribution. Tomo 2. kingi given below refer to the subdefinitive plumage. Pane I. lidilorial Sucre. Caracas, 317 pp. Descriptions of structural colors arc unusually sub- Prager, E. M,. & A. C. Wilson. 1975. Slow evolution- jective, as color seen by the observer varies according ary loss of lhe potential tor interspecific hybrid- to I he angle of inspection and direct ion of light. Tor ization in birds: a manifestation of slow rcgu- this tea son I use general color descriptions. I.Hi.iv evolution Proceedings or llic Virtual The dark bluish-green crown of young kingi is re- Academy of Science USA 72:200-204. placed tfrom anterior to posterior) by an ovate crown Salvin. O 1892. Catalogue of the birds in the British patch composed of brilliant bluish-green feathers. The Museum, Vol. 16, t-ondon. 703 pp. hindncek, back, and rump arc dark green: feathers arc Sibiey. C. G. & B. L. Monroe. Jr. 1990. Distribution gray, tipped with green. Upper-tail coverts are bluish and taxonomy of birds of the world. Yale Uni- green, A few while leathers form an iudisiinei patch versity Press. New Haven, Connecticut, 111] on the lower hack. pp. The dorsal plumage of tyriantbina is dark dusky Weller, A.-A., & K.-L. Schuchmann. 1997. The hybrid green, brighter on the crown, and with coppery high- origin of a Venezuelan Trochilid, Amazilia dis- lights on the lower back and rump, heathers are gray, tant Wet more & Phelps 1956.—Omithologia handed subterminally with coppery-green, and tipped Neotropical 8:107-112- broadly with dark green When viewed head-on in di- Wilkinson. I... 1989. SYSTAT I he system for statistics. rect light, plumage posterior to the miderown region SYSTAT, Inc.. Evanston, Illinois. 822 pp. appears sooty black. Immature tyriantbina lack a con- Zimmer. J. T 1952. Studies of Peruvian birds, No. fi2 trasting rump parch The hummingbird genera Patagana, Suppho. Under a diffuse light source, the dorsum of the hy- Polyonytmis, Ramphomlcron, Metollura, Chat- brid is a rich greenish-blue (paler on the crown), a cestlgma, Taphrolesbia, and Aglaiocercus.— color that is distinctly different from that of the pre- American Museum Novilates 1595:1-29. sumed parental species. Feathers on the left side of the 2usi, R. 1... & O. D. Bentz. 1982. Variation of a muscle foreerown are discolored, possibly by a preservative in hummingbirds and swifts and its systematic chemical. Dorsal feathers are dark gray, tipped with implications.—Proceedings of the Biological greenish-blue. Crown feathers are not modified as in Society of Washington 95:412-420. adull kingi. When light is reflected obliquely (>9() from the observer), the dorsal plumage appears purple; Appendix 1 when viewed head-on the hindcrown. back, and rump appear black, A few rump feathers are tipped with Species of hummingbirds that occur in Esiado Me~- huff rida, Venezuela: Campylaptenu falaOus, Collbri ihat- The ventral plumage of kingi is medium green ex- assirtUS, C niranuni, Kltiix guimeii, Lophnrnis dclat- hibiting subdued iridescence. A lew small shining tret. L. stictolophus, ChlerfStts nouttuSt Chlomstilbon green disks occur on the throat of more mature indi- tnettisugus, C. poormumi, Thaluronia fitrcota, H\la- viduals. The harhs of ventral feathers are narrowly chaiis cyitnus, Chrysuronia oenone, Amazilia versi- tipped with buff or grayish-buff, especially along the color, A. fimhtiala, A. viridiguxler. Chalyburii btiffimii. m id I me of the abdomen. Some males in juvenile and Heliodoxa letidbeateri. Stemoctyta cyanopectux, Crte- subdefinitive plumage (e.g.. AMNH 484067) have a ligena coeiigena, Ocreatus underwootSi, Aglaiocercus white or buffy-whitc line extending from the base of kingi, Ueliomuswr Umgiruxtrix. ('ii<.irloii-rtu\ jintrda- the hill posterior to below the eye. Downy vent feath- nii. ers are dark gray tipped with while or pale gray. Un- deiiail covens are dark green broadly edged with buff. Appendix 2 Tibial feathers are short (extending half way to rhe Comparative description of plumages of male hallux from the tibiotarsal joint), dark olive-gray and Agluiocrrrus kingi ctitidtitus, MetaUuru lyritinthittu or- narrowly tipped with scattered grayish-buff barbs. eopota, and their presumed hybrid. AMNH 146645 The venter of tyritmthitut is dark dull green: feathers The molts and plumages of male Aglaiocercus spp. are are lipped with buff or grayish-brown, especially aiong incompletely known. Young males f>6 months?) ac- the midline. Feather tipping imparts a mottled appear- quire a plumage that differs from the definitive plum- ance to the underparts, A narrow ovate gorgcl extends age of adull males. This subdefinitive plumage is char- from the chin to the upper breast in tyrianthinn. When acterized by shorter outer reetrices, an incompletely view head-on in direct light, the nuriculars and sides developed crown patch (8 of 20 examined), and rem- of the throat appear matte black and contra si greatly nants of a while rump patch {sec Zimmer 1952). One with the brilliant green gorget, Suhadull males have a quarter (5 of 20) of the males in subdefinitive plumage huffy line extending from the bill to below the eye. 520 PROCI-l-DINCiS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

Vent feathers are dark pray tipped with while or pale The tail of kingi is deeply forked. The basal portions gray. Under-tail coverts are buff with a large broadly of the reel rices that are obscured in the folded tail are lanceolate spot (coppery- or bronzy-red) along the black. The exposed dorsal sections of the inner reclri- midline. Tibial feathers (dark gray lipped wiib buff) ces (1-4) are deep brilliant purple tipped wilh hluish- extend to ihe bus of the hallux. green. The outermost rectrices (5) lack bluish-green The venter of 'be hybrid is similar in pattern 10 kin- tips, are great fy elongated, narrow (5-6 mm wide, 25 %i, but with the green portions replaced with bluish- mm from lip), and bowed in cross-section. Inner rec- green (paler than do is u ml. A small brilliant leather irices are smoothly tapered; rectrix 5 is bluntly tipped. occurs along (he midline at mid throat (deep blue up Venlrally. the vanes arc dull purplish-black. The ra- separated from (he gray ba.se by narrow blending bands chises in fdngi are blackish-brown dorsally. medium of light blue and coppery-gold). A buffy-white stripe brown venlrally. extends from the bill to below the eye (Fig. 2). Vent The tail of tyriamhina is shallowly forked. Rectrices feathers are dark gray lipped with while. I'ndertail co- are wide (10-! I mm), nearly flat in cross-section, verts are hull with a lanceolate subterminal .spot (pur- abruptly truncate at the lip. and metallic coppery-red above and below. Raehises are dark brown above and ple) near the midline. Tibial leathers (dark gray broad- below, ly tipped with bull) of the hybrid extend about halfway The tail of the hybrid is moderately forked. Feather to the hallux, but may have been damaged by knotting size and shape are intermediate between kingi and r.v- of the specimen label string. rhmihina. The outermost rectrices (5) are slightly The remiges of Aimfi are brownish-black faintly tint- bowed in cross-section (ca. 9.1 mm at widest point). ed with purple. The outer vanes of the primary coverts Rectrices are metallic reddish-purple, above and be- and the innermost secondaries are edged with shining low, the innermost < 1-2) are diffusely tipped with pur- green or bluish-green: secondary coverts are broadly ple. This iridescence, especially from the ventral sur- lipped with bluish-green. The remiges of ryriumhina faces, is similar in visual essence lo thai of tyriumhiiui. are very similar in color bul faintly linled wilh bronze Raehises are dark brown above, medium brown ven- or olive. Wing coverts and the innermost secondaries lrally. are bronzy-green. Bill color is black in kingi, tyrhwthiiut. and hybrid. The remiges of the hybrid resemble those of kingi. In dorsal profile, the bills of both parental species are Wing coverts and innermost secondaries are purple abruptly tapered, more so in kingi. The hill profile of tipped with dark bluish-green. (he hybrid is similar to thai of tyriantliinti.