Syndactyla dimidiata (Pelzeln, 1859) Anabates dimidiatus Pelzeln, 1859 Philydor dimidiatus (Pelzeln, 1859) Berlepsch & Hellmayr, 1905 Pseudoxenops dimidiatus (Pelzeln, 1859) Pinto, 1932 Xenoctistes mirandae Snethlage, 1928 Syndactyla mirandae (Snethlage, 1928) Novaes, 1953 Philydor baeri Hellmayr, 1911 Pseudoxenops baeri (Hellmayr, 1911) Pinto, 1938 Syndactyla dimidiata baeri (Hellmayr, 1911) Robbins & Zimmer, 2005 Range. Syndactyla dimidiata is widely distributed throughout the Brazilian Cerrado, with confirmed records from southern Mato Grosso, Mato Grosso do Sul, Goiás, southern Bahia, Minas Gerais, and western Paraná, also reaching northeastern Paraguay (Figure 4), with elevations ranging from 150 m to 1250 m a.s.l. Several doubtful or erroneous records of this species exist: 1) Although it is said to occur in the Pantanal (Dubs 1992), there is no confirmed record of S. dimidiata for this region (Tubelis & Tomas 2003); 2) A record from the Rio Tibagi valley, Paraná, where this species has been considered rare (Anjos & Schuchmann 1997; Anjos et al. 1997), cannot be accepted either. Although this region has extensive grasslands, harboring some species typical of open Cerrado, birds restricted to gallery forests of the Cerrado have not been found there, where forests are dominated by the gymnosperm Araucaria angustifolia (Anjos & Graf 1993). This record has also not been accepted by Straube (2004), because it is from a region with “phytophysiognomy, climate and topography unsuitable with the range and ecological requirements” of this species. Furthermore, it seems that some kind of typographical error occurred, because Anjos & Schuchmann (1997) cited Anjos & Graf (1993) as the source of the record of S. dimidiata for Paraná, which is not verified in that publication. For other inconsistencies on the natural history data presented by Anjos and co-authors, see Vallejos & Morimoto (2012); 3) Peters (1951) included the "Fazenda Transvaal, [municipality of] Rio Verde" in the range of S. d. baeri, and Pinto (1978) also included "Rio Verde" in its range, probably following Peters (1951). The Fazenda Transvaal was surveyed by Walter Garbe from March to June 1940 and from April to November 1941 (Pinto 1945). Given that no specimen of S. dimidiata from this locality is housed in MZUSP or in MCZ, where the material collected by Garbe is deposited (Pinto 1945), we consider the occurrence of S. dimidiata at this locality to be invalid; 4) The record in Straube (2004) for Rio Paracaí is an error (Straube in litt.). Habitat. Syndactyla dimidiata is a locally uncommon species, being absent from many apparently suitable riparian habitats. It was not possible to identify clear habitat preferences in the three localities where this species has been recorded by the senior author, as follows: 1) In the Fazenda Água Limpa, one pair was observed in a preserved gallery forest about 150–200 m wide, with a canopy about 10–15 m tall, immersed in a matrix of open Cerrado vegetation. This forest bordered the Córrego Capetinga, which flowed between high banks about 1–3 m tall that evidently protect this forest from flooding. Botanical information about this gallery forest can be found elsewhere (Seabra et al. 1991; Ribeiro & Felfili 2009); 2) In the Fazenda Monte Carmelo, two specimen were collected in a seasonally flooded gallery forest about 15–20 m tall, with open understory. Although the forest was well preserved, probably pristine, all the adjacent savannas and grasslands have been replaced by eucalyptus plantations; 3) In Santa Marta, where the species was not collected, the gallery forest was narrow (ca. 50 m wide) and about 8–10 m tall. The understory was dominated by vine tangles. Patches of the original vegetation on the surroundings of the gallery forest have been logged and replaced by subsistence crops, but otherwise the matrix was covered by open cerrado vegetation. Information available in the literature or on specimen labels also did not help to understand the habitat requirements of this species. In Minas Gerais and Paraguay, Robbins & Zimmer (2005) found it in dense vinetangle dominated understory of semi-humid forests. These authors also observed this species in Brasília, where it inhabited semi-humid gallery forests. The specimens BMNH 1984.1.6 and MNRJ 33235 were collected at the border of a “flooded gallery forest”. We wrote to several colleagues who have had previous experience with this species in an attempt to gather further information on its habitat requirements. Marcelo Ferreira de Vasconcelos (pers. com.) reported that the specimens collected by him in the Brejinho das Ametistas (Vasconcelos et al. 2012) were obtained in a well preserved unflooded gallery forest, about 15–20 m tall and 50 m wide. The adjacent matrix was covered by reasonably well preserved open cerrado. Records from the Chapada do Catuni (Vasconcelos et al. 2006) were obtained in a gallery forest, parts of which flood seasonally (M.F. Vasconcelos, pers. com.). Gustavo Malacco (pers. com.) informed us that this species is more frequently found in well preserved seasonally flooded gallery forests in the Triângulo Mineiro region, extreme western of Minas Gerais. Wagner Nogueira observed this species in dense vine-tangle dominated understory of semi-humid forests. He noted that it is generally found in the wettest portions of the forests, generally on or near permanently flooded forests or even “veredas” (Mauritia palm swamps). Therefore, it seems that S. dimidiata prefers well preserved gallery forests with dense undergrowth, especially in areas that flood at least seasonally. This species seems to tolerate some degree of disturbance of the matrix within which gallery forests occur, but it apparently is not tolerant to severe disturbances in the gallery forests where it lives. Maybe this is the reason why we failed to locate this species in the disturbed gallery forests briefly sampled by us in the municipalities of Ipameri, São João da Aliança and Planaltina, where it was known to occur in the past (Appendix 1). We still have much to learn about this species’ natural history in order to better understand its rarity and local occurrence. Conservation. Although S. dimidiata has been previously considered a globally Near-threatened species (Collar et al. 1994), it is presently a “Least Concern” species (BirdLife International 2013). In Brazil it is also a “Least Concern” species, even though it is Endangered in Minas Gerais (COPAM 2010) and Critically Endangered in Paraná (Straube et al. 2004), where it had been listed as probably extinct by previous authors (Scherer-Neto & Straube 1995; Straube & Bornschein 1995). In Paraguay, where this species is rare, it is considered Endangered (Guyra Paraguay 2005). It has been recorded in several Brazilian protected areas, namely Fazenda Água Limpa (Motta-Junior 1991), Reserva Ecológica do IBGE (Motta-Junior 1991), Parque Nacional de Brasília (Ridgely & Tudor 1994), Estação Ecológica de Águas Emendadas (Bagno 1998), Parque Nacional das Emas (Pacheco 1995), and Parque Nacional da Serra da Canastra (Silveira 1998). The species also occurs in two protected areas in Paraguay: Parque Nacional Cerro Corá (Hayes & Scharf 1995) and Parque Nacional Serranía San Luis (Robbins et al. 1999). However, S. dimidiata does not seem to be common in any of the above cited conservation units. The small number of specimens found in ornithological collections, and the scarcity of records in the literature or in sound archives (e.g. Macaulay Library, http://macaulaylibrary.org/), are strong evidence of the natural rarity of the species. This hypothesis is supported by comparing the number of records of this species published in the WikiAves (http://en.wikiaves.com) with those of Hylocryptus rectirostris, another Cerrado endemic furnariid restricted to gallery forests. There are records of S. dimidiata for 14 municipalities in three Brazilian states, while there are records of H. rectirostris for 102 municipalities in eight states (data accessed on 16 July 2013). The WikiAves portal is a free tool for online publication of photos and sounds of Brazilian birds, and has proved to be an invaluable source of records of rare and elusive species (e.g., Lopes & Gonzaga 2013). Furthermore, S. dimidiata is endemic to the Cerrado, one of the Brazilian biogeographic domains suffering major anthropogenic impacts (Klink & Machado 2005). Although the most recent land cover mapping of the Cerrado revealed that 61% of its area was still covered by natural vegetation, non-anthropogenic areas are concentrated in its northern portion, in the states of Tocantins, Maranhão and Piauí (Sano et al. 2010), where the species does not occur. Only 15 % of the southern Cerrado, where the range of the species is centered, remains intact (Sano et al. 2010). Marini et al. (2009b) investigated future climate-driven change on the distribution of 26 broad-range birds from the Cerrado, and found that climate change will not equally affect all Cerrado birds. These authors found that S. dimidiata is one of the three species predicted to suffer the most from climate change, with a range loss of 69% by 2099 under a full dispersion scenario, and a range shift of about 300 km to the southeast. The direction of range change is toward the most developed and populated region of Brazil (Marini et al. 2009b), where patches of natural vegetation are small and highly fragmented (Sano et al. 2010). A few large reserves (e.g. Emas, Brasília, and Serra da Canastra National Parks) should protect some populations, but there are still large gaps in protected areas for this species (Marini et al. 2009a). Given the reasons above, we suggest considering this species as globally Vulnerable under the criteria C2ai of IUCN (2001), which means “continuing decline, observed, projected, or inferred, in numbers of mature individuals” and “no subpopulation estimated to contain more than 1000 mature individuals”.

Published as part of Lopes, Leonardo Esteves & Gonzaga, Luiz Pedreira, 2014, Taxonomy, distribution, natural history and conservation of the Russet-mantled Foliage-gleaner Syndactyla dimidiata (Pelzeln, 1859) (Aves: Furnariidae), pp. 435-449 in Zootaxa 3754 (4) on pages 442-444, DOI: 10.11646/zootaxa.3754.4.5, http://zenodo.org/record/226673