Art-Entstehung bei einigen Waldvögeln Amazoniens

Zusammenfassung

Die Annahme ungestörter Entwicklung der Wald-Fauna Amazoniens über lange geologische Zeiträume hinweg ist wahrscheinlich unzutreffend. Der Wechsel von ariden und humiden Klimaphasen während des Quartär führte zu wiederholtem Schrumpfen und nachfolgender Ausdehnung der Regenwälder. Diese überdauerten trockene Perioden in stark reduzierten humiden Gebieten, die der Wald-Fauna als „Refugien“ dienten. Hier entstanden zahlreiche Arten und Unterarten von isolierten und reduzierten Populationen amazonischer Waldvögel, die bei neuerlicher Ausdehnung der Wälder mit den Populationen anderer Wald-Refugien in sekundären Kontakt kamen. Die Differenzierung der Vogelfauna Amazoniens ist geologisch sehr jung und erfolgte relativ rasch, was wahrscheinlich auch für die anderen Wirbeltierfaunen dieses Gebietes gilt.

Als Beispiele für die Entstehung amazonischer Waldvögel werden Verbreitung, geographische Variation und historische Differenzierung mehrerer Artengruppen und Superspezies diskutiert: (1) DiePionopsitta caica Artengruppe mit 6 Arten, von denenP. vulturina früher in die monotypische Gattung „Gypopsitta“ gestellt wurde, (2) diePhoenicircus carnifex Superspezies mit 2 Arten, (3) dieXipholena punicea Superspezies mit 3 Arten, (4) diePipra aureola Superspezies mit 3 Arten, von denenP. filicauda früher in die monotypische Gattung “Teleonema„ gestellt wurde, (5) diePipra erythrocephala Superspezies mit 5 Arten, (6) diePipra serena Superspezies mit 7 Arten und (7) dieEuphonia cayennensis Superspezies mit 3 Arten.

Die Arten der Superspezies amazonischer Vögel stammen wahrscheinlich von je einem gemeinsamen Vorfahren ab, dessen Areal während arider Klimaphasen in mehrere Teilstücke zersplittert wurde. Einzelheiten dieser Entwicklung, insbesondere die Zuordnung bestimmter Differenzierungsphasen zu einzelnen Trocken- bzw. Isolationsperioden, bleiben noch unbekannt, da wir weder über die Florengeschichte Südamerikas noch über die variable Evolutionsrate der verschiedenen Populationen in den Wald-Refugien sichere, detaillierte Aussagen machen können.

Zahlreiche Glieder der Superspezies amazonischer Vögel schließen einander trotz Fehlens ökologischer Schranken geographisch strikte aus ohne zu hybridisieren, obwohl sie über weite Gebiete in direktem Kontakt stehen. Diese Formen sind anscheinend schon sexuell isoliert, konkurrieren aber noch ökologisch miteinander, wodurch ein sympatrisches Vorkommen verhindert wird.

Die breiten Flüsse Amazoniens haben wahrscheinlich — von Ausnahmen abgesehen — nicht die Art-Entstehung bei Waldvögeln im zentralen Südamerika verursacht, sondern lediglich die Ausbreitung der Waldvögel von den Wald-Refugien modifiziert oder begrenzt.

Summary

The alternating humid and dry climatic periods of the Pleistocene and post-Pleistocene probably caused the repeated shrinkage and later expansion of the Amazonian forest. During arid phases forest animals were restricted to a number of isolated small humid “refugia” where the forests survived. Many new species and subspecies probably developed in these forest refugia from isolated populations of Amazonian forest birds. The latest differentiation of the Amazonian bird fauna at the genus and species level is, geologically speaking, very recent and occurred rather rapidly. This seems also to be true for the other vertebrate forest faunas of the Amazonian region.

The distribution, geographical variation and presumed historic differentiation of several species groups and superspecies of Amazonian forest birds are discussed.

• (1) Among the forest parrots thePionopsitta caica species group (Fig. 3 and 4) comprises the following species:caica, barrabandi, vulturina (previously considered to form the monotypic genus “Gypopsitta”),pyrilia, pulchra, andhaematotis. Thes forms are characterized by a very conservative color pattern of the wing and tail which is altered or simplified only in the peripheral speciespulchra, haematotis andcaica. On the other hand, the color of the head is a very plastic character within this superspecies: It is black and yellow (with or without feathers) in the cis-Andean species and pure yellow or brownish and pink in the species inhabiting the trans-Andean lowlands. The various species exclude each other geographically where their ranges adjoin (possibly as a result of ecological competition). Hybridization is not known to occur. Only two species may locally be sympatric in NW-Colombia which, however, needs confirmation.

• (2) The black and red cotingidsPhoenicircus nigricollis andP. carnifex are characteristic inhabitants of the upper und lower Amazonian forests, respectively (Fig. 5). Both species meet north and south of the central Amazon river, apparently without hybridizing along the zone of secondary contact.

• (3) The three species of the dark purplish-red cotingidsXipholena punicea, X. lamellipennis andX. atropurpurea form a group of closely related species (Fig. 6 and 7). The Guiana formX. punicea has extended its range into upper Amazonia and came into contact withX. lamellipennis between the Madeira and Tapajós rivers as well as in the Serra do Cachimbo.

• (4) The colorful manakins of thePipra aureola superspecies (Fig. 8 to 11) are widely distributed and rather common inhabitants of the forest interior east of the Andes. These species arePipra aureola, P. fasciicauda, andP. filicauda. The latter form is currently placed in the monotypic genus “Teleonema” because of its filiform rectrices and appears in modern lists widely separated from the other two species. However, the extremely similar coloration (Fig. 8) and the parapatric distribution pattern (Fig. 11) indicate that “Teleonema”filicauda is nothing but the upper Amazonian representative of theaureola-fasciicauda group and should be placed in the genusPipra. The members of theaureola superspecies strictly exclude each other geographically where their ranges adjoin (presumably as a result of ecologic competition). Hybridization apparently is very rare along the zones of secondary contact (e. g. “Pipra heterocerca”).

• (5) The species of thePipra erythrocephala superspecies inhabit the lower and middle levels of the Amazonian forest.P. erythrocephala, P. rubrocapilla, P. chloromeros are typical cis-Andean forest birds, whileP. mentalis occupied the trans-Andean forest north to Mexico.P. cornuta inhabits the montane forests of the south Venezuelan highlands. These species exclude each other geographically where their ranges adjoin (presumably as a result of ecologic competition). Hybridization is not known to occur.

• (6) ThePipra serena superspecies comprises a number of geographical representatives of quite different appearance (Fig. 14):P. serena, P. coronata, P. nattereri, P. vilasboasi, P. iris, P. isidorei, andP. caeruleocapilla. The males of the species living north of the Amazon and west of the Andes are predominantly black, while those of the forms occuring south of this river are green with a yellow belly. The upper head and, in some cases, the rump and upper tail coverts are blue, yellow or white. The green and black groups apparently hybridize extensively south of the upper Amazon river along a zone of secondary contact. A number of hybrid populations have been described from this region as geographical “subspecies”:P. coronata “circumpicta”,P. c. “arimensis”,P. c. “chloromelaena”,P. c. “hoffmannsi” (Fig. 15).

  On several table mountains of southern Venezuela,Pipra serena occupies the montane forests above the lowland habitat ofP. coronata. This “pseudo-overlap” of the ranges of these two species should not be interpreted as sympatry of the forms involved. These species as well as the other forms of this superspecies exclude each other geographically along the zone of secondary contact (presumably as a result of ecologic competition). Hybridization is only known in the instance mentioned above.

• (7) The species comprising theEuphonia cayennensis superspecies (Fig. 16 and 17) inhabit upper Amazonia(E. rufiventris), lower Amazonia(E. cayennensis), and the forests of southeastern Brazil(E. pectoralis). Details on the zone of contact of the two Amazonian forms are not yet known.

  The component species of each superspecies of Amazonian birds probably originated from a common ancestor whose range was split into a number of small isolated portions during dry climatic periods of the Quaternary. The sister populations inhabiting the various “forest refugia” (Fig. 1) deviated from their ancestor and from each other by selection and chance, in this way developing the present characteristics of the various species. Upon the return of more humid conditions and the expansion of the Amazonian forest newly developed forms came into secondary contact with varying results: more or less extensive hybridization (this situation is probably more widespread than presently known), geographic exclusion without appreciable hybridization (here belong the examples discussed in this article) or more or less extensive range overlap if sexual isolation as well as ecologic compatibility had been reached during the previous period of geographic isolation. These zones of secondary contact of Amazonian forest birds reveal striking faunal discontinuities in a continuous forest environment. The contact zones are still very poorly known and should be studied in detail in the field.

  The above interpretation of the differentiation of Amazonian forest birds should be considered merely as a working model which in many instances will certainly turn out to oversimplify very complex historic events. This is particularly true since we still do not know the details of the vegetational history of Amazonia and are completely ignorant of the rôle of extinction and the varying rate of differentiation of isolated bird populations.

  Although the lower Amazon river and the wide lower portions of its tributaries probably represent barriers to the dispersal of a number of birds of the dark forest interior (antbirds, manakins), the significance of these rivers for the differentiation of the various species discussed seems to have been only rather slight. Most species that are separated by the broad lower portions of the Madeira, Tapajós or Xingú river may be in direct contact in the headwater region where these forms readily crossed the narrower river courses. In a number of mediumsized and ecologically incompatible Amazonian forest bird species that came into secondary contact the border of their ranges stabilized along broad river courses (which in these cases constitute partial barriers to dispersal). In this way ecologic competition is avoided.