21 March 19V0 PROC. BIOL. SOC. WASH. 103(1). 1990, pp. 6-25 SYSTEMATICS OF THE "GREEN-THROATED " (AVES: TROCHILIDAE): VALID TAXA OR HYBRIDS?

Gary R. Graves

Abstract. — Four of", Heliangelus squamigularis Gould, 1871, Heliangelus barrati Mulsanl & Vcrreaux, 1872, Heliotrypha speciosa Salvin, 1891, and Heliotrypha simoni Boucard, 1892, were described from 19th century commercial "Bogota" collections. The systematic status of these taxa, which I collectively refer to as "green-throated sunangeis" (GTS), is unresolved, but they have been variously treated as one or more valid species or as hybrids. I examined three systematic hypotheses—thai GTS specimens represent (1) one or more valid species; (2) genetic variants of other species; or (3) hybrids. Plumage and mensural characters of GTS (n = 14) suggest they represent hybrids of Heliangelus amethyst icollis x Erioenemis cupreoventris from the Eastern Cordillera of the Colombian . Alternate hypotheses of hybridity are discussed. Discrimination of hybrids and identifying their parental species depends upon an efficient "hybrid diagnosis." The current method of diagnosing hybridity is often insufficient in that the materials, methods, and results are not properly documented. 1 address these problems and suggest guidelines for hybrid diagnoses.

Untold thousands of skins of 19th century specimens. Of more general were exported from northwestern South interest, I examine the assumptions, ma- America in the 19th century for the milli- terials, and methods of the hybrid diagnosis nery trade and collectors of natural history in avian . specimens. Systematists sorted through some of the massive shipments of "Bogota" Taxonomy of the "Green-throated trade skins and described dozens of new Sunangeis" species, a few from unique specimens (e.g., Boucard 1892). Most were subsequently Four species of hummingbirds, that I col- verified by the discovery of populations; lectively refer to as "green-throated sun- others were determined to be of hybrid or- angels" (hereafter abbreviated as GTS), were igin (Berlioz & Jouanin 1944). However, the described from 19th century "Bogota" col- validity of more than a dozen taxa remains lections: Heliangelus squamigularis Gould, indeterminate (Morony et al, 1975). These 1871; Heliotrypha banalI Mulsant & Ver- represent some of the most challenging reaux, 1872; Heliotrypha speciosa Salvin, problems in avian taxonomy. Resolving 1891; and Heliangelus simoni Boucard, their systematic status depends on the me- 1892. Heliotrypha Gould. 1853 is now con- chanics of discriminating avian hybrids from sidered a junior synonym of Heliangelus valid biological species. Gould, 1848. This paper has two aims that are ad- Taxonomic uncertainty within the group dressed concurrently. I evaluate the system- began with Gould (1871). who was initially- atic status of an enigmatic group of hum- inclined to consider the type specimen of mingbird taxa known only from a handful H. squamigularis a sport or variant of some VOLUME 103, NUMBER other Heliangelus species, but who after fur- plied that the description was based on two ther investigation characterized it as a new or more syntypes. Cory (1918) and Lord species related to Heliangelus c.xonis and Rothschild (Hartert 1922) each obtained a //. amethysticollis. H barrali and //. squa- syntype of H. simoni, presumably from migularis were not compared with one Boucard before his death in 1905. The la- another before being described. Salvin bels of both specimens (FMNH 46294. (1892) considered these taxa as identical. AMNH 483683) are marked, "Heliotrypha but distinct from his newly described //. simoni. £, Typical specimen. ." in speciosa. In the first review of all four taxa, what appears to be Boucard's handwriting Boucard (1895) followed Salvin's synony- (fide Greenway 3 978). Simon (1921) stated my of//, barrali and H. squamigttiaris, but in a footnote that he could not find the type treated //. simoni (Boucard 1892) and H. of//, simoni in Boucard's collection, which speciosa as valid species, while noting the suggests that only two syntypes existed, both possibility that both were varieties of H, of which were sold or exchanged by Boucard squamigttiaris. Cory (1918) lumped //. spe- to other museums, Berlioz (1936), however, ciosa and H. simoni and initialed the two- argued that a specimen in the Boucard Col- species taxonomy for the group adopted by lection labeled "Heliotrypha speciosa. 6, Co- Simon (1921). Peters (1945), and provi- lombia" (now deposited in MNHN. Paris), sionally by Morony et al. (1975). Hartert was in (act the type and only existing spec- (1922). perhaps cued by Boucard (1895), imen of //. simoni (sec Berlioz & Jouanin proposed that H. simoni and H. speciosa 1944. Jouanin 1950). He apparently based were aberrations of a single valid GTS his conclusion on the close resemblance of species (-H. squamigularis). The possible the specimen to Boucard's description of H, hybrid origin of GTS was first raised by simoni. and on the fact that Boucard (1895) Berlioz (1936), who suggested thai a spec- did not mention possessing a specimen of imen in Paris, which had previously been //, speciosa. identified as //. simoni (discussed later), Several explanations are possible for this represented a hybrid of Heliangelus e.xortis discrepancy. Assuming that there were orig- x Haphphaedia aitrcliae. This opinion was inally three examples of//, simoni. Boucard endorsed by Jouanin (1950) and Greenway may have attached a new label to the re- (1978), but Berlioz & Jouanin (1944) were maining syntype after disposing of the other less conclusive, stating simply that GTS were two. Other possibilities arc that Boucard hybrids between either //. exortis or //. (1895) obtained the specimen of//, speciosa amethysticollis and some species of luioc- after the publication of his monograph or nemis (including the closely related that the specimen was labeled by Boucard Haphphaedia), Meyer dc Schauensee (1949) as //. speciosa before he described //. simoni at first doubted the notion of hybridity but and never relabeled afterwards. In any event, later agreed with Berlioz & Jouanin (1944) only specimens designated by Boucard Mu- and supposed that GTS were hybrids of Er- seum labels as "H. simoni" should be re- iocnemis sp. * Heliangelus sp. (Meyer de garded as syntypes. Schauensee 1966). Hilly & Brown (1986) listed H. speciosa and H. squamigularis as Materials and Methods presumed hybrids without mentioning pa- The type specimens of GTS arc deposited rental species. in three different museums and are not available for loan. This prevented me from Syntypes of Heliotrypha simoni comparing all type specimens simulta- Boucard's (1892) use of the plural "spec- neously. 1 examined three type specimens: imens" in his description of//, simoni im- (1) the type of Heliotrypha barrali (AMNH PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

Fig. 1. Ventral view of "green-throated sunangels" deposited in the American Museum of Natural History (from left to right; 483680, 483681, 483678,483684, 483682, syntype of "Heliotrypha simoni" [483683], type oVHeUotrypha barralr [37655]).

37655) and (2) a syntype of Heliotrypha si- tional D'Histoirc Nalurellc (MNHN). Paris; moni (AMNH 483683) in the American the types of H. squamigularis (BM Museum of Natural History (see Hartert 88.7.25.178) and H. speciosa (BM 1922, Greenway 1978): and (3) a syntype 87.3.22.889), and an unnumbered speci- of Heliotrypha simoni (FMNH 46294) in men of//, harraii (photographs only) in the the Field Museum of Natural History. The British Museum of Natural History (see Ap- AMNH types were compared directly with pendix), GTS specimens and color transpar- five additional specimens (AMNH 483678, encies were compared with series of all 483680,483681,483682.483684) that have hummingbird species in the National Mu- been variously identified as one or more of seum of Natural History (USNM) and the the GTS taxa (Fig. I). The FMNH type was American Museum of Natural History. In compared directly with FMNH 46286 addition to the GTS specimens examined {identified as //. barraii on its Boucard Mu- in this study, at least one other specimen seum label). I examined one additional GTS exists (Berlioz 1964). Color comparisons of specimen (ANSP 160344 [formerly AMNH specimens were made under Examolites® 483679]}, These specimens were compared (Macbeth Corp.). with mensural data and color transparencies Measurements (wing chord, tail from in- of the following specimens: one labeled H. sertion of centra] rectrices to tip ofthe outer speciosa (considered by Berlioz [1936] as a and innermost rectrices, and culmen from syntype of//, simoni) in the Museum Na- anterior extension of feathers) were taken VOLUME 103, NUMBER

Table 1. —Measurements (mm) of "green-throated sunangels.'

Outermost **f Wing chord ,.::;. Central iretm < 'uinicn

TVpe specimens (1) sqitamigittam (BM) adult 63.5" 43.2- 33.0" 16.6' (2) barruli (AMNH) immature 63.1 42.8 34.5 17.3 (3) speciosaOlM) adult 63.5- 40.6- 36.8" 19.4' (4) .«'mom'(AMNH) immature 60.4 43.4 33.6 18.5 (5) simoni (FMNH) adult 64.5 43.3 34.7 16.9 Other specimens (fi) "sppntmT (MNHN) adult 64.0" 42.5" 32.0" 19.0* (7) FMNH 46286 immature 64.5 43.9 35.7 16.8 (8) AMNH 48367 = adult 64.6 43.9 35.9 17.1 (9) AMNH 483680 i in in ,II u re 64.5 43.8 35.6 15.7 (10) AMNH 483681 adult 63.7 414 33.6 16.3 (11) AMNH 483682 adult 56.5 — — — (12) AMNH 483684 adult 62.3 40.8 — 16.8 (13) ANSP 160344 immature 60.5 39.8 33.6 16.8 • Immaturcs have corrugations on the ramphothecum of the upper jaw. " Measurement from Salvin 1892. ' Measurement courtesy of J. Becker. " Measurement courtesy of C. Jouanin, with digital calipers and rounded lo the card 1895, Cory 1918, Peters 1945. Morony nearest 0.1 mm (Table 1). et al. 1975). I used principal components analysts Do green-throated sunangels represent rare (PCA) on unlransformed variables to re- genetic variants of other species?—-Several duce the dimensionality of data and lo fa- examples of inlra-population variation in cilitate the analysis of morphology in two plumage are known in Heliangelus species. dimensions. Unrotated principal compo- Polymorphism in the number of iridescent nents were extracted from correlation ma- gorget feathers in females has been well doc- trices (SYSTAT). umented, particularly in Heliangelus exortis (Chapman 1917; Zimmer 1951; Bleiweiss Systematic Status of 1985a. b). and melanism involving part or Green-throated Sunangels the entire plumage is known in a number Investigations of GTS have engendered a of Andean genera (Hartert 1922, Greenway remarkable variety of systematic opinions. 1978, Graves, pcrs. obs.). Intrasexual color If nothing else, this strongly suggests that polymorphism, however, does not appear multiple hypotheses of origin must be ad- to be significantly correlated with size. GTS dressed. Accordingly. I considered three closely resemble some species of Helian- possibilities. GTS may represent one or gelus (e.g., H. exortis), but differ in body more of the following entities: (1) rare ge- proportions from all species and by having netic variants of other Heliangelus species lengthened libiat plumes and a green or sil- (Gould 1871): (2) hybrids (Berlioz 1936; very-green gorget in combination with bril- Berlioz & Jouanin 1944: Meyer deSchauen- liantly reflective plumage on the posterior see 1949. 1966; Jouanin 1950; Green way part of the body. These qualitative char- 1978; Hilty & Brown 1986), or (3) popu- acters indicate that GTS are not plumage lation samples of one or more valid biolog- variants of any other species of humming- ical species (Gould 1871. Salvin 1892. Bon- . II) PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

Hybrids or species?— As demonstrated by hybrids? How were characters defined and a century of equivocal taxonomy, it is dif- apomorphies identified? ficult to determine whether GTS are hybrids 3. Documentation of results: Can the hy- or valid species. This is due primarily lo pothesis ofhybridity be rejected? If not, how two factors. GTS specimens were collected were the parental species identified to the in the 19th century and are unaccompanied exclusion of all others? How were alternate by ecological, sexual, or locality data. They hypotheses (e.g., valid taxon: genetic or de- are believed to have originated from the velopmental variant) rejected? Andean region of northwestern South America, an area of high species diversity Hybrid Diagnosis where new species of hummingbirds are still being discovered (e.g.. mxrabi- Potential parental species.— For any hy- lis). The large number of GTS specimens (n brid of unknown parentage, the pool of po- •» 15), presumably collected in a biotically tential parental species (species hypotheti- diverse but poorly-known region, favors the cal ly or actually available for hybridization) valid species hypothesis. On the other hand, can be defined taxonomically and geograph- the plumage color and morphology of GTS ically. Inierordinal hybridization is un- are variable and intermediate between sun- known in (Gray 1958); interfamilial angels (Heliangelus) and pufllegs {Erioc- hybridization has been reported in captivity nemis and Haplophai'did). This suggests that (e.g., turkey x guinea fowl) but is unknown hybridization is involved. As hybrids have in nature. Thus, the taxonomic pool can be no standing in zoological nomenclature, the narrowed considerably if the hybrid can be burden of proof is on taxonomists to reject identified to a particular family-level group the hybrid origin of GTS conclusively be- (e.g., hummingbird or duck), which is al- fore conferring species status on them. ways the case. The taxonomic pool may be The process of discriminating avian hy- further restricted to a subfamily, genus, or brids and their parental species can be a single pair of species when the rationale termed the "hybrid diagnosis," Most tax- for doing so can be vigorously supported. onomists consider the pathways of hybrid For example, Parkes (1984) properly re- diagnosis to be self-evident and the docu- stricted the pool of potential species of a mentation of methods and diagnostic as- hybrid cuckoo collected in Pennsylvania to sumptions to be unnecessary. However, the only pair of Coecyzus species that occur omissions of these crueial data obscure the sympatrically in North America north of diagnoses of all but the most obvious cases the Gulf coast. In the interest of compre- ofhybridity. Beyond calling attention to an hensiveness, however, he could have also "unusual" specimen, a hybrid report based addressed the six other species of Coccyzus on an incomplete diagnosis is of little value in a few sentences in much the same way a to taxonomists and evolutionary biologists. taxonomist would mention other species in As a minimum, the following points (not the differential diagnosis of a new species. mutually exclusive), should be explicitly ad- As suggested by the cuckoo example, the dressed in hybrid diagnoses. pool of potential parental species can be 1. Potential parental species: What species limited geographically. The degree of limi- were considered as possible parental species tation depends on knowledge of the migra- and why? tory habits of the potential parental species 2. Diagnostic assumptions of character and the geographic origin of the hybrid. For analysis: What operational assumptions instance, the taxonomic pool of potential were made concerning the inheritance of parental species of a hybrid hummingbird plumage and morphological characters of is defined by the family Trochilidae (345 + VOLUME 103, NUMBER 1 II species). A hybrid hummingbird originating C. anna), but is lacking in the single known from Arizona could have no more than 20 C. anna x Archilochus aicxandri hybrid potential parental species (190 species com- (Banks & Johnson 1961). This suggests either binations). On the other hand, a hybrid from that few genes control the color of coronal an unspecified area of northwestern South plumage or that phenotypic expression is America could have 150+ potential paren- controlled by a modifer in these species. The tal species (10.440 species combinations). shape of gorget feathers, rectrices, and re- Clearly, the difficulty of hybrid diagnosis is miges of hybrids is generally intermediate directly proportional to taxonomic species between those of the parental species, re- diversity and geographic scope. flecting a polygenic mode of inheritance. Diagnostic assumptions of character anal- Banks & Johnson (1961) assumed that ysis.—-In diagnosing putative hybrids. 1 as- hybridization in hummingbirds does not sumed that mensural characters, such as produce trails of species or genera other than wing and bill length, were poly genie and those involved in the particular cross. This additive and that the morphology of hybrids assumption rules out the possibility of ata- does not exceed that of the parental species vistic characteristics—those not found in (Falconer 1981). Plumage characters in hy- either parental species but which reflect a brids may resemble a mosaic of the parental pattern postulated to be ancestral or the re- species or be inherited intact from one par- sult of mixed alleles encoding polygenic ent, depending on the number of encoding traits. Although atavism is well known in genes and their interaction (Hull 1949. certain anseriform hybrids (e.g., Harrison &. Buckley 1982), Hypothetically, hybrids may Harrison 1963), it has not been documented exhibit a wide range of plumage phenotypes. in hybrid hummingbirds. The major pigments in bird plumage, mel- This study was geared toward the iden- anins, carotenoids. and porphyrins, appear tification of apomorphic character states in to be under separate genetic control and mu- putative hybrids. The mosaic expression of tually independent between feather tracts. parental autapomorphies in a number of The inheritance of structural colors, which characters is the best indicator of hybridity dominate the plumage of hummingbirds, is of a unique specimen and provides the only poorly understood (Fox & Vevers 1960, Lu- direct evidence of parentage. However, be- cas & Stettenheim 1972), but the complex- cause many plumage characters are poly- ity of color-producing structures suggests a genic, the expression of parental apomor- polygenic mode of inheritance. phies may be obscured in hybrids. When What little is known about inheritance in parental apomorphies are not identifiable, hybrid hummingbirds is summarized by the parentage of a hybrid may be indicated, Banks & Johnson (1961) and Short & Phil- although less conclusively, by the expres- lips (1966). Strongly contrasting patterns of sion of a combination of plesiomorphic non-structural color (e.g.. rufous and black characters unique to a single pair of parental rectrices of Selasphorus sp.) arc expressed species. in some fashion in all crosses. There is rea- son to doubt that the same is always true, Results however, for plumage characters exhibiting brilliant structural color. For example, the The original labels of GTS specimens are coronal iridescence found in species of Ca- marked "Columbia" (sic), "Colombia," lypteis evident in five examples of C. anna "New Grenada," or "Bogota Collection." x Selasphorus sasin and one specimen of Thus, the geographic pool can only be de- C. anna x Stcttula calliope (although it is fined in general terms. Berlioz & Jouanin never as extensive on the hybrids as it is on (1944) showed that the vast majority of skins 12 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

Fig. 2 Feel and tarsi of "green-throated sunangels" (a = FMNH 46286; b = FMNH 46294), Hetiangelus ameihysiicoltis (e, male), and Eriocnemis cupreovemris (d, male). Note downy leg puffs on tibias of a, b, and d.

Fig. 3. Ventral view or "green-throated sunangels": (a) type of "Heliangelus squamigularis": (b) FMNH 46286; (c) syntype of "Helionypha simotti" (FMNH 46294); (d) "Heliompha speciosa" (MNHN). VOLUME 103, NUMBER 1 13

•

1

l*H - / VOLUME 103. NUMBER I I? prepared in the "Bogota" method were col- parental species to five of the 61 genera of lected in the northern Andes and adjacent Colombian hummingbirds. lowlands, a region roughly encompassed by Reduction of the species pool is support- the present boundaries of Colombia. In ihe ed by comparison of general morphology. absence of unequivocal locality data for any The remiges, rectrices, body plumage, and of the "green-throated sunangels," the geo- bill of GTS are un specialized and lack many graphic pool of potential parental species of the elaborations that are common within must initially include all hummingbirds re- trochiline hummingbirds. Assuming poly- corded from Colombia, a total of 143 species gen ic inheritance of these structures, if GTS in 61 genera (Hilty & Brown 1986). are hybrids, then their parental species must There are two major lineages of hum- be morphologically unspecialized. Sexual mingbirds, the Phaelhornitriinae and Tro- dichromatism within the series of GTS chilinae. As Heliangelus and related genera specimens, if any, is minor (Fig. 1). Collec- belong to the Trochilinae. 1 treat the tions of sexually dichromatic species of subfamily Phaethomilhinae as an outgroup. hummingbirds from Bogota collections are I identified all species of hummingbirds that often sexually skewed in favor of brightly occur in Colombia that shared with GTS colored adult males {Graves, in prep.). one or both of the following characters that Plumage of immature males ofthese species are apomorphic with respect to species in may resemble that of females. Because im- the outgroup: (I) lengthened downy tibial mature GTS specimens do not differ sig- plumes (leg puffs) (Fig. 2); and (2) a brilliant nificantly in appearance from adults (which gorget thai contrasts with adjacent plumage suggests that the sexes are similar), sexual and extends from the chin posteriorly to the bias in collecting due to appearance of GTS upper breast (Figs, 3, 4). With the exception specimens is unlikely. Assuming a 1:1 sex of GTS, no taxon exhibits both of these ratio the probability of finding only males characters. It follows that if GTS are hy- or females in a random sample of 13 (num- brids, then one parental species contributed ber examined) individuals is P < 0.0002 leg puffs and the other the brilliant gorget. (Binomial test). However, as pointed out by Species representing four genera of hum- Haldane(1922), the heterogametic sex (9 in mingbirds, Boissonneaua, Eriocrtemis, birds) may be rare or absent in FL hybrids. , and Ocreatus. have downy Therefore, for diagnostic purposes 1 enter- tibial plumes that exceed those of GTS in tained the possibility that the sample of GTS length. Narrowing the pool of potential pa- specimens was exclusively male. rental contributors of the brilliant gorget is By structural criteria alone, 106 species more difficult. Including taxa with brilliant- (47 genera including Ocreatus and Boisson- ly reflective throats that do not contrast with neaua) may be eliminated from the pool of adjacent plumage, species representing most potential parental species. These include of genera of the subfamily Trochilinae could species with specialized bills (e.g.. Ensifera be the gorgeted parent. However, only ensijera, Schisies geoffwyi), remiges (e.g., species of Heliangelus have gorgets that are Campylopterus falcatus, Agtaeaais cupri- similar in structure, shape, and size to those pennis), rectrices (e.g., Ocreatus underwoo- of GTS. Thus, a first review based on two dii, Acestrura mulsant), and body plumage apormophic characters limits the potential (e.g., Coiibri coruscans. Lophornis stictolo-

Fig. 4. Type of "llelioirypha speciosa": {a) ventral view; (b) enlargement of upper breast showing white pectoral hand: (c) side view of head showing sloping profile forehead and thick bill. 16 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pha). External morphology oft he remaining tural brilliance found in many species of 37 species representing 14 genera {KlaiS, (e.g., Eriocnemis vestitus, E. luciani, I Chlorestes, Leptdopyga. Chrysuronia, Gold- E. cupreoventris, E. mirabilis, E. alinae, E. mania, Goethalsia, Atnazilia, Adelomyia, derbyi). Undertail coverts of the potential Anthocephala, Urosticte, Phlogophiius. He- Heliangelus parental species vary from pure liangelus. Eriocnemis. Haplophaedia) is rel- white to gray with white margins. Contrary atively unspecialized. These bear further to Berlioz's reasoning, the hybrid progeny scrutiny as potential parental species of GTS. of Eriocnemis sp. x Heliangelus sp. might GTS have unpatterned rectricesand uni- be expected to have green undertail coverts formly dark bills. This suggests that species with white or grayish-white margins, simi- with spotted or patterned rectrices (e.g., lar to those of GTS. Atnazilia sp., Anthocephala floriccps. Ade- One character of GTS that has not re- lomyia meianogenys, Phlogophiius hemi- ceived mention by previous investigators is leucurus) or markedly bicolor lower ram- the extensive distribution of brilliant green photheca (e.g.. Atnazilia spp.) can be reflections from the body plumage. These eliminated, leaving species from three gen- reflections extend posteriorly to the vent and era (He/iangelus. Eriocnemis, Haplophae- upper tail coverts and are particularly ap- dia) as potential parental species of GTS. parent when specimens are viewed head-on Excepting these, all other species may be in direct light. Barbule modifications of this rejected from the species pool by two or type are well-developed in several species more criteria (available from the author). In of Eriocnemis (e.g.. E. vestitus. E. cupreo- sum, rejection of species whose distinctive ventris) but are weakly developed or lacking characters (some of which are apomorphic) in both species of Haplophaedia. Several are not found in GTS, reduces the species species of Hcliangelus (e.g., H. exortis, H. pool to nearly the same subset of species amethysticoltis, H. viola) exhibit brilliant that share apomorphic characters with GTS. reflections on the upper breast but lack them Berlioz (1936) noted the downy tibial posteriorly when viewed head-on in direct plumes of the MNHN specimen, Heliotry- light. The presence of brilliantly reflective pha speciosa, and suggested that some plumage (similar to that of GTS specimens) species of Colombian puff eg {Eriocnemis on the flanks and abdomen of hypothetical spp., Haplophacdia spp.) was one of its par- Haplophaedia spp. x Hcliangelus spp. hy- ents. He concluded that the entirely green brids would represent a clear case of "ata- body plumage and non-brilliant undertail vism," a phenonemon thai has not been coverts of this specimen seemed to preclude demonstrated in trochiiine hybrids. the possibility of a cross with a species of Several other characters contradict Ber- Eriocnemis (with brilliant violet or blue un- lioz's hypothesis that Haplophaedia aure- dertail covens) and thai only one hybrid liae is involved in the parentage of GTS. combination was possible. Haplophaedia. Both sexes of H. aureliae have bronze aureliae x Hcliangelus exortis. Berlioz's crowns and uppcrtail coverts that contrast statement on undertail coverts may be ex- with the green back and rump. The dorsum tended to other GTS specimens—none of of GTS specimens lacks such contrast, and, the specimens 1 examined exhibit the struc- in fact, some specimens are brightest on the

Fig. 5. Head profiles of "green-throated sunange Is" (a = FMNH 46286, b = FMNH 46294) and Haplophaedia aureliae (c). Nasal operculum of H. aureliae is more inflated and exposed than in "green-throated sunangels" and Heliangelus spp. VOLUME 103, NUMBER I 17 IS PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

Fig. 6 Bivariate plol ol" Principal Component factor scores of "green-throated sunangcls" and some of their potential parental species. Diamonds = Hcliangclus exonis. Circles = Eriocnemis cupreovertiris. Triangles = Haplophaedia aureliae. Squares = IFeliangeltu amethysticollis. Hollow and solid symbols represent males and females, respectively. Numbers represent "green-throated " specimens from Table 1. Lines envelop groups of males and females of Heliangetus timclhysiicotiis % Eriocnemis eupreownlr/.t.

uppertail coverts. Northern races of Hap- bill of GTS specimens (Filzpatrick et al. lophaedia aureliae (e.g.. //. a. aureliae, H. 1979) (Fig. 5). In Heliangelus exortis and a. caucensis), especially females, have white H. amethysticollis, feathering extends an- or grayish white abdomens and lower breasts teriorly to the distal edge of the nasal oper- speckled peripherally with green. Feathers culum but does not cover it. Feathering in of the ventral midlines of females and im- Haplophaedia spp. does not reach the distal mature males of Colombian species ofHe- edge of the operculum, which is inflated and liangetus have wide buffy margins. Conse- exposed. In Eriocnemis. feathers extend to quently, the venters of hypothetical female the distal edge of the nasal operculum, or hybrids of//, aureliae x Hcliangetus spp. slightly beyond, imparting a sloped appear- would be extensively buffy, not green as in ance to the forehead in profile. Feathering GTS. and forehead profile of GTS is somewhat Another important character is the rela- intermediate between that found in Helian- tionship between the nasal operculum and gelus exortis or //. amethysticollis and sev- the anterior extension of feathering on the eral species of Eriocnemis, but differs from VOLUME 103, NUMBER I 1" a

Fig. 7. Males of the two most probable parental species of "green-throated sunangels" examined in this study: (a) Me/iangelus ametfiysticollis; (b) Eriocnemis cupreoventris. that of//, aureliae or H. iugens. This and (see Bleiweiss [1988] for measurements) and the other characters mentioned above in- E. htciani have deeply forked tails and are dicate that both species of Ilaplophaedia significantly larger than GTS specimens or may be rejected as potential parental species any of its possible Heliangekis parents. Nei- of GTS and support the hypothesis that the ther of these pulfleg species occur in the puilleg parent is some species of Eriocne- Eastern Cordillera in sympatry with !ie- mis. liangetus ameihysticoihs and hypothetical Of the seven species of Eriocnemis (ex- hybrids between these species and H. exoriis cluding E. godini, which is of uncertain sta- would have tails more deeply forked than tus) that occur north of , all but E. in any GTS specimen. In addition, GTS lack cupreoventris can be rejected as a parental white bases to throat feathers found in those species of GTS. Both sexes of E. mosquera species and the greenish outer rectrices of 20 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

Table 2. —Ranges of measurements (mm) of potential parents! species of "green-throated sunangels." Spec- imens from a number of localities were chosen in order to incorporate the range of intraspecific variation found in Colombian populations.

Species Sex = Winy churd Outermost rcL'lrin Innermost reelrix Oilmen

Heliangelus exortis i 10 64.0-67.6 44.8-50.0 32.1-34.5 13.3-15.0 S 10 56.9-60.9 38.9-41.3 29.7-33.9 14.2-16.8

Heliangelus amethysticoltis -• 10 65.1-71.8 43.6-48.8 41.4-46.0 14.8-16.0 s L0 60.5-65.6 35.9-42.4 36.9-39.6 15.5-17.1 Eriocnemis cupreoventris $ 15 58.8-65.2 40.6^15.3 24.3-28.4 17.8-19.6 9 7 58.2-60.6 38.7-42.6 23.2-26.3 16.7-17.9 Haphphaedia aureliae 3 10 59.1-62.8 35.4-41.3 32.6-37.6 17.3-19.6 ; 10 55.5-58.9 32.8-37.8 32.3-36.1 16.2-20.1

E. mosquera. E. mirabilis and E. alinae are H. strophianus, and H. spencei possess apo- small, have restricted ranges in the northern morphies not found in GTS.) Both species Andes, and possess several apomorphic occur sympatrically with E. cupreoventris in characters that are not expressed in GTS. the Eastern Cordillera (see Hilly & Brown E. derbyi is sexually dichromatic and di- 1986). H. amethysticollis differs from H. ex- vergent in tail structure and plumage (e.g., ortisprimarily in havinga well-defined white black leg puffs) and can be conclusively re- pectoral band below the throat of both sexes, jected as a possible parent of GTS. E. ves- a larger, more extensive gorget in males, and titus is moderately sexually dichromatic but a less deeply forked tail. Intergeneric hy- both sexes possess a small violet (male) or brids involving these species may best be blue (female) gorget on the upper throat. distinguished by mensural characters of the Hypothetical Heliangelus spp. x E. vestitus parental species (Table 2). Additional de- hybrids of both sexes would probably have scriptions of potential parental species can a small gorget of some shade of purple or be found in Salvin (1892), Zimmer (1951), pink. Female hybrids would be buffier and Bleiweiss (1985a, b), and Hilly & Brown less green on the breast and lower belly than (1986). are GTS. Measurements of the most probable pa- E. cupreoventris exhibits weak sexual di- rental species of GTS (H. exortis, H. ame- chromatism, lacks a contrasting gorget, and thysticollis, E. cupreoventris-. H. aureliae in- is similar in size to both GTS and possible cluded for comparison) overlap extensively Heliangelus parents. The nasal operculum (Table 2). I compared the measurements of is partially covered with feathers in both these species and GTS specimens. Under sexes. When viewed head-on in direct light, the diagnostic assumptions used in this the body plumage of adult males reflect a study, a GTS specimen could not be the brilliant golden-green anteriorly changing to hybrid progeny of a pair of species if the bluish-green on the upper tail coverts and measurements of the specimen occurred to coppery-gold on the breast and belly. Fe- outside the cumulative range (±0.5 mm for males are slightly duller below. These char- wing and tail; ±0.2 mm for culmen) of their acters make E. cupreoventris the most prob- measurements. Statistics were not per- able parental pulfieg species of GTS. formed because the reference samples were The probable sun angel parent can be lim- chosen to maximize ranges of measure- ited to the only two species with extensive ments. This procedure is conservative be- distributions in the Colombian Andes. H. cause it assumes that the inheritance of exortis and //. amethysticollis. (H. mavors. quantitative characters is mutually inde- VOLUME 103. NUMBER ] 21

Table 3.—Comparison of "green-throated sunangel" (GTS) measurements with the cumulative range of measurements for combinations of polen I lal parental species (Heliatigelu.'i ametftysticollis, H. exortis, Ehocnemis cupreoventris, Haptophaedia aureliae), A male or female symbol indicates that all measurements of a particular "green-throated sunangel" specimen fall within the range of measurements (by sex) for that combination of species. Numbers in parentheses refer to specimens listed in Table I for which all measurements were available. Ratios at the bottom of each column denote the minimum number of males and maximum number of females possible assuming that all "green-throated sunangels" represented hybrids of those species. Binomial f-values are given, assuming a 1:1 sex ratio.

GTS f-f time!} r * E. t mm. Jf. artKth] \ X H. ultrctutf ft txards * E. otfHVifr fl i'.\t>r>'ix * /'/ iiiiri'UtlC (1) £ s a t (2) 3! <5S s • {3} 6 M — 4 <4) i s s t (5) t i s a (6) i i a a (7) a s — s (8) * $ - 6 (9) i $ - i (10) ss £2 a s (13) ! 39 % at 3:S 8:3 6:5 6:1 10:1 p < 0.08 p > 0.20 P <:0.05 P S 0.005

pendent. Unfortunately, this procedure rules velope. However, as noted previously with out few hybrid possibilities (Table 3). None univariate comparisons, if GTS are the of the possible pairs of parental species can progeny of H. exortis x H. aureliae, they be rejected for 7 of 11 of the GTS speci- would be predominately male. This fact, and mens. a variety of plumage characters previously Multivariate morphological relationships discussed suggest that GTS are not hybrids (Table 4) of potential parental species and of the latter two species. The scatter of GTS GTS specimens are illustrated by the first factor scores within the //. amethysticoilis two axes of a Principal Components Anal- x E. cupreoventris envelope shows that an ysis in Fig. 6. Inspection of factor scores even sex ratio is possible. With the excep- revealed thai most GTS specimens are clus- tion of//, speciosa. plumage characters and tered near the center of the bivariate plot. measurements of GTS specimens are con- Only one specimen (type of H. speciosa) falls within the envelope outlining the factor Table 4.—Factor loadings for the first two principal scores for H. aureliaeandH. amethystkoUis components from analysis of "green-throated sunan- and none occurs in the H. exortis x E. cu- gels" and potential parental species (sec Fig, 6). preoventris envelope. Assuming that the in- 1 II heritance of polygenic size and shape char- acters is reflected in the spread of factor Variable scores, these pairs of species are not in- Wing chord 0.91 0.10 volved in the parentage of GTS, with the Outermost rcclrix 0.85 -0.39 Central rcclrix 0.67 0.72 possible exception of the type of//, speciosa. Culmen -0.72 0.32 All GTS specimens fall within the factor Variance explained score envelope of //. amethysticoilis x E. Percent 62.5 19.6 cupreoventris, and ten of eleven specimens Cumulative 62.5 82.1 fall within the H. exortis x ff. aureliae en- 22 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON sistent with the hypothesis that they rep- jected, GTS cannot be considered as valid resent hybrids of H, amethysticollis and E. taxa. Additionally, the data (especially the cupreoventrt's (Fig. 7). Note, however, that variability and inconsistency of plumage the wing chord of AMNH 483682 (Table characters) do not support the hypothesis 1), which is probably a female, is signifi- that GTS, taken as a whole, represent pop- cantly shorter than any specimen in the ulation samples of one or more valid species. sample of these two species. Geographic origin.—II. amethysticollis The position of H. speciosa on the bivar- and E. cupreoventris are sympatric in An- iate plot is well removed from the other dean forests and shrublands (2000-3000 m GTS specimens. H. speciosa differs from elevation) in the northern half of the Eastern other GTS specimens in having a broad Cordillera. Thus, GTS specimens could ac- white pectoral band (Fig. 4) instead of a few tually have been collected in the environs semi-concealed spots, a shall owly-forked of "Bogota." Mutsant & Verreaux (1872) tail, and a slightly longer bill. The presence reported that the type ofHe/iotrypha hurrah of a white pectoral band and the confor- was collected on the Rio Saldana, Depart- mation of the gorget of H. speciosa indicates ment of To lima, in the Central Cordillera that a white-banded species of He! i an gel us of the Colombian Andes. However, as pre- (e.g., H. amethysticollis) is one of the pa- viously mentioned, the original labels of//. rental species. The well-developed leg puffs harrali and other specimens lack specific lo- of II. speciosa (fide M. P. Wallers) indicate cality data and in the absence of corrobo- that the other parental species is a puffieg. rating evidence, the possible Central Cor- The anterior extension of feathering over dilleran origin of the type of//, barrali can the nasal operculum of H. speciosa. how- be dismissed. ever, indicates that the other parent could Nomenclature. — Hybrids are individuals not be H. aureliae which is similar to it in and not taxa. Thus, the names Heliangelus size and shape. H. speciosa lies within the squamigularis Gould, 1871, Heliotrypha PCA envelope for //. cupreoventiis x //. barrali Mulsant & Verreaux, 1872, Heli- amethysticollis. Despite the difference in ap- otrypha speciosa Salvin. 1891. and Heli- pearance of //. speciosa from other GTS otrypha simoni Boucard. 1892, are available specimens, it seems probable that (hey rep- only for the purposes of homonymy in tax- resent the same hybrid cross. Additional onomy and should not be used in the pop- study of the specimen may be required to ular literature. For the purposes of field verify this fact. If true, then H. speciosa rep- guides, these hybrids may be referred to col- resents an extreme hybrid phenotype that lectively as "green-throated sunangels." resembles its sunangel parent much more than its puffleg parent. Discussion Hybrid diagnoses can be simple or ex- Conclusions tremely complex depending on circum- With the possible exception of //. spe- stances. Factors that affect the success of ciosa, GTS specimens examined in this hybrid diagnoses include: (I) the number of study, for which measurements were avail- hybrid individuals and their age, sex, and able, are probably hybrids of Ileiiangetus hybrid composition (e.g., F,, backcross); (2) amethysticollis x E. cupreoventiis. Berlioz's the number ofdistinctive plumage and mor- (1936) hypothesis that the MNHN speci- phological characters on the hybrid; (3) the men and perhaps others were hybrids of number of species in the putative hybrid's Haplophaedia aureliae and Hettangelus ex- taxonomic group; (4) and knowledge of the ortis is not supported by the data. Because hybrid's taxonomic group and of the re- the hypothesis of hybridity cannot be re- gional avifauna where the hybrid original- VOLUME 103, NUMBER ] 23 ed. (The most challenging diagnoses are of curators and staff of the American Museum hybrids represented by a unique, unsexed, of Natural History and the Academy of Nat- possibly immature specimen without spe- ural Sciences of Philadelphia (ANSP) for cific locality data, which belongs to a spe- providing access to collections and the Field ciose, dull-plumaged, and poorly known Museum of Natural History for loaning GTS taxonomic group from a poorly collected specimens. Lastly, I would like to thank R. region of high species diversity!) L, Zusi for patiently answering untold thou- Ideally, parental species are identified with sands of my questions about hummingbird certainty, but failing this, what result jus- systematics over the past few years. tifies the considerable effort expended in the Museum research was supported by grams average hybrid diagnosis? In terms of value from the Smithsonian Research Opportu- to future researchers, it is far better to have nities Fund and the Frank M. Chapman a short list of species that includes the cor- Memorial Fund of the American Museum. rect pair of parents, than an exact deter- I dedicate this paper to Jacques Berlioz mination of parental species that may be and Christian Jouanin in recognition of their wrong (errors of this sort are frequently per- pioneering work on hummingbird hybrids. petuated in the literature; see Graves 1988). Rejection of any species from the pool of potential parental species must be based on Literature Cited the unequivocal violation of diagnostic as- Banks, R. C, & N. K. Johnson. 1961. A review of sumptions, and there is no logical reason North American hybrid hummingbirds. —Con- for reducing the species pool beyond the dor 63:3-28. limits suggested by the data. Conclusive Berlioz, J. I 936. Note sur 1'idcntile probable du type d'Heliolrypha simoni Boucard (Troehilides).— knowledge of what species or species com- Bulletin du Museum National D'Hisloire Na- binations are not parents, as well as those lurelle. Paris 14:233-236. that might be parents of a hybrid is far more -. 1964. La collection de Troehilides A. L. But- valuable than perhaps recognized by most ler.-Otseau 34:91-105. taxonomists. especially when the majority , and C. Jouanin. 1944. Listc de Troehilides (rouves dans les collections commercials de of species fall into the former category. Bogota—Oi seau 14:126-155. Species in the latter category constitute the Bleiweiss, R. 1985a. Variation and population struc- nucleus for future analyses. ture of the Tourmaline Sunangel. lleliangelus exortis exortis (Aves, Trochilidae).— American Acknowledgments Museum Novitates 2811:1-14. . 1985b, Iridescent polychromatism in a fe- R. Banks, M. R. Browning, P. Cannell, F. male hummingbird: is it related to feeding stral- Gill, S. Olson, and R. Zusi commented on egies?-Auk 102:701-713. . 19S8. Systematics and geographic variation an earlier version of the manuscript. Banks, in the Golden-breasted Purfleg Eriocnemis mas- R. Bleiweiss, T. Schulenberg, and Zusi re- quers (Aves. Trochilidae). —American Museum viewed a later incarnation. I thank them for Novitates 2913:1-8. their suggestions and criticism. I thank J. Boucard. A. 1892. A complete list up to date of the Becker, H. James, C. Jouanin (MNHN), M. hummingbirds found in Columbia, with de- scriptions of several supposed new species.— Robbing (ANSP), and M. Walters (BM) for The Humming Bird 2:73-87. providing measurements or answering de- . !895. Genera of humming birds. Vol. 5, Part tailed questions about specimens, Becker 1. Bournemouth, London. and Jouanin for providing excellent color Buckley. P. A. 1982, Avian genetics. Pp. 21-1 10 in transparencies of specimens for my use, and M. Petrak, ed.. Diseases of cage and aviary birds. 2nd ed. Lea and Febiger. Philadelphia. 680 pp. V. Krantz for preparing illustration prints. Chapman, F. M. 1917. The distribution of bird-life M. Farmer deserves special thanks for in Colombia. —Bulletin of the American Mu- translating French. I am also grateful to the seum of Natural History 36:1-729. 24 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

Cory, C. B. 1918. Catalogue of birds oflhe Americas. Parkes, K.. C 1984. An apparent hybrid Black-billed Part 2, No. I.—Field Museum of Natural His- x Yellow-billed Cuckoo.-Wilson Bulletin 96: tory Zoological Series 13:1-313. 294-296. Falconer, D. S. 1981. Introduction to quantitative Peters, J. 1945. Check-list of birds of the world. Vol. genetics. 2nd ed. Longman. New York, 340 pp. 5. Museum of Comparative Zoology, Cam- Fitzpatrick, J. W.. D. E. Willard. and J. W. Terborgh. bridge, Massachusetts, 306 pp. 1979. A new species of hummingbird from Salvin. O. 1891. Descriptions of new species of Upu- .-Wilson Bulletin 91:177-186. pae and Trochili in the collection of the British Fox, H. M. and G. Vevers. I960. The nature of an- Museum. —The Annals and Magazine of Nat- imal colours. Macmillan, New York, 246 pp. ural History (6th series) 7:374-378, Gould, J. 1871. Descriptions of six new humming- . 1892. Catalogue of birds in the British Mu- birds.—Proceedings of the Zoological Society of seum, Vol. 16. London 1871:503-505. Short. L, L„ and A. R. Phillips. 1966. More hybrid Graves, G. R. 1988. Evaluation of Vermtvura x Opo- hummingbirds from the United States. —Auk ronis hybrid wood-warblers.—Wilson Bulletin 83:253-265. 100:285-289. Simon, E. I92L Histoirc naturelle des Trochilidae Gray, A. P. 1958. Bird hybrids. Commonwealth Ag- (synopsis el catalogue). Paris. 416 pp. ricultural Bureaux, Bucks, England, 390 pp. Zimmer, J. T. 1951. Studies of Peruvian Birds. No. Green way, J. C Jr. 1978. Type specimens of birds 61. The genera Aglaeacus, iMfresnaya, Pteroph- in the American Museum of Natural History. anes. Boissonneawz. Heliangehts. Eriocnemis, Part 2. —Bulletin of the American Museum of Haplophaedia. Ocreatus, and Leshia. — Ameri- Natural History 161:1-305. can Museum Novitates 1540:1-55. Haldane. J. B, S. 1922. Sex-ratio and unisexual ste- rility in hybrid .—Journal of Genetics Department of Vertebrate Zoology, Na- 12:101-109. Harrison. J. M., & J. G. Harrison. 1963. Comments tional Museum of Natural History. Smith- on a hybrid Red Shoveler * Northern Shovcl- sonian Institution, Washington, DC. 20560. er. —Bulletin of the British Ornithological Club 83:21-25. Hartert. E. 1922. Types of birds in the Tring Mu- Appendix seum. B. Types in the general collection. Trochi- Comparative Description of li. — Novitates Zoologicae 29:403-412. "Green-throated Sunangels" Hilly, S. L.. & W. L. Brown. 1986. A guide to the birds of Colombia. Princeton University Press. Descriptions of structural colors are unusually sub- Princeton, New Jersey, 836 pp. jective and actual color varies with the angle of in- Hult,F.B. 1949. Genetics orthc fowl. McGraw-Hill. spection and direction of light. For this reason I use New York, 590 pp. general color descriptions. Numbers in parentheses re- Jouanin, C 1950. Catalogue systemaliquc des types fer to specimens in Table I. dc Trochilidcs du Museum National d'Histoire The crown, nape. back, and rump are medium green. Nature He de Paris. —Bulletin du Museum Na- Upperlail coverts are medium green to bluish-green. tional d'Histoire Naturelle de Paris 22:1-27. There is no contrast between crown and back. When Lucas, A. M.,& P. R, Slettenheim. 1972. Avian anat- viewed head-on in direct light, scattered leathers on omy—integument. Agricultural handbook 362. crown (5). back, wings, and upper tail coverts show United Stales Department of Agriculture, brilliant golden-green to green reflections. A brilliant Washington. DC. green frontlet, variable in intensity and definition, is Meyer de Schaucnscc, R M. 1949. Birds of the Re- found in most adults (absent in 10), but is faint or public of Colombia. Pan 2.—Caldasia 23:381- lacking in immatures. When present, the frontlet is 644. small (1, 2,4,6, 8, 11, 12), similar in size to that found . 1966. The species of birds of South America. in //. exortis or /•/. amethysticoUis. Prominence of the Livingstone Press. Narberth, Pennsylvania, 577 frontlet is affected by variations in skin preparation. pp. Lores, auriculars, and neck at the sides of the throat Morony, J. J.. Jr., W. J. Bock, &J. Farrand, Jr. 1975. are medium green, but appear much darker when viewed Reference list oflhe birds of the wo rid. Amer- head-on. A small while postocular spot is present. A ican Museum of Natural History, New York. brilliant gorget is found in all specimens but is variable Mulsant, E„ & J. Verrcaux. 1872, Description dune in size, color, and degree of contrast with adjacent cspece nouvelle d'oiseau moche. — Annales So- plumage. Gorget margins are somewhat irregular and ciete Linneenne de Lvon 18:106-107. indistinct in all adult specimens. Gorget color is vari- VOLUME 103. NUMBER 25 able and can be characterized (viewed head-on) as mostly subterminal. are present on two to ten leathers hiuish-green (4) and silvery-green (2, '). 13) to silvery of the upper pectoral area on most specimens (1.2,7, bluish-green (7) in im matures, and from bluish-green 8. 9, 10. II, 13). Llndcrlail coverts are variable in (5) and pale green (12) to silvery-green (1.3.6,8. 10. length (not exceeding half the length of the tail) and II) in adults. In indirect light, gorgets of some speci- are medium green or bluish-green with narrow to broad mens emit fairtl coppery or pinkish rellections. Color while or grayish white margins: basal barbs in some variation appears to he Fairly continuous and color specimens are long and downy. Central rectrices are characterizations arc arbitrary. For example, "silvery- dark green to bronzy green; outer ret rices are bluish- green" includes various shades of pale metallic green hlack. Depth of tail fork varies from 3.8 to 10.5 mm. (= "leaden" of Sal vin 1892). Gorgets of immaturcs are Outer web of outermost rectrix is well-developed ( • 'A oval in shape, wider posteriorly and may he surround- widlfi of inner web). The outermost reclrix ranges from ed hy dull, lax plumage (7) Gorgets of adults arc larger, !S (i in •'. "> mi « idth and is mote acuminate in shape contrast less with adjacent plumage, and may have than in // extra ix at II. amethyxticotlis. Tibia! leathers irregular margins (5). Brilliant gorget feathers are (leg pulls) arc white or bufly (2), variable in length (2.5 rounded and about the size of those in male Hetian- to 8 mral, and more or less downy in texture. (The get us sp. (e.g., //. exums). but become progressively presence of leg puffs on (1) cannot be determined from smaller toward the chin and malar regions in adults. photographs of specimens or published descriptions]. The upper and lower breast, abdomen, and Hanks (ex- The remiges are unmodified (e.g.. not emarginaic) and cept 2) are medium green. Feathers along the midline dull dark brown in color. Outer webs of primaries of ofimmatures may have narrow'buffy margins. Brilliant some specimens have a faint bronzy-green sheen. Hills gulden-green or green reflections (or faint coppery in are unmodified and straight and dark blackish-brown. 5) arc scattered over the breast, abdomen, and flanks The tipper mandible of (3) is broken (Fig. 4). Feathering when viewed head-on. Plumage ofimmatures is duller extends anteriorly on the bill to the distal edge of the than in adults. The type of//, sficcii.ua 1.1) differs from nasal llange (not inflated!, somewhat obscuring it. Feel other GTS specimens in having a large white pectoral are dark brown or dark blackish brown. patch spotted with green discs and darker underparts Additional notes on plumage can be found in Berlioz with few brilliant rejections. White or huffy spots, 119361,