The Systematic Position of the Cocos Flycatcher
Total Page：16
File Type：pdf, Size：1020Kb
The Condor 86~42-47 0 The Cooper Omthological Society 1984 THE SYSTEMATIC POSITION OF THE COCOS FLYCATCHER WESLEY E. LANYON ABSTRACT.-The Cocos Flycatcher (Nesotriccusridgwuyz), traditionally allied with Myiarchus, was recently placed in another subfamily, Fluvicolinae, near Ernpidonax and Cnemotriccus.A comparative study of the skull and syrinx dem- onstrates that the nearest relatives of Nesotriccusare Phaeomyias murina and Capsiempisflaveoluin the subfamily Elaeniinae, sinceall three generashare unique derived character states of the nasal septum and of the supporting elements of the syrinx. In 1891 Charles Townsend collected a new I had decided on the basis of external mor- flycatcher on Cocos Island, which he later phology alone to exclude Nesotriccus from named Nesotriccusridgwayi (Townsend 1895). Myiarchus. I doubted that nestsof Nesotriccus, He called attention to the zoogeographicand, when discovered, would be located within tree inferentially, the morphological relationship cavities and lined with fur and feathers, the between his new monotypic genusfrom Cocos, derived nesting behavior that defines the over 500 kilometers off the Pacific coast of myiarchine flycatchers(Lanyon 1982, Lanyon Costa Rica, and an equally interestingflycatch- and Fitzpatrick, 1983). Subsquently I exam- er in the GalapagosIslands, magnirostris, which ined the skull of Nesotriccusand, contrary to Ridgway (1893) had recently removed from the findings of Warter (1965) found it to be Myiarchus and placed in its own genus, Eri- very distinct from the skull of Myiarchus. My bates. Hellmayr (1927) similarly was im- interest in Nesotriccuswas revived with the pressedwith the affinity of Nesotriccusand Er- descriptions of the first known nests (Sherry, ibatesand placed both close to Myiarchus, the in press)and the availability of alcoholic spec- widespreadand successfulmainland genus.As imens taken by Sherry. I was certain that a recently as 1955, Eisenmann followed Hell- careful study of the skull and syrinx would mayr in placing Nesotriccusnext to Myiarchus provide meaningful clues to the nearest rela- in a list of speciesof Middle American birds. tives of Nesotriccus.This paper presents the Subsequent recommendations (Swarth 193 1; results of that study. Lanyon 1978; Zimmer, unpubl. notes, Amer. Mus. Nat. Hist.) that magnirostris should be METHODS returned to Myiarchus left unresolved the sys- Research on the skull (Warter 1965) and the tematic position of Nesotriccus. syrinx (Ames 197 1) has demonstrated the The allegedaffinity between Nesotriccusand utility of these structures in determining re- Myiarchus was rejected by Swarth (193 1) and lationships within the suboscines.I have been Zimmer (unpubl. notes) on the basis of exter- influenced greatly by both of these important nal morphology. In a study of cranial anatomy, studies and I follow the terminology of these Warter (1965) concluded that Nesotriccus“ ap- authors. Two complexes in particular have pears intermediate in structure between the proven effective in diagnosinggroups of tyrant Myiarchinae (s.s.) and Todirostrum, but oth- flycatchers: (1) the nasal septum, and notably erwise is probably more closely related to the the degree to which there is ossification and former.” Ames (197 1) reported the syrinx of buttressingwith internal and/or laterally pro- Nesotriccusto be unlike the syringesof Myiar- jecting supporting elements, and (2) the mor- thus, Empidonax, Contopus, and Cnemotric- phology of the bony and cartilaginous sup- cus, but made no recommendation as to how porting elementsin the syrinx, and in particular Nesotriccusshould be classifiedon the basisof the number, shape,and position of the internal syringeal morphology. In the only definitive cartilages(Lanyon 1982, Lanyon and Fitzpat- classification of the Tyrannidae since Hell- rick 1983). mayr (1927) Traylor (1977) removed Neso- Thus far I have examined the skulls of 83 triccusfrom its traditional position among the genera and the syringesof 86 genera of the 9 1 myiarchine flycatchers and placed it in the genera of tyrant flycatchers (sensu Traylor subfamily Fluvicolinae, between Empidonax 1977) in a larger, ongoing study of the phy- and Cnemotriccus, on the basis of external logeny of the Tyrannidae. In addition to the morphology. anatomical collections at the American Mu- At the time that I was completing my re- seum of Natural History (AMNH), material vision of the genusMyiarchus (Lanyon 1978) hasbeen borrowed from the Carnegie Museum 1421 COCOS FLYCATCHER 43 Nesotriccus Phoeomyias and Capskmpis’ Septa / Y Fluvicoline Septa -v- FIGURE I. Photographsof the skulls of selected tyrant flycatchers (ventral aspect; anterior end to the left) and diagrammatic representationsof cross sectionsof the correspondingnasal septa (projecting downward from the roof of the nasal capsule): I, 2 Nesotriccus ridgwuyi (AMNH 92 IO), I5 x and 20 x ; 3, Phaeomyias murina (AMNH 7306), 20 x ; 4, Capsiempisflaveola (UMM 200869), 20 x ; 5, Empidonax euleri (AMNH 6934), 15x ; 6, Cnemotriccusfuscatus (AMNH 6685), I5 x ; 7, Myiozetetes similis (AMNH 7 l73), IO x ; 8, Myiarchus tuberculifer (AMNH 7 l75), 10 x . Arrows indicate presenceof trabecular plates on the nasal septa of all skulls except those of the two tyrannines (7 and 8). ofNatural History in Pittsburgh,the Field Mu- gerkus and Uhler 1977) to facilitate the study seum of Natural History in Chicago (FMNH), of the supportingelements and to differentiate the Museum of Zoology at Louisiana State between bony and cartilaginous tissue. University in Baton Rouge (LSU), the Mu- seum ofZoology at the University of Michigan RESULTS AND DISCUSSION in Ann Arbor (UMMZ), the National Museum of Natural History in Washington, D.C., and NASAL SEPTUM the Peabody Museum of Natural History, Yale The nasal septum of Nesotriccusis character- University, in New Haven. Specimens cited ized by having (1) a well-developed internal here are identified to collection by the abbre- supporting rod that presumably servesto but- viations given above. Syringes were stained tress the septum (“r” in photographs 1 and 2, with alcian blue and alizarin red (after Din- Fig. l), and (2) a trabecular plate in the form 44 WESLEY E. LANYON of short and narrow lateral projections from The syringesof the fluvicolines with which the septum, elevated above the ventral edge Traylor (1977) associated Nesotriccus, i.e., of the septum and located near the ventral end Empidonax and Cnemotriccus, have a very of the internal supporting rod (indicated by different morphology: (1) the tracheal elements arrows in photographs 1 and 2 and the cor- are not fused into a drum, (2) the pessulusis responding cross-sectionaldiagram in Fig. 1). cartilaginous,narrow, and not well-developed, The posterior ends of these lateral projections and (3) the internal cartilages are large, very extend caudally in closeproximity to, but more broad, and nearly straight (photographs4 and or less free from, the septum. 5 in Fig. 2). Since none of the genera in Tray- Only two genera of tyrant flycatchers other lor’s Fluvicolinae has a well-developed drum, than Nesotriccushave the unusual character- as seenin Nesotriccus,I was puzzled by Ames’s istics of the nasal septum described above. (1971:61) description of a drum in the syrinx These are the monotypic Phaeomyias and of Cnemotriccuspoecilurus (=Knipolegus poe- Capsiempis,both placed by Traylor (1977) in cilurus in Traylor 1977). I reexamined the the subfamily Elaeniinae. Their nasal septa, specimen(AMNH 6743) that he had dissected illustrated in photographs 3 and 4, Figure 1, and found that it had been mislabeled; it was are remarkably similar to that of Nesotriccus. actually Mecocerculus leucophrys,which be- Traylor (1977) placed Nesotriccus in his longs in the Elaeniinae and has a well-devel- subfamily Fluvicolinae, between Empidonax oped drum. and Cnemotriccus. All of the genera in this The syrinx of Myiarchus and near relatives subfamily, except Nesotriccus,have a large, (Lanyon 1982, Lanyon and Fitzpatrick 1983) conspicuoustrabecular plate located along the differs from that of Nesotriccusin (1) the lack ventral edge of the nasal septum, i.e., not el- of fusion of tracheal elements into a drum, (2) evated above the ventral edge (indicated by the absenceof a well-devloped bony pessulus, arrows in photographs 3 and 4 and the cor- and (3) the presence of two pairs of internal responding cross-sectionaldiagram in Fig. 1). cartilages, the larger of which attaches to the Moreover, the plate in the fluvicolines is no- ventral portion of the trachea-bronchial junc- ticeably concave when viewed in crosssection. tion and is L-shaped (photograph 6 in Fig. 2). All of the genera in Traylor’s (1977) third subfamily, the Tyranninae (which includes TAXONOMIC CONCLUSIONS Myiarchus and relatives), have nasalsepta that The first conclusionto be drawn from this study lack a trabecular plate. When viewed from be- of the nasal septum and the syrinx of Neso- low, the ventral edge of the septum in all tyr- triccusis that the notion of close relationship annines is thin and almost knife-like as illus- with the myiarchine flycatchers, through trated in photographs 7 and 8, Figure 1, and Myiarchus magnirostris of the Galapagos Is- there are no lateral projections of any kind. lands, should be put to rest once and for all. There simply is no morphological support for SYRINX this old contention, and therefore Traylor For the purposesof this study, the significant (1977) was correct in removing Nesotriccus characteristics of the supporting